LinuxMusicians

When I said "the signal" I meant the visually evident drops to 0 (zero) that were easy to see in the snapshots of your .wav file. This, in my opinion is the source of the "static" that is heard and extends over the entire time of the piece of music.

Locate the source of the "static" and eliminate it.

That 'signal' I refered to is what you are hearing as awful noise/intereference/clicks/whatever...

If I were sitting in front of your machine, I could probably track this down. But I am a little stumped.

Do you have any other JACK apps you could patch into the insert to see if the static is there?

What about if you take the insert, and just patch the OUTS to the INS to see what happens?

Anyone else have any ideas? I'm running out!

I have an idea.

How about recording quiescence?

Silence, but with a mic on and at least one guitar or something?

Amplify, then analyze.

??

If you can see the noise you might be able to identify it.

Ricardus wrote:If I were sitting in front of your machine, I could probably track this down. But I am a little stumped.

Do you have any other JACK apps you could patch into the insert to see if the static is there?

What about if you take the insert, and just patch the OUTS to the INS to see what happens?

Anyone else have any ideas? I'm running out!

I use hydrogen drum app in their with no static. I attached the current routing. I do recall an issue when I first started messing with Ardour where I was getting static turned out to be a routing problem. I am still semi routing stupid, but hope to change this within the next few days once and for all. As far as patching it any other way, you have to step me thru it. See below for how I have it and if it is the issue:

The insert is correct, but we don't see all the audio routing. I don't think you have this wrong but just in case, show a screenshot of patchage with all the audio (blue) connections.

And what if you increase the periods per buffer in jack. You don't need very low latency when editing/mixing/mastering.

Attached!

Try disconnecting SYSTEM CAPTURE 1/2 from JAMIN TEST IN 1/2.

Here is the correct layout with everything connected as you guys set me up. This is different from the first one I showed you.

I tried removing the System Capture and everything slows down and skips like really bad like with static. I actually then had to exit out of Jack even and reopen it and Ardour to get rid of that as I switched to another session it was still there.

Check out this new attachment.

Why are you running two instances of Ardour when you are trying to isolate and locate the source of irritating noise?

You said:

...I tried removing the System Capture and everything slows down and skips like really bad like with static....

"Removing the System Capture" is a simple rerouting and should have no such effect as "everything slows down and skips like really bad like with static."

Using Catia (or even top, KSysGuard(System Monitor), or something similar), what kind of DSP load/CPU usage are you seeing when you produce the static?

I attached the system monitor file under load of a JAMin piped song. I beleive it reads about 30% average.

Is that the CPU usage when you actually create the "static?"

Or is that playing back a file recorded with the "static" already in it?

Once the noise is in a file, it's there. Any tests or changes you make after the noise is in a file will be unlikely to help uncover the original source of the noise. Find the source of the noise.

To me, your original "static"-bearing audio looks like dropouts due to buffer problems. Are you seeing any xruns while you create the noise?

What happened when you increased the buffer space as Pablo suggested above:

...And what if you increase the periods per buffer in jack. You don't need very low latency when editing/mixing/mastering....

luster wrote:Why are you running two instances of Ardour when you are trying to isolate and locate the source of irritating noise?

I don't think that's really two instances of ardour.

luster wrote: Using Catia (or even top, KSysGuard(System Monitor), or something similar), what kind of DSP load/CPU usage are you seeing when you produce the static?

Also, what does Ardour report as DSP useage when you're just playing the audio?

What is "Monsanto"?

Ricardus wrote:

luster wrote:Why are you running two instances of Ardour when you are trying to isolate and locate the source of irritating noise?

I don't think that's really two instances of ardour.

Yes, it is only one instance of Ardour. I was in a rush and didn't notice...

I did take some time to look more closely at the noise. It is very regular throughout the piece, with the drop-outs occurring nearly exactly 260 samples apart. Are all your sample rates and bit depths the same?

Additionally, all the noise I inspected (many, many drop-outs, throughout the piece) is also regular in shape: exactly 4 samples wide, and dropping up or down to zero. It could be 3 samples wide--or even 5 samples wide--I am unsure about how to specify/figure the width. But each of the many dropouts I inspected is exactly as wide as this: