ShimadaHaruka
New Member
Log Analysis: https://obsproject.com/tools/analyzer?log_url=https://obsproject.com/logs/rNIahbOmW3ZOOIFk
Full Log: https://obsproject.com/logs/rNIahbOmW3ZOOIFk
I will detail my testing process:
I have included a lengthy log with all the different tests that I ran. First I tested NVENC H264. I tested once on P1 quality and once on P7 since in the past there were issues based on the quality setting. With both settings I was able to adjust the bitrate while the stream was live without issue. I could go up and down freely regardless of the starting bitrate. (manual mid-stream bitrate changes don't seem to be reflected in the log.) Dynamic bitrate worked perfectly (tested at the end of the log.)
When switching to NVENC HEVC I could manually adjust the bitrate as long as I didn't go below the bitrate that I started at. P1 or P7 made no difference. I could not lower the bitrate mid-stream lower than the bitrate that I started at. It would briefly drop, then bounce back up to the bitrate the stream started at and stay there.
With those results in mind I first tested dynamic bitrate starting from a low bitrate of 2000 and then raised it mid-stream to 8000 hoping that it would work but it did not. The log doesn't even show any attempts at bitrate changes despite me saturating my limited upload bandwidth (10megabit) with speedtests. On youtube the stream stayed in a buffering state and gave the message, "Error: YouTube is not receiving enough video to maintain smooth streaming. As such, viewers will experience buffering." I tried another test starting from 8000 with similar results, although this time the log does indicate that attempts were made to lower the bitrate.
Lastly I tested Dynamic Bitrate with h.264 which worked perfectly, no buffering, no problems.
I think that covers what I did. I would not have gone to all this length except that I didn't see anything online about this problem so it must be something on my end, otherwise I'm sure someone else would have mentioned it. There's no issues on github about HEVC and dynamic bitrate as far as I can tell, so it's probably isolated just to me and not a problem with YouTube or OBS. But I don't know why it worked with h264 and not HEVC.
This is not a dealbreaker for me. It would be nice if it worked, but for now if I lower my bitrate to around 4500kbps then it works pretty good even without dynamic bitrate. I just think that dynamic bitrate is a great feature that I would love to use. But I'm not seeking advice on how to work around the issue, nor am I seeking general advice regarding OBS settings or livestreaming. Also the settings that I used for the test may not reflect the settings that I use for livestreaming (for example, I don't use display capture when streaming).
I've tried to do my due diligence before coming here with the issue. I don't think it's a network issue as dynamic bitrate works well with h264. I've included the log of all the tests that I ran. The Analyzer doesn't indicate any problems, although I used many different settings during the course of the test and I think it only analyzes the last settings used. This dynamic bitrate issue is probably caused by some boneheaded thing that I did that screwed it up somehow.
Full Log: https://obsproject.com/logs/rNIahbOmW3ZOOIFk
I will detail my testing process:
I have included a lengthy log with all the different tests that I ran. First I tested NVENC H264. I tested once on P1 quality and once on P7 since in the past there were issues based on the quality setting. With both settings I was able to adjust the bitrate while the stream was live without issue. I could go up and down freely regardless of the starting bitrate. (manual mid-stream bitrate changes don't seem to be reflected in the log.) Dynamic bitrate worked perfectly (tested at the end of the log.)
When switching to NVENC HEVC I could manually adjust the bitrate as long as I didn't go below the bitrate that I started at. P1 or P7 made no difference. I could not lower the bitrate mid-stream lower than the bitrate that I started at. It would briefly drop, then bounce back up to the bitrate the stream started at and stay there.
With those results in mind I first tested dynamic bitrate starting from a low bitrate of 2000 and then raised it mid-stream to 8000 hoping that it would work but it did not. The log doesn't even show any attempts at bitrate changes despite me saturating my limited upload bandwidth (10megabit) with speedtests. On youtube the stream stayed in a buffering state and gave the message, "Error: YouTube is not receiving enough video to maintain smooth streaming. As such, viewers will experience buffering." I tried another test starting from 8000 with similar results, although this time the log does indicate that attempts were made to lower the bitrate.
Lastly I tested Dynamic Bitrate with h.264 which worked perfectly, no buffering, no problems.
I think that covers what I did. I would not have gone to all this length except that I didn't see anything online about this problem so it must be something on my end, otherwise I'm sure someone else would have mentioned it. There's no issues on github about HEVC and dynamic bitrate as far as I can tell, so it's probably isolated just to me and not a problem with YouTube or OBS. But I don't know why it worked with h264 and not HEVC.
This is not a dealbreaker for me. It would be nice if it worked, but for now if I lower my bitrate to around 4500kbps then it works pretty good even without dynamic bitrate. I just think that dynamic bitrate is a great feature that I would love to use. But I'm not seeking advice on how to work around the issue, nor am I seeking general advice regarding OBS settings or livestreaming. Also the settings that I used for the test may not reflect the settings that I use for livestreaming (for example, I don't use display capture when streaming).
I've tried to do my due diligence before coming here with the issue. I don't think it's a network issue as dynamic bitrate works well with h264. I've included the log of all the tests that I ran. The Analyzer doesn't indicate any problems, although I used many different settings during the course of the test and I think it only analyzes the last settings used. This dynamic bitrate issue is probably caused by some boneheaded thing that I did that screwed it up somehow.