DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . Claims 1-20 are pending under this Office action.
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the claims at issue are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); and In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on a nonstatutory double patenting ground provided the reference application or patent either is shown to be commonly owned with this application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The USPTO internet Web site contains terminal disclaimer forms which may be used. Please visit http://www.uspto.gov/forms/. The filing date of the application will determine what form should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to http://www.uspto.gov/patents/process/file/efs/guidance/eTD-info-I.jsp.
Claim 1-20 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1-20 of U. S. Patent No. 12,198,230. Although the claims at issue are not identical, they are not patentably distinct from each other, see the following mapping table.
Application No. 18/957,693 (Instant Application)
U.S. Patent No. 12,198,230
1. A method of applying anisotropic texture filtering to a texture using a texture filtering unit, the method comprising:
determining whether an input amount of anisotropy is above a maximum amount of anisotropy;
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy, performing a sampling operation to sample texels of the texture to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy; and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy:
performing a plurality of sampling operations to sample texels of the texture to determine a respective plurality of intermediate filtered texture values; and
combining the plurality of intermediate filtered texture values to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy.
1. A method of applying anisotropic texture filtering to a texture using a texture filtering unit, wherein the texture filtering unit is configured to apply anisotropic filtering using a filter kernel which can be adapted to apply different amounts of anisotropy up to a maximum amount of anisotropy, the method comprising:
receiving an indication of an input amount of anisotropy and an input direction of anisotropy for filtering the texture; determining whether the input amount of anisotropy is above the maximum amount of anisotropy;
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy:
configuring the filter kernel to apply the input amount of anisotropy; and
performing a sampling operation to sample texels of the texture using the filter kernel to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy and the input direction of anisotropy; and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy:
configuring the filter kernel to apply an amount of anisotropy that is not above the maximum amount of anisotropy;
performing a plurality of sampling operations to sample texels of the texture using the filter kernel to determine a respective plurality of intermediate filtered texture values; and
combining the plurality of intermediate filtered texture values to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy and the input direction of anisotropy.
2. The method of claim 1, wherein the texture filtering unit is configured to apply anisotropic filtering using a filter kernel which can be adapted to apply different amounts of anisotropy up to the maximum amount of anisotropy.
1. A method of applying anisotropic texture filtering to a texture using a texture filtering unit, wherein the texture filtering unit is configured to apply anisotropic filtering using a filter kernel which can be adapted to apply different amounts of anisotropy up to a maximum amount of anisotropy, the method comprising:
3. The method of claim 2, further comprising:
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy, configuring the filter kernel to apply the input amount of anisotropy for use in said performing a sampling operation to sample texels of the texture to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy; and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy, configuring the filter kernel to apply an amount of anisotropy that is not above the maximum amount of anisotropy for use in said performing a plurality of sampling operations to sample texels of the texture to determine a respective plurality of intermediate filtered texture values.
1. if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy: configuring the filter kernel to apply the input amount of anisotropy; and
performing a sampling operation to sample texels of the texture using the filter kernel to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy and the input direction of anisotropy; and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy:
configuring the filter kernel to apply an amount of anisotropy that is not above the maximum amount of anisotropy;
performing a plurality of sampling operations to sample texels of the texture using the filter kernel to determine a respective plurality of intermediate filtered texture values.
4. The method of claim 1, wherein said combining the plurality of intermediate filtered texture values comprises performing a weighted sum of the intermediate filtered texture values.
2. The method of claim 1 wherein said combining the plurality of intermediate filtered texture values comprises performing a weighted sum of the intermediate filtered texture values.
5. The method of claim 4, wherein the weights of the weighted sum are non-negative and represent a normalised filtering function.
3. The method of claim 2 wherein the weights of the weighted sum are non-negative and represent a normalised filtering function.
6. The method of claim 2, wherein the filter kernel is configured to apply anisotropic filtering in an input direction of anisotropy.
4. The method of claim 1 wherein the filter kernel is configured to apply anisotropic filtering in the input direction of anisotropy.
7. The method of claim 1, wherein said plurality of sampling operations sample respective subsets of texels of the texture, wherein the respective subsets of texels are displaced with respect to each other in the texture space of the texture in accordance with an input direction of anisotropy.
5. The method of claim 1 wherein said plurality of sampling operations sample respective subsets of texels of the texture, wherein the respective subsets of texels are displaced with respect to each other in the texture space of the texture in accordance with the input direction of anisotropy.
8. The method of claim 2, wherein the filter kernel can be adapted to apply different amounts of anisotropy between a minimum amount of anisotropy and the maximum amount of anisotropy, wherein the minimum amount of anisotropy corresponds to an anisotropic ratio of 1 and the maximum amount of anisotropy corresponds to an anisotropic ratio of 2.
6. The method of claim 1 wherein the filter kernel can be adapted to apply different amounts of anisotropy between a minimum amount of anisotropy and the maximum amount of anisotropy, wherein the minimum amount of anisotropy corresponds to an anisotropic ratio of 1 and the maximum amount of anisotropy corresponds to an anisotropic ratio of 2.
9. The method of claim 2, wherein the texture is represented with a mipmap comprising a plurality of levels, wherein each level of the mipmap comprises an image representing the texture at a respective level of detail, wherein the texture filtering unit has minimum and maximum limits on an amount by which it can alter the level of detail when it uses the filter kernel to filter texels from an image of a single level of the mipmap, wherein the range of level of detail between the minimum and maximum limits defines an intrinsic region of the texture filtering unit, and wherein levels of detail outside of the range of level of detail between the minimum and maximum limits define an extrinsic region of the texture filtering unit, wherein the method comprises:
receiving an input level of detail for filtering the texture;
determining whether the received input level of detail is in an intrinsic region or an extrinsic region of the texture filtering unit;
if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit:
reading texels from a single mipmap level of the mipmap; and
filtering the read texels from the single mipmap level with the filter kernel of the texture filtering unit to determine a filtered texture value representing part of the texture at the input level of detail; and
if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit:
reading texels from two mipmap levels of the mipmap; and
processing the read texels from the two mipmap levels with the texture filtering unit to determine a filtered texture value representing part of the texture at the input level of detail.
7. The method of claim 1 wherein the texture is represented with a mipmap comprising a plurality of levels, wherein each level of the mipmap comprises an image representing the texture at a respective level of detail, wherein the texture filtering unit has minimum and maximum limits on an amount by which it can alter the level of detail when it uses the filter kernel to filter texels from an image of a single level of the mipmap, wherein the range of level of detail between the minimum and maximum limits defines an intrinsic region of the texture filtering unit, and wherein levels of detail outside of the range of level of detail between the minimum and maximum limits define an extrinsic region of the texture filtering unit, wherein the method comprises:
receiving an input level of detail for filtering the texture;
determining whether the received input level of detail is in an intrinsic region or an extrinsic region of the texture filtering unit;
if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit:
reading texels from a single mipmap level of the mipmap; and
filtering the read texels from the single mipmap level with the filter kernel of the texture filtering unit to determine a filtered texture value representing part of the texture at the input level of detail; and
if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit:
reading texels from two mipmap levels of the mipmap; and
processing the read texels from the two mipmap levels with the texture filtering unit to determine a filtered texture value representing part of the texture at the input level of detail.
10. The method of claim 9, wherein the maximum amount of anisotropy that the filter kernel can apply depends upon the amount by which the level of detail is altered when the texture filtering unit uses the filter kernel to filter texels.
8. The method of claim 7 wherein the maximum amount of anisotropy that the filter kernel can apply depends upon the amount by which the level of detail is altered when the texture filtering unit uses the filter kernel to filter texels.
11. The method of claim 9, wherein a first of the two mipmap levels is associated with a first intrinsic region of the texture filtering unit, wherein a second of the two mipmap levels is associated with a second intrinsic region of the texture filtering unit, and wherein said processing the read texels from the two mipmap levels with the texture filtering unit comprises:
filtering the read texels from the first of the two mipmap levels with the texture filtering unit to determine a first intermediate filtered texture value at a first intermediate level of detail within the first intrinsic region of the texture filtering unit;
filtering the read texels from the second of the two mipmap levels with the texture filtering unit to determine a second intermediate filtered texture value at a second intermediate level of detail within the second intrinsic region of the texture filtering unit; and
determining a filtered texture value representing part of the texture at the input level of detail by using the input level of detail to interpolate between the first intermediate filtered texture value at the first intermediate level of detail and the second intermediate filtered texture value at the second intermediate level of detail.
9. The method of claim 7 wherein a first of the two mipmap levels is associated with a first intrinsic region of the texture filtering unit, wherein a second of the two mipmap levels is associated with a second intrinsic region of the texture filtering unit, and wherein said processing the read texels from the two mipmap levels with the texture filtering unit comprises:
filtering the read texels from the first of the two mipmap levels with the texture filtering unit to determine a first intermediate filtered texture value at a first intermediate level of detail within the first intrinsic region of the texture filtering unit;
filtering the read texels from the second of the two mipmap levels with the texture filtering unit to determine a second intermediate filtered texture value at a second intermediate level of detail within the second intrinsic region of the texture filtering unit; and
determining a filtered texture value representing part of the texture at the input level of detail by using the input level of detail to interpolate between the first intermediate filtered texture value at the first intermediate level of detail and the second intermediate filtered texture value at the second intermediate level of detail.
12. The method of claim 9, wherein the texture filtering unit has a plurality of intrinsic regions and extrinsic regions, wherein there is an intrinsic region and an extrinsic region for each of a plurality of the mipmap levels of the mipmap.
10. The method of claim 7 wherein the texture filtering unit has a plurality of intrinsic regions and extrinsic regions, wherein there is an intrinsic region and an extrinsic region for each of a plurality of the mipmap levels of the mipmap.
13. The method of claim 9, wherein said determining whether the received input level of detail is in an intrinsic region or an extrinsic region of the texture filtering unit comprises:
determining a first indication of a mipmap level,
PNG
media_image1.png
14
16
media_image1.png
Greyscale
PNG
media_image1.png
14
16
media_image1.png
Greyscale
, such that
PNG
media_image2.png
17
89
media_image2.png
Greyscale
PNG
media_image2.png
17
89
media_image2.png
Greyscale
; and
determining a second indication of a mipmap level,
PNG
media_image3.png
14
17
media_image3.png
Greyscale
PNG
media_image3.png
14
17
media_image3.png
Greyscale
, such that
PNG
media_image4.png
16
92
media_image4.png
Greyscale
PNG
media_image4.png
16
92
media_image4.png
Greyscale
;
wherein
PNG
media_image5.png
14
15
media_image5.png
Greyscale
PNG
media_image5.png
14
15
media_image5.png
Greyscale
is the received input level of detail,
PNG
media_image6.png
16
24
media_image6.png
Greyscale
PNG
media_image6.png
16
24
media_image6.png
Greyscale
is the minimum limit and
PNG
media_image7.png
15
26
media_image7.png
Greyscale
PNG
media_image7.png
15
26
media_image7.png
Greyscale
is the maximum limit,
wherein the received input level of detail is in an intrinsic region if
PNG
media_image8.png
14
49
media_image8.png
Greyscale
PNG
media_image8.png
14
49
media_image8.png
Greyscale
, wherein the received input level of detail is in an extrinsic region if
PNG
media_image9.png
14
49
media_image9.png
Greyscale
PNG
media_image9.png
14
49
media_image9.png
Greyscale
; and
wherein
PNG
media_image10.png
19
134
media_image10.png
Greyscale
PNG
media_image10.png
19
134
media_image10.png
Greyscale
.
11. The method of claim 7, wherein said determining whether the received input level of detail is in an intrinsic region or an extrinsic region of the texture filtering unit comprises:
determining a first indication of a mipmap level,
d
h
i
, such that
d
h
i
=
d
l
-
p
m
i
n
λ
; and
determining a second indication of a mipmap level,
d
l
o
, such that
d
l
o
=
d
l
-
p
m
a
x
λ
;
wherein
d
l
is the received input level of detail,
p
m
i
n
λ
is the minimum limit and
p
m
a
x
λ
is the maximum limit, and
wherein the received input level of detail is in an intrinsic region if
d
h
i
=
d
l
o
, wherein the received input level of detail is in an extrinsic region if
d
h
i
≠
d
l
o
;
a
n
d
wherein
0
≤
p
m
i
n
λ
≤
p
m
a
x
λ
<
1
.
14. The method of claim 13, wherein if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit:
texels are read from the single mipmap level of the mipmap indicated by
PNG
media_image1.png
14
16
media_image1.png
Greyscale
PNG
media_image1.png
14
16
media_image1.png
Greyscale
and
PNG
media_image3.png
14
17
media_image3.png
Greyscale
PNG
media_image3.png
14
17
media_image3.png
Greyscale
; and
a parameter
PNG
media_image11.png
15
13
media_image11.png
Greyscale
PNG
media_image11.png
15
13
media_image11.png
Greyscale
is set to determine the amount by which the level of detail is altered from the single mipmap level with the texture filtering unit, wherein
PNG
media_image12.png
16
193
media_image12.png
Greyscale
PNG
media_image12.png
16
193
media_image12.png
Greyscale
,
wherein
p
h
i
λ
and
p
l
o
λ
also indicate the amount by which the level of detail is altered from the single mipmap level with the texture filtering unit.
12. The method of claim 11, wherein if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit:
texels are read from the single mipmap level of the mipmap indicated by
d
h
i
and
d
l
o
; and
a parameter
p
λ
is set to determine the amount by which the level of detail is altered from the single mipmap level with the texture filtering unit, wherein
p
λ
=
d
l
-
d
h
i
=
d
l
-
d
l
o
=
p
h
i
λ
=
p
l
o
λ
.
wherein
p
h
i
λ
and
p
l
o
λ
also indicate the amount by which the level of detail is altered from the single mipmap level with the texture filtering unit.
15. The method of claim 13, wherein if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit:
texels are read from the two mipmap levels of the mipmap indicated by
PNG
media_image1.png
14
16
media_image1.png
Greyscale
PNG
media_image1.png
14
16
media_image1.png
Greyscale
and
PNG
media_image3.png
14
17
media_image3.png
Greyscale
PNG
media_image3.png
14
17
media_image3.png
Greyscale
;
a parameter
PNG
media_image13.png
16
16
media_image13.png
Greyscale
PNG
media_image13.png
16
16
media_image13.png
Greyscale
is set to determine the amount by which the level of detail is altered from a first of the two mipmap levels indicated by
PNG
media_image1.png
14
16
media_image1.png
Greyscale
PNG
media_image1.png
14
16
media_image1.png
Greyscale
with the texture filtering unit, wherein
PNG
media_image14.png
16
57
media_image14.png
Greyscale
PNG
media_image14.png
16
57
media_image14.png
Greyscale
; and
a parameter
PNG
media_image15.png
16
17
media_image15.png
Greyscale
PNG
media_image15.png
16
17
media_image15.png
Greyscale
is set to determine the amount by which the level of detail is altered from a second of the two mipmap levels indicated by
PNG
media_image3.png
14
17
media_image3.png
Greyscale
PNG
media_image3.png
14
17
media_image3.png
Greyscale
with the texture filtering unit, wherein
p
l
o
λ
=
p
m
i
n
λ
.
13. The method of claim 11, wherein if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit:
texels are read from the two mipmap levels of the mipmap indicated by
d
h
i
and
d
l
o
;
a parameter
p
h
i
λ
is set to determine the amount by which the level of detail is altered from a first of the two mipmap levels indicated by
d
h
i
with the texture filtering unit, wherein
p
h
i
λ
=
p
m
a
x
λ
; and
a parameter
p
l
o
λ
is set to determine the amount by which the level of detail is altered from a second of the two mipmap levels indicated by
d
l
o
, with the texture filtering unit, wherein
p
l
o
λ
=
p
m
i
n
λ
.
16. The method of claim 14, further comprising:
determining a first anisotropic filtering indication
PNG
media_image16.png
14
16
media_image16.png
Greyscale
PNG
media_image16.png
14
16
media_image16.png
Greyscale
, wherein if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit then
PNG
media_image17.png
14
71
media_image17.png
Greyscale
PNG
media_image17.png
14
71
media_image17.png
Greyscale
, and wherein if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit then
PNG
media_image18.png
16
118
media_image18.png
Greyscale
PNG
media_image18.png
16
118
media_image18.png
Greyscale
; and
determining a second anisotropic filtering indication
PNG
media_image19.png
14
17
media_image19.png
Greyscale
PNG
media_image19.png
14
17
media_image19.png
Greyscale
, wherein if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit then
PNG
media_image20.png
14
74
media_image20.png
Greyscale
PNG
media_image20.png
14
74
media_image20.png
Greyscale
, and wherein if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit then
PNG
media_image21.png
17
117
media_image21.png
Greyscale
PNG
media_image21.png
17
117
media_image21.png
Greyscale
;
setting a first anisotropic filtering parameter
p
h
i
μ
to determine how much anisotropic filtering, in terms of additional level of detail, the texture filtering unit is to apply to the texels read from the mipmap level indicated by
d
h
i
, wherein if
μ
h
i
≤
m
a
x
p
h
i
λ
,
p
m
a
x
μ
then
p
h
i
μ
=
μ
h
i
, otherwise
p
h
i
μ
=
m
a
x
p
h
i
λ
,
p
m
a
x
μ
; and
setting a second anisotropic filtering parameter
p
l
o
μ
to determine how much anisotropic filtering, in terms of additional level of detail, the texture filtering unit is to apply to the texels read from the mipmap level indicated by
d
l
o
, wherein if
μ
l
o
≤
m
a
x
p
l
o
λ
,
p
m
a
x
μ
then
p
l
o
μ
=
μ
l
o
, otherwise
p
l
o
μ
=
m
a
x
p
l
o
λ
,
p
m
a
x
μ
;
wherein
t
l
is the received indication of an input amount of anisotropy, and
p
m
a
x
μ
represents the maximum amount of anisotropy that the filter kernel can apply in terms of additional level of detail.
14. The method of claim 12, further comprising:
determining a first anisotropic filtering indication
PNG
media_image16.png
14
16
media_image16.png
Greyscale
PNG
media_image16.png
14
16
media_image16.png
Greyscale
, wherein if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit then
PNG
media_image17.png
14
71
media_image17.png
Greyscale
PNG
media_image17.png
14
71
media_image17.png
Greyscale
, and wherein if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit then
PNG
media_image18.png
16
118
media_image18.png
Greyscale
PNG
media_image18.png
16
118
media_image18.png
Greyscale
; and
determining a second anisotropic filtering indication
PNG
media_image19.png
14
17
media_image19.png
Greyscale
PNG
media_image19.png
14
17
media_image19.png
Greyscale
, wherein if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit then
PNG
media_image20.png
14
74
media_image20.png
Greyscale
PNG
media_image20.png
14
74
media_image20.png
Greyscale
, and wherein if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit then
PNG
media_image21.png
17
117
media_image21.png
Greyscale
PNG
media_image21.png
17
117
media_image21.png
Greyscale
;
setting a first anisotropic filtering parameter
p
h
i
μ
to determine how much anisotropic filtering, in terms of additional level of detail, the texture filtering unit is to apply to the texels read from the mipmap level indicated by
d
h
i
, wherein if
μ
h
i
≤
m
a
x
p
h
i
λ
,
p
m
a
x
μ
then
p
h
i
μ
=
μ
h
i
, otherwise
p
h
i
μ
=
m
a
x
p
h
i
λ
,
p
m
a
x
μ
; and
setting a second anisotropic filtering parameter
p
l
o
μ
to determine how much anisotropic filtering, in terms of additional level of detail, the texture filtering unit is to apply to the texels read from the mipmap level indicated by
d
l
o
, wherein if
μ
l
o
≤
m
a
x
p
l
o
λ
,
p
m
a
x
μ
then
p
l
o
μ
=
μ
l
o
, otherwise
p
l
o
μ
=
m
a
x
p
l
o
λ
,
p
m
a
x
μ
;
wherein
t
l
is the received indication of an input amount of anisotropy, and
p
m
a
x
μ
represents the maximum amount of anisotropy that the filter kernel can apply in terms of additional level of detail.
17. A texture filtering unit configured to apply anisotropic texture filtering to a texture, the texture filtering unit being configured to:
determine whether an input amount of anisotropy is above a maximum amount of anisotropy;
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy, perform a sampling operation to sample texels of the texture to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy; and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy:
perform a plurality of sampling operations to sample texels of the texture to determine a respective plurality of intermediate filtered texture values; and
combine the plurality of intermediate filtered texture values to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy.
15. A texture filtering unit configured to apply anisotropic texture filtering to a texture, using a filter Kernel which can be adapted to apply different amounts of anisotropy up to a maximum amount of anisotropy, the texture filtering unit being configured to:
receive an indication of an input amount of anisotropy and an input direction of anisotropy for filtering the texture;
determine whether the input amount of anisotropy is above the maximum amount of anisotropy;
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy:
configure the filter kernel to apply the input amount of anisotropy; and
perform a sampling operation to sample texels of the texture using the filter kernel to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy and the input direction of anisotropy; and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy:
configure the filter kernel to apply an amount of anisotropy that is not above the maximum amount of anisotropy;
perform a plurality of sampling operations to sample texels of the texture using the filter kernel to determine a respective plurality of intermediate filtered texture values; and
combine the plurality of intermediate filtered texture values to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy and the input direction of anisotropy.
18. The texture filtering unit of claim 17, wherein the texture is represented with a mipmap comprising a plurality of levels, wherein each level of the mipmap comprises an image representing the texture at a respective level of detail, wherein the texture filtering unit has minimum and maximum limits on an amount by which it can alter the level of detail when it filters texels from an image of a single level of the mipmap, wherein the range of level of detail between the minimum and maximum limits defines an intrinsic region of the texture filtering unit, and wherein levels of detail outside of the range of level of detail between the minimum and maximum limits define an extrinsic region of the texture filtering unit, wherein the texture filtering unit is configured to:
receive an input level of detail for filtering the texture;
determine whether the received input level of detail is in an intrinsic region or an extrinsic region of the texture filtering unit;
if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit:
read texels from a single mipmap level of the mipmap; and
filter the read texels from the single mipmap level with the texture filtering unit to determine a filtered texture value representing part of the texture at the input level of detail; and
if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit:
read texels from two mipmap levels of the mipmap; and
process the read texels from the two mipmap levels with the texture filtering unit to determine a filtered texture value representing part of the texture at the input level of detail.
16. The texture filtering unit of claim 15, wherein the texture is represented with a mipmap comprising a plurality of levels, wherein each level of the mipmap comprises an image representing the texture at a respective level of detail, wherein the texture filtering unit has minimum and maximum limits on an amount by which it can alter the level of detail when it uses the filter kernel to filter texels from an image of a single level of the mipmap, wherein the range of level of detail between the minimum and maximum limits defines an intrinsic region of the texture filtering unit, and wherein levels of detail outside of the range of level of detail between the minimum and maximum limits define an extrinsic region of the texture filtering unit, wherein the texture filtering unit is configured to:
receive an input level of detail for filtering the texture;
determine whether the received input level of detail is in an intrinsic region or an extrinsic region of the texture filtering unit;
if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit:
read texels from a single mipmap level of the mipmap; and
filter the read texels from the single mipmap level with the filter kernel of the texture filtering unit to determine a filtered texture value representing part of the texture at the input level of detail; and
if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit:
read texels from two mipmap levels of the mipmap; and
process the read texels from the two mipmap levels with the texture filtering unit to determine a filtered texture value representing part of the texture at the input level of detail.
19. A graphics processing unit, comprising the texture filtering unit as set forth in claim 17, wherein the graphics processing unit is configured to use the filtered texture value determined by the texture filtering unit to render an image of a scene in which the texture is applied to a surface in the scene.
19. A graphics processing unit comprising the texture filtering unit of claim 15, wherein the graphics processing unit is configured to use the filtered texture value determined by the texture filtering unit to render an image of a scene in which the texture is applied to a surface in the scene.
20. A non-transitory computer readable storage medium having stored thereon an integrated circuit definition dataset that, when processed in an integrated circuit manufacturing system, configures the integrated circuit manufacturing system to manufacture a texturing filtering unit which is configured to apply anisotropic texture filtering to a texture, the texture filtering unit being configured to:
determine whether an input amount of anisotropy is above a maximum amount of anisotropy;
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy, perform a sampling operation to sample texels of the texture to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy; and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy:
perform a plurality of sampling operations to sample texels of the texture to determine a respective plurality of intermediate filtered texture values; and
combine the plurality of intermediate filtered texture values to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy.
20. A non-transitory computer readable storage medium having stored thereon an integrated circuit definition dataset that, when processed in an integrated circuit manufacturing system, configures the integrated circuit manufacturing system to manufacture a texturing filtering unit which is configured to apply anisotropic texture filtering to a texture, using a filter kernel which can be adapted to apply different amounts of anisotropy up to a maximum amount of anisotropy, the texture filtering unit being configured to:
receive an indication of an input amount of anisotropy and an input direction of anisotropy for filtering the texture;
determine whether the input amount of anisotropy is above the maximum amount of anisotropy;
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy:
configure the filter kernel to apply the input amount of anisotropy; and
perform a sampling operation to sample texels of the texture using the filter kernel to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy and the input direction of anisotropy; and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy:
configure the filter kernel to apply an amount of anisotropy that is not above the maximum amount of anisotropy;
perform a plurality of sampling operations to sample texels of the texture using the filter kernel to determine a respective plurality of intermediate filtered texture values; and
combine the plurality of intermediate filtered texture values to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy and the input direction of anisotropy.
Claim 1 of the instant application is drawn to a method of applying anisotropic texture filtering to a texture using a texture filtering unit, the method comprising: determining whether an input amount of anisotropy is above a maximum amount of anisotropy; if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy, performing a sampling operation to sample texels of the texture to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy; and if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy: performing a plurality of sampling operations to sample texels of the texture to determine a respective plurality of intermediate filtered texture values; and combining the plurality of intermediate filtered texture values to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy.
While the exact wordings of claim 1 of the ‘230 patent may not bot the same as that of claim 1 of the instant application, but there is no significant difference in scope between the claim 1 of the instant application and the claim 1 of the patent ‘230. Therefore, claim 1 of the instant application cannot be considered patentably distinct over claim 1 of the ‘230 patent.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-8, 17, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Toksvig, etc. (US 7372468 B1) in view of Donovan, etc. (US 7221371 B2).
Regarding claim 1, Toksvig teaches that a method of applying anisotropic texture filtering to a texture using a texture filtering unit, the method (See Toksvig: Fig. 6, and Col. 13 Lines 11-27, “FIG. 6 is a block diagram of an exemplary embodiment of a respective computer system, generally designated 600, and including a host computer 610 and a graphics subsystem 607 in accordance with one or more aspects of the present invention. Computing system 600 may be a desktop computer, server, laptop computer, palm-sized computer, tablet computer, game console, portable wireless terminal such as a PDA (personal digital assistant) or cellular telephone, computer based simulator, or the like. Host computer 610 includes host processor 614 that may include a system memory controller to interface directly to host memory 612 or may communicate with host memory 612 through a system interface 615. System interface 615 may be an I/O (input/output) interface or a bridge device including the system memory controller to interface directly to host memory 612. An example of system interface 615 known in the art includes Intel.RTM. Northbridge”) comprising:
determining whether an input amount of anisotropy (See Toksvig: Figs. 2A-C, and Col. 5 Lines 25-44, “FIG. 2A illustrates an anisotropic footprint, footprint 201, in accordance with one or more aspects of the present invention. Footprint 201 has a major axis, major axis 202 of length 8.75 and a minor axis, minor axis 203, of length 1.75. The anisotropic ratio is 5 and the LOD computed as log base 2 of minor axis 203 is 0.807. LODfrac is also 0.807. Ideally, the texture samples used to produce an anisotropically filtered texture value for footprint 201 are from a mip map for a LOD of 0.807. Unfortunately, texture mapping systems typically only include prefiltered LOD mip maps for integer values. However, good filtering results may be achieved by using texture samples that are spaced by one texel for both the coarse and fine LOD mip maps, as shown in FIGS. 2B and 2C, respectively, to approximate an LOD mip map of 0.807. Spacing the bilinear texture samples by one texel produces a filtered texture value without over or under sampling, minimizing undesirable visual artifacts. Methods described in conjunction with FIGS. 3A, 5A, and 5D may be used to space the texture samples by one texel to produce a filtered texture value”; and Fig. 6 Col. 15 Lines 41-50, “Texture unit 690 includes an anisotropic unit 700. At a high level, anisotropic unit 700 computes anisotropic texture mapping parameters such as LODt, the logratio and/or anisotropic ratio, angle of anisotropy, and the axis of anisotropy, using techniques known to those skilled in the art. As previously described, these anisotropic texture mapping parameters are used to determine the number, spacing (spread), and mip map level weights of bilinear texture samples to perform anisotropic texture filtering for a fragment”. Note that the anisotropic ratio is mapped to the input anisotropic amount; and the anisotropy parameters are used to determine the spacing, sample counts and mip map levels, and this inherently supports a practical “maximum” anisotropy amounts but a secondary art will be used to teach this maximum anisotropy concept) is above a maximum amount of anisotropy;
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy, performing a sampling operation to sample texels of the texture to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy (See Toksvig: Figs. 1A-E, and Col. 2 Lines 19-32, “FIG. 1D illustrates a prior art application of ten bilinear samples, bilinear samples 140, that are positioned along major axis 130 to approximate an elliptical footprint for a coarse LOD mip map, such as footprint 115. Each bilinear sample corresponds to an isotropically filtered texture sample for an LOD of a texture map that is computed using conventional bilinear isotropic filtering. Each bilinear sample of bilinear samples 140 is spaced less than one texel apart. In some conventional systems, the bilinear samples in the coarse LOD are spaced by 0.5 to 1.0 texels apart. Therefore the coarse LOD mip map is oversampled, possibly introducing visual artifacts and requiring more computations and texel reads than if the spacing were one texel apart”. Note that for lower anisotropy, close to isotropic, bilinear samples are adjusted by the anisotropic ratio/LOD, and this maps to a (single effective) sampling operation tuned to the anisotropy amount); and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy (See Toksvig: Figs. 2A-C, and Col. 5 Lines 25-44, “FIG. 2A illustrates an anisotropic footprint, footprint 201, in accordance with one or more aspects of the present invention. Footprint 201 has a major axis, major axis 202 of length 8.75 and a minor axis, minor axis 203, of length 1.75. The anisotropic ratio is 5 and the LOD computed as log base 2 of minor axis 203 is 0.807. LODfrac is also 0.807. Ideally, the texture samples used to produce an anisotropically filtered texture value for footprint 201 are from a mip map for a LOD of 0.807. Unfortunately, texture mapping systems typically only include prefiltered LOD mip maps for integer values. However, good filtering results may be achieved by using texture samples that are spaced by one texel for both the coarse and fine LOD mip maps, as shown in FIGS. 2B and 2C, respectively, to approximate an LOD mip map of 0.807. Spacing the bilinear texture samples by one texel produces a filtered texture value without over or under sampling, minimizing undesirable visual artifacts. Methods described in conjunction with FIGS. 3A, 5A, and 5D may be used to space the texture samples by one texel to produce a filtered texture value”; and Figs. 1-E, and Col. 1 Lines 63-67 ~ Col. 2 Lines 1-7, “FIG. 1C illustrates footprint 115 including a minor axis 125 that is significantly shorter than a major axis 130. Texture samples along major axis 130, the axis of anisotropy, are read from one or more mipmap levels and are blended to produce a pixel color. The level from which the samples are read is determined using a level of detail (LOD) value which is nominally the log base 2 of the length of minor axis 125. The number of texture samples read from the texture map is determined based on the ratio of the major axis to the minor axis, the anisotropic ratio, with more texture samples needed as the ratio increases, i.e. as the ellipse becomes more elongated”. Note that when the anisotropy is large, multiple bilinear sampling operations are performed along the major axis):
performing a plurality of sampling operations to sample texels of the texture to determine a respective plurality of intermediate filtered texture values (See Toksvig: Figs. 2A-C, and Col. 5 Lines 45-58, “FIG. 2B illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a coarse LOD mip map in accordance with one or more aspects of the present invention. Specifically six bilinear samples, bilinear samples 204 are spaced by one texel along major axis 202 to cover major axis 202. In comparison with FIG. 1D, in which ten bilinear samples are used from the coarse LOD mip map, the number of bilinear texture samples is reduced. Although the texture samples, bilinear samples 204, are labeled as bilinear texture samples, in other embodiments of the present invention texture samples may be used that are produced using other reconstruction filtering techniques known to those skilled in the art”; Col. 5 Lines 59-67, “The length of major axis 202 in the coarse LOD mip map is 4.375, so 4.375 bilinear texture samples of bilinear samples 204 cover major axis 202. Bilinear samples 204 may be filtered to produce a filtered texture value for the coarse LOD mip map. The filtered texture value for the coarse LOD mip map is then weighed by the coarse mip map level weight (LODfrac) and combined with a filtered texture value for the fine LOD mip map to produce the filtered texture for footprint 201”; and Col. 6 Lines 1-14, “FIG. 2C illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a fine LOD mip map in accordance with one or more aspects of the present invention. Specifically ten bilinear samples, bilinear samples 205 are spaced by one texel along major axis 202 to cover major axis 202. The length of major axis 202 in the fine LOD mip map is 8.750, so 8.750 bilinear texture samples of bilinear samples 205 cover major axis 202. Bilinear samples 205 may be filtered to produce a filtered texture value for the fine LOD mip map. The filtered texture value for the fine LOD mip map is then weighed by the fine mip map level weight (1-LODfrac) and combined with the filtered texture value for the coarse LOD mip map to produce the filtered texture for footprint 201”. Note that both coarse and fine mip map are filtered samples and they are mapped to the intermediate filtered texture values); and
combining the plurality of intermediate filtered texture values to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy (See Toksvig: Figs. 2A-C, and Col. 6 Lines 1-14, “FIG. 2C illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a fine LOD mip map in accordance with one or more aspects of the present invention. Specifically ten bilinear samples, bilinear samples 205 are spaced by one texel along major axis 202 to cover major axis 202. The length of major axis 202 in the fine LOD mip map is 8.750, so 8.750 bilinear texture samples of bilinear samples 205 cover major axis 202. Bilinear samples 205 may be filtered to produce a filtered texture value for the fine LOD mip map. The filtered texture value for the fine LOD mip map is then weighed by the fine mip map level weight (1-LODfrac) and combined with the filtered texture value for the coarse LOD mip map to produce the filtered texture for footprint 201”. Note that both coarse and fine mip map are filtered samples and they are combined to produce the final filtered texture values).
However, Toksvig fails to explicitly disclose that (whether an input amount of anisotropy) is above a maximum amount of anisotropy.
However, Donovan teaches that (whether an input amount of anisotropy) is above a maximum amount of anisotropy (See Donovan: Fig. 2, and Col. 4 Lines 6-26, “FIG. 2 illustrates an embodiment of a method of determining a number of texture samples for use in an anisotropic texture map filtering computation in accordance with one or more aspects of the present invention. In step 205 the base-two logarithm (log) of the minor axis length is computed to produce a logminor value. In step 210 the base-two log of the major axis length is computed to produce a logmajor value. In step 215 a logratio value is computed by subtracting the logmajor value from the logminor value. The logratio value is equivalent to the base-two log of the ratio value. In embodiments supporting a maximum anisotropy of 1/16, the logratio value ranges from -4 to 0, where a logratio value of 0 indicates an isotropic footprint. Increasing the logratio value decreases the number of samples averaged over a range of logratios. The amount of processing needed to compute an anisotropically filtered texture value is proportional to the number of samples needed. Furthermore, performing computations in log space simplifies the computations and may be more efficient; for example, subtraction in log space is used instead of division. Likewise, addition is used instead of multiplication”; Figs. 3A-B, and Col. 6 Lines 46-42, “In step 325 the first-modified logratio value is clamped to a value between -1024 and zero when a maximum of 16 samples are used for anisotropic filtering. Alternatively, the first-modified logratio is clamped to another value based on a different maximum number of samples supported for anisotropic filtering”; and Col. 8 Lines 1-7, “In step 365 the second-modified logratio value is clamped to a value between -1024 and zero when a maximum of 16 samples are used for anisotropic filtering, to produce the first-modified logratio value. Alternatively, the second-modified logratio value is clamped to another value based on a different maximum number of texture samples supported for anisotropic filtering”. Note that maximum anisotropy is explicitly used to control the sample counts, making the conditional “if above max” clear and obvious in anisotropic filtering).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention was effectively filed to modify Toksvig to have (whether an input amount of anisotropy) is above a maximum amount of anisotropy as taught by Donovan in order to minimize image quality degradation for desired level of filtering performance (See Donovan: Fig. 2, and Col. 2 Lines 1-4, “Accordingly, there is a need to balance performance of anisotropic texture sample filtering with image quality to minimize image quality degradation for a desired level of anisotropic texture sample filtering performance”). Toksvig teaches a method and system that may modify the number of texture samples used to produce an anisotropically filtered texture mapped pixel and improve texture mapping performance. When the number of texture samples is reduced, fewer texels are read and fewer filtering computations are needed to produce a texture value for an anisotropic footprint; while Donovan teaches a system and method that may reduce the number of texture samples anisotropically filtered to determine a texture value associated with a graphics fragment by anisotropically filtering the textures based on the anisotropy comparison with the maximum anisotropy and filtering differently. Therefore, it is obvious to one of ordinary skill in the art to modify Toksvig by Donovan to compare the anisotropic amount with the maximum anisotropy and filter the texture differently based on the comparison results. The motivation to modify Toksvig by Donovan is “Use of known technique to improve similar devices (methods, or products) in the same way”.
Regarding claim 2, Toksvig and Donovan teach all the features with respect to claim 1 as outlined above. Further, Donovan teaches that the method of claim 1, wherein the texture filtering unit is configured to apply anisotropic filtering using a filter kernel (See Donovan: Fig. 4, and Col. 8 Lines 27-34, “Texture Unit 400 includes an Anisotropic Unit 405. A Control Unit 420 within Anisotropic Unit 405 processes the program instructions, such as instructions to load one or more programmable knob values, e.g., the first knob value, the second knob value, the third knob value, or the like, into a Logratio Modification Unit 415. Logratio Modification Unit 415 includes storage elements, e.g., registers, to store the one or more knob values”; and Col. 8 Lines 48-63, “Sample Location Unit 425 outputs sample locations, weights, LODfine, and LODcoarse to an Address Computation Unit 450. Address Computation Unit 450 uses texture parameters (texture ID, and the like) received by Texture Unit 400 to determine addresses for reading texture samples (the first number of texture samples or the first and the second number of texture samples) from memory. Address Computation Unit 450 outputs the addresses to a Read Interface 460. Read Interface 460 outputs the addresses and a read request to a memory, e.g., cache, RAM, ROM, or the like. Texture samples read from memory are received from the memory by a Texture Filter Unit 470. Texture Filter Unit 470 receives the weights from Address Computation Unit 450 and filters the texture samples read from memory using bilinear interpolation, trilinear interpolation, or anisotropic filtering to produce filtered texture samples. The filtered texture samples are output to a shader unit, described further herein, to compute a color for each fragment”. Note that an Anisotropic Unit 405 is mapped to the filter kernel) which can be adapted to apply different amounts of anisotropy (See Donovan: Fig. 4, and Col. 8 Lines 35-47, “Parameters produced by the rasterizer are received by a Logratio Computation Unit 410 within Anisotropic Unit 405. Logratio Computation Unit 410 computes the logratio value as previously described in conjunction with FIG. 2. The logratio value is output by Logratio Computation Unit 410 to Logratio Modification Unit 415. Logratio Modification Unit 415 modifies the logratio value to produce the first-modified logratio value and when LODfrac is not equal to zero, Logratio Modification Unit 415 also modifies the logratio value to produce the second-modified logratio value. The operations performed by Logratio Modification Unit 415 serve to reduce the number of texture samples read and filtered to produce the filtered texture sample”. Note that the anisotropic unit modifying various values with one or more programmable knob values is mapped to “adapted to”) up to the maximum amount of anisotropy (See Donovan: Fig. 2, and Col. 4 Lines 6-26, “FIG. 2 illustrates an embodiment of a method of determining a number of texture samples for use in an anisotropic texture map filtering computation in accordance with one or more aspects of the present invention. In step 205 the base-two logarithm (log) of the minor axis length is computed to produce a logminor value. In step 210 the base-two log of the major axis length is computed to produce a logmajor value. In step 215 a logratio value is computed by subtracting the logmajor value from the logminor value. The logratio value is equivalent to the base-two log of the ratio value. In embodiments supporting a maximum anisotropy of 1/16, the logratio value ranges from -4 to 0, where a logratio value of 0 indicates an isotropic footprint. Increasing the logratio value decreases the number of samples averaged over a range of logratios. The amount of processing needed to compute an anisotropically filtered texture value is proportional to the number of samples needed. Furthermore, performing computations in log space simplifies the computations and may be more efficient; for example, subtraction in log space is used instead of division. Likewise, addition is used instead of multiplication”; Figs. 3A-B, and Col. 6 Lines 46-42, “In step 325 the first-modified logratio value is clamped to a value between -1024 and zero when a maximum of 16 samples are used for anisotropic filtering. Alternatively, the first-modified logratio is clamped to another value based on a different maximum number of samples supported for anisotropic filtering”; and Col. 8 Lines 1-7, “In step 365 the second-modified logratio value is clamped to a value between -1024 and zero when a maximum of 16 samples are used for anisotropic filtering, to produce the first-modified logratio value. Alternatively, the second-modified logratio value is clamped to another value based on a different maximum number of texture samples supported for anisotropic filtering”. Note that maximum anisotropy is explicitly used to control the sample counts).
Regarding claim 3, Toksvig and Donovan teach all the features with respect to claim 2 as outlined above. Further, Donovan teaches that the method of claim 2, further comprising:
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy, configuring the filter kernel to apply the input amount of anisotropy for use in said performing a sampling operation to sample texels of the texture to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy (See Donovan: Fig. 4, and Col. 8 Lines 27-34, “Texture Unit 400 includes an Anisotropic Unit 405. A Control Unit 420 within Anisotropic Unit 405 processes the program instructions, such as instructions to load one or more programmable knob values, e.g., the first knob value, the second knob value, the third knob value, or the like, into a Logratio Modification Unit 415. Logratio Modification Unit 415 includes storage elements, e.g., registers, to store the one or more knob values”; and Fig. 2, and Col. 4 Lines 43-61, “In step 225 the logratio value is modified to produce a first-modified logratio value based on one or more of the knob values, as described further herein in conjunction with FIG. 3A, potentially shortening the footprint. In step 230 a first number of texture samples based on the first-modified logratio value is determined. In one embodiment, a first ratio is computed: first ratio=2.sup.(first-modified logratio). The first number of texture samples is computed using the first ratio. When the first ratio is 1 or more, the first number of texture samples is 1. Otherwise, the following equation is used: first number of texture samples=2*ceil(0.5/(first ratio)). Alternatively, a different equation based on the first ratio is used to determine the first number of texture samples”. Note that when the anisotropy is small, the first modified logratio is used to determine the first number of filtered samples, and this is mapped to anisotropy less than the maximum amount of anisotropy); and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy, configuring the filter kernel to apply an amount of anisotropy that is not above the maximum amount of anisotropy for use in said performing a plurality of sampling operations to sample texels of the texture to determine a respective plurality of intermediate filtered texture values (See Donovan: Fig. 2, and Col. 4 Lines 6-26, “FIG. 2 illustrates an embodiment of a method of determining a number of texture samples for use in an anisotropic texture map filtering computation in accordance with one or more aspects of the present invention. In step 205 the base-two logarithm (log) of the minor axis length is computed to produce a logminor value. In step 210 the base-two log of the major axis length is computed to produce a logmajor value. In step 215 a logratio value is computed by subtracting the logmajor value from the logminor value. The logratio value is equivalent to the base-two log of the ratio value. In embodiments supporting a maximum anisotropy of 1/16, the logratio value ranges from -4 to 0, where a logratio value of 0 indicates an isotropic footprint. Increasing the logratio value decreases the number of samples averaged over a range of logratios. The amount of processing needed to compute an anisotropically filtered texture value is proportional to the number of samples needed. Furthermore, performing computations in log space simplifies the computations and may be more efficient; for example, subtraction in log space is used instead of division. Likewise, addition is used instead of multiplication”; Figs. 3A-B, and Col. 6 Lines 46-42, “In step 325 the first-modified logratio value is clamped to a value between -1024 and zero when a maximum of 16 samples are used for anisotropic filtering. Alternatively, the first-modified logratio is clamped to another value based on a different maximum number of samples supported for anisotropic filtering”; and Col. 8 Lines 1-7, “In step 365 the second-modified logratio value is clamped to a value between -1024 and zero when a maximum of 16 samples are used for anisotropic filtering, to produce the first-modified logratio value. Alternatively, the second-modified logratio value is clamped to another value based on a different maximum number of texture samples supported for anisotropic filtering”; and Fig. 2, and Col. 5 Lines 16-22, “In step 250 the first number of texture samples in the mip map specified by LODfine are read from memory. In step 250, when LODfrac does not equal zero, the second number of texture samples in the mip map specified by LODcoarse are also read from memory. In step 255 texture samples read from memory in step 250 are anisotropically filtered to produce a filtered texture sample”. Note that when the anisotropy is large, the first modified logratio is used to determine the first number of filtered samples, the second modified logratio is used to determine the second number of filtered samples, both the first and second number of samples are used to generate the filtered texture samples, and this is mapped to filtering the texture with anisotropy above the maximum amount of anisotropy).
Regarding claim 4, Toksvig and Donovan teach all the features with respect to claim 1 as outlined above. Further, Toksvig teaches that the method of claim 1, wherein said combining the plurality of intermediate filtered texture values comprises performing a weighted sum of the intermediate filtered texture values (See Toksvig: Figs. 2A-C, and Col. 5 Lines 59-67, “The length of major axis 202 in the coarse LOD mip map is 4.375, so 4.375 bilinear texture samples of bilinear samples 204 cover major axis 202. Bilinear samples 204 may be filtered to produce a filtered texture value for the coarse LOD mip map. The filtered texture value for the coarse LOD mip map is then weighed by the coarse mip map level weight (LODfrac) and combined with a filtered texture value for the fine LOD mip map to produce the filtered texture for footprint 201”; and Col. 6 Lines 1-14, “FIG. 2C illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a fine LOD mip map in accordance with one or more aspects of the present invention. Specifically ten bilinear samples, bilinear samples 205 are spaced by one texel along major axis 202 to cover major axis 202. The length of major axis 202 in the fine LOD mip map is 8.750, so 8.750 bilinear texture samples of bilinear samples 205 cover major axis 202. Bilinear samples 205 may be filtered to produce a filtered texture value for the fine LOD mip map. The filtered texture value for the fine LOD mip map is then weighed by the fine mip map level weight (1-LODfrac) and combined with the filtered texture value for the coarse LOD mip map to produce the filtered texture for footprint 201”).
Regarding claim 5, Toksvig and Donovan teach all the features with respect to claim 4 as outlined above. Further, Toksvig teaches that the method of claim 4, wherein the weights of the weighted sum are non-negative and represent a normalised filtering function (See Toksvig: Figs. 2A-C, and Col. 5 Lines 4-24, “The target level of detail, LODt, is base-two log of the length of the minor axis of a footprint. Other techniques known to those skilled in the art may be used to compute an LOD. Generally, when performing anisotropic filtering with trilinear interpolation between mipmap levels for each isotropic tap, the two LODs should be calculated as follows. The fine texture map LOD, LODfine, is set to the integer portion of LODt, and the coarse level LOD, LODcoarse, is set to LODfine+1. Interpolation between the two levels is performed according to the fractional part of LODt, LODfrac. LODfrac is a mip map level weight for the coarse LOD and 1-LODfrac is a mip map level weight for the fine LOD. When performing trilinear interpolation for an anisotropic footprint, the bilinear texture samples for the fine LOD mip map are combined to produce a fine LOD texture value and the bilinear texture samples for the coarse LOD mip map are combined to produce a coarse LOD texture value. The fine LOD texture value and the coarse LOD texture value are then interpolated according to the fractional part of LODfrac to produce a trilinearly filtered texture value for the anisotropic footprint”; and Fig. 7, and Col. 15 Lines 51-64, “In some embodiments of the present invention, anisotropic unit 700 outputs unnormalized texture coordinates, such as u, v, and p, that are dependent on mipmap dimensions for LODcoarse and LODfine and that are typically represented in a fixed point format. In other embodiments of the present invention, anisotropic unit 700 outputs normalized texture coordinates, such as s, t, and r. Anisotropic unit 900 also outputs bilinear texture sample positions, mip map level weights, LODfrac, LODfine, and LODcoarse”. Note that the weighted sum is fractional, and they are non-negative, and summed to 1 or normalized).
Regarding claim 6, Toksvig and Donovan teach all the features with respect to claim 2 as outlined above. Further, Donovan teaches that the method of claim 2, wherein the filter kernel is configured to apply anisotropic filtering in an input direction of anisotropy (See Donovan: Figs. 1A-D, and Col. 1 Lines 23-46, “Classic mip maps are isotropically filtered, i.e. filtered symmetrically in the horizontal and vertical directions using a square filter pattern. Isotropically filtered mip maps result in high quality images for surfaces with major and minor texture axes that are similar in length. However, when an isotropically filtered texture is applied to a receding surface viewed "on edge", aliasing artifacts (blurring) become apparent to a viewer as the texture is effectively "stretched" in one dimension, the receding direction, as the texture is applied to the surface. A Footprint 115 is a pixel footprint in texture space, with a Position 135 being the pixel center. FIG. 1B illustrates a prior art application of Texture Map 101 applied to pixels of a Surface 140 that is receding in image space. When viewed in image space, Footprint 115 (an ellipse) appears as Footprint 116 (a circle). While isotropic filtering of texture samples within a pixel footprint that forms a circle in texture space results in a high-quality image, isotropic filtering of texture samples within a pixel footprint that forms an ellipse, such as Footprint 115, results in an image with aliasing artifacts. In contrast to isotropic filtering, anisotropic filtering uses a rectangular shaped filter pattern, resulting in fewer aliasing artifacts for footprints with major and minor axes that are not similar in length in texture space”).
Regarding claim 7, Toksvig and Donovan teach all the features with respect to claim 1 as outlined above. Further, Toksvig teaches that the method of claim 1, wherein said plurality of sampling operations sample respective subsets of texels of the texture, wherein the respective subsets of texels are displaced with respect to each other in the texture space of the texture in accordance with an input direction of anisotropy (See Toksvig: Figs. 2A-C, and Col. 5 Lines 45-58, “FIG. 2B illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a coarse LOD mip map in accordance with one or more aspects of the present invention. Specifically six bilinear samples, bilinear samples 204 are spaced by one texel along major axis 202 to cover major axis 202. In comparison with FIG. 1D, in which ten bilinear samples are used from the coarse LOD mip map, the number of bilinear texture samples is reduced. Although the texture samples, bilinear samples 204, are labeled as bilinear texture samples, in other embodiments of the present invention texture samples may be used that are produced using other reconstruction filtering techniques known to those skilled in the art”. Note that samples along the major axis is mapped to sample subsets in user input direction).
Regarding claim 8, Toksvig and Donovan teach all the features with respect to claim 2 as outlined above. Further, Toksvig teaches that the method of claim 2, wherein the filter kernel can be adapted to apply different amounts of anisotropy between a minimum amount of anisotropy and the maximum amount of anisotropy, wherein the minimum amount of anisotropy corresponds to an anisotropic ratio of 1 (See Toksvig: Figs. 1A-E, and Col. 2 Lines 24-41, “The region in texture space corresponding to a pixel is called the pixel's "footprint". A pixel can be approximated with a circle in screen space. For texture mapping of two-dimensional textures, the corresponding footprint in texture space can be approximated by an ellipse. In classic use of mipmaps, a mipmap level is chosen so that the footprint when scaled to that level is about 1 texel (texture pixel) in diameter. Then a bilinear filter is used to interpolate between the values of four texels forming a 2.times.2 square around the footprint center, to produce a bilinear texture sample. This is called isotropic filtering, because it filters equally in the two texture space dimensions, u and v. Although the filter yielding excellent image quality, the ideal filter, has an approximately elliptical shape, isotropic filtering approximates the ellipse with a circle, to simplify the texture sampling and filtering computations. Therefore, portions of the footprint are not sampled, resulting in visual artifacts caused by undersampling”. Note that anisotropy is mapped to the anisotropy ratio of 1) and the maximum amount of anisotropy corresponds to an anisotropic ratio of 2 (See Toksvig: Figs. 2A-C, and Col. 5 Lines 25-44, “FIG. 2A illustrates an anisotropic footprint, footprint 201, in accordance with one or more aspects of the present invention. Footprint 201 has a major axis, major axis 202 of length 8.75 and a minor axis, minor axis 203, of length 1.75. The anisotropic ratio is 5 and the LOD computed as log base 2 of minor axis 203 is 0.807. LODfrac is also 0.807. Ideally, the texture samples used to produce an anisotropically filtered texture value for footprint 201 are from a mip map for a LOD of 0.807. Unfortunately, texture mapping systems typically only include prefiltered LOD mip maps for integer values. However, good filtering results may be achieved by using texture samples that are spaced by one texel for both the coarse and fine LOD mip maps, as shown in FIGS. 2B and 2C, respectively, to approximate an LOD mip map of 0.807. Spacing the bilinear texture samples by one texel produces a filtered texture value without over or under sampling, minimizing undesirable visual artifacts. Methods described in conjunction with FIGS. 3A, 5A, and 5D may be used to space the texture samples by one texel to produce a filtered texture value”. Note that the anisotropic ratio 8.75/1.75 = 5 means that Toksvig teaches the anisotropy ratio can be from 1 (isotropic) to 5, and the current cited limitation of anisotropic ratio 2 is obvious for less computational power hardware, and not patentable).
Regarding claim 17, Toksvig and Donovan teach all the features with respect to claim 1 as outlined above. Further, Toksvig and Donovan teach that a texture filtering unit configured to apply anisotropic texture filtering to a texture, the texture filtering unit being (See Toksvig: Fig. 6, and Col. 13 Lines 11-27, “FIG. 6 is a block diagram of an exemplary embodiment of a respective computer system, generally designated 600, and including a host computer 610 and a graphics subsystem 607 in accordance with one or more aspects of the present invention. Computing system 600 may be a desktop computer, server, laptop computer, palm-sized computer, tablet computer, game console, portable wireless terminal such as a PDA (personal digital assistant) or cellular telephone, computer based simulator, or the like. Host computer 610 includes host processor 614 that may include a system memory controller to interface directly to host memory 612 or may communicate with host memory 612 through a system interface 615. System interface 615 may be an I/O (input/output) interface or a bridge device including the system memory controller to interface directly to host memory 612. An example of system interface 615 known in the art includes Intel.RTM. Northbridge”) configured to:
determine whether an input amount of anisotropy (See Toksvig: Figs. 2A-C, and Col. 5 Lines 25-44, “FIG. 2A illustrates an anisotropic footprint, footprint 201, in accordance with one or more aspects of the present invention. Footprint 201 has a major axis, major axis 202 of length 8.75 and a minor axis, minor axis 203, of length 1.75. The anisotropic ratio is 5 and the LOD computed as log base 2 of minor axis 203 is 0.807. LODfrac is also 0.807. Ideally, the texture samples used to produce an anisotropically filtered texture value for footprint 201 are from a mip map for a LOD of 0.807. Unfortunately, texture mapping systems typically only include prefiltered LOD mip maps for integer values. However, good filtering results may be achieved by using texture samples that are spaced by one texel for both the coarse and fine LOD mip maps, as shown in FIGS. 2B and 2C, respectively, to approximate an LOD mip map of 0.807. Spacing the bilinear texture samples by one texel produces a filtered texture value without over or under sampling, minimizing undesirable visual artifacts. Methods described in conjunction with FIGS. 3A, 5A, and 5D may be used to space the texture samples by one texel to produce a filtered texture value”; and Fig. 6 Col. 15 Lines 41-50, “Texture unit 690 includes an anisotropic unit 700. At a high level, anisotropic unit 700 computes anisotropic texture mapping parameters such as LODt, the logratio and/or anisotropic ratio, angle of anisotropy, and the axis of anisotropy, using techniques known to those skilled in the art. As previously described, these anisotropic texture mapping parameters are used to determine the number, spacing (spread), and mip map level weights of bilinear texture samples to perform anisotropic texture filtering for a fragment”. Note that the anisotropic ratio is mapped to the input anisotropic amount; and the anisotropy parameters are used to determine the spacing, sample counts and mip map levels, and this inherently supports a practical “maximum” anisotropy amounts but a secondary art will be used to teach this maximum anisotropy concept) is above a maximum amount of anisotropy (See Donovan: Fig. 2, and Col. 4 Lines 6-26, “FIG. 2 illustrates an embodiment of a method of determining a number of texture samples for use in an anisotropic texture map filtering computation in accordance with one or more aspects of the present invention. In step 205 the base-two logarithm (log) of the minor axis length is computed to produce a logminor value. In step 210 the base-two log of the major axis length is computed to produce a logmajor value. In step 215 a logratio value is computed by subtracting the logmajor value from the logminor value. The logratio value is equivalent to the base-two log of the ratio value. In embodiments supporting a maximum anisotropy of 1/16, the logratio value ranges from -4 to 0, where a logratio value of 0 indicates an isotropic footprint. Increasing the logratio value decreases the number of samples averaged over a range of logratios. The amount of processing needed to compute an anisotropically filtered texture value is proportional to the number of samples needed. Furthermore, performing computations in log space simplifies the computations and may be more efficient; for example, subtraction in log space is used instead of division. Likewise, addition is used instead of multiplication”; Figs. 3A-B, and Col. 6 Lines 46-42, “In step 325 the first-modified logratio value is clamped to a value between -1024 and zero when a maximum of 16 samples are used for anisotropic filtering. Alternatively, the first-modified logratio is clamped to another value based on a different maximum number of samples supported for anisotropic filtering”; and Col. 8 Lines 1-7, “In step 365 the second-modified logratio value is clamped to a value between -1024 and zero when a maximum of 16 samples are used for anisotropic filtering, to produce the first-modified logratio value. Alternatively, the second-modified logratio value is clamped to another value based on a different maximum number of texture samples supported for anisotropic filtering”. Note that maximum anisotropy is explicitly used to control the sample counts, making the conditional “if above max” clear and obvious in anisotropic filtering);
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy, perform a sampling operation to sample texels of the texture to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy (See Toksvig: Figs. 1A-E, and Col. 2 Lines 19-32, “FIG. 1D illustrates a prior art application of ten bilinear samples, bilinear samples 140, that are positioned along major axis 130 to approximate an elliptical footprint for a coarse LOD mip map, such as footprint 115. Each bilinear sample corresponds to an isotropically filtered texture sample for an LOD of a texture map that is computed using conventional bilinear isotropic filtering. Each bilinear sample of bilinear samples 140 is spaced less than one texel apart. In some conventional systems, the bilinear samples in the coarse LOD are spaced by 0.5 to 1.0 texels apart. Therefore the coarse LOD mip map is oversampled, possibly introducing visual artifacts and requiring more computations and texel reads than if the spacing were one texel apart”. Note that for lower anisotropy, close to isotropic, bilinear samples are adjusted by the anisotropic ratio/LOD, and this maps to a (single effective) sampling operation tuned to the anisotropy amount); and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy (See Toksvig: Figs. 2A-C, and Col. 5 Lines 25-44, “FIG. 2A illustrates an anisotropic footprint, footprint 201, in accordance with one or more aspects of the present invention. Footprint 201 has a major axis, major axis 202 of length 8.75 and a minor axis, minor axis 203, of length 1.75. The anisotropic ratio is 5 and the LOD computed as log base 2 of minor axis 203 is 0.807. LODfrac is also 0.807. Ideally, the texture samples used to produce an anisotropically filtered texture value for footprint 201 are from a mip map for a LOD of 0.807. Unfortunately, texture mapping systems typically only include prefiltered LOD mip maps for integer values. However, good filtering results may be achieved by using texture samples that are spaced by one texel for both the coarse and fine LOD mip maps, as shown in FIGS. 2B and 2C, respectively, to approximate an LOD mip map of 0.807. Spacing the bilinear texture samples by one texel produces a filtered texture value without over or under sampling, minimizing undesirable visual artifacts. Methods described in conjunction with FIGS. 3A, 5A, and 5D may be used to space the texture samples by one texel to produce a filtered texture value”; and Figs. 1-E, and Col. 1 Lines 63-67 ~ Col. 2 Lines 1-7, “FIG. 1C illustrates footprint 115 including a minor axis 125 that is significantly shorter than a major axis 130. Texture samples along major axis 130, the axis of anisotropy, are read from one or more mipmap levels and are blended to produce a pixel color. The level from which the samples are read is determined using a level of detail (LOD) value which is nominally the log base 2 of the length of minor axis 125. The number of texture samples read from the texture map is determined based on the ratio of the major axis to the minor axis, the anisotropic ratio, with more texture samples needed as the ratio increases, i.e. as the ellipse becomes more elongated”. Note that when the anisotropy is large, multiple bilinear sampling operations are performed along the major axis):
perform a plurality of sampling operations to sample texels of the texture to determine a respective plurality of intermediate filtered texture values (See Toksvig: Figs. 2A-C, and Col. 5 Lines 45-58, “FIG. 2B illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a coarse LOD mip map in accordance with one or more aspects of the present invention. Specifically six bilinear samples, bilinear samples 204 are spaced by one texel along major axis 202 to cover major axis 202. In comparison with FIG. 1D, in which ten bilinear samples are used from the coarse LOD mip map, the number of bilinear texture samples is reduced. Although the texture samples, bilinear samples 204, are labeled as bilinear texture samples, in other embodiments of the present invention texture samples may be used that are produced using other reconstruction filtering techniques known to those skilled in the art”; Col. 5 Lines 59-67, “The length of major axis 202 in the coarse LOD mip map is 4.375, so 4.375 bilinear texture samples of bilinear samples 204 cover major axis 202. Bilinear samples 204 may be filtered to produce a filtered texture value for the coarse LOD mip map. The filtered texture value for the coarse LOD mip map is then weighed by the coarse mip map level weight (LODfrac) and combined with a filtered texture value for the fine LOD mip map to produce the filtered texture for footprint 201”; and Col. 6 Lines 1-14, “FIG. 2C illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a fine LOD mip map in accordance with one or more aspects of the present invention. Specifically ten bilinear samples, bilinear samples 205 are spaced by one texel along major axis 202 to cover major axis 202. The length of major axis 202 in the fine LOD mip map is 8.750, so 8.750 bilinear texture samples of bilinear samples 205 cover major axis 202. Bilinear samples 205 may be filtered to produce a filtered texture value for the fine LOD mip map. The filtered texture value for the fine LOD mip map is then weighed by the fine mip map level weight (1-LODfrac) and combined with the filtered texture value for the coarse LOD mip map to produce the filtered texture for footprint 201”. Note that both coarse and fine mip map are filtered samples and they are mapped to the intermediate filtered texture values); and
combine the plurality of intermediate filtered texture values to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy (See Toksvig: Figs. 2A-C, and Col. 6 Lines 1-14, “FIG. 2C illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a fine LOD mip map in accordance with one or more aspects of the present invention. Specifically ten bilinear samples, bilinear samples 205 are spaced by one texel along major axis 202 to cover major axis 202. The length of major axis 202 in the fine LOD mip map is 8.750, so 8.750 bilinear texture samples of bilinear samples 205 cover major axis 202. Bilinear samples 205 may be filtered to produce a filtered texture value for the fine LOD mip map. The filtered texture value for the fine LOD mip map is then weighed by the fine mip map level weight (1-LODfrac) and combined with the filtered texture value for the coarse LOD mip map to produce the filtered texture for footprint 201”. Note that both coarse and fine mip map are filtered samples and they are combined to produce the final filtered texture values).
Regarding claim 19, Toksvig and Donovan teach all the features with respect to claim 17 as outlined above. Further, Toksvig teaches that a graphics processing unit, comprising the texture filtering unit as set forth in claim 17, wherein the graphics processing unit is configured to use the filtered texture value determined by the texture filtering unit to render an image of a scene in which the texture is applied to a surface in the scene (See Toksvig: Figs. 6-7, and Col. 14 Lines 64-67 ~ Col. 15 Lines 1-12, “Fragment shader 655 outputs the shaded fragment data, e.g., color and depth, and codewords generated from shader program instructions to raster operations unit 665. Raster operations unit 665 includes a read interface and a write interface to memory controller 620 through which raster operations unit 665 accesses data stored in local memory 640 or host memory 612. Raster operations unit 665 optionally performs near and far plane clipping and raster operations, such as stencil, z test, blending, and the like, using the fragment data and pixel data stored in local memory 640 or host memory 612 at a pixel position (image location specified by x,y coordinates) associated with the processed fragment data. The output data from raster operations unit 665 is written back to local memory 640 or host memory 612 at the pixel position associated with the output data and the results, e.g., image data are saved in graphics memory”; Col. 15 Lines 21-40, “FIG. 7 is a block diagram of an exemplary embodiment of texture unit 690 from FIG. 6 in accordance with one or more aspects of the present invention. In some embodiments, Texture unit 690 receives data, e.g., program instructions and parameters associated with fragments (texture identifiers, texture coordinates such as s, t, and r, and the like) from a rasterizer, such as rasterizer 650. Texture coordinates s, t, and r are typically represented in a floating point format such as a 32 bit format (1 bit sign, 23 bit mantissa, and 8 bit exponent). A fragment is formed by the intersection of a pixel and a primitive. Primitives include geometry, such as points, lines, triangles, quadrilaterals, meshes, surfaces, and the like. A fragment may cover a pixel or a portion of a pixel. Likewise, a pixel may include one or more fragments. Texture unit 690 receives texture coordinates from rasterizer 650, uses the texture coordinates to perform anisotropic texture filtering of texels read from memory, and then outputs a filtered texture value corresponding to a set of texture coordinates associated with a fragment, e.g., a textured fragment”; and Col. 16 Lines 51-64, “Conventionally, the same number of texture samples is used to sample the coarse LOD and the fine LOD. Therefore, the coarse LOD is oversampled and more texels are read and processed than are needed to produce an image of acceptable quality. The number of texture samples may be reduced based on a mip map level weight to produce each anisotropically filtered texture value, possibly improving system performance. The spacing between each of the texture samples may also be modified, in turn reducing or increasing the number of texture samples. Additionally, the spacing between each of the texture samples may be modified to improve the image quality or caching performance for reading texels without changing the number of texture samples”. Note that primitives are processed and images are generated, and this is mapped to the current limitation of generating a scene).
Regarding claim 20, Toksvig and Donovan teach all the features with respect to claim 1 as outlined above. Further, Toksvig and Donovan teach that a non-transitory computer readable storage medium having stored thereon an integrated circuit definition dataset that, when processed in an integrated circuit manufacturing system, configures the integrated circuit manufacturing system to manufacture a texturing filtering unit which is configured to apply anisotropic texture filtering to a texture, the texture filtering unit being (See Toksvig: Fig. 6, and Col. 13 Lines 11-27, “FIG. 6 is a block diagram of an exemplary embodiment of a respective computer system, generally designated 600, and including a host computer 610 and a graphics subsystem 607 in accordance with one or more aspects of the present invention. Computing system 600 may be a desktop computer, server, laptop computer, palm-sized computer, tablet computer, game console, portable wireless terminal such as a PDA (personal digital assistant) or cellular telephone, computer based simulator, or the like. Host computer 610 includes host processor 614 that may include a system memory controller to interface directly to host memory 612 or may communicate with host memory 612 through a system interface 615. System interface 615 may be an I/O (input/output) interface or a bridge device including the system memory controller to interface directly to host memory 612. An example of system interface 615 known in the art includes Intel.RTM. Northbridge”) configured to:
determine whether an input amount of anisotropy (See Toksvig: Figs. 2A-C, and Col. 5 Lines 25-44, “FIG. 2A illustrates an anisotropic footprint, footprint 201, in accordance with one or more aspects of the present invention. Footprint 201 has a major axis, major axis 202 of length 8.75 and a minor axis, minor axis 203, of length 1.75. The anisotropic ratio is 5 and the LOD computed as log base 2 of minor axis 203 is 0.807. LODfrac is also 0.807. Ideally, the texture samples used to produce an anisotropically filtered texture value for footprint 201 are from a mip map for a LOD of 0.807. Unfortunately, texture mapping systems typically only include prefiltered LOD mip maps for integer values. However, good filtering results may be achieved by using texture samples that are spaced by one texel for both the coarse and fine LOD mip maps, as shown in FIGS. 2B and 2C, respectively, to approximate an LOD mip map of 0.807. Spacing the bilinear texture samples by one texel produces a filtered texture value without over or under sampling, minimizing undesirable visual artifacts. Methods described in conjunction with FIGS. 3A, 5A, and 5D may be used to space the texture samples by one texel to produce a filtered texture value”; and Fig. 6 Col. 15 Lines 41-50, “Texture unit 690 includes an anisotropic unit 700. At a high level, anisotropic unit 700 computes anisotropic texture mapping parameters such as LODt, the logratio and/or anisotropic ratio, angle of anisotropy, and the axis of anisotropy, using techniques known to those skilled in the art. As previously described, these anisotropic texture mapping parameters are used to determine the number, spacing (spread), and mip map level weights of bilinear texture samples to perform anisotropic texture filtering for a fragment”. Note that the anisotropic ratio is mapped to the input anisotropic amount; and the anisotropy parameters are used to determine the spacing, sample counts and mip map levels, and this inherently supports a practical “maximum” anisotropy amounts but a secondary art will be used to teach this maximum anisotropy concept) is above a maximum amount of anisotropy (See Donovan: Fig. 2, and Col. 4 Lines 6-26, “FIG. 2 illustrates an embodiment of a method of determining a number of texture samples for use in an anisotropic texture map filtering computation in accordance with one or more aspects of the present invention. In step 205 the base-two logarithm (log) of the minor axis length is computed to produce a logminor value. In step 210 the base-two log of the major axis length is computed to produce a logmajor value. In step 215 a logratio value is computed by subtracting the logmajor value from the logminor value. The logratio value is equivalent to the base-two log of the ratio value. In embodiments supporting a maximum anisotropy of 1/16, the logratio value ranges from -4 to 0, where a logratio value of 0 indicates an isotropic footprint. Increasing the logratio value decreases the number of samples averaged over a range of logratios. The amount of processing needed to compute an anisotropically filtered texture value is proportional to the number of samples needed. Furthermore, performing computations in log space simplifies the computations and may be more efficient; for example, subtraction in log space is used instead of division. Likewise, addition is used instead of multiplication”; Figs. 3A-B, and Col. 6 Lines 46-42, “In step 325 the first-modified logratio value is clamped to a value between -1024 and zero when a maximum of 16 samples are used for anisotropic filtering. Alternatively, the first-modified logratio is clamped to another value based on a different maximum number of samples supported for anisotropic filtering”; and Col. 8 Lines 1-7, “In step 365 the second-modified logratio value is clamped to a value between -1024 and zero when a maximum of 16 samples are used for anisotropic filtering, to produce the first-modified logratio value. Alternatively, the second-modified logratio value is clamped to another value based on a different maximum number of texture samples supported for anisotropic filtering”. Note that maximum anisotropy is explicitly used to control the sample counts, making the conditional “if above max” clear and obvious in anisotropic filtering);
if it is determined that the input amount of anisotropy is not above the maximum amount of anisotropy, perform a sampling operation to sample texels of the texture to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy (See Toksvig: Figs. 1A-E, and Col. 2 Lines 19-32, “FIG. 1D illustrates a prior art application of ten bilinear samples, bilinear samples 140, that are positioned along major axis 130 to approximate an elliptical footprint for a coarse LOD mip map, such as footprint 115. Each bilinear sample corresponds to an isotropically filtered texture sample for an LOD of a texture map that is computed using conventional bilinear isotropic filtering. Each bilinear sample of bilinear samples 140 is spaced less than one texel apart. In some conventional systems, the bilinear samples in the coarse LOD are spaced by 0.5 to 1.0 texels apart. Therefore the coarse LOD mip map is oversampled, possibly introducing visual artifacts and requiring more computations and texel reads than if the spacing were one texel apart”. Note that for lower anisotropy, close to isotropic, bilinear samples are adjusted by the anisotropic ratio/LOD, and this maps to a (single effective) sampling operation tuned to the anisotropy amount); and
if it is determined that the input amount of anisotropy is above the maximum amount of anisotropy (See Toksvig: Figs. 2A-C, and Col. 5 Lines 25-44, “FIG. 2A illustrates an anisotropic footprint, footprint 201, in accordance with one or more aspects of the present invention. Footprint 201 has a major axis, major axis 202 of length 8.75 and a minor axis, minor axis 203, of length 1.75. The anisotropic ratio is 5 and the LOD computed as log base 2 of minor axis 203 is 0.807. LODfrac is also 0.807. Ideally, the texture samples used to produce an anisotropically filtered texture value for footprint 201 are from a mip map for a LOD of 0.807. Unfortunately, texture mapping systems typically only include prefiltered LOD mip maps for integer values. However, good filtering results may be achieved by using texture samples that are spaced by one texel for both the coarse and fine LOD mip maps, as shown in FIGS. 2B and 2C, respectively, to approximate an LOD mip map of 0.807. Spacing the bilinear texture samples by one texel produces a filtered texture value without over or under sampling, minimizing undesirable visual artifacts. Methods described in conjunction with FIGS. 3A, 5A, and 5D may be used to space the texture samples by one texel to produce a filtered texture value”; and Figs. 1-E, and Col. 1 Lines 63-67 ~ Col. 2 Lines 1-7, “FIG. 1C illustrates footprint 115 including a minor axis 125 that is significantly shorter than a major axis 130. Texture samples along major axis 130, the axis of anisotropy, are read from one or more mipmap levels and are blended to produce a pixel color. The level from which the samples are read is determined using a level of detail (LOD) value which is nominally the log base 2 of the length of minor axis 125. The number of texture samples read from the texture map is determined based on the ratio of the major axis to the minor axis, the anisotropic ratio, with more texture samples needed as the ratio increases, i.e. as the ellipse becomes more elongated”. Note that when the anisotropy is large, multiple bilinear sampling operations are performed along the major axis):
perform a plurality of sampling operations to sample texels of the texture to determine a respective plurality of intermediate filtered texture values (See Toksvig: Figs. 2A-C, and Col. 5 Lines 45-58, “FIG. 2B illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a coarse LOD mip map in accordance with one or more aspects of the present invention. Specifically six bilinear samples, bilinear samples 204 are spaced by one texel along major axis 202 to cover major axis 202. In comparison with FIG. 1D, in which ten bilinear samples are used from the coarse LOD mip map, the number of bilinear texture samples is reduced. Although the texture samples, bilinear samples 204, are labeled as bilinear texture samples, in other embodiments of the present invention texture samples may be used that are produced using other reconstruction filtering techniques known to those skilled in the art”; Col. 5 Lines 59-67, “The length of major axis 202 in the coarse LOD mip map is 4.375, so 4.375 bilinear texture samples of bilinear samples 204 cover major axis 202. Bilinear samples 204 may be filtered to produce a filtered texture value for the coarse LOD mip map. The filtered texture value for the coarse LOD mip map is then weighed by the coarse mip map level weight (LODfrac) and combined with a filtered texture value for the fine LOD mip map to produce the filtered texture for footprint 201”; and Col. 6 Lines 1-14, “FIG. 2C illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a fine LOD mip map in accordance with one or more aspects of the present invention. Specifically ten bilinear samples, bilinear samples 205 are spaced by one texel along major axis 202 to cover major axis 202. The length of major axis 202 in the fine LOD mip map is 8.750, so 8.750 bilinear texture samples of bilinear samples 205 cover major axis 202. Bilinear samples 205 may be filtered to produce a filtered texture value for the fine LOD mip map. The filtered texture value for the fine LOD mip map is then weighed by the fine mip map level weight (1-LODfrac) and combined with the filtered texture value for the coarse LOD mip map to produce the filtered texture for footprint 201”. Note that both coarse and fine mip map are filtered samples and they are mapped to the intermediate filtered texture values); and
combine the plurality of intermediate filtered texture values to determine a filtered texture value which has been filtered in accordance with the input amount of anisotropy (See Toksvig: Figs. 2A-C, and Col. 6 Lines 1-14, “FIG. 2C illustrates a number of bilinear samples spread along major axis 202 to approximate elliptical footprint 201 of FIG. 2A for a fine LOD mip map in accordance with one or more aspects of the present invention. Specifically ten bilinear samples, bilinear samples 205 are spaced by one texel along major axis 202 to cover major axis 202. The length of major axis 202 in the fine LOD mip map is 8.750, so 8.750 bilinear texture samples of bilinear samples 205 cover major axis 202. Bilinear samples 205 may be filtered to produce a filtered texture value for the fine LOD mip map. The filtered texture value for the fine LOD mip map is then weighed by the fine mip map level weight (1-LODfrac) and combined with the filtered texture value for the coarse LOD mip map to produce the filtered texture for footprint 201”. Note that both coarse and fine mip map are filtered samples and they are combined to produce the final filtered texture values).
Allowable Subject Matter
Claims 9-16 and 18 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. The best arts searched do not teach the cited limitations of “the method of claim 2, wherein the texture is represented with a mipmap comprising a plurality of levels, wherein each level of the mipmap comprises an image representing the texture at a respective level of detail, wherein the texture filtering unit has minimum and maximum limits on an amount by which it can alter the level of detail when it uses the filter kernel to filter texels from an image of a single level of the mipmap, wherein the range of level of detail between the minimum and maximum limits defines an intrinsic region of the texture filtering unit, and wherein levels of detail outside of the range of level of detail between the minimum and maximum limits define an extrinsic region of the texture filtering unit, wherein the method comprises: receiving an input level of detail for filtering the texture; determining whether the received input level of detail is in an intrinsic region or an extrinsic region of the texture filtering unit; if it is determined that the received input level of detail is in an intrinsic region of the texture filtering unit: reading texels from a single mipmap level of the mipmap; and filtering the read texels from the single mipmap level with the filter kernel of the texture filtering unit to determine a filtered texture value representing part of the texture at the input level of detail; and if it is determined that the received input level of detail is in an extrinsic region of the texture filtering unit: reading texels from two mipmap levels of the mipmap; and processing the read texels from the two mipmap levels with the texture filtering unit to determine a filtered texture value representing part of the texture at the input level of detail.”
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GORDON G LIU whose telephone number is (571)270-0382. The examiner can normally be reached Monday - Friday 8:00-5:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Devona E Faulk can be reached at 571-272-7515. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/GORDON G LIU/Primary Examiner, Art Unit 2618
Prosecution Insights