RobustEnigma
New Member
I've been trying to chase this problem for a while now, with no real success.
I play mostly FPS games (Call of Duty: WWII, and Black Ops 4)
The latter of the two kicks the desync far more frequently than WWII.
I've tried so many things, I'm not sure where to look next.
I've tried reducing Webcam Size, Audio Bitrates, Sample Rates, Enabling Buffering, and so much more.
I monitored my CPU usage during the desync and it remains stable during the kick.
Here is the log during Test Stream/Recording: https://obsproject.com/logs/vvkfQNdyLIo1gnY0
Here is the log after stopping the Test Stream/Recording: https://obsproject.com/logs/mPijOr6D-O8EaIIx
The audio is fine during the start of games, until there's too much sound going on at once.
For instance, it will be lined up, until I get a headshot, or an explosion happens nearby.
It seems like it just gets overwhelmed, and can't keep up, so it drops out by about a full second.
I can fix it on the fly by right clicking on my Elgato Source, going to Properties, and switching the Audio Output Mode from WaveOut to any other option, and back again to WaveOut.
Using Direct Sound as an Audio Output causes severe graphical stuttering.
My PS4 is running Linear PCM.
My earbuds run through my Blue Yeti, but the problem was persistent with my wireless headset (before it broke) and running through my actual sound card as well.
If you have any suggestions, please, feel free to suggest.
My gut assumption is my dated processor is just not keeping up with the encoding, but I can't be certain.
Additionally, as a pre-emptive, I'm going to run another test, deleting extra Scenes, Deletings a few extra Sources, and changing Audio Bitrate from 128 to 160, and downscaling Output to 720.
Here's that log During: https://obsproject.com/logs/UPwiiZKdAXFv3gUt
Here's that log After: https://obsproject.com/logs/fmDidCO5j2BbGR6F
Unsure if I just got lucky in game, but it seemed like it took about twice as long for the audio to desync this time around than it did at the previous settings.
But still at about one desync during match.
Here's to hoping it's something simple, and thanks in advance for the support.
--Edit// My goal for tomorrow, is to start streaming only at 720 output, reduce my webcam down from 1080, and see if there are even more sources I can eliminate. I'm assuming that old ones I don't use are still queuing up despite not being visible.
Additionally I saw in the logs that there are still some missing graphical assets I've yet to remove from sources. Going to try and address that.
I've kept my webcam at a shrunk 1080, as I use a full screen face cam when chatting with viewers, and at 480 the fuzz was apparent, but I think even 720 will be an improvement. I figure, even if it doesn't fix it, added stability is good either way. Each of my attempts at solving this issue I've reset the variables as I've tried each one, so maybe the key is dialing everything in at once.
Lastly, I think setting the audio nitrate to 160 during the previous test was counter productive. My thought was, maybe I was trying to squeeze too much through the 128 bitrate, and upping to 160 would help, but I believe my stressed fatigued brain thought about this problem in the wrong light.
Also, I noticed
and figure that might be something worth addressing, as well as not running OBS Studio as Admin. Not sure, but it's all a start.
Either way, I will let you know when I run another test tomorrow.
Fingers crossed. //Edit--
I play mostly FPS games (Call of Duty: WWII, and Black Ops 4)
The latter of the two kicks the desync far more frequently than WWII.
I've tried so many things, I'm not sure where to look next.
I've tried reducing Webcam Size, Audio Bitrates, Sample Rates, Enabling Buffering, and so much more.
I monitored my CPU usage during the desync and it remains stable during the kick.
Here is the log during Test Stream/Recording: https://obsproject.com/logs/vvkfQNdyLIo1gnY0
Here is the log after stopping the Test Stream/Recording: https://obsproject.com/logs/mPijOr6D-O8EaIIx
The audio is fine during the start of games, until there's too much sound going on at once.
For instance, it will be lined up, until I get a headshot, or an explosion happens nearby.
It seems like it just gets overwhelmed, and can't keep up, so it drops out by about a full second.
I can fix it on the fly by right clicking on my Elgato Source, going to Properties, and switching the Audio Output Mode from WaveOut to any other option, and back again to WaveOut.
Using Direct Sound as an Audio Output causes severe graphical stuttering.
My PS4 is running Linear PCM.
My earbuds run through my Blue Yeti, but the problem was persistent with my wireless headset (before it broke) and running through my actual sound card as well.
If you have any suggestions, please, feel free to suggest.
My gut assumption is my dated processor is just not keeping up with the encoding, but I can't be certain.
Additionally, as a pre-emptive, I'm going to run another test, deleting extra Scenes, Deletings a few extra Sources, and changing Audio Bitrate from 128 to 160, and downscaling Output to 720.
Here's that log During: https://obsproject.com/logs/UPwiiZKdAXFv3gUt
Here's that log After: https://obsproject.com/logs/fmDidCO5j2BbGR6F
Unsure if I just got lucky in game, but it seemed like it took about twice as long for the audio to desync this time around than it did at the previous settings.
But still at about one desync during match.
Here's to hoping it's something simple, and thanks in advance for the support.
--Edit// My goal for tomorrow, is to start streaming only at 720 output, reduce my webcam down from 1080, and see if there are even more sources I can eliminate. I'm assuming that old ones I don't use are still queuing up despite not being visible.
Additionally I saw in the logs that there are still some missing graphical assets I've yet to remove from sources. Going to try and address that.
I've kept my webcam at a shrunk 1080, as I use a full screen face cam when chatting with viewers, and at 480 the fuzz was apparent, but I think even 720 will be an improvement. I figure, even if it doesn't fix it, added stability is good either way. Each of my attempts at solving this issue I've reset the variables as I've tried each one, so maybe the key is dialing everything in at once.
Lastly, I think setting the audio nitrate to 160 during the previous test was counter productive. My thought was, maybe I was trying to squeeze too much through the 128 bitrate, and upping to 160 would help, but I believe my stressed fatigued brain thought about this problem in the wrong light.
Also, I noticed
Code:
[CoreAudio encoder]: CoreAudio AAC encoder not installed on the system or couldn't be loadedEither way, I will let you know when I run another test tomorrow.
Fingers crossed. //Edit--
Last edited: