corrupted recordings, glitches on playback

I got two problems, possibly related, but not sure:

1. Some recorded takes are corrupted. This happens independent of project size, buffer size and CPU load. I could not find a reason/repetitive pattern yet. When it happens, all further recordings are also corrupted until Sonar is restarted, then it works fine, until ... it happens again ... next hour, next day ??? (last time it happened in a 2 track project recording MIDI and audio from an external piano)
2. On playback I get occasional audio glitches/crackles (independent of buffer sizes). In most projects it happens rarely (short crackles), but I found one project that repeatedly does it (at what appears random times i.e. about once in a 3 min playback, but at different times in each playback); here it is more like a buzzing noise of 0.5 - 2 sec.

 I did a number of tests:

• My first idea was HDD issues / pending HDD failure so I did some bechmarks tests. HDD bechmarks show read/write speed at about 110-120 MB/s; not spectacular but sufficient as the trouble making project lists about 15 MB/s in the task manager at peak reading times (some 40 tracks @ 96 kHz, 24 bit). So disk throughput does not seem to an issue. Additionally, I found enough space on the somewhat faster system drive to move the project there for a test. Same audio glitching issues.
• When the record corruption happens, it is not audible while direct monitoring via the interface (i.e. no problem at the A/D conversion, perfect audio quality there), only at playback it shows the corruption (sometimes totally garbled, sometimes a number of crackles), like it gets corrupted in the USB transfer or in Sonar (??)
• Malwarebytes found 0 detections. CCleaner nothing really relevant (but I let it clean anyway). This is what I expected as it is a dedicated DAW, which is online only for DAW software updates.
• LatencyMon reports a supersmooth system if run over extended times. Nothing recorded there, even when run with the troublemaking project until there is an audio glitch.
• Sonar itself is (probably) not to blame (?) - at least it happens with 2016.08, 2016.09, 2016.10
• Interestingly, playback glitching does not happen with an older audio interface (Roland Octa-Capture), but it also did not happen with the new (MOTU 1248 AVB) when I extensively tested it 2 months ago. I tried MOTU drivers from 2 months ago, but same story.
• I swapped USB cables, USB ports, same situation.
• BTW, win10 Pro, all updates applied.

 
I have the feeling that some component is on it's way out (but I can't fully exclude Sonar from the list). I will do more tests, but that requires cabling up different hardware ...
 
Any suggestions welcome.
 
 
 

Outside chance Rob but for point 2, in the project that has random noise, you don't have a plugin that has reverted to demo mode by any chance? I can't recall which plugin it was but I had a similar issue with a project/plugin after a windows update.

I did have an issue when I updated the CA2A on my Studio rig.
As it is not connected to the internet the "new" CA2A was indeed in DEMO mode.
"Cheesy" is right, but easy to fix.
Click the down arrow on the upper right hand corner of the plug and authorize.
Of course if you haven't updated the CA2A or are connected to the internet disregard this message.
And of course, this a$$umes you have a CA2A inserted in the project.
 
T

DeeringAmps
I did have an issue when I updated the CA2A on my Studio rig.
As it is not connected to the internet the "new" CA2A was indeed in DEMO mode.
"Cheesy" is right, but easy to fix.
Click the down arrow on the upper right hand corner of the plug and authorize.
Of course if you haven't updated the CA2A or are connected to the internet disregard this message.
And of course, this a$$umes you have a CA2A inserted in the project.
 
T

does ca-2a still "phone home" even after activation? that could explain a bunch of the other "ca-2a/vst scan failing" posts, if ca-2a is receiving an unexpected response from a server... just a thought

"does ca-2a still "phone home" even after activation?"
I don't think so, its fine on the Studio rig after I did the "offline" authorization.
T

I think I'm having the same problem... You can check my latest post further down labelled Sonar Recording Static. I get problems with my MOTU 828x, connected with USB or with Thunderbolt. My old Roland UA25-EX records fine with Sonar and the MOTU 828x records fine with Harrison MixBus, but the MOTU and Sonar seem to have major problems randomly from time to time. I am considering getting a RME interface and scrapping the MOTU to see if that runs more stable. I don't really want to scrap Sonar because I like it and have two albums in the go right now.
-Mark.

Hi Again,
 
Cakewalk support confirmed with me that the "DC offset" is a bug that is not fixed yet... make sure it is Un-Checked... that fixed my recording on input 1 and 2 to two separate tracks problem. 
 
Now my problem is random with just a simple mono track... I even encountered it on the most basic session with only one audio track routed straight to the master bus with no plugins at all.  A few restarts and re-trys got the same session and track to start recording fine again... no other changes were made to my session or the rest of the computer/hardware.  Very frustrating... I can't fully trust Sonar for tracking at the moment.
 
I have contacted MOTU, but because the problem doesn't happen on Harrsion MixBus (or Digital Performer, for that matter) they say it is Sonar's fault and they can't do anything about it.
 
I love Sonar Platinum, but I really don't want to have to record a bunch of tracks in one piece of software, and then import everything into Sonar for mixing, and then back again for possibly adding more tracks... and so on.
 
A couple of times I even had a blue screen crash where the error message included the MOTU driver as the problem (right in the middle of playing 2 different 64 bit soft synths). MOTU said it wasn't the driver's fault and had me contact Microsoft... Microsoft said it wasn't their problem and had me contact ASUS... the best ASUS could do was have me do a disk check... the disks were fine.  I am pretty fed up with contacting Tech support now.
 
I have setup my computer with all of the audio optimizations I could find out there and still can't reach 100% stability.  I don't mind the odd bug that I can work around, but recording tracks with random crackles is not acceptable.  The best I can come up with is that Sonar Platinum and MOTU drivers do not play nicely together.  I have even tried switching from ASIO to WDM... no improvement.  My Computer has been running rock solid otherwise and LatencyMon shows no problems when run for long periods... even with Sonar playing a big session with lots of plugins in the background - I don't run LatencyMon when recording or doing real work ;-)
 
My old Cakewalk UA25-EX interface did have the same "DC offset" problem as the MOTU, but once that was unchecked, the old UA25-EX has been totally solid for playback and recording... I can even get it to run at smaller buffer sizes than the MOTU without glitches, although I am running more than 16 outputs out of the MOTU and just stereo out of the UA25-EX.
 
So, for me, it might come to either replacing the MOTU, or replacing Sonar... don't really want to do either.  PLEASE HELP Cakewalk ;-)
 
Thanks,
Mark

Hi
 
I tried a few more things.
 
Playback glitching also happens when using X3, so definitely not something that Platinum introduced.
 

ryecatchermark
Cakewalk support confirmed with me that the "DC offset" is a bug that is not fixed yet... make sure it is Un-Checked... that fixed my recording on input 1 and 2 to two separate tracks problem. 

 
It turned out that I have never used this opton. I checked all my systems. Hence, this cannot be the reason for the record corruptions I had.
 

ryecatchermark 
Just an update... I think the use of MOTU's CueMix software (and possibly the MOTU Audio Console software as well) can trigger the crackles to start in Sonar, especially when opening the MOTU software AFTER Sonar is already open.  If you get your buffer settings and monitor mix set before opening Sonar, it might help.  When I contacted MOTU and asked why their 828x drivers for Windows haven't been updated for over a year, they said it's because they were working perfectly, but more recent updates of Windows 10 may be causing some problems???

Well, I had this suspicion myself but so far could not prove anything. The replacement of CueMix for the AVBs now runs in a web interface; it tried many things but could not deliberately reproduce any glitches ...
 
 
I started checking Sonar's audio options by toggling on/off each one to see whether any may have any impact. Interestingly, the playback seems to be better when unchecking "Use MMCSS" i.e. when checked I can get the project to produce an audible glitch at least once in the 3 min playback (of that one particular project!), if unchecked, I so far could not get that ... so I need to keep observing this and read up on MMCSS ... unfortunately, documentation is very sparse, like
 
Use MMCSS. This option gives real-time applications such as SONAR higher priority for resource scheduling (thread scheduling) under Windows 7 and later (leave it checked).
 
I love these non-descriptions that basically say "you wouldn't understand anyway, so just do it" ...
 
Mark, could you try whether this "MMCSS" makes any difference in your case?

Never mind... the same crackle/static problem came back, even with MMCSS un-checked.  This time I think it may be when using the MOTU with other applications and then coming back to Sonar.  I completely closed Sonar down, went on the web and played audio through YouTube, then closed the web browser and played a couple of songs on iTunes, then closed iTunes and went back to Sonar... as soon as I put Sonar in input mode to record, the static started.  I closed Sonar and re-opened it and the static was gone.  I tried re-creating the process again... I couldn't re-create the problem.  So I tried it again and this time disabled my wi-fi before opening Sonar... the problem came back.  I put wi-fi back on and opened and closed Sonar a couple of times and the problem went away again??? 
 
Problem not solved yet, although maybe I can work around it???
 
Mark

Hi Klaus,
 
Thanks for your input. I came across this information today when doing a web research and was surprised that the top audio interface manufacturer in the Sonar world gives these recommendations.  Unfortunately the info dates back to 2013 and I could not find anything similar more recent or directly related to our brand of interfaces. Neither do I know whether any of this MMCSS stuff was updated with win10 (as it dates back to the Vista days)
 
Anyway, the problem at hand is clearly an unusual issue in which software and driver developer will point the finger at the other one ... so any straw to cling onto is worthwhile (so pls keep the suggestions coming) ...
 
My first "MMCSS off" tests were OK, but the recording corruption issues are so infrequent that it will take days before I can tell whether they are gone or not ...
 

K.
 
Download the free HD-TUNE and do a surface check of the drive you use for audio with it.
if it's a SSD they fail in the most peculiar ways. Today's off the shelf mechanical disks are crap.
Defrag the drive you use for audio with the free defraggler.
 

Hi Bitman,
 
I am recording to a SSD drive... Samsung 950 Pro.  I use my spinner drive only to store samples.  Windows 10 is running off a separate SSD drive.  I heard that it is not safe to defrag a SSD drive???
 
Mark

ryecatchermark
I heard that it is not safe to defrag a SSD drive???

Not actually unsafe unless the disk is nearing the "wear out point". I have seen stuff claiming Windows does defrag an SSD once a month if system restore is switched on, but everything else I've seen says Windows doesn't automatically defrag an SSD.

Whether you can defrag an SSD "manually" from a command prompt I don't know. Never tried it and not going to. As SuperG says thanks to how an SSD (and the TRIM function) work fragmentation basically isn't an issue. If you think it might be worth a try copying all the data from the SSD to another drive, fast-formating the SSD then copying the data back should achieve pretty much the same thing without the lots of write operations defraggers use as they shuffle the bits around.

tlw

ryecatchermark
I heard that it is not safe to defrag a SSD drive???

Not actually unsafe unless the disk is nearing the "wear out point". I have seen stuff claiming Windows does defrag an SSD once a month if system restore is switched on, but everything else I've seen says Windows doesn't automatically defrag an SSD.

Whether you can defrag an SSD "manually" from a command prompt I don't know. Never tried it and not going to. As SuperG says thanks to how an SSD (and the TRIM function) work fragmentation basically isn't an issue. If you think it might be worth a try copying all the data from the SSD to another drive, fast-formating the SSD then copying the data back should achieve pretty much the same thing without the lots of write operations defraggers use as they shuffle the bits around.

Fragmentation is not a concept that is actually relevant in the realm of SSD's.
 
By comparison, in traditional spinning HDD's, the read/write heads have to seek out the positions of the data on the  physical spinning disk platter to do their thing.  Overall, this time adds up, and contributes to the physical limits on HDD data throughput.
 
For the SSD's, it takes nanoseconds to grab the fragments, so it becomes a non-issue

it's a strange world. I can repro that when I do NOT enable MMCSS in Sonar, playback does NOT glitch; if I turn it back on, I get at least one serious almost buzz like glitch per song playback (different place everytime, independent of buffers and other settings)