All times are UTC+01:00




Post new topic  Reply to topic  [ 166 posts ]  Go to page Previous 111 12 13 14 1517 Next
Author Message
PostPosted: Sat Apr 20, 2019 3:15 am 

Joined: Sat Nov 10, 2012 2:16 am
Posts: 106
Location: Australia
Hi Hans

Windows 10 64 Bit, Stereo Tool v9.24 B022 Exception

1. Exception Issue Logged
Attachment:
ST923B022_64b-SA-Exception.jpg
ST923B022_64b-SA-Exception.jpg [ 25.94 KiB | Viewed 7715 times ]
Caught during overnight test run (was not doing anything on the machine console). Closing it did not produce any audio loss/issues.

2. Running ST normally for a while, switch to bypass, switch back. Major instability on FM and Normal Out due to buffer issues.
They are maxed out (200%) and not back to normal 100%. Audio is speed up to bring back into sync which depending on configuration can take several seconds to a minute to stabilise. Over several turn-off tests, the FM Out is typically 200% but many a time the Normal out is at <10%?
MicroMPX stereo image seems to get corrupted (has an oscillating pattern in the image level) - only way to fix it is to turn the IP flow to the endpoint off and back on again, or restart ST.

3. May I ask if you thought of looking into licencing the DANTE Virtual Audio card and code it into Stereo Tool to provide native support? (www.audinate.com/products/dante-enabled/partners)

Regards

Ross


Top
   
PostPosted: Sat Apr 20, 2019 9:24 am 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11185
Hi,

The points under 2 - are they related? So, the MicroMPX issue is also caused by going to bypass? Can you reproduce it?

Since you're specifically asking for Dante: Is this with a Dante input sound card? I'm asking because we have tested turning bypass on/off on normal sound cards just a few weeks ago and haven't seen any weird issues. Also, how big are your buffer settings? (Actually, it might help to get your full settings).

About 3: Since it's a virtual sound card it should work like any other sound card, without any need from us to support anything special. What is going wrong on your end, or what are you missing?


Top
   
PostPosted: Sat Apr 20, 2019 2:30 pm 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11185
Quote:
Windows 10 64 Bit, Stereo Tool v9.24 B022 Exception

1. Exception Issue Logged

ST923B022_64b-SA-Exception.jpg

Caught during overnight test run (was not doing anything on the machine console). Closing it did not produce any audio loss/issues.
Ok, I found where this is, but I think it must have been caused by something else. Do you have a directory C:\temp on your system? If so, please check if there's a file C:\temp\StereoTool_Exceptions_Log.txt and if so, send it to me. I expect that there is an earlier exception reported that in turn caused this one.


Top
   
PostPosted: Sat Apr 20, 2019 3:18 pm 

Joined: Sat Nov 10, 2012 2:16 am
Posts: 106
Location: Australia
Quote:
Ok, I found where this is, but I think it must have been caused by something else. Do you have a directory C:\temp on your system? If so, please check if there's a file C:\temp\StereoTool_Exceptions_Log.txt and if so, send it to me. I expect that there is an earlier exception reported that in turn caused this one.
Hi Hans

Contents of log file:

---
EXCEPTION OCCURRED IN ProcessingChain Thread (Exception 0xc0000005 at address 0x7ff6af1154fd, read/write=0, r/w location=0x0). Offset-info: 0x7ff6b1580889
EXCEPTION OCCURRED IN release_ss: ss_inuse[index]
EXCEPTION OCCURRED IN ProcessingChain Thread (Exception 0xc0000005 at address 0x7ff6b0e94d59, read/write=1, r/w location=0x0). Offset-info: 0x7ff6b1580889

Regards

Ross


Top
   
PostPosted: Sat Apr 20, 2019 3:21 pm 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11185
Ah. That makes more sense indeed. Will try to figure out what happened...

Edit: Ok, this has happened before on your end! I found the code that's causing it and it says:
Code:
//#warning CRASH HERE - LIVEWIREAUS. Oddly enough in Bypass mode!!!!
I haven't been able to figure out what was wrong before (and the Bypass mode made no sense, but I guess I can ignore that part of the comment now).


Just to make sure: Are there no errors in the file BEFORE this one? Since it still looks like a weird place to suddenly crash.\\\


Edit: I searched a bit further and I really need to have your *full* settings file (SAVE -> EXPORT ALL) to figure out what's happening here. There are too many possible paths through the code to verify all of them.


Top
   
PostPosted: Sun Apr 21, 2019 2:08 am 

Joined: Sat Nov 10, 2012 2:16 am
Posts: 106
Location: Australia
Quote:
Ah. That makes more sense indeed. Will try to figure out what happened...
I haven't been able to figure out what was wrong before (and the Bypass mode made no sense, but I guess I can ignore that part of the comment now).
This was before I started running in Bypass mode and found the other issue reported. The service was left running unattended overnight running, ASIO In, Out, LQLL, FM feeding Breakaway Pipeline @ 192k and running microMPX in v2/512k mode. Came back to the machine with that error shown on the console. It was still processing fine so I grabbed and logged the error in case it was something important to review.
Quote:
Just to make sure: Are there no errors in the file BEFORE this one? Since it still looks like a weird place to suddenly crash.\\\
Edit: I searched a bit further and I really need to have your *full* settings file (SAVE -> EXPORT ALL) to figure out what's happening here. There are too many possible paths through the code to verify all of them.
Yes - I generally archive the file after a new version, or playing around with my normal config so I don't end up with items from previous versions. I thought the --- line indicated the last item written or when it physically restarted.
Will arrange to send you the exported .ini file shortly.

Ross


Top
   
PostPosted: Sun Apr 21, 2019 2:31 am 

Joined: Sat Nov 10, 2012 2:16 am
Posts: 106
Location: Australia
Quote:
Hi,
The points under 2 - are they related? So, the MicroMPX issue is also caused by going to bypass? Can you reproduce it?
Yes, I was able reproduce the error. Everytime I placed into Bypass and returned, I was able to reproduce it 50% of the time.
FM Out buffer was brought back high (full), Normal out was under 20%. In this mode, in most cases the buffers were brought back into line (FM Out you could hear the buffer being reduced by speed changes). microMPX did not suffer straight away - came back online and after about 5 seconds, the stereo image intensity/level started oscillating noticeably. I have the microMPX decoder running into a 30w RVR transmitter running very low power so I can hear the output of it back in my home office. The level variations could be heard across 3 different FM tuners (I thought it was my Denon unit playing up). FM Output I can also hear locally - I feed it into MPXTool for monitoring and status.
I reported both scenarios (Buffer/microMPX) as it appeared the common trigger was the return from Bypass.
Quote:
Since you're specifically asking for Dante: Is this with a Dante input sound card? I'm asking because we have tested turning bypass on/off on normal sound cards just a few weeks ago and haven't seen any weird issues. Also, how big are your buffer settings? (Actually, it might help to get your full settings).
It was more of a query of expanding the current AES67 option and supporting DANTE natively without the need of another virtual sound card driver (ie having it fully integrated). I will also admit I have not been able to get the current AES67/DANTE config to work for me either...

Regards

Ross


Top
   
PostPosted: Sun Apr 21, 2019 4:48 am 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11185
Ok,
1- I've received your settings file, will try to find out what's going on.
2- AES67/Dante should work with what's there now, but we haven't tested Dante yet so it might now. Do you get no signal at all? Note that some of these protocols need to be set to AES67 mode specifically... AES67 is independant of Dante, LiveWire, WheatNet, Ravenna etc, so we should not need to support any specifics of any of those protocols.
3- Just for clarity: Are you using the built-in MicroMPX encoder or a separate one via the Breakaway Pipeline? And is the stereo from the direct FM output ok? (I can't think of any reason why it wouldn't be).

Anyway I have your settings now so let's see if it reproduces...


Edit: Ah, FM buffer problem reproduces here. And the cause is clear: We just stop sending any audio to it in Bypass mode, and don't re-open the sound card when you disable Bypass. If FM had also been running in ASIO-mode it would have been ok, but without it, not. And your resampling settings are quite extreme so yes you will hear that.

Edit #2: Crash code can be reached in this preset from Auto EQ, Bass EQ and Multiband 2.

Edit #3: I think I got it. An assert is failing when I load your settings. The cause appears to be related to a timing issue with threading (if I move the mouse above the ST window it fails faster). And you are using 7 threads with subthreads, so 14 processing threads in total. Which is probably more than the number of cores that you have. Now, obviously that shouldn't crash. Will look into it - this could be causing all kinds of weird things.


Top
   
PostPosted: Sun Apr 21, 2019 7:22 am 

Joined: Sat Nov 10, 2012 2:16 am
Posts: 106
Location: Australia
Quote:
3- Just for clarity: Are you using the built-in MicroMPX encoder or a separate one via the Breakaway Pipeline? And is the stereo from the direct FM output ok? (I can't think of any reason why it wouldn't be).
Yes - I am using the internal MicroMPX encoder and correct - the FM Out had issues as well. While the non ASIO FM out I can understand causing issues, I thought the ASIO Normal out would not be affected, but it was as well...
Quote:
Edit #3: I think I got it. An assert is failing when I load your settings. The cause appears to be related to a timing issue with threading (if I move the mouse above the ST window it fails faster). And you are using 7 threads with subthreads, so 14 processing threads in total. Which is probably more than the number of cores that you have. Now, obviously that shouldn't crash. Will look into it - this could be causing all kinds of weird things.
Test PC running a Intel(R) Core(TM) i7-5960X CPU @ 3.00GHz - specs in operation with ST below -:

Base speed: 3.00 GHz
Sockets: 1
Cores: 8
Logical processors: 16
Virtualisation: Disabled
Hyper-V support: Yes
L1 cache: 512 KB
L2 cache: 2.0 MB
L3 cache: 20.0 MB

Utilisation 45%
Speed 3.48 GHz
Up time 7:04:00:46
Processes 232
Threads 2819
Handles 121230

Regards

Ross


Top
   
PostPosted: Sun Apr 21, 2019 11:09 am 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11185
Ok, I was wrong. I did find a bug (that was only present when using Spectral Balance) and I've now fixed it. But it cannot have caused exceptions, only a small offset in compression levels occasionally. I'm not seeing any problems anymore now. I will let your preset run in debug mode for a while now.

Edit (half a day later): No problems so far. And in debug mode, they really should show up. No idea yet what's happening on your end...


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 166 posts ]  Go to page Previous 111 12 13 14 1517 Next

All times are UTC+01:00


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