cancel
Showing results for 
Search instead for 
Did you mean: 

General Discussions

benman2785
Big Boss

720p50 3500kbps twitch.tv OBS

fetokis

here are my settings for 720p50@3500kbps in OBS Studio with AMD AMF Encoder

pastedImage_2.png

have in mind:
0. i use 720p50 because you have more bitrate per frame as 720p60
1. 3500kbps is not much for 720p50 (3500kbps is ok-ish for 720p30)(720p60 would need minimum 4500kbps)
1b. what is your ISPs upload bandwidth?
2. VBV Buffer Initial Fullness HAS to be 0 (otherwise you will have pixel mash all 2s)
3. usually "Balanced" "Main" gives best results
4. Coding Type could also be CABAC (better compression = 1% better quality; so not much but hey)
5. a better quality stream of 720p50 needs 4200kbps with a Buffer of 256kb
5b. 720p60 to look good would need 5000kbps with a Buffer of 288kb

here a video of the setttings you see: http://hoffmeister.tk/obs/xaymar/vids/Fetokis_720p50_3500kbps.mp4 

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS
369 Replies

@Kucur42 

you can do 900p60 - you need 5 SMT Cores for that
bitrate and settings are he same. you maybe have to use "main" or "baseline" instead of "high"

why 50fps (PAL) and not 60fps

1080p60@6000kbps = (6000/60*1000)/(1920x1080) = 0.048225
1080p50@6000kbps = (6000/50*1000)/(1920x1080) = 0.057870
900p60@6000kbps = (6000/60*1000)/(1600x900) = 0.069444
900p50@6000kbp = (6000/50*1000)/(1600x900) = 0.083333
720p60@4500kbp = (4500/60*1000)/(1280x720) = 0.081380
720p60@5000kbps = (5000/60*1000)/(1280x720) = 0.090422
720p50@4500kbps  = (4500/50*1000)/(1280x720) = 0.097656
720p50@5000kbps  = (5000/50*1000)/(1280x720) = 0.108507

the higher the number the better the quality

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

Hi!

net.JPG

My connection speeds changed!

Your suggesstions best for me for x264 with NotCPUcores,

But I cannot play my 100% competitive with x264   (fells like mouse input lag)

Can you help me with AMD settings please? 

 

0 Likes

@Kucur42 
8Mbits iis good - now you can stream 900p50/60 @6000kbps + 1250kb buffer

but mouse-lag isnt normal - please show me how you set up NotCPUCores (do screenshots)

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

Here,

1.JPG

2.JPG

FPS doesn't chance that much.

I have two monitors, thats changes something?

0 Likes

@Kucur42 

1. NotCPUCores is set up correct
2. i also use two monitors (AOC G2460PF + AOC G2460VQ6) on a RX 580 4G AND DONT HAVE THAT PROBLEM

A: what game does have this issue?
B: do you have VSYNC enabled?
C: do you have "enhanced Sync" enabled?

EDIT:
PLEASE open Taskmanager and check the load on the last remaining SMT core? - it should be around 10-25%; if it is more check what programm is causing it and then close that programm (its possible that very high load on last SMT core will choke your GPU-Driver)

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

A) CS:GO 

B) No

C) Disabled

Should I use same x264 settings you gave me?

0 Likes

CS:GO should not behave this way

yes use my x264 settings

or use:

benman2785_0-1612478079740.png

+ c264 options:
level=4.2 cabac=1 threads=0 sliced-threads=0 partitions=p8x8,b8x8,i8x8,i4x4 8x8dct=1 ref=3 bframes=2 rc-lookahead=30 me=hex merange=8 subme=8 deblock=-3:2:0 direct=spatial direct-8x8=-1 sync-lookahead=0 aq-mode=1 aq-strength=0.7 trellis=0 deadzone-inter=32 deadzone-intra=16 weightp=1 weightb=1 cqm4i=7,16,22,24,16,22,24,28,18,22,27,33,22,24,32,47 cqm4p=13,15,17,18,15,17,18,20,17,18,21,22,18,20,22,25 cqm8i=7,13,16,18,22,22,24,28,13,13,18,20,22,24,28,31,16,18,22,22,24,28,28,32,18,18,22,22,24,28,31,33,18,22,22,24,27,29,33,40,22,22,24,27,29,33,40,48,22,22,24,28,32,38,47,57,22,24,29,32,38,47,57,69 cqm8p=13,14,15,16,17,17,18,19,14,15,16,17,17,18,19,20,15,16,17,17,18,19,20,21,16,17,17,18,19,20,21,22,17,17,18,19,21,22,22,23,17,18,19,20,22,22,23,25,18,19,20,22,22,23,25,26,19,20,21,22,23,25,26,27

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

Tried different game.

PUBG, x264 

1612522256638.jpg

And this is your last x264 settings results. (I took the photo after ended the stream)

1612521917695.jpg

PUBG FPS droped 130 to 90FPS.

0 Likes

@Kucur42 

looks better - but there is a lot of load on last SMT core
you should close most programms you dont need when streaming

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

Hi!

I've considering changing my CPU for gaming and streaming.

My options are 3700X or 5600X. Which one would you recommend? 

 

0 Likes

@Kucur42 
depends: what is your GPU?

if you have a RTX 2000 or 3000 you can go with the 5600X - because you will use NVenc (as it gives 98% quality of x264)
if you have an AMD GPU you should go with 3700X - as you have to use x264 for encoding in best quality (because AMF has 93% quality of x264 - and maybe 85% in hard to encode scenes like with lots of vegetation(grass or trees))

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

@benman2785 

What you can tell me about settings to stream apex legends , can you help me?

 

0 Likes

@annihalator 

what specs?
btw - it doesnt matter what game: its either the AMF settings or the x264 settings - depending on your hardware

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

@benman2785 

Ryzen 5 2600 Stock

Rx 570 4gb 

RAM 16gb 2800mhz

0 Likes

@annihalator 

use x264 for 720p50/60 (4500kbps + 9000kb buffer) or 1080p50/60 (as in picture)

benman2785_0-1616600850139.png

custom x264:
level=4.2 cabac=1 8x8dct=1 partitions=p8x8,b8x8,i8x8,i4x4 ref=4 bframes=2 b-adapt=1 fast-pskip=0 rc-lookahead=30 sync-lookahead=30 me=hex merange=16 subme=6 deblock=-3:2:0 direct=spatial aq-mode=2 trellis=0 deadzone-inter=11 deadzone-intra=11 psy-rd=0.1:0.0 weightp=1 weightb=1


& use NotCPUCores (github)

benman2785_1-1616600928845.pngbenman2785_2-1616600954013.png

and hit assign when game runs

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS
0 Likes

@benman2785 

Thx , i Will try that options and see if i have problems , thx

0 Likes

@benman2785 

did you mean 9000kbs buffer or 900? and for 720p50 cabal level is not 3.2?

0 Likes

@annihalator 

9000kb buffer
3.2 is minimum for 720p50/60 - but anything higher is also ok // so when you prefer to stream in 1080p50 some times you can leave it at 4.2


 

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

Hello again, which one to use for OBS? ncc.png

broadcaster cores or remaining cores?

and another question. I turn on the cpu monitor in the game.when i start recording or streaming my cpu not stable (static) it jump from 35-40-50-35--60% etc .. its normal? Without recording, streaming same but less % around 11-33% its normal cpu not show static % p.s. game + stream = 40-60%cpu normal?And cpu temp 45-50c? ( settings 720p/60, veryfast + customs ) ryzen 2600 non OC.

0 Likes

@rursi 

in my testing i figured out that "broadcaster cores" doesnt limit x264 performance that strong (only some 1-3fps - yet still over 60fps) 
SOME games (like MetroExodus) dont like it when you use "ProcessLasso" or "NotCPUCores" or manually assign cores to exe via taskmanager (or start it with assigned cores via batch) - you will notice that very quickly // in those uncommon cases you should simply force OBS where you had "YOUR-GAME.EXE" to "non-physical cores" with high prio

about your second question:
it is normal that your CPU has higher (and "unstable") load when you stream and play at the same time
so 40-95% is still fine - also temps lower than 75°C are fine (on Ryzen) (better keep them in the low 60s)

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

@Kucur42 

the post you kudoed isnt for you - this was for a dude who needed AMD AMF // but YOU need x264 (and are capable of doing so)

also - check my last post - there is info about bitrate and buffer (set it correctly); also set OBS as i said, also with NotCPUCores

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS
davidred
Journeyman III

I have also Capture card Aver media live gamer HD 2 uncompressed, zero-lag. My Pc recognize the card and I installed software also everything work fine, but when i am cloning with my gpu nothing change in my stream there is no reaction.

0 Likes

@davidred 

i never worked with CaptureCards - so i cant help you with that...

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS
davidred
Journeyman III

For months now I've been playing with the settings in OBS for a better quality on twitch to absolutely no avail. My issue isn't so much game quality, it's the webcam quality. The game usually looks okay no matter what settings I use but when I encode using the hardware AVC/AMF encoder, my webcam is completely and totally pixelated while any slight movement is happening in the game. I've gone through and tried these settings you've given to others and they don't work. I usually stream in 720p 60fps.

0 Likes

@davidred 

what is your CPU? // AMD AMF Encoder is only usefull with 4C/8T or lower CPUs...

what is your Webcam? what are its settings? (my webcam is 720p - but i run it 424x240)
check https://drive.google.com/drive/folders/1ov1aU0SkWV1UCiI6FTED23qtRUYJl17S - 720p50 if webcam looks ok for you

also: do you use a green-screen?

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS
SoloMattu
Adept I

Hi, i has tried a lot of settings to make my stream look better, with some settings i need more sharpness and with some others it looks sharp but there is a pixelation every 2 or 3 seconds. Also, when i up the resolution to 900p it looks sharper than 720p, but it uses more pc.

I want to have a nice quality stream with 720p if i can, but if that isn't possible i can use 900p, what settings do you recomend me? i play all type of games (League of Legends, The Witcher 3, Dark Souls 3, Dead By Daylight, etc ) and this is my pc:

CPU: i3 6100

GPU: RX 570 4GB

RAM: 8GB 2133MHZ

 

(Sorry if i wrote something bad, im from Argentina)

0 Likes

@SoloMattu 

mh, this is hard

your CPU is not capable of x264
your GPU is capable of AMD AMFenc

do you stream on twitch.tv or YT?

for 720p60 use

benman2785_1-1613476195261.png

if you get "encoder overload":
1. reduce ingame quality settings (from ultra to medium or low) (AMFenc needs "headroom" of GPU)
2. if still "overload" use "constrained high" or "main" Profile

NOW if streaming twitch.tv in 900p60
use:

benman2785_2-1613476332957.png


have in mind: AMD AMFenc is not as good as x264 or NVenc - you will have more pixelation than with the other encoders
also: your CPU is very weak - even my old Ryzen 5 2500U 4C/8T Laptop CPU performs 90% better in Multitasking - and for streaming you need "raw horse power"

so again: if you get "encoder overload" you should lower ingame settings - if you still have problems with encoder you should lower "profile" for AMD AMF

btw - NVenc is also no option for you (so you dont need to buy an Nvidia GPU) as Nvidia "cheats" with their NVenc as it uses 3x more CPU than AMD AMFenc - and CPU is already your bottleneck with i3 6100

sorry i couldnt help more



PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

Hello again, i tried the settings you said for 720p60, this is how it looks:

Benman obs settings test rx570 - YouTube

(The video doesn't have sound)

Some months ago i had a gtx 750ti and tried nvenc, that encoder looks better with less bitrate than this rx570, i changed to amd because of the price/quality for gaming, but seeing that the encoder is not good is not worth buying an amd card, i can't get a better quality with another settings? also, i stream on Twitch, not in youtube.

0 Likes

1. this is not 720p60
2. i need the orginal file
3. yes NVenc is better - but it also uses more CPU (and you dont have any more CPU left)
4. Youtube alters the videofile: just check the file in next post

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

@SoloMattu 

EDIT:
had issues with OBS - it used VBR instead CBR in my last videos
have fixed that now - videos comming soon

EDIT2:

here two videos:
AMF 720p60@4500kbps
http://benmanshafen.de/obs/720p60-1613517813.flv

x264 900p60@6000kbps (NVenc will look 2-3% worse)
http://benmanshafen.de/obs/900p50-1613517942.flv

Witcher 3 is hard to encode - thats why even with x264 you will have problems with the leaves on the trees etc

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

best i could get for Witcher 3 in short time was this:

benman2785_0-1613518614736.png

http://benmanshafen.de/obs/720p60-1613518426.flv

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

Thanks, it looks better now, but i have some question about three parameters, how works the VBV Buffer Size and the VBV Buffer Initial Fullness, how i know what to put in each one? and my other question is if the Maximum Reference Frames are like b-frames on Nvenc/x264 or are other type of frames?

VBV Buffer is simply the Buffer
the higher the buffer the higher the max peak 
sometimes a frame needs a bigger bitrate - thats what the buffer is for
"standard" for 4500kbps is 4500kb - but in CBR this will create peaks to up to 8000kbps and therefore will result in a very high bitrate drop to in worst case down to 500kbps
so you set a more strict buffer - like 1250kb or 1500kb for 4500kbps - this will "lower maximal quality of a single frame - but maintain a better quality overall" because your max peak will be around 5500kbps and therefore it will drop in worst case to 3500kbps

buffer initial fullness - is actually only important when the buffer is flushed or at the start of the stream - yet it can also create higher peaks when set to 100% // usually anthing between 60-75% will provide a more "constant quality"

no reference frames are not b-frames
ref frames are the number of "last frames" that a new frame can refer to.
b-frames on the other hand are frames that can refer in both ways: previous frame and upcoming next frame
afaik AMD AMFenc only supports b-frames higher than 1 for their R9 380(X) and R9 385 GPUs (they can do up to 3 b-frames)

ps glad i could help a little
pps i dont know how good AMFenc is for RDNA2 (as i am waiting until the prices are "normal" again)

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

benman2785 hello! Are u try new obs plugin StreamFX 0.10.0a2 ( AMD AMF - AVC) via ffmpeg.

Maby u can give good ffmpeg settings.

 

i don't know I write correct settings ?

amddd.png

 

but quality much beter then old AMF/AVC

0 Likes

i havent used this - but i will try now

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

impatient waiting for your tests results & settings.

0 Likes

@rursi 

so far i can say i am not able to set custom ffmpeg options xD - but i already asked Xaymar how to set them properly - so i also have to wait for an anwser

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

ok here are the possible options for StreamFX ffmpeg

h264_amf AVOptions:
  -usage             <int>        E..V...... Encoder Usage (from 0 to 3) (default transcoding)
     transcoding     0            E..V...... Generic Transcoding
     ultralowlatency 1            E..V...... 
     lowlatency      2            E..V...... 
     webcam          3            E..V...... Webcam
  -profile           <int>        E..V...... Profile (from 66 to 257) (default main)
     main            77           E..V...... 
     high            100          E..V...... 
     constrained_baseline 256          E..V...... 
     constrained_high 257          E..V...... 
  -level             <int>        E..V...... Profile Level (from 0 to 62) (default auto)
     auto            0            E..V...... 
     1.0             10           E..V...... 
     1.1             11           E..V...... 
     1.2             12           E..V...... 
     1.3             13           E..V...... 
     2.0             20           E..V...... 
     2.1             21           E..V...... 
     2.2             22           E..V...... 
     3.0             30           E..V...... 
     3.1             31           E..V...... 
     3.2             32           E..V...... 
     4.0             40           E..V...... 
     4.1             41           E..V...... 
     4.2             42           E..V...... 
     5.0             50           E..V...... 
     5.1             51           E..V...... 
     5.2             52           E..V...... 
     6.0             60           E..V...... 
     6.1             61           E..V...... 
     6.2             62           E..V...... 
  -quality           <int>        E..V...... Quality Preference (from 0 to 2) (default speed)
     speed           1            E..V...... Prefer Speed
     balanced        0            E..V...... Balanced
     quality         2            E..V...... Prefer Quality
  -rc                <int>        E..V...... Rate Control Method (from -1 to 3) (default -1)
     cqp             0            E..V...... Constant Quantization Parameter
     cbr             1            E..V...... Constant Bitrate
     vbr_peak        2            E..V...... Peak Contrained Variable Bitrate
     vbr_latency     3            E..V...... Latency Constrained Variable Bitrate
  -enforce_hrd       <boolean>    E..V...... Enforce HRD (default false)
  -filler_data       <boolean>    E..V...... Filler Data Enable (default false)
  -vbaq              <boolean>    E..V...... Enable VBAQ (default false)
  -frame_skipping    <boolean>    E..V...... Rate Control Based Frame Skip (default false)
  -qp_i              <int>        E..V...... Quantization Parameter for I-Frame (from -1 to 51) (default -1)
  -qp_p              <int>        E..V...... Quantization Parameter for P-Frame (from -1 to 51) (default -1)
  -qp_b              <int>        E..V...... Quantization Parameter for B-Frame (from -1 to 51) (default -1)
  -preanalysis       <boolean>    E..V...... Pre-Analysis Mode (default false)
  -max_au_size       <int>        E..V...... Maximum Access Unit Size for rate control (in bits) (from 0 to INT_MAX) (default 0)
  -header_spacing    <int>        E..V...... Header Insertion Spacing (from -1 to 1000) (default -1)
  -bf_delta_qp       <int>        E..V...... B-Picture Delta QP (from -10 to 10) (default 4)
  -bf_ref            <boolean>    E..V...... Enable Reference to B-Frames (default true)
  -bf_ref_delta_qp   <int>        E..V...... Reference B-Picture Delta QP (from -10 to 10) (default 4)
  -intra_refresh_mb  <int>        E..V...... Intra Refresh MBs Number Per Slot in Macroblocks (from 0 to INT_MAX) (default 0)
  -coder             <int>        E..V...... Coding Type (from 0 to 2) (default auto)
     auto            0            E..V...... Automatic
     cavlc           2            E..V...... Context Adaptive Variable-Length Coding
     cabac           1            E..V...... Context Adaptive Binary Arithmetic Coding
  -me_half_pel       <boolean>    E..V...... Enable ME Half Pixel (default true)
  -me_quarter_pel    <boolean>    E..V...... Enable ME Quarter Pixel (default true)
  -aud               <boolean>    E..V...... Inserts AU Delimiter NAL unit (default false)
  -log_to_dbg        <boolean>    E..V...... Enable AMF logging to debug output (default false)

 

i dont know if we can use other options like "me_range=" etc

so the only useful options would be
-quality=2 -profile=100 -level=42 -rc=1 -preanalysis=true -coder=1

but actually you would only need: -coder=1
as you can set the other options via gui

if i have more time i will test "standard" ffmpeg options to squeeze out more quality

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS

what do you think - or quality will be better with  plugin?

0 Likes

depends:

if some settings run on ffmpeg (CPU) and some on AMF - yes i can improve
if all is run solely on AMF than MAYBE - but just a slight quality improvement

here a test video (download it or click it):
http://benmanshafen.de/obs/900p60_sFX.mp4
using:

benman2785_0-1613744380784.png

with custom settings:
-usage=2 -quality=0 -profile=100 -level=52 -rc=1 -preanalysis=true -coder=1 -qmin=25 -qmax=46 -me_range=48 -subq=9 -me_half_pel=true -me_quarter_pel=true -vbaq=true -trellis=2 -intra_refresh_mb=170

-usage=2                    - enables "low-latency"-mode (should be used for streaming)
-quality=0                   - "Balanced" as it usually gives best SSIM
-profile=100               - "high" should be used for +HD encoding
-level=52                     - "Level 5.2" alows more macroblocks than 4.2 - but actually 4.2 (or 42) would be enough
-rc=1                            - enables CBR
-preanalysis=true     - enables "lookahead"
-coder=1                     - enables "CABAC"
-qmin=25                    - min QP 25 (default = 22)
-qmax=46                   - max QP 46 (default =46)
-merange=48             - should force MotionEstimation to 48 (default = 36)(dont know if this works)
-subq=9                       - subme 9 (dont know if it works)
-me_half_pel=true     - MotionEstimation - i dont know if this is already enabled with next parameter
-me_quarter_pel=true      - MotionEstimation - "best" AMD AMF can do
-vbaq=true                   - Variable Adaptive Quantization - can improve image quality (depending on source)
-trellis=2                       - i dont know if this is "active" with AMFenc
-intra_refresh_mb=170      - max number of Intra-Refresh per Macroblock (default is "0" or "255") 

PC: R7 2700X @PBO + RX 580 4G (1500MHz/2000MHz CL16) + 32G DDR4-3200CL14 + 144hz 1ms FS P + 75hz 1ms FS
Laptop: R5 2500U @30W + RX 560X (1400MHz/1500MHz) + 16G DDR4-2400CL16 + 120Hz 3ms FS