New Computer, Occasional Slight A/V Stutter During Recording

R

ReadyToLearn

New Member
Greetings!

I have a brand new (few months old) desktop with pretty decent specs, but I am still seeing some slight stutter while recording and so I'm trying to figure out where the bottleneck is. Note that I am recording-only, not streaming, and these are just "talking head" videos of myself facing the camera (head and shoulders, with some hand gesturing as I talk). Log (and Analyzed log) below and nothing stands out to me as problematic (though I see there has now been a minor OBS update). When setting up I used the OBS wizard and went with the recommended settings. The only thing I can think of is that I have my Blue Yeti X microphone configured as the mic within the Logitech c920 webcam settings, as opposed to configuring the Blue Yeti X as a separate audio source. I would not think that should make a difference unless somehow it is forcing the Yeti mic audio through the Logitech webcam hardware somehow.

Any thoughts, suggestions, recommendations are very appreciated!
:)


Analyzer | OBS

OBS (Open Broadcaster Software) is free and open source software for video recording and live streaming. Stream to Twitch, YouTube and many other providers or record your own videos with high quality H264 / AAC encoding.
obsproject.com obsproject.com

My Hardware/Software

Custom Built computer (AVADirect), December 2023
Motherboard: Asus Pro WS W680-ACE
CPU: Intel Core i5-14600K 14 (6P+8E) Cores 2.6 - 5.3GHz Turbo, LGA 1700, 181W
Ram: 64GB (2x 32GB KSM48E40BD8KM-32HM, DDR5 4800MT/s, CL40, 2Rx8, ECC Unbuffered DIMM Memory)
Video Card: GeForce RTX 4060 DUAL-RTX4060-O8G, 2505 - 2535MHz, 8GB GDDR6, Graphics Card
Sound Card: Onboard Sound
Hard Drive: Samsung 1TB 990 PRO, 7450 / 6900 MB/s, V-NAND 3-bit MLC, PCIe 4.0 x4 NVMe 2.0, M.2 2280 SSD

Monitor Max Resolution: 1680x1050

Windows 11 Pro x64

Webcam: Logitech c920 (max resolution: 1920 x 1080)
Microphone: Blue Yeti X

 
R

rockbottom

Active Member
The log is incomplete, there's a sample rate mismatch & few minor things but there's no rendering or encoding lag. Try without Norton 360 mucking up the works.
 
L

Lawrence_SoCal

Active Member
Are you sure stuttering is with recording, and not the playback (ie video player)?

also beware mismatch between screen resolution and canvas size. Is that right, your is NOT a 1080p monitor?
17:34:43.155: size={1680, 1050}

17:34:43.282: video settings reset:
17:34:43.282: base resolution: 1920x1080
17:34:43.282: output resolution: 1920x1080
 
R

ReadyToLearn

New Member
rockbottom said:
The log is incomplete, there's a sample rate mismatch & few minor things but there's no rendering or encoding lag. Try without Norton 360 mucking up the works.
Click to expand...
Here's another log -- perhaps this one is more complete?
And thanks for the suggestion regarding Norton 360. What aspect of Norton can be problematic (eg, Firewall somehow; or just "Suspend Background Activity"; or something else)?

Lawrence_SoCal said:
Are you sure stuttering is with recording, and not the playback (ie video player)?

also beware mismatch between screen resolution and canvas size. Is that right, your is NOT a 1080p monitor?
17:34:43.155: size={1680, 1050}

17:34:43.282: video settings reset:
17:34:43.282: base resolution: 1920x1080
17:34:43.282: output resolution: 1920x1080
Click to expand...
That is actually an interesting question (regarding is the stuttering in recording or playback)? I don't actually know. I've always played back in VLC, but I just tested playback in a few other programs and I am seeing/hearing the slight stutter/clipping in the same places in each program. So yeah, I guess it does seem to be in the recorded file itself, as far as I can tell.

You are correct regarding the monitor -- it is 1680 x 1050. I think when I was first setting things up in OBS last year on a prior computer I couldn't figure out how to have an Output resolution of 1920x1080 unless I made the Base resolution match. Is there a better way I could be going about this? My main goals are I want to be able to see myself on the screen, nearly fullscreen, and be recording output to 1080p. And I would really prefer not to get a larger monitor, for various reasons.

Thank you both of you for your replies!
 
R

rockbottom

Active Member
Norton came pre-installed on a POS Gateway I bought back in 2005. I nuked it in 3 days. It sucked then & don't think anything has changed since.

No lag in the 2nd log either but Norton is still in play & the sample rate mis-match (Yeti) is still there as well. Anyway, now that there is a complete log I can see your video timings look A-OK but your audio is lagging & needs improvement (sped up).
 
R

ReadyToLearn

New Member
rockbottom said:
Norton came pre-installed on a POS Gateway I bought back in 2005. I nuked it in 3 days. It sucked then & don't think anything has changed since.

No lag in the 2nd log either but Norton is still in play & the sample rate mis-match (Yeti) is still there as well. Anyway, now that there is a complete log I can see your video timings look A-OK but your audio is lagging & needs improvement (sped up).
Click to expand...
Haha, well I've actually chosen to use Norton for a number of years and have been content with it. I am curious what you use instead (I know there are some good options out there)?

That makes sense what you're saying about the audio lagging. What are my options for resolving that? Could having the Blue Yeti X microphone configured as the mic within the Logitech c920 webcam settings, as opposed to configuring the Blue Yeti X as a separate audio source, be somehow problematic?
 
R

rockbottom

Active Member
Defender

The steps above to start. I would also go to my mobo's support website & check to see if there has been a new audio driver released.
 
R

ReadyToLearn

New Member
rockbottom said:
The steps above to start. I would also go to my mobo's support website & check to see if there has been a new audio driver released.
Click to expand...
Cool, sure, I can check for newer mobo audio drivers (though the computer was built only a few months ago).
And by "steps", do you just mean trying a recording with Norton disabled (ie, Silent mode), or was there something else you're alluding to which I am not spotting in our correspondence (or not comprehending)?
 
L

Lawrence_SoCal

Active Member
I was a happy Symantec AV customer for many years (large enterprise)... but Symantec went downhill a while ago, and Norton has always been less stable/reliable .
I don't trust MS to adequately monitor itself, but MS has made it very hard for 3rd parties to do a decent job...

I'd recommend testing at least with Norton 360 (or whichever version) disabled... just in case.

There have been BIOS, CPU, and GPU firmware and driver issues all associated with 'stuttering', as well as mismatch in FPS and output refresh rates (doesn't appear to apply to your situation)

As for 1080p recording... the issue is that you do NOT have 1080p content, you have 1050p... re-scaling is going to lower quality and may cause artefacts, or other issues (or not). If you want to Record at 1080p, then you need a 1080p or higher resolution monitor... there are virtual monitor type workaround approaches (I think, maybe), but not and also see yourself at same time.
 
R

ReadyToLearn

New Member
Lawrence_SoCal said:
I was a happy Symantec AV customer for many years (large enterprise)... but Symantec went downhill a while ago, and Norton has always been less stable/reliable .
I don't trust MS to adequately monitor itself, but MS has made it very hard for 3rd parties to do a decent job...

I'd recommend testing at least with Norton 360 (or whichever version) disabled... just in case.

There have been BIOS, CPU, and GPU firmware and driver issues all associated with 'stuttering', as well as mismatch in FPS and output refresh rates (doesn't appear to apply to your situation)

As for 1080p recording... the issue is that you do NOT have 1080p content, you have 1050p... re-scaling is going to lower quality and may cause artefacts, or other issues (or not). If you want to Record at 1080p, then you need a 1080p or higher resolution monitor... there are virtual monitor type workaround approaches (I think, maybe), but not and also see yourself at same time.
Click to expand...
Thank you so much for all of this info! I apologize for taking so long to respond -- a lot of life happened, and I've only just recently been able to do another recording. Per you and @rockbottom I added OBS files/folders to Norton Whitelists so Norton should not be interfering. I also checked and I do have the latest audio drivers for my mobo.

Here is my latest log, and this video still has the same issues (slight stutter, etc):

My hunch is that the issue is the re-scaling you talked about, and you've also helped me better understand how OBS works. I had been thinking that my source is my 1080p video camera content, so there should be no issue with sending that straight to 1080p output. But you have helped me understand that the actual content is determined by monitor size because that is the physical limitation for the Scene (if I am using my terms correctly). This makes sense in retrospect because OBS is more intended for being able to add overlays and other complex layouts and then streaming/recording that content. For my purposes I just want to pipe my 1080p video camera content straight to a 1080p recording, while being able to view myself fullscreen (on my 1050p monitor) during the recording process. So it seems likely that I either need to get a 1080p monitor (which I'm not super keen on for various reasons, though it isn't entirely out of the question), or probably consider a different software solution (unless I can figure out how to do the virtual monitor thing effectively AND still be able to view myself during recording).

One question, though: Even with a 1080p monitor, will there still be some interpolation issues (and thus potential lag/delay), given that the Scene window doesn't seem to fill the screen, or perhaps that is only happening for me because I have a 1050p monitor and my Scene is set to 1080p?
 
R

rockbottom

Active Member
The log is incomplete, can't tell if the audio timing improved with the new drivers. You're on your own with Norton. OBS may be whitelisted but that crap is still running in the background wasting resources.

There's no lag but for the next test, set your monitor @ 60HZ, disable Game Bar & Game DVR (Windows). Run OBS as Admin.
 
R

ReadyToLearn

New Member
rockbottom said:
The log is incomplete, can't tell if the audio timing improved with the new drivers. You're on your own with Norton. OBS may be whitelisted but that crap is still running in the background wasting resources.

There's no lag but for the next test, set your monitor @ 60HZ, disable Game Bar & Game DVR (Windows). Run OBS as Admin.
Click to expand...
Ah, sorry. Here is the complete log:

My computer specs are (I think) quite decent, so I don't think Norton in the background is an issue. I run Process Explorer in my system tray and I'm not observing issues from Norton regarding resource hogging, etc.

Per your suggestion I've looked into disabling the Game Bar, etc since I don't use that anyway, but it seems like they've removed the ability to turn it off in Windows 11.

Windows says my monitor is running at 59.95Hz already (with no option to change it).

rockbottom said:
Also need to get rid of the sample rate mismatch. If needed run OBS @ 44k so it matches the mic.
Click to expand...
Can you say more about how I do this? Do you mean set the Audio Bitrate to 44k, because it looks like it only goes down to 64? Oh wait, you probably mean Audio->General->Sample Rate->change to 44.1k instead of 48k. Is there anything downside to not running at 48k?
 
L

Lawrence_SoCal

Active Member
I got a 1440p monitor as I needed to capture a 1080p window, without rescaling... and I wanted room to crop Application Window title and bottom bar...etc. A 1200p monitor would have worked as well, but 1440p is more common, and more importantly a better deal at the time.
 
R

rockbottom

Active Member
Cool, timings all look good, no lag anywhere.

That's it. Not really & if the mic can't be changed to 48k, it what has to be done.

With the monitor being the way it is (59.95), I would try running OBS video @ 59.94 instead of 30.

Settings > Video
 
L

Lawrence_SoCal

Active Member
I am in the do NOT run OBS Studio as Admin camp, unless all else fails and specific reason to do so (usually gaming related GPU resource scheduling.... which doesn't seem to be your issue.. but ??)

As mentioned above... regarding Audio sampling rate, the biggest issue is to use a common sampling rate... some devices can be changed, others can't... so 1st step is to find common ground. If you have a choice of 44.1KHz and 48, I'd go with 48 as no need to match old CD sampling rate (if I'm recalling correctly)... instead, does your stream destination have a recommendation... I'd be inclined to use that..
 
R

ReadyToLearn

New Member
Lawrence_SoCal said:
I am in the do NOT run OBS Studio as Admin camp, unless all else fails and specific reason to do so (usually gaming related GPU resource scheduling.... which doesn't seem to be your issue.. but ??)

As mentioned above... regarding Audio sampling rate, the biggest issue is to use a common sampling rate... some devices can be changed, others can't... so 1st step is to find common ground. If you have a choice of 44.1KHz and 48, I'd go with 48 as no need to match old CD sampling rate (if I'm recalling correctly)... instead, does your stream destination have a recommendation... I'd be inclined to use that..
Click to expand...
Hmm, well, my Yeti mic is limited to 44.1k, but I upload the recorded files to YouTube, which requires a sample rate of 96khz or 48khz. So then potentially I should replace my mic (for one that is 48k)?
 
R

ReadyToLearn

New Member
Okay, actually my Blue Yeti X is supposed to be capable of 48k, and when I look in my Windows 11 settings it is set at 48k. I did some experimenting, and it looks like that by setting the Yeti to be an audio source within the webcam config it was limiting it to 44k. But if I add the Yeti as a totally separate audio device (like in the following log) it appears to actually show up as 48k. I did a brief test in this log and it seems like maybe this resolves the problem, but I'll need to see when I am doing a more extensive recording if that is actually the case. Note that you'll also see another audio device (Shokz OpenComm) which I was also testing out, but which was muted for this test.

I'll post again with an update when I have had a chance to do a longer recording. Thanks again, both of you, for all of your help!
 
You must log in or register to reply here.
Top