Closed GoogleCodeExporter closed 8 years ago
Built and tested with current head on Linux, the encoding went ok. Further
tests are being done on windows, results pending.
Original comment by ya...@google.com
on 31 Jul 2013 at 5:09
Original comment by ya...@google.com
on 31 Jul 2013 at 5:10
Tested with windows build: cygwin+vs2012. The encoding has gone through to
frame 250 without any issue.
Original comment by ya...@google.com
on 31 Jul 2013 at 2:40
Frank, can you let me know how you built vpxenc.exe? It could have been the
issue is specific to the build you had. Thanks.
Original comment by ya...@google.com
on 31 Jul 2013 at 2:42
# master
cd ~/on2
rm -rf libvpx
git clone http://git.chromium.org/webm/libvpx.git
rm -rf win64
mkdir win64
cd win64
CROSS=x86_64-w64-mingw32- ../libvpx/configure --target=x86_64-win64-gcc
--enable-experimental --enable-static-msvcrt --disable-install-docs
--disable-unit-tests --disable-docs
make -i V=1
Original comment by fbarch...@google.com
on 31 Jul 2013 at 8:07
Removing --enable-experimental a crash still occurs
e:\mediatests\garden>timex ..\vpxenc -w 4096 -h 2304 --fps=24000/1001
--target-bitrate=6000 garden.4096x2304_24Hz_P420.yuv -o garden0.tmp.vp9.
webm --fpf=garden_vp9.fpf -p 2 --pass=2 --codec=vp9 --good --cpu-used=0
--lag-in-frames=25 --min-q=0 --max-q=63 --end-usage=vbr --auto-alt-ref
=1 --kf-max-dist=9999 --kf-min-dist=0 --drop-frame=0 --static-thresh=0
--bias-pct=50 --minsection-pct=0 --maxsection-pct=2000 --arnr-maxframes
=7 --arnr-strength=5 --arnr-type=3 --sharpness=0 --undershoot-pct=100 -v --psnr
Pass 2/2 frame 45/29 6116893B 2322940 ms 52794.09 ms/f [ETA 83:07:37]
38.140 38.099 37.729 38.785 275107F
This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.
Pass 2/2 frame 59/34 9988545B 4347467 ms 73685.89 ms/f [ETA 93:45:46]
Assertion failed!
Program: c:\bin\vpxenc_jul31.exe
File: ../libvpx/vp9/common/vp9_entropymv.c, Line 95
Expression: 0
It does not occur if I turn off alt ref.
It does not occur with cpu used of 1 or 2, but does with 0.
Original comment by fbarch...@google.com
on 2 Aug 2013 at 5:09
Yaowu is this something you are working on? Can we close this?
Original comment by jimbankoski@google.com
on 22 Aug 2013 at 1:27
Issue no longer reproduces with vpxenc built aug 24 VP9 Encoder
v1.2.0-4110-g66ccf5d
Original comment by fbarch...@google.com
on 27 Aug 2013 at 4:21
Thank you very much for the update, Frank.
There were a number of bug fixes merged, it appears that they, (I suspect the
ones limit mv search range into borders), might have fixed the underline issue
for your particular build.
Let's close this for now.
Original comment by ya...@google.com
on 27 Aug 2013 at 4:40
Please re-open... I've repro'ed it with cpu-used=5 with aug 24 version:
vpxenc -w 4096 -h 2304 --fps=24000/1001 --target-bitrate=6000
garden.4096x2304_24Hz_P420.yuv -o garden0.tmp1.vp9
.webm --fpf=garden_vp9.fpf -p 2 --pass=2 --codec=vp9 --good --cpu-used=5
--lag-in-frames=25 --min-q=0 --max-q=63 --end-usage=vbr --auto-alt-re
f=1 --kf-max-dist=9999 --kf-min-dist=0 --drop-frame=0 --static-thresh=0
--bias-pct=50 --minsection-pct=0 --maxsection-pct=2000 --arnr-maxframe
s=7 --arnr-strength=5 --arnr-type=3 --sharpness=0 --undershoot-pct=100 -v --psnr
Pass 2/2 frame 70/45 12873967B 5728594 ms 81837.06 ms/f [ETA 94:08:28]
←[KAssertion failed!611 38.643 189836F
Program: c:\bin\vpxenc_aug24.exe
File: ../libvpx/vp9/common/vp9_entropymv.c, Line 95
Expression: 0
This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.
Original comment by fbarch...@chromium.org
on 27 Aug 2013 at 6:18
The command line appears to have only the 2nd pass, can you double check and
make sure that the first pass is running on same length of input video as the
second pass?
Original comment by ya...@google.com
on 28 Aug 2013 at 12:23
The issue appears to be fixed in sept 6 version, VP9 Encoder
v1.2.0-4213-gd1268c5
Pass 2/2 frame 681/656 38272025B 38372958 ms 56347.48 ms/f [ETA 33:43:49]
45.707 44.610 48.638 50.376 13869F
Original comment by fbarch...@google.com
on 10 Sep 2013 at 5:11
Original issue reported on code.google.com by
fbarch...@google.com
on 30 Jul 2013 at 12:42