Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. |
18th October 2018, 15:58 | #6442 | Link |
Pig on the wing
Join Date: Mar 2002
Location: Finland
Posts: 5,733
|
The average QP remained pretty much the same, differences were around 0.01-0.03 units between 1:1 - -3:-3. Based on my tests, the higher values did soften the image more at least in places I checked, so maybe the bits were allocated elsewhere in the image.
__________________
And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon... |
18th October 2018, 16:40 | #6443 | Link | |
Moderator
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,770
|
Quote:
Picojoules per pixels is the more relevant metric here. Heat produced is equal to power draw. Better coolers can get the heat away from the CPU power better of course, but watts to the CPU is going to be the same as watts of heat to dissipate. |
|
18th October 2018, 16:45 | #6444 | Link | ||
Moderator
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,770
|
Quote:
Quote:
It IS surprising that you wouldn’t see bitrate OR QP change, even with a significant loss of detail. Not that using 6:6 is something anyone is likely to do in practice. But reduced detail should result in lower bitrate and/or QP. With lowered detail (including grain/gain noise), prediction should be more efficient... Maybe log your repro for this as an issue for MCW? |
||
18th October 2018, 16:48 | #6445 | Link | |
Moderator
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,770
|
Quote:
Playing back at 1/4 speed is generally okay to see temporal artifacts better while still keeping temporal coherence. You can still detect discontinuities between IDR/i/P/B/b that way. Testing at a higher CRF, like maybe 28, can also make differences at lot more obvious. Working at a quality where most things look pretty good can make differences much harder to detect. |
|
18th October 2018, 17:24 | #6446 | Link | |
Registered Developer
Join Date: Mar 2010
Location: Hamburg/Germany
Posts: 10,348
|
Quote:
With overclockable CPUs, you can control the AVX512 offset and make it useful for x265, but on Xeons or the like, you probably don't get that level of control, and the downclock during AVX512 might offset the advantages it offers. Additionally, x265 is a pretty "light" AVX512 load. If you change the offset for a lower downclock and then run a strong AVX512 load (like pure math, ie. FFTs, for a prolonged time), your system may become unstable, due to the extreme density of all the heat and power draw. So its really hard to balance.
__________________
LAV Filters - open source ffmpeg based media splitter and decoders |
|
19th October 2018, 07:47 | #6447 | Link | ||
Registered User
Join Date: Aug 2014
Posts: 28
|
Quote:
My chip is OC'd to 4.5ghz with a -4 offset so that it drops to 4.1 when using avx. I guess I should have tried to compare encoding speed with avx on and off instead of just avx-512 enabled and disabled. I have no throttling issues and the load is a pretty consistent 100% with the occasional 1 second dip down to 87% every 30 seconds or so. What is the command to disable avx entirely? Quote:
--preset slower --crf 17 --profile main10 --me 3 --subme 5 --psy-rd 1.5 --psy-rdoq 5.0 --rdoq-level 1 --qcomp 0.8 --deblock -1:-1 --no-sao --repeat-headers --hdr-opt --range limited --colorprim 9 --transfer 16 --colormatrix 9 --master-display "G(13250, 34500)B(7500, 3000)R(34000, 16000)WP(15635, 16450)L(10000000, 1)" Last edited by DotJun; 19th October 2018 at 07:51. |
||
19th October 2018, 12:21 | #6448 | Link | ||
RipBot264 author
Join Date: May 2006
Location: Poland
Posts: 7,815
|
Quote:
Quote:
__________________
Windows 7 Image Updater - SkyLake\KabyLake\CoffeLake\Ryzen Threadripper Last edited by Atak_Snajpera; 19th October 2018 at 12:23. |
||
19th October 2018, 13:53 | #6449 | Link |
Registered User
Join Date: Feb 2015
Posts: 326
|
--asm sse4
Main asm levels are: --asm no --asm sse2 --asm ssse3 --asm sse4 --asm avx2 --asm avx512 --asm avx512 works a bit different -- it only enable possibility to use AVX-512 in auto-detection of CPU capabilities. For hard use of AVX-512 code (without checking) please use --asm avx,avx512 |
19th October 2018, 19:36 | #6450 | Link |
Lost my old account :(
Join Date: Jul 2017
Posts: 326
|
Not sure if there is anyone that works on x265 is active in this thread anymore, but I've seen some discussion regarding --CTU and --merange both in this thread in the past and recently in a few threads.
From what I’ve read here and from what I’ve experienced from my testing is that using a CU size of 64 is overkill for 1080p and bellow, I see both a speed increase and a multithread increase when lowering it together with merange with no apparent loss in compression. I also saw some posts way back in this thread that suggested that the default CTU size should be based on resolution, wasn’t this implemented for any specific reason or is it just that no one has committed a patch for it? I also have a question for merange, it says in the docs that the value of 57 is based on CTU-size and search method, but the value is set to 57 for all presets even though CTU-size and search method vary. How come? Last edited by excellentswordfight; 19th October 2018 at 19:49. |
19th October 2018, 20:37 | #6451 | Link |
Pig on the wing
Join Date: Mar 2002
Location: Finland
Posts: 5,733
|
I finished my short tests today, and these are the settings that I found to suit my requirements.
1080p: Code:
--deblock -3:-3 --no-strong-intra-smoothing --merange 44 --no-sao --qcomp 0.75 --aq-mode 3 --aq-strength 0.8 --ctu 32 --max-tu-size 32 --qg-size 8 --tu-inter-depth 4 --tu-intra-depth 4 --limit-tu 4 --limit-refs 3 --max-merge 3 --rd-refine --ref 6 --bframes 10 --crf 20.5 Code:
--deblock -3:-3 --no-strong-intra-smoothing --merange 38 --no-sao --qcomp 0.8 --aq-mode 3 --aq-strength 0.8 --ctu 16 --max-tu-size 16 --qg-size 8 --tu-inter-depth 3 --tu-intra-depth 3 --limit-tu 4 --limit-refs 3 --max-merge 3 --rd-refine --ref 6 --bframes 10 --crf 19.5 I did try finding out differences between deblock 1:1 and -3:-3 in motion but couldn't tell. If someone had ABX'd me, I probably wouldn't have been able to say which one is which. Mind you, the scene I used contained quite a lot of motion and some fast cuts so maybe they were just not visible there. Anyway, I chose a low value as I think it will retain detail and sharpness better. I watch everything from about 3-3.5 meters anyway, so any small blocking won't be that visible. Edit: both encodes using the preset "veryslow".
__________________
And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon... Last edited by Boulder; 20th October 2018 at 16:19. |
20th October 2018, 00:30 | #6452 | Link |
Registered User
Join Date: Aug 2014
Posts: 50
|
I did a 4k down to 1080 encode the other day, with an accidental 0:0 deblock setting. It looks good in a "slick," "photoshopped" way, but it definitely loses a lot of detail, at least at CRF 23.
That's too much deblocking for me. Though I do put db at 0:0 when increasing CRF and dropping resolution a lot. |
20th October 2018, 05:53 | #6453 | Link | |
Registered User
Join Date: Aug 2014
Posts: 28
|
x265 HEVC Encoder
Quote:
Thanks, I’ll put up three test runs of default, sse4 and avx512 to see what changes there are in FPS and efficiency. Will medium or fast preset be ok to use or it has to be slower or higher? Will it be ok to use “asm sse4” or do I have to specify sse4.2 like my log file shows? Last edited by DotJun; 20th October 2018 at 06:11. |
|
20th October 2018, 10:42 | #6454 | Link | ||
Pig on the wing
Join Date: Mar 2002
Location: Finland
Posts: 5,733
|
Quote:
__________________
And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon... |
||
20th October 2018, 22:39 | #6455 | Link | |
Moderator
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,770
|
Quote:
AFAIK it works by reducing signaling overhead without changing actual pixels. There are several bitstream options that do things like that. This is the one likely to have the most impact, as WP signaling can take place MANY times per frame. The others are per frame or per GOP. Sent from my iPhone using Tapatalk |
|
21st October 2018, 07:23 | #6456 | Link |
Registered User
Join Date: Aug 2014
Posts: 28
|
I did a few more test runs with the same parameters I stated before using 50k frames test clip. Here are the results.
Normal Preset: sse4 2.19fps avx2 2.89fps 512 3.18fps They all had the exact same kbps of 23463. Slower Preset: sse4 0.71fps avx2 0.89fps 512 0.95fps All of them ended up with the exact same kbps of 22868. My previous test from the other day seems to be a failure since I didn't use correct the correct switches for --asm. |
21st October 2018, 13:21 | #6457 | Link | |
Pig on the wing
Join Date: Mar 2002
Location: Finland
Posts: 5,733
|
Quote:
Without the optimization: Optimization enabled: In higher detailed area, the optimized encode looked better. I think I need to find a scene with some still background like sky, and see which one looks better in motion. The banding in flat areas can be quite eye-catching once you notice it.
__________________
And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon... |
|
21st October 2018, 18:54 | #6458 | Link | |
Moderator
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,770
|
Quote:
Was there any measurable perf difference? |
|
21st October 2018, 19:37 | #6459 | Link | |
Pig on the wing
Join Date: Mar 2002
Location: Finland
Posts: 5,733
|
Quote:
I also checked how the optimization works in normal playback, and my eyes didn't like the result. The flat areas suffered a bit too much, there was a short scene with a nice, slightly noisy but flat coloured background which was lit by some flickering candlelight. The normal encode was slightly better looking there, there was not as much swimming blocks effect as there was with the optimized version.
__________________
And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon... |
|
21st October 2018, 20:35 | #6460 | Link | ||
Moderator
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,770
|
Quote:
Quote:
It would be interesting to see what the difference was in a stream analyzer. Or just looking at the log-level 2 csv files. |
||
|
|