Need help to find good settings for my source [Archive]

View Full Version : Need help to find good settings for my source

juGGaKNot

6th May 2009, 14:59

Hi again

First of all i would like to say thnx to all of you, i have learned a lot of "stuff" here, usefull or less usefull that is another thing to talk about :P but nevertheless THNX.

I want to find good setting for my source, it being always uncompressed ( that i make ) so not the general movies one would find here ( rip or otherwise )

I need to make my profile final because i am thinking of making a gui so here goes :

The Source Part 1, (http://www.mediafire.com/download.php?x5znyijzfou)Part 2, (http://www.mediafire.com/download.php?munymdyrtzu)Part 3, (http://www.mediafire.com/download.php?hqjnnmymzoy)Part 4, (http://www.mediafire.com/download.php?zoqgzjwrymd)Part 5, (http://www.mediafire.com/download.php?njgzmdgimme)Part 6 (http://www.mediafire.com/download.php?522jttqjtew)

Things that i want : PS3/XBOX/Standalone players/blu-ray compatibility for the .264 stream

Most of the resolutions used are in the range 800-600/1024x576/1280x800, i will use 1920x1080 later on so i am thinking of :

profile 4.0, 3r frames, 3b frames with b-piramid ( b-piramid problems with hardware decoding are gone if using <3 refs and <3 b frames no ?

What vfv buffer and maxrate should i use for compatibility ?

As for the rest, you have the source, encode at will and post your results, 20 MB/minute - 30 MB/Minute is the general range for these type of movies.

Things to look for : in the upper right corner you have a kill text, at low bitrate you will see a lot of noise there.

If you have time please post your log/x264 version/encoded video and what you think will help quality for my source.

THNX in advance, cheers.

buzzqw

6th May 2009, 17:35

with b-piramid
i prefer 4 ref , a little higher subme, and NO b pyramid

BHH

juGGaKNot

6th May 2009, 20:23

i prefer 4 ref , a little higher subme, and NO b pyramid

BHH

I see, right now i don't use b-pyramid i was just asking

The problem with the source ( related to this thread (http://forum.doom9.org/showthread.php?t=146747) ) is that older builds of x264 give better quality than new ones :

x264 core 66 r1114M a933a3e

E:\>"D:\juGGaKNot & Friends\RipBot264v1.13.3\tools\x264\x264.exe" "E:\temp\RipBot264temp\job1\job1.avs" --pass 1 --bitrate 2700 --stats "E:\temp\RipBot264temp\job1\job1.stats" --sar 1:1 --level 4.0 --aud --nal-hrd --vbv-bufsize 25000 --vbv-maxrate 25000 --filter 0,0 --ref 3 --mixed-refs --bframes 3 --b-adapt 2 --weightb --direct auto --subme 9 --trellis 2 --partitions all --8x8dct --me tesa --threads auto --thread-input --progress --no-psnr --no-ssim --partitions none --subme 3 --me dia --output "E:\temp\RipBot264temp\video.264"

avis [info]: 1024x576 @ 30.00 fps (600 frames)
x264 [info]: using SAR=1/1
x264 [info]: using cpu capabilities: MMX2 SSE2Slow Slow_mod4_stack
x264 [info]: profile High, level 4.0
x264 [info]: slice I:14 Avg QP:22.56 size: 25578
x264 [info]: slice P:285 Avg QP:25.10 size: 14657
x264 [info]: slice B:301 Avg QP:31.19 size: 7899
x264 [info]: consecutive B-frames: 17.7% 30.0% 33.8% 18.4%
x264 [info]: mb I I16..4: 28.5% 53.8% 17.7%
x264 [info]: mb P I16..4: 37.0% 0.0% 0.0% P16..4: 40.1% 0.0% 0.0% 0.0% 0.0% skip:22.8%
x264 [info]: mb B I16..4: 13.3% 0.0% 0.0% B16..8: 34.0% 0.0% 0.0% direct:13.9% skip:38.8% L0:43.6% L1:45.4% BI:11.1%
x264 [info]: final ratefactor: 24.80
x264 [info]: 8x8 transform intra:4.7% inter:57.6%
x264 [info]: direct mvs spatial:98.0% temporal:2.0%
x264 [info]: ref P L0 77.4% 13.6% 9.0%
x264 [info]: ref B L0 80.9% 19.1%
x264 [info]: kb/s:2765.2

encoded 600 frames, 10.25 fps, 2770.68 kb/s

E:\>"D:\juGGaKNot & Friends\RipBot264v1.13.3\tools\x264\x264.exe" "E:\temp\RipBot264temp\job1\job1.avs" --pass 2 --bitrate 2700 --stats "E:\temp\RipBot264temp\job1\job1.stats" --sar 1:1 --level 4.0 --aud --nal-hrd --vbv-bufsize 25000 --vbv-maxrate 25000 --filter 0,0 --ref 3 --mixed-refs --bframes 3 --b-adapt 2 --weightb --direct auto --subme 9 --trellis 2 --partitions all --8x8dct --me tesa --threads auto --thread-input --progress --no-psnr --no-ssim --output "E:\temp\RipBot264temp\video.264"

avis [info]: 1024x576 @ 30.00 fps (600 frames)
x264 [info]: using SAR=1/1
x264 [info]: using cpu capabilities: MMX2 SSE2Slow Slow_mod4_stack
x264 [info]: profile High, level 4.0
x264 [info]: slice I:14 Avg QP:23.30 size: 24542
x264 [info]: slice P:285 Avg QP:25.92 size: 13321
x264 [info]: slice B:301 Avg QP:31.19 size: 8202
x264 [info]: consecutive B-frames: 17.7% 30.0% 33.8% 18.4%
x264 [info]: mb I I16..4: 25.6% 67.2% 7.2%
x264 [info]: mb P I16..4: 8.8% 19.0% 1.3% P16..4: 32.5% 11.0% 4.4% 0.1% 0.1% skip:22.8%
x264 [info]: mb B I16..4: 1.0% 3.0% 0.2% B16..8: 47.3% 2.2% 2.6% direct:10.5% skip:33.3% L0:46.4% L1:49.6% BI: 4.0%
x264 [info]: 8x8 transform intra:66.1% inter:84.0%
x264 [info]: direct mvs spatial:95.3% temporal:4.7%
x264 [info]: ref P L0 78.6% 13.2% 8.2%
x264 [info]: ref B L0 83.3% 16.7%
x264 [info]: kb/s:2643.5
encoded 600 frames, 1.67 fps, 2649.03 kb/s

x264 core 47 svn-531

avis [info]: 1024x576 @ 30.00 fps (600 frames)
x264 [info]: using cpu capabilities MMX MMXEXT SSE SSE2 3DNow!
x264 [info]: slice I:9 Avg QP:25.22 size: 23823:00
x264 [info]: slice P:388 Avg QP:27.85 size: 14847
x264 [info]: slice B:203 Avg QP:30.03 size: 4823
x264 [info]: mb I I16..4: 36.5% 53.3% 10.3%
x264 [info]: mb P I16..4: 17.6% 30.7% 4.1% P16..4: 27.3% 5.6% 0.8% 0.1% 0
.0% skip:13.8%
x264 [info]: mb B I16..4: 1.3% 7.6% 1.1% B16..8: 20.5% 1.2% 1.6% direct:
5.0% skip:61.6%
x264 [info]: final ratefactor: 25.09
x264 [info]: 8x8 transform intra:59.9% inter:56.3%
x264 [info]: ref P 76.2% 14.6% 9.2%
x264 [info]: ref B 73.1% 16.3% 10.6%
x264 [info]: kb/s:2781.7

encoded 600 frames, 2.93 fps, 2782.11 kb/s
avis [info]: 1024x576 @ 30.00 fps (600 frames)
x264 [info]: using cpu capabilities MMX MMXEXT SSE SSE2 3DNow!
x264 [info]: slice I:9 Avg QP:25.00 size: 21971:00
x264 [info]: slice P:388 Avg QP:27.97 size: 13948
x264 [info]: slice B:203 Avg QP:28.44 size: 5896
x264 [info]: mb I I16..4: 40.0% 51.8% 8.2%
x264 [info]: mb P I16..4: 19.0% 31.7% 2.7% P16..4: 26.6% 6.1% 0.9% 0.1% 0
.0% skip:12.9%
x264 [info]: mb B I16..4: 1.4% 8.0% 1.3% B16..8: 22.6% 1.5% 2.1% direct:
6.5% skip:56.8%
x264 [info]: 8x8 transform intra:60.5% inter:54.3%
x264 [info]: ref P 76.0% 14.6% 9.4%
x264 [info]: ref B 72.9% 16.4% 10.6%
x264 [info]: kb/s:2722.6

encoded 600 frames, 3.10 fps, 2723.09 kb/s

New build better ratefactor but less quality ( more noise around killtext )

So i am trying to find a good cfg, some people use trellis 0/psy trellis 0 and aq 0 and get good quality for ex.

juGGaKNot

8th May 2009, 09:59

http://www.mediafire.com/download.php?qlzvnywrz4y

Writing library : x264 core 67 r1145M d2e1e1c
Encoding settings : cabac=1 / ref=5 / deblock=1:0:0 / analyse=0x3:0x133 / me=tesa / subme=7 / psy_rd=1.0:0.0 / mixed_ref=1 / me_range=32 / chroma_me=1 / trellis=0 / 8x8dct=1 / cqm=0 / deadzone=21,11 / chroma_qp_offset=-2 / threads=3 / nr=0 / decimate=0 / mbaff=0 / bframes=3 / b_pyramid=0 / b_adapt=2 / b_bias=0 / direct=2 / wpredb=1 / keyint=250 / keyint_min=25 / scenecut=40 / rc=2pass / bitrate=3100 / ratetol=2.0 / qcomp=0.60 / qpmin=5 / qpmax=51 / qpstep=6 / cplxblur=20.0 / qblur=0.5 / vbv_maxrate=25000 / vbv_bufsize=25000 / ip_ratio=1.40 / pb_ratio=1.30 / aq=0

http://www.mediafire.com/download.php?dehnmaeznzn

Writing library : x264 core 47 svn-531
Encoding settings : cabac=1 / ref=3 / deblock=1:0:0 / analyse=0x3:0x133 / me=umh / subme=6 / brdo=0 / mixed_ref=0 / me_range=24 / chroma_me=1 / trellis=0 / 8x8dct=1 / cqm=0 / chroma_qp_offset=0 / slices=1 / nr=0 / decimate=1 / bframes=3 / b_pyramid=0 / b_adapt=1 / b_bias=0 / direct=1 / wpredb=1 / bime=0 / keyint=250 / keyint_min=25 / scenecut=40 / rc=2pass / bitrate=3100 / ratetol=1.0 / rceq='blurCplx^(1-qComp)' / qcomp=0.60 / qpmin=4 / qpmax=51 / qpstep=6 / cplxblur=20.0 / qblur=0.5 / ip_ratio=1.40 / pb_ratio=1.30

The 2'nd one looks better, check the noise @ killtext ..

So, any takers ? the uncompressed audio and video is in the first post.

juGGaKNot

12th May 2009, 07:27

Last up, no takers ?

poisondeathray

12th May 2009, 16:19

Disable psy-rd for a cleaner, less noisy image

juGGaKNot

12th May 2009, 21:37

Disable psy-rd for a cleaner, less noisy image

K, noted, any other things ?

juGGaKNot

14th May 2009, 14:24

Disable psy-rd for a cleaner, less noisy image

r531 Download (http://www.mediafire.com/download.php?qqlnhr2nmuq)
r699 Download (http://www.mediafire.com/download.php?ykn3mgagxem)
r1148 Download (http://www.mediafire.com/download.php?odlwkojmyot)

See samples, r1148 have a lot of noise ( right corner ) and qp is in general bigger by ~3 when compared to old 531 and 699.

r531 : 1 pass : avg qp:18.31 avg qp:22.92 avg qp:22.77 1.61 fps, 2897.02 kb/s final ratefactor: 20.73
r531 : 2 pass : avg qp:19.00 avg qp:22.20 avg qp:22.60 1.66 fps, 3385.05 kb/s

--pass 1 --bitrate 3400 --level 4.1 --ref 3 --mixed-refs --no-fast-pskip --bframes 3 --b-rdo --bime --weightb --direct temporal --subme 6 --trellis 2 --analyse p8x8,b8x8,i4x4,i8x8 --8x8dct --qpmin 3 --me umh --merange 28 --thread-input --sar 1:1 --cqm "jvt"
--pass 2 --bitrate 3400 --level 4.1 --ref 3 --mixed-refs --no-fast-pskip --bframes 3 --b-rdo --bime --weightb --direct temporal --subme 6 --trellis 2 --analyse p8x8,b8x8,i4x4,i8x8 --8x8dct --qpmin 3 --me umh --merange 28 --thread-input --sar 1:1 --cqm "jvt"

r699 : 1 Pass : avg qp:19.24 avg qp:22.03 avg qp:21.76 2.05 fps, 2920.06 kb/s final ratefactor: 20.18
r699 : 2 Pass : avg qp:19.36 avg qp:21.54 avg qp:22.22 2.16 fps, 3378.46 kb/s

--pass 1 --bitrate 3400 --level 4.1 --ref 3 --mixed-refs --bframes 3 --b-rdo --bime --weightb --direct temporal --subme 6 --trellis 2 --analyse p8x8,b8x8,p4x4 --b-bias 1 --me esa --thread-input
--pass 2 --bitrate 3400 --level 4.1 --ref 3 --mixed-refs --bframes 3 --b-rdo --bime --weightb --direct temporal --subme 6 --trellis 2 --analyse p8x8,b8x8,p4x4 --b-bias 1 --me esa --thread-input

r1148 : 1 Pass : Avg QP:20.25 Avg QP:20.78 Avg QP:21.85 3.50 fps, 2776.20 kb/s final ratefactor: 23.31
r1148 : 2 Pass : Avg QP:20.16 Avg QP:20.68 Avg QP:21.76 3.61 fps, 3310.72 kb/s

--pass 1 --bitrate 3400 --level 4.1 --ref 3 --mixed-refs --no-fast-pskip --bframes 3 --b-adapt 2 --weightb --direct temporal --subme 7 --trellis 2 --psy-rd 0:0 --partitions p8x8,b8x8,i4x4,i8x8 --8x8dct --qpmin 5 --ipratio 1.1 --pbratio 1.1 --vbv-bufsize 24000 --vbv-maxrate 24000 --qcomp 1 --me umh --merange 24 --thread-input --sar 1:1 --no-dct-decimate --no-psnr --no-ssim
--pass 2 --bitrate 3400 --level 4.1 --ref 3 --mixed-refs --no-fast-pskip --bframes 3 --b-adapt 2 --weightb --direct temporal --subme 7 --trellis 2 --psy-rd 0:0 --partitions p8x8,b8x8,i4x4,i8x8 --8x8dct --qpmin 5 --ipratio 1.1 --pbratio 1.1 --vbv-bufsize 24000 --vbv-maxrate 24000 --qcomp 1 --me umh --merange 24 --thread-input --sar 1:1 --no-dct-decimate --no-psnr --no-ssim

Using --subme 9--me tesa --merange 28 i got :

r1148 : 1 Pass : Avg QP:20.03 Avg QP:20.41 Avg QP:21.67 1.22 fps, 2792.96 kb/s final ratefactor: 23.06
r1148 : 2 Pass : Avg QP:19.83 Avg QP:20.42 Avg QP:21.58 1.23 fps, 3307.90 kb/s

r531 :

avis [info]: 1024x576 @ 30.00 fps (1800 frames)
x264 [info]: Using sar=1/1
x264 [info]: Using cpu capabilities mmx mmxext sse sse2 3dnow!
X264 [info]: Slice i:26 avg qp:18.31 size: 35428 psnr mean y:48.66 u:51.99
v:51.95 avg:49.44 global:46.99
x264 [info]: Slice p:1124 avg qp:22.92 size: 15040 psnr mean y:44.99 u:49.70
v:49.60 avg:46.01 global:43.11
x264 [info]: Slice b:650 avg qp:22.77 size: 6001 psnr mean y:44.99 u:49.48
v:49.24 avg:45.93 global:42.18
x264 [info]: Mb i i16..4: 3.9% 82.0% 14.1%
x264 [info]: Mb p i16..4: 2.9% 33.7% 1.5% p16..4: 44.7% 9.9% 2.2% 0.0% 0
.0% skip: 5.0%
x264 [info]: Mb b i16..4: 0.2% 3.0% 0.4% b16..8: 55.8% 1.5% 2.2% direct:
0.7% skip:36.0%
x264 [info]: final ratefactor: 20.73
x264 [info]: 8x8 transform intra:87.7% inter:82.3%
x264 [info]: Ref p 76.5% 14.3% 9.3%
x264 [info]: Ref b 78.2% 13.9% 7.9%
x264 [info]: Psnr mean y:45.043 u:49.655 v:49.500 avg:46.029 global:42.785 kb/s:
2896.83

encoded 1800 frames, 1.61 fps, 2897.02 kb/s

avis [info]: 1024x576 @ 30.00 fps (1800 frames)
x264 [info]: Using sar=1/1
x264 [info]: Using cpu capabilities mmx mmxext sse sse2 3dnow!
X264 [info]: Slice i:26 avg qp:19.00 size: 32634 psnr mean y:48.44 u:51.75
v:51.73 avg:49.23 global:47.42
x264 [info]: Slice p:1124 avg qp:22.20 size: 17631 psnr mean y:45.49 u:49.96
v:49.87 avg:46.49 global:44.53
x264 [info]: Slice b:650 avg qp:22.60 size: 7262 psnr mean y:45.23 u:49.55
v:49.30 avg:46.17 global:43.69
x264 [info]: Mb i i16..4: 4.1% 85.8% 10.0%
x264 [info]: Mb p i16..4: 2.8% 34.6% 1.5% p16..4: 42.2% 11.1% 2.5% 0.0% 0
.0% skip: 5.2%
x264 [info]: Mb b i16..4: 0.3% 3.3% 0.4% b16..8: 54.3% 2.0% 2.5% direct:
1.0% skip:36.3%
x264 [info]: 8x8 transform intra:88.3% inter:84.1%
x264 [info]: Ref p 76.0% 14.5% 9.4%
x264 [info]: Ref b 77.0% 14.6% 8.4%
x264 [info]: Psnr mean y:45.435 u:49.837 v:49.693 avg:46.412 global:44.238 kb/s:
3384.85

encoded 1800 frames, 1.66 fps, 3385.05 kb/s

r699 :

avis [info]: 1024x576 @ 30.00 fps (1800 frames)
x264 [info]: using cpu capabilities: MMX MMXEXT SSE SSE2 SSE3 3DNow!
x264 [info]: slice I:25 Avg QP:19.24 size: 34647 PSNR Mean Y:47.34 U:50.97
V:51.14 Avg:48.19 Global:45.48
x264 [info]: slice P:1179 Avg QP:22.03 size: 14622 PSNR Mean Y:44.97 U:49.39
V:49.44 Avg:45.97 Global:43.03
x264 [info]: slice B:596 Avg QP:21.76 size: 6365 PSNR Mean Y:45.24 U:49.09
V:49.02 Avg:46.09 Global:42.08
x264 [info]: mb I I16..4: 48.2% 0.0% 51.8%
x264 [info]: mb P I16..4: 41.1% 0.0% 0.0% P16..4: 37.0% 11.8% 2.0% 0.2% 0
.1% skip: 7.9%
x264 [info]: mb B I16..4: 3.1% 0.0% 0.0% B16..8: 57.2% 2.1% 2.7% direct:
1.7% skip:33.0%
x264 [info]: final ratefactor: 20.18
x264 [info]: ref P 77.0% 13.5% 9.5%
x264 [info]: ref B 78.6% 13.2% 8.2%
x264 [info]: SSIM Mean Y:0.9685625
x264 [info]: PSNR Mean Y:45.093 U:49.311 V:49.326 Avg:46.037 Global:42.715 kb/s:
2919.85

encoded 1800 frames, 2.05 fps, 2920.06 kb/s

avis [info]: 1024x576 @ 30.00 fps (1800 frames)
x264 [info]: using cpu capabilities: MMX MMXEXT SSE SSE2 SSE3 3DNow!
x264 [info]: slice I:25 Avg QP:19.36 size: 33199 PSNR Mean Y:47.39 U:50.94
V:51.16 Avg:48.24 Global:46.78
x264 [info]: slice P:1179 Avg QP:21.54 size: 17167 PSNR Mean Y:45.39 U:49.66
V:49.74 Avg:46.38 Global:44.59
x264 [info]: slice B:596 Avg QP:22.22 size: 7159 PSNR Mean Y:45.11 U:49.04
V:48.95 Avg:46.01 Global:43.56
x264 [info]: mb I I16..4: 48.6% 0.0% 51.4%
x264 [info]: mb P I16..4: 41.0% 0.0% 0.0% P16..4: 34.6% 12.6% 2.0% 0.2% 0
.1% skip: 9.5%
x264 [info]: mb B I16..4: 3.2% 0.0% 0.0% B16..8: 54.3% 2.5% 2.8% direct:
1.8% skip:35.4%
x264 [info]: ref P 76.3% 13.9% 9.8%
x264 [info]: ref B 78.0% 13.6% 8.4%
x264 [info]: SSIM Mean Y:0.9763089
x264 [info]: PSNR Mean Y:45.326 U:49.472 V:49.496 Avg:46.284 Global:44.238 kb/s:
3378.24

encoded 1800 frames, 2.16 fps, 3378.46 kb/s

r1148 :

avis [info]: 1024x576 @ 30.00 fps (1800 frames)
x264 [info]: using SAR=1/1
x264 [info]: using cpu capabilities: MMX2 SSE2Slow
x264 [info]: profile High, level 4.1
x264 [info]: slice I:24 Avg QP:20.25 size: 25881
x264 [info]: slice P:726 Avg QP:20.78 size: 15023
x264 [info]: slice B:1050 Avg QP:21.85 size: 8849
x264 [info]: consecutive B-frames: 5.7% 32.4% 41.6% 20.3%
x264 [info]: mb I I16..4: 18.0% 68.2% 13.7%
x264 [info]: mb P I16..4: 12.5% 36.4% 3.9% P16..4: 32.0% 7.1% 2.9% 0.0% 0
.0% skip: 5.2%
x264 [info]: mb B I16..4: 5.3% 10.8% 1.1% B16..8: 54.6% 2.1% 1.7% direct:
3.8% skip:20.7% L0:43.2% L1:47.2% BI: 9.6%
x264 [info]: final ratefactor: 23.31
x264 [info]: 8x8 transform intra:67.0% inter:83.1%
x264 [info]: coded y,uvDC,uvAC intra:57.8% 50.5% 7.1% inter:28.1% 24.9% 1.8%
x264 [info]: ref P L0 80.8% 11.5% 7.8%
x264 [info]: ref B L0 85.2% 14.8%
x264 [info]: kb/s:2776.0

encoded 1800 frames, 3.50 fps, 2776.20 kb/s

avis [info]: 1024x576 @ 30.00 fps (1800 frames)
x264 [info]: using SAR=1/1
x264 [info]: using cpu capabilities: MMX2 SSE2Slow
x264 [info]: profile High, level 4.1
x264 [info]: slice I:24 Avg QP:20.16 size: 26594
x264 [info]: slice P:726 Avg QP:20.68 size: 17131
x264 [info]: slice B:1050 Avg QP:21.76 size: 11194
x264 [info]: consecutive B-frames: 5.7% 32.4% 41.6% 20.3%
x264 [info]: mb I I16..4: 17.3% 71.7% 11.0%
x264 [info]: mb P I16..4: 12.7% 36.8% 3.5% P16..4: 28.8% 7.2% 3.2% 0.0% 0
.0% skip: 7.8%
x264 [info]: mb B I16..4: 5.2% 11.1% 1.3% B16..8: 52.5% 2.8% 2.0% direct:
3.7% skip:21.5% L0:41.1% L1:45.4% BI:13.6%
x264 [info]: 8x8 transform intra:67.5% inter:84.4%
x264 [info]: coded y,uvDC,uvAC intra:58.8% 51.8% 7.4% inter:31.1% 23.1% 1.5%
x264 [info]: ref P L0 80.4% 12.1% 7.6%
x264 [info]: ref B L0 85.5% 14.5%
x264 [info]: kb/s:3310.5

encoded 1800 frames, 3.61 fps, 3310.72 kb/s

r1148x :

avis [info]: 1024x576 @ 30.00 fps (1800 frames)
x264 [info]: using SAR=1/1
x264 [info]: using cpu capabilities: MMX2 SSE2Slow
x264 [info]: profile High, level 4.1
x264 [info]: slice I:24 Avg QP:20.03 size: 25710
x264 [info]: slice P:735 Avg QP:20.41 size: 14825
x264 [info]: slice B:1041 Avg QP:21.67 size: 9061
x264 [info]: consecutive B-frames: 6.4% 32.7% 40.9% 20.0%
x264 [info]: mb I I16..4: 18.4% 67.8% 13.8%
x264 [info]: mb P I16..4: 12.5% 34.2% 3.9% P16..4: 31.7% 8.7% 3.1% 0.0% 0
.0% skip: 5.9%
x264 [info]: mb B I16..4: 4.5% 9.5% 1.0% B16..8: 56.6% 2.4% 1.9% direct:
3.8% skip:20.2% L0:42.9% L1:47.6% BI: 9.5%
x264 [info]: final ratefactor: 23.06
x264 [info]: 8x8 transform intra:66.3% inter:82.5%
x264 [info]: coded y,uvDC,uvAC intra:58.4% 49.8% 6.9% inter:29.8% 26.0% 2.0%
x264 [info]: ref P L0 79.8% 12.1% 8.1%
x264 [info]: ref B L0 85.3% 14.7%
x264 [info]: kb/s:2792.8

encoded 1800 frames, 1.22 fps, 2792.96 kb/s

avis [info]: 1024x576 @ 30.00 fps (1800 frames)
x264 [info]: using SAR=1/1
x264 [info]: using cpu capabilities: MMX2 SSE2Slow
x264 [info]: profile High, level 4.1
x264 [info]: slice I:24 Avg QP:19.83 size: 26864
x264 [info]: slice P:735 Avg QP:20.42 size: 16742
x264 [info]: slice B:1041 Avg QP:21.58 size: 11390
x264 [info]: consecutive B-frames: 6.4% 32.7% 40.9% 20.0%
x264 [info]: mb I I16..4: 17.3% 71.7% 11.0%
x264 [info]: mb P I16..4: 12.8% 34.8% 3.4% P16..4: 28.4% 8.5% 3.6% 0.0% 0
.0% skip: 8.5%
x264 [info]: mb B I16..4: 4.5% 9.7% 1.1% B16..8: 54.2% 3.4% 2.3% direct:
3.7% skip:21.0% L0:40.9% L1:45.7% BI:13.3%
x264 [info]: 8x8 transform intra:67.0% inter:84.2%
x264 [info]: coded y,uvDC,uvAC intra:58.7% 50.7% 6.7% inter:32.7% 24.3% 1.6%
x264 [info]: ref P L0 79.5% 12.5% 8.1%
x264 [info]: ref B L0 85.5% 14.5%
x264 [info]: kb/s:3307.7

encoded 1800 frames, 1.23 fps, 3307.90 kb/s

Dark Shikari

14th May 2009, 14:42

juGGaKNot

14th May 2009, 18:41

Of course the QP is bigger because of adaptive quantization and so forth.

You can't compare two different versions of x264 when you're using totally different options between them.

If you love a video full of messy blocks and blurred-out details, feel free to turn off AQ and/or psy-RD. Don't roll back to an ancient x264.

I tried tons of profiles settings, none of them produce the same quality/filesize as those old builds

This is why i've added the source files, so others can test.

I do not see "a video full of messy blocks and blurred-out details" in the old one ( and it was encoded at low bitrate for this type of file ) but when i look at new ones ( as said, tried alot of settings ) its a noisefest @ killtext ( upper right corner ) and bigger filesize.

Don't know where to go from here, this i why i uploaded the source and made this thread ....

Sharktooth

14th May 2009, 23:06

newer x264 buillds produce better quality than older ones. there is NOT EVEN A SHADOW OF A DOUBT about it.
if you like the old behaviour get a NEWER build and set --aq-strength to 0 and --psy-rd to 0:0 ...

---- AND ----

COMPARE with the same settings without messing with settings and stupid values like, for example, --no-dct-decimate or --merange 28 !!!

juGGaKNot

15th May 2009, 07:18

juGGaKNot

15th May 2009, 08:20

Tested, looks like i can lower the bitrate by ~500 for the same quality as old builds, solved, samples tomorrow, porting to bat.

I'm using the default ps3/xbox profile with filter -3,-3, refs 4 and merange 24 ( or 32 )

--pass 1 --bitrate 3100 --stats "D:\x264\@temp\movie.stats" --level 4.1 --ref 4 --mixed-refs --bframes 3 --b-adapt 2 --weightb --direct auto --deblock -3:-3 --subme 7 --trellis 2 --psy-rd 0:0 --partitions p8x8,b8x8,i4x4,i8x8 --8x8dct --qpmin 3 --ipratio 1.1 --pbratio 1.1 --vbv-bufsize 24000 --vbv-maxrate 24000 --qcomp 0.5 --me umh --merange 24 --threads auto --thread-input --aq-mode 0 --sar 1:1 --progress --no-psnr --no-ssim
--pass 2 --bitrate 3100 --stats "D:\x264\@temp\movie.stats" --level 4.1 --ref 4 --mixed-refs --bframes 3 --b-adapt 2 --weightb --direct auto --deblock -3:-3 --subme 7 --trellis 2 --psy-rd 0:0 --partitions p8x8,b8x8,i4x4,i8x8 --8x8dct --qpmin 3 --ipratio 1.1 --pbratio 1.1 --vbv-bufsize 24000 --vbv-maxrate 24000 --qcomp 0.5 --me umh --merange 24 --threads auto --thread-input --aq-mode 0 --sar 1:1 --aud --progress --no-psnr --no-ssim

Is the --aud in the second pass needed ? i saw a thread where it lead to some problems with playback on a player

About refs, should i go with 4 or the maximum for the resolution, any hardware decoding problems if i use many.

thnx, cheers.

Sharktooth

15th May 2009, 13:55

--aud is srtictly needed only for raw .avc output. if the encoder output is mp4 or mkv you can remove it (megui does automatically).
refs are limited by the level.

juGGaKNot

15th May 2009, 14:37

--aud is srtictly needed only for raw .avc output. if the encoder output is mp4 or mkv you can remove it (megui does automatically).
refs are limited by the level.

I use bat with raw output .264, mp4box mux

Refs : yes i know the formula, i want to make a calculator in the bat

Just asking if more than 5 helps, same for b-frames, i see that my source can use ~8 b-frames ( b-adapt 2 ) before their all 0.0 ( tested with 16 )

All this while being ps3/xbox compatible! now i use 4 refs with 4 b-frames.

This is the bitrate "automation" script i made, from it i will do the refs :

if %myres% GEQ 160 GOTO A1
:A1
if %myres% LEQ 639 GOTO A2
if %myres% GEQ 640 GOTO B1
:B1
if %myres% LEQ 799 GOTO B2
if %myres% GEQ 800 GOTO C1
:C1
if %myres% LEQ 959 GOTO C2
if %myres% GEQ 960 GOTO D1
:D1
if %myres% LEQ 1023 GOTO D2
if %myres% GEQ 1024 GOTO E1
:E1
if %myres% LEQ 1279 GOTO E2
if %myres% GEQ 1280 GOTO F1
:F1
if %myres% LEQ 1919 GOTO F2
if %myres% GEQ 1920 GOTO G1
:G1
if %myres% LEQ 2560 GOTO G2
:A2
SET CQ=2500
GOTO END
:B2:
SET CQ=2700
GOTO END
:C2:
SET CQ=2900
GOTO END
:D2
SET CQ=3100
GOTO END
:E2
SET CQ=3400
GOTO END
:F2
SET CQ=4000
GOTO END
:G2
SET CQ=4500
ECHO CQ
:END

Sharktooth

15th May 2009, 14:41

more than 5 refs or b-frames is generally useless but it also depends on the source.
sane settings would be no more than 3 refs and 3 b-frames.

juGGaKNot

15th May 2009, 14:47

more than 5 refs or b-frames is generally useless but it also depends on the source.
sane settings would be no more than 3 refs and 3 b-frames.

Hmm, k i'll save myself 30 minutes and leave it at 4r and 4b

Source seems to like it, i got almost no noise @ killtext compared to 3r 3b.