distortion in SAC only with SAW recording

[gvieth]

I have recently installed SAWStudio on our SAC machine and every time I hit record there is distortion heard in SAC.
I have confirmed that all the resolution, sample rate etc. is the same in both SAC & SAW.
I'm sure it's something simple that I'm missing I just can't figure out what it is.
Any ideas?

Thank you,

Gary Vieth
Odessa, TX

[Guitarkeys.com]

Same problem not long ago.

Change you tap point in SAW... problem solved.

[gvieth]

Is this the input chan play tap at SRC/PRE-PATCH/PST-FDR option?
I thought that was only for sending back to SAC.
There isn't anything going to SAC while in record mode is there?

Thanks,

Gary

[gvieth]

I have changed the tap points and no improvement.

I have another system with SAC & SAWStudio and it was pretty much plug and play. Works great. Just can't figure this one out.

[Mattseymour]

Is sac dropping buffers?

It sounds like maybe the disk you're recording to is causing some conflict with the sound card.

[gvieth]

Yes it drops output buffers.
A lot of them.
Using RME Raydat cards.

[gvieth]

Oh, I am trying to record to a second hard drive (not the primary)

[Butch Bos]

I have had NO problems at all recording and I am using Raydats
What Buffer size? I run at 2 64 most of the time
I have had 6 different systems over the last 4 years using SAC and SAW lite and SAW full all using 1 or 2 Raydats non with any problems recording
If you are using 2 raydats are they sync'd correctly ?

Butch

[Trackzilla]

On my old rig, there would be a burst of dropped buffers and accompanying distortion right when I would first hit record. It would not recur if I stopped & restarted. It was due to the extra overhead from the files being initialized. Never bothered to figure out what wasn't keeping up, I just learned to to set-up everything for recording during soundcheck, hit & then stop record, cycle SAC to clear the dropped buffer count, and then know I was ready to record on a whim without further issues.

[Mattseymour]

Check for irq sharing between the raydat card and the interface your second disk is connected to.

[Bob L]

With multiple cards make sure they are synced correctly using the internal cable and make sure the clocks are set in the proper order... the master card using internal or master clock and the second card set to sync from the cable connection.

Bob L

[gvieth]

Some info
Card 1 clocked internal
Sample rate 48000
Buffer 64. WMD 18
Card 2 clocked sync in
Sample rate 48000
Buffer 64. WMD 18

SAC
Sample 48000
Resolution 24 bits no dither

Audio set up asio
Preload buffer 2 size 64
Audio driver Asio protocol. (32 bit int)

SAW
Sample 48000
Resolution 24 bits dither off

Audio set up asio
Preload buffer 6 Size 64
Out. Buffer 4.

I actually was set to recording to main hard drive
That's not what I want to do but is set that way right now

[RBIngraham]

You are using the SAW-SAC link right? If you are then the audio driver settings in SAW should make not make any difference. Only the sample rate should have to match I think.

[gvieth]

Yes I am using SAC-link.
SAC works great. No problems.
Everything seems right in SAWStudio but the second I hit record SAC starts clicking. Even when I solo in to a channel in SAC.
I'm sure it's probably something I have over looked, just don't know what it would be.

[Mattseymour]

Did you check the irq sharing as suggested?

Another thing to try is creating a load of disk I/o (copy some big files from one disk to the other) with just sac running. Does this cause the same problem?

Try recording on saw to an external USB disk. What happens then?

[gvieth]

I am not sure what the irq is.
But I found that it is actually recording to main hard drive.
I'll get a flash drive and try.
Thank you

[RBIngraham]

What are the computer specs? You are not just running out of CPU power or the like are you? How many Inputs and Outputs do you have active in SAC and/or SAW when this happens? If you have a lot of inputs going, what happens if you just try one single input?

[Mattseymour]

Actually a flash drive is a bad idea. It will almost certainly be too slow and that will cause problems itself. A USB2 external HD will handle about 24 tracks of audio usually. Many USB sticks will struggle with 2.

Sac/saw needs the data to flow freely for everything to work. If you have a system that's too slow, disks which are too slow or irq conflicts, basically anything that stops the data flowing freely, then you'll get dropped buffers as the system fails to keep up with realtime.

The suggestions made by Richard and me will hopefully help rule out some of these things.

You need to know what causes the system to struggle and that requires logically going through these ideas one at a time so you can track down the issue.

To find irqs, go to control panel, system, device manager then view by connection type. Expand the irq list and you'll see them all. What irq do your raydats use, what else (if anything) is on the same irq and what is you HD interface using, and sharing?

[Bob L]

Just try placing SAWStudio in REC/RDY mode... this will pass the data but not write to the disk... if everything is ok there, then it would seem something is loading down the system when the disk is engaged... check your device manager and make sure your disk is not bogging down... if it is an older ide drive, make sure it is not stuck in PIO mode... make sure it gets up into DMA mode.

Bob L

[dbarrow]

A USB2 external HD will handle about 24 tracks of audio usually.

Is that correct, Bob?

[Bob L]

I have done many sessions directly to usb external harddrives... the track limit is dependent on many things that can affect the overall performance of the harddrive and entire system... each one is different.

Check the harddrive performance using my HDSpeedTest program found in the misc downloads of the SAWStudio site... see what kind of results you get.

Bob L

[gvieth]

The problem I am experiencing happens in rec/rdy mode also.

Specs,

Windows 7 Professional Service Pack 1
Intel i5-3570K 3.4 GHz
4 GB RAM
32-bit

I don't know what I am looking at on the IRQ stuff so I have tried to attach an image.
Thanks,

Gary

[Mattseymour]

All this is assuming half decent modern hardware but yes 24 channels is fine with the average USB 2 HD.

So I've assumed 44.1k sample rate and 24-bit which is a bitrate of 1058.4kbps
24 channels would be: 25401.6kbps which is a little over 3MB/s
Given even a cheapo USB2 sata chipset will manage 25MB/s you're fine with this track count.

As Bob says, it isn't all that simple and interface overhead, disk performance and a whole host of things come into play that means this doesn't necessarily scale up to the theoretical limit of the USB2 interface or even the practical limit of the chipset you're using.
However, if your system can't handle 24 channels to a USB 2 external HD (not any flash key) there's a problem. The last big thing I recorded had about 38 channels all being tracked to an external drive.

[Mattseymour]

The problem I am experiencing happens in rec/rdy mode also.

Specs,

Windows 7 Professional Service Pack 1
Intel i5-3570K 3.4 GHz
4 GB RAM
32-bit

I don't know what I am looking at on the IRQ stuff so I have tried to attach an image.
Thanks,

Gary

My (limited) understanding is that in theory all those things on IRQ 16 and 17 (which includes your raydats) are on virtual interrupts that are assigned by the bios. Because they aren't physical interrupt lines again, in theory, everything should play nicely.

But in my experience that doesn't always work. One of my systems puts a USB interface on the same IRQ number as the MOTU 424 card and I cannot use that interface. Disable it and all is well.

Personally, I'd experiment with using just one raydat at a time. Remove the clock slave card and see what happens with just the one. Then move that one into the slow previously occupied by the slave and see what happens.

[gvieth]

Didn't get a chance to test with one raydat etc but did find out that it does not have a problem with up to 16 channels. Anything over and it drops buffers.

[gvieth]

Alex Stroke recommended selecting "force single CPU" in SAW and that seems to have fixed it. Embarrassed I didn't try that sooner.
I won't know for sure until tonight but I will post the results.

[joeyb]

don't know if you fixed it. But I had the same thing today , not sure what your doing but here's what happened. I use SAW studio at home and use SAC/SAWstudio live PA (RME)on a notebook (just works super) came into school to record teachers singing (overdub) to a wav track in SAW studio set up SEC for mics used link tried =distortion . long stoy short went into SEC and changed the input selection to SS1:01 (play back wave is on track1 ) distortion gone works great

[gvieth]

Seems like the force single CPU fixed the problem.
Gonna try saving to the second hard drive Sunday

[gvieth]

Well, I guess I may have spoke too soon.
No problems that one time, have not been able to get it to work again.
I'll try knocking down the channel count back down to 16 and see if that helps.

[MikeDee]

Recheck your "Force Single CPU" setting...if you did not save your Preferences previously, you most likely are no longer in that mode...therefore, your issue will have resurfaced.

HTH,

[gvieth]

That was the first thing I checked.
Sunday night I tried again and got an oscillation like an output in SAW was fed back to an input in SAC.
I guess I just don't have SAW setup right.
When I checked all the resolution etc and did the import from SAC and arm last import I just assumed everything would work.
(It did on my other rig)
Just haven't had much time to work with it.

[gvieth]

[/QUOTE]long stoy short went into SEC and changed the input selection to SS1:01 (play back wave is on track1 ) distortion gone works great[/QUOTE]

What do you mean by SEC?

[PhaseShifter]

What do you mean by SEC?

I think he means SAC.

[operationwhat]

long stoy short went into SEC and changed the input selection to SS1:01 (play back wave is on track1 ) distortion gone works great[/QUOTE]

What do you mean by SEC?[/QUOTE]

What is your CPU usage level? Be sure it's not over 50%...