Suddenly getting PCI Master Overflow

[Spindrift]

Recently got a Xite 1-D, and it has been working perfectly for several weeks now, but today when starting, on first boot I got an error message "cannot communicate with Scope hardware" or something to that effect.
After rebooting everything seemed fine until I opened the DAW, then I get this:


The project loaded fine yesterday, and have not installed or updated anything since then. Tried with a minimal project with just IO, and same error as soon as I open Bitwig.

Anyone seen the same issue?
I'm guessing my hardware is faulty and have to start RMA

[garyb]

did you try another cable?
did you try reseating the PCIe card?

[Spindrift]

Yes, everything was seated properly, but reseated and now tried another cable, still no luck.

[Spindrift]

Tried reducing number of ASIO channels, and it seems that 8 in and 8 out works, but more than that and the message appears. In existing projects that worked fine yesterday I have 14 in 14 out.

[dante]

What sample rate ?

[Spindrift]

44.1k

[garyb]

PCI overload is usually a motherboard issue, but who knows? it could be a failed graphics card or a bad power supply or?
is the bios still intact?
how many masterverbs? they should only be limited by the dsp resources.

do you have another computer for testing?

[Spindrift]

Motherboard has been working perfectly with both PCI cards and Xite. No graphics card, and PSU is also only a few weeks old and reported voltages are good. Everything seems normal with the machine otherwise. Bios settings are intact.

Seems a bit flakey what I can load. Managed to load two masterverb with 8 ASIO in and out once, but trying again I get error on first. With 4 in and out I managed to load 10.

[Spindrift]

I guess I could try installing Scope on the gaming rig I have...but I really would rather not.
In your experience does it happen that a motherboard suddenly get reduced PCI bandwidth overnight?
But I guess based on your response that you have not heard about an Xite developing this issue suddenly either?

[dawman]

I got those messages and found out my CPU fan had quit working and the CPU was overheating.
Haven’t seen those messages since replacing it.

So many variables it’s to the point when everything works fine, I stop trying for the ultimate this or that.
Scope runs 100% stable once it gets situated right. The minute I try for more tighter timings and fiddling around I fuck shit up so I’m fixing something that’s working fine.

[Spindrift]

It is not the config or any changes made...like I said it worked yesterday, and has been no changes or updates since then.
I have been running stress tests and checking voltages and temps, and board seems to be working flawlessly.

[garyb]

yes, parts can die, it's just not an XITE kind of issue. you can always get an RMA, but i don't think that's a solution. checking with the other rig might tell.

cpu fan might be an issue, too. there is definitely something wrong because so few masterverbs work. it sounds like something died in the computer. do you have a graphics card to try?

[garyb]

it worked yesterday doesn't mean much. it doesn't work now.

[Spindrift]

Like I said, have been running stress tests, and cannot find any issue with the board and temps are fine. I cannot see how the CPU temp can be fine while running stress test for 15 minutes but still CPU fan would be the cause of communication error with scope right away on first boot.
Same goes with the internal graphics, I find it very unlikely it will pass a stress test without issues but fail specifically when running with Xite.

That it worked yesterday and then is dead after a reboot without any updates and changes was in response to dawman who started taking about making changes. I agree with him that when set up right with compatible hardware Scope just works. The only time in my over 20 years with PCI cards I have seen communication errors was when caps on one of my Pulsar cards went last year.

Xite is what I added last to the setup, the error is coming from Xite, and there is no indication of issues otherwise with the system, so I'm quite confident that it is not the mobo, PSU or CPU that needs replacing.

But if you never ever heard of an Xite failing with similar symptoms either, I guess I have to find time to get another machine to try it out on

[garyb]

15 minutes might tell you a lot. if you need an RMA, then you will know.

[Spindrift]

Yeah, if it would been 15 minutes I would have done it right away, but a bit more tricky with this computer, hence my hesitation.

But tested it now, and getting the same error on that computer, so it seems like an RMA unfortunately.

[Spindrift]

So this strange issue is not over yet it seems.

S|C decided to send me a unit for lending until RMA was handled, but when plugging in my unit back in it was magically working again.
Really unsure what to write it up to since I tested in on two different machines getting the same error, but I told them to cancel the delivery and that I would keep using the unit and see if the problem would reappear. Now I have been using it for a few months and it has been working great, but today I have the same issue again with PCI overflow as soon as I activate many ASIO channels.
I did notice an issue earlier where turning off/on one of my speakers caused scope to error. So I get the feeling the unit is not faulty and it might be related to a power issue.

It is an old building and could be an issue with wiring, and guess I have to get an electrician to check it out. But that could be a big project, so hoping I can find a solution before that. So current immediate plan it to order a Furman and see if that helps the situation.

[valis]

Ground noise or lift, bad isolation on ground....etc

[dante]

Can you go to another house - friend or relative - and test the rig there ?

[valis]

Look up 'star configuration' and octopus in relation to running studio power, and try doing the test with as few grounded components as possible (computers, usb devices, wall warts etc) with all power for xite/pc/etc coming from the same outlet. Also beware that your audio cabling can be a major factor here, especially if you're mixing balance & unbalanced cables. The shielding in that case is the primary concern, but unbalanced devices & USB stuff (soundcards etc) seem especially prone to issues with poorly grounded internals on gear.

Anyway first strip things down to the simplest usage cases u can and report any difference(s).