All times are UTC+02:00




Post new topic  Reply to topic  [ 707 posts ]  Go to page Previous 14 5 6 7 871 Next
Author Message
PostPosted: Wed Nov 30, 2022 4:01 pm 
User avatar

Joined: Tue Mar 17, 2009 2:56 pm
Posts: 4161
Quote:
The only change in 008 that the block size behavior is the same as before, but with slightly lower latency, it might be easier to make it misbehave. However, M at 2.0000 isn't neccessarily a problem - did it cause issues or not get out of it anymore?
What did you set the latency to? I think you set it really low, correct? Since the latency is 1 block lower than in the old framework, that might explain this, if it was already on the edge before.
I was using 40ms bufffersize. Same as in beta7. i think i can use even lower in beta7 like 30 and it still will be good. Upon ST start it all looks and work ok, then at some random point after some time "M" fastly goes to 2.000 and audio starts to stutter (same as beta4) If i dont touch it it will eventually stop audio stutter and come back to 1.0000 until again for no reason go again to 2.0000 after 5 minutes or 1 hour.
Maybe we now need to use blocksize? but how to fastly discover wich one.

_________________
control point
control point2


Top
   
PostPosted: Wed Nov 30, 2022 4:04 pm 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11211
Quote:
Quote:
The only change in 008 that the block size behavior is the same as before, but with slightly lower latency, it might be easier to make it misbehave. However, M at 2.0000 isn't neccessarily a problem - did it cause issues or not get out of it anymore?
What did you set the latency to? I think you set it really low, correct? Since the latency is 1 block lower than in the old framework, that might explain this, if it was already on the edge before.
I was using 40ms bufffersize. Same as in beta7. i think i can use even lower in beta7 like 30 and it still will be good. Upon ST start it all looks and work ok, then at some random point after some time "M" fastly goes to 2.000 and audio starts to stutter (same as beta4) If i dont touch it it will eventually stop audio stutter and come back to 1.0000 until again for no reason go again to 2.0000 after 5 minutes or 1 hour.
Maybe we now need to use blocksize? but how to fastly discover wich one.
The block size was hard coded to 1024. So if you set it to that it should behave like it did in 007, but that shouldn't be needed. Could you try what happens if you set the buffer a bit higher? And, just to make sure: I assume that this setting worked fine with the old I/O layer?


Top
   
PostPosted: Wed Nov 30, 2022 4:24 pm 
User avatar

Joined: Tue Mar 17, 2009 2:56 pm
Posts: 4161
Quote:
Quote:
Quote:
The only change in 008 that the block size behavior is the same as before, but with slightly lower latency, it might be easier to make it misbehave. However, M at 2.0000 isn't neccessarily a problem - did it cause issues or not get out of it anymore?
What did you set the latency to? I think you set it really low, correct? Since the latency is 1 block lower than in the old framework, that might explain this, if it was already on the edge before.
I was using 40ms bufffersize. Same as in beta7. i think i can use even lower in beta7 like 30 and it still will be good. Upon ST start it all looks and work ok, then at some random point after some time "M" fastly goes to 2.000 and audio starts to stutter (same as beta4) If i dont touch it it will eventually stop audio stutter and come back to 1.0000 until again for no reason go again to 2.0000 after 5 minutes or 1 hour.
Maybe we now need to use blocksize? but how to fastly discover wich one.
The block size was hard coded to 1024. So if you set it to that it should behave like it did in 007, but that shouldn't be needed. Could you try what happens if you set the buffer a bit higher? And, just to make sure: I assume that this setting worked fine with the old I/O layer?
I don't think i ever saw this in old I/O - "M" to go to 2.0000. Will try with 100ms.
Ideal Block size for what i found is usually multiply of samplerate, so at 48kHz ideal should be 960 or 1920 sometimes 480, but it's not rule and not same for input and output.

_________________
control point
control point2


Top
   
PostPosted: Wed Nov 30, 2022 4:29 pm 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11211
Quote:
Quote:
Quote:

I was using 40ms bufffersize. Same as in beta7. i think i can use even lower in beta7 like 30 and it still will be good. Upon ST start it all looks and work ok, then at some random point after some time "M" fastly goes to 2.000 and audio starts to stutter (same as beta4) If i dont touch it it will eventually stop audio stutter and come back to 1.0000 until again for no reason go again to 2.0000 after 5 minutes or 1 hour.
Maybe we now need to use blocksize? but how to fastly discover wich one.
The block size was hard coded to 1024. So if you set it to that it should behave like it did in 007, but that shouldn't be needed. Could you try what happens if you set the buffer a bit higher? And, just to make sure: I assume that this setting worked fine with the old I/O layer?
I don't think i ever saw this in old I/O - "M" to go to 2.0000. Will try with 100ms.
Ideal Block size for what i found is usually multiply of samplerate, so at 48kHz ideal should be 960 or 1920 sometimes 480, but it's not rule and not same for input and output.
Yes, and that's what the new code tends to do (still depending on the sound card). But I saw it open the output at 441 at 44.1 kHz since the update, same as in the old code. Since everything *should* be the same I'm a bit surprised about this 2.000 thing... Unless there's a wrapping issue (maybe it dropped below 0 and then went to 2?).


Top
   
PostPosted: Wed Nov 30, 2022 4:37 pm 
User avatar

Joined: Tue Mar 17, 2009 2:56 pm
Posts: 4161
Quote:
Quote:
Quote:

The block size was hard coded to 1024. So if you set it to that it should behave like it did in 007, but that shouldn't be needed. Could you try what happens if you set the buffer a bit higher? And, just to make sure: I assume that this setting worked fine with the old I/O layer?
I don't think i ever saw this in old I/O - "M" to go to 2.0000. Will try with 100ms.
Ideal Block size for what i found is usually multiply of samplerate, so at 48kHz ideal should be 960 or 1920 sometimes 480, but it's not rule and not same for input and output.
Yes, and that's what the new code tends to do (still depending on the sound card). But I saw it open the output at 441 at 44.1 kHz since the update, same as in the old code. Since everything *should* be the same I'm a bit surprised about this 2.000 thing... Unless there's a wrapping issue (maybe it dropped below 0 and then went to 2?).
hmm not to 0 goes to when it's around 0.9000 then it's like it tries to sync and passes 1.0000 goes to 1.1000 and sudenly in 3 seconds goes to 2.0000.
But again not a rule. When i start ST it starts with M at 0.9 and slowly goes to 1.0000 and it's fine for some time, then for no obvious reason i hear audio stutters and M to 2.0000.

_________________
control point
control point2


Top
   
PostPosted: Wed Nov 30, 2022 5:35 pm 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11211
Quote:
Quote:
Quote:

I don't think i ever saw this in old I/O - "M" to go to 2.0000. Will try with 100ms.
Ideal Block size for what i found is usually multiply of samplerate, so at 48kHz ideal should be 960 or 1920 sometimes 480, but it's not rule and not same for input and output.
Yes, and that's what the new code tends to do (still depending on the sound card). But I saw it open the output at 441 at 44.1 kHz since the update, same as in the old code. Since everything *should* be the same I'm a bit surprised about this 2.000 thing... Unless there's a wrapping issue (maybe it dropped below 0 and then went to 2?).
hmm not to 0 goes to when it's around 0.9000 then it's like it tries to sync and passes 1.0000 goes to 1.1000 and sudenly in 3 seconds goes to 2.0000.
But again not a rule. When i start ST it starts with M at 0.9 and slowly goes to 1.0000 and it's fine for some time, then for no obvious reason i hear audio stutters and M to 2.0000.
If the buffer filling drops below 0, it reports that the buffer is maximally full - that's wrong and I'll fix that now. Not sure if this is what was causing the issue but it might be - once it thinks that the buffer is too full it will start to play faster and only make the issue worse. Internally, the buffer is bigger so I can distinguish between slightly below 0 vs slightly above 2. And I thought that we did that - but apparently not.

Edit: Oh, there also was a place where, if the buffer filling was exactly 0, synchonisation would switch off. Note that these issues aren't new in version 008 or 009, they were there since 004. Anyway, both issues are fixed, so I hope that the behavior will be much better in the next build.


Top
   
PostPosted: Thu Dec 01, 2022 9:00 pm 

Joined: Wed Apr 03, 2013 10:20 pm
Posts: 64
Tried an easy stream in/out thingi in regualar Beta 10.02-009 and in latest "New Interface preview Beta v10.02-007"... VLC all new installed. And an "All New Win 10" too.

The URL I give ST is real but it wont take it. And it will not stop nagging about it either. I have to kill ST and delete the .ini to start over, and no-way can I make it read a stream that older versjons read perfectly. Am I all into some new stuff I dont know about or is there a new bug I havent cought yet :) (I never got the covid one either).

It never gives me the option to correct URL, it just spawns numbers it tried to connect but it failes.

BTW, I do love the new setups, it even sounds better. At the moment I'm trying to update clients presets....but hehe no stream in no fun out :)

Greate work on the new stuff, take care and stay healthy
Attachment:
Screen Shot 2022-12-01 at 8.28.06 PM.png
Screen Shot 2022-12-01 at 8.28.06 PM.png [ 24.87 KiB | Viewed 1853 times ]
Attachment:
Screen Shot 2022-12-01 at 8.44.35 PM.png
Screen Shot 2022-12-01 at 8.44.35 PM.png [ 12.02 KiB | Viewed 1853 times ]
Regards,

MrDonT


Top
   
PostPosted: Thu Dec 01, 2022 9:05 pm 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11211
Quote:
Tried an easy stream in/out thingi in regualar Beta 10.02-009 and in latest "New Interface preview Beta v10.02-007"... VLC all new installed. And an "All New Win 10" too.

The URL I give ST is real but it wont take it. And it will not stop nagging about it either. I have to kill ST and delete the .ini to start over, and no-way can I make it read a stream that older versjons read perfectly. Am I all into some new stuff I dont know about or is there a new bug I havent cought yet :) (I never got the covid one either).

It never gives me the option to correct URL, it just spawns numbers it tried to connect but it failes.

BTW, I do love the new setups, it even sounds better. At the moment I'm trying to update clients presets....but hehe no stream in no fun out :)

Greate work on the new stuff, take care and stay healthy
Screen Shot 2022-12-01 at 8.28.06 PM.png
Screen Shot 2022-12-01 at 8.44.35 PM.png

Regards,

MrDonT
I just happened to run into a similar issue today, but only with too new libVLC versions (and on both the old and new Stereo Tool frameworks). Are you really sure that it works with the old version on the same system?


Top
   
PostPosted: Thu Dec 01, 2022 9:16 pm 

Joined: Wed Apr 03, 2013 10:20 pm
Posts: 64
I've got a 9.92 on the same pc reading stream in and giving it out with out problems, and I also using a Sonobus on the as a backup. It works.
Its kinda weird, but its good to know that its not just me then :)

MrDonT


Top
   
PostPosted: Thu Dec 01, 2022 9:22 pm 

Joined: Wed Apr 03, 2013 10:20 pm
Posts: 64
The fun note on the side is that I can show them "my demos" with brand new layout and theire all melting like caramels....but to give them their own stations fails and thats hehe funny in my mind )(yea my stuff werks, yers dont heheh)

Back to the serious kind of things, I do still miss the opportunity to use a VLC stream as a backup input (audio 2).

MrDonT


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 707 posts ]  Go to page Previous 14 5 6 7 871 Next

All times are UTC+02:00


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Limited