PT8HD & 6031 Errors

[tunetown]

Man, talk about grief. I've had this for about a month or so. Cannot get a stable system. I keep getting these 6031 error messages. As many of you know, I've been running N4 on an i7 system for ages. I've had my share of probs with UAD2 quad but these have pretty been much resolved.

I'm not the only one. Digi are looking into it however they are saying it's system specific and not related to PT. It's an interrupt problem. Apparently all those with this issue had stable and working rigs up until v7.4. I've gone through all their system tweaks, changed pci-e card slots, updated all drivers and software and still get this error.

It improved once i reduced my RTAS core from 8 to 3. seems I can use more RTAS plugs with less power :icon_rolleyes: but error still returns.

I'm certainly no newbie in these things but feel like it here. Oh, and Digi are also deleting posts at DUC on this issue.

Anyone here got any ideas on other things I can try.

Thanks guys.

[tunetown]

Also, as my system is unsupported, Vista64, It's not Digi's fault. I haven't installed W7 yet. I'm also running a 4 slot chassis.

Cheers

[LEX]

http://duc.digidesign.com/showthread.php?t=262212&highlight=6031

Biggest problem is RTAS. RTAS is a complete POS. Digi needs to rewrite it from the ground up.

Until then, I'd avoid it if you can.

Try the thread above.

LEX

[tunetown]

Thanks Lex. I've read that thread. When I have some time I'll do a full reinstall and see if that sorts it out. PIA.

Was RTAS working well on v7? All my 3rd party plugs and instruments require RTAS. The only TDM I have are the ones that came with PT8. If this is the problem then it's a real deal breaker for me.

I hope they get this sorted

Cheers

[tunetown]

These errors can also randomly occur without any RTAS plugs engaged. Just on playback and record. They go away when increasing buffer size.

Frustrating.

[LEX]

These errors can also randomly occur without any RTAS plugs engaged. Just on playback and record. They go away when increasing buffer size.

Frustrating.

Sounds like a system issue. I'd remove all the cards and start with just the PT card.
Find the slot that doesn't give the error.

Then, start adding other cards in one by one until you find the card that is give the problem.

That's the one you'll have to move around.

LEX

[nikki-k]

Hi!
Ah, PT HD on Windows. I busted my butt on coming up with a stable system back several years ago.. hence my thread on PT HD Win systems being a sticky. TDM is a VERY interesting monster. It simply hates anything interfering with activity on it's PCI buss.

What motherboard? How much and type of RAM?
Biggie: Have you ever run the exact system you have with XP x86?
If yes, did you experience any problems at any time?

PT can be very sensitive, so unfortunately it can take baby steps to resolve an issue properly.

[TAFKAT]

Hey Peter,

PT on any flavour of x64 is a minefield , as much as you are going to hate hearing this, XP will probably be the best bet.

Digi are miles behind the curve on anything to do with x64 driver wise. I managed to get a HD system up on XPx64 for a client, and the same tap-dance ensued with idiotic errors as you are reporting, on the same system with XPSP3 it was rock solid.

I have done numerous PTHD systems on Windows over the years, always on XP, and they have always been pretty solid.

You can also duel boot that system of yours into OSx86 if you are game. I am not going to give you any further detail of that online tho, drop me a PM or give me a call.

:009:

[tunetown]

Thanks guys. I've got a dual boot xp system. I was also getting these errors on xp. Although, since ive changed the pcie slots I haven't revisited the xp install. I'll give that a try. I have a PT session this thursday.

I just find it hard to imagine it being this hard. My system is rock solid with N4 and 2 RMEhdsp cards. The rme cards are on the pci slots of the gigabyte ex58 ud5 mobo. 6 gigs of ram.

I'll remove the uad2 quad and see if that helps. They are sharing irq's. I'll also revisit the xp install and update all drivers to see if thats stable. The thursday session is all audio so no need for massive memory from 64bit.

Vin, I'll give you a call Wed for a chat. I appreciate your offer.

I'll keep testing and also post my Sandra specs to the DUC and see what the digi support has to say. Unfortunately, based on what they've already said, it's my issue. Although many users who had stable systems are now experiencing this issue tells me that digi need to take this more seriously.

Cheers

[nikki-k]

Ahh...
Motherboards are VERY sensitive with PT HD cards. One of the early suggested boards by Digi.. one they approved.. would produce errors like mad under certain circumstances with more than 2 cards.

I would suggest...
Machine off, HD Core (all cards are PCIe, correct?) installed, I/O unit connected. Keep it simple initially.. only ht eone card, and one I/O.
Boot to XP.
Check the plug-ins directory. To begin, remove any 3rd party plugs.. only the plugs installed with a fresh, PT HD only install in there for now. You can place the others in the plug-ins (unused) directory (IIRC, that is the name).

Launch PT.
Create a new session on your audio drive, and then save.
Create a single audio track, and then hit play.
Any errors?
Once a single card and I/O are running *properly* under XP, then it is simply a matter of adding to that until it buggers up again. Do you have an HD1? HD2? HD3? I am assuming it is Accel, since you mentioned PCIe.

I did notice you have Apogee I/O's.. are you using those with the HD cards, or Digi I/O units?
I will check over on the DUC for your post if you want.. I can also mention to a friend to give it a once over as well. The Windows side is small, but GREAT group of users over there.

[tunetown]

Thanks Nikki,

Yep, all Apogee. 1x AD16x ad and 2x da16x's. No digi interfaces. These feed my API console. Works great with the native RME interfaces.

I will give your advice a try, however I will limp through my next session until I can give this more examination. I'll keep you posted.

Cheers

[tunetown]

Oh, and yes, HD1 core PCIe. No accel cards. Its in a Magma 4 card expansion chassis.

[TAFKAT]

Oh, and yes, HD1 core PCIe. No accel cards. Its in a Magma 4 card expansion chassis.

Hey Peter,

Magma chassis..., you are a glutton for punishment.. LOL

Before anything else, try plugging the card into one of the onboard PCIe slots , if the problem persists then we can safely rule out the Magma.

[Animus]

Man, talk about grief. I've had this for about a month or so. Cannot get a stable system. I keep getting these 6031 error messages. As many of you know, I've been running N4 on an i7 system for ages. I've had my share of probs with UAD2 quad but these have pretty been much resolved.

I'm not the only one. Digi are looking into it however they are saying it's system specific and not related to PT. It's an interrupt problem. Apparently all those with this issue had stable and working rigs up until v7.4. I've gone through all their system tweaks, changed pci-e card slots, updated all drivers and software and still get this error.

It improved once i reduced my RTAS core from 8 to 3. seems I can use more RTAS plugs with less power :icon_rolleyes: but error still returns.

I'm certainly no newbie in these things but feel like it here. Oh, and Digi are also deleting posts at DUC on this issue.

Anyone here got any ideas on other things I can try.

Thanks guys.

Guess you will be coming back to Nuendo eh? :D

[LEX]

Unfortunately on the PC side, Digi only qualifies certain computers and if you don't use them, you are SOL.

That is usually why MAC and PT go together.

LEX

[nikki-k]

Hmm... just verifying...
The HD Cire is PCIe? The PCIe Core cards are all Accel, while PCI HD Core are non-Accel. What Magma model? Unfortunately, Magma + non-Digi I/O + non-Digi DSP in same system + Chassis can be rough to tame in a "construct, connect, off to the races!" scenario. Unless there is a previous run by another user (or Digi) with the (close to) exact setup. "Tempermentaldesign" might have been a good name before Avidesign.. lol...

Trying to recall the exact version of the last known working PT HD that *allowed* 3rd party I/O. There have been releases where Apogee, Aurora, etc have had to *tweak* post release... most times they found a way to stay a step ahead tho ;)

[tunetown]

Hey Peter,

Magma chassis..., you are a glutton for punishment.. LOL

Before anything else, try plugging the card into one of the onboard PCIe slots , if the problem persists then we can safely rule out the Magma.
Yeh, probably. It's this model http://www.magma.com/store/expressbox4.html . Not endorsed by Digi however Magma specifically states it's use is for PT core cards. It's Pcie to Pcie. Never had a magma problem before. I bought this because the PCIe core card doesn't fit in my box. I can also expand to HD4 when required.

I'll remove the UAD card today and give the xp install a run. As I'm only tracking I should be able to complete my Thursday session without grief. The session is being sent back to LA for mixing so no need for plugs.

Thanks for all your advice guys. I'll get to the bottom of this. OR, maybe Digi will sort it out. Something in their code has changed. PT7.4 was rock solid on some systems that are now also having this problem.

Cheers

[TAFKAT]

Yeh, probably. It's this model http://www.magma.com/store/expressbox4.html . Not endorsed by Digi however Magma specifically states it's use is for PT core cards. It's Pcie to Pcie. Never had a magma problem before. I bought this because the PCIe core card doesn't fit in my box. I can also expand to HD4 when required.

Hey Peter,

No doubt spec wise the Magma should be fine, just trying to dot the i's.

I have seen my share of issues with PTHD and Magma before, I remember one time watching a local so called PT/MAC guru loose his mind for a week trying to settle a MBP and Magma/PTHD live rig , I stepped in while he was at lunch, swapped the cable between the PCMCIA and Chassis for another cable we had from a S/H Chassis lying around , and bling... resolved.

:rotfl:

[tunetown]

Ok, for you gluttons who want to navigate this minefield here are some results of a days testing and changes.

I narrowed this down to the UAD card. Without it the XP install ran flawless and I could load up almost limitless RTAS plugs. Insert the UAD2 Quad and load a couple of Wrapped UAD plugs and boom, 6031 error. This was with the UAD in the no# 2 16x slot. Sharing IRQ with the HD card in Magma on PCIx8-1 slot. This on the Gigabyte EX-58 UD5. Video card on 16x slot 1 is ATI Radeon HD4670.

Changed the UAD to PCIx4_1 slot and I gained a heap of stability and and allowed more UAD instances. Still not great. Still sharing IRQ with HD but much smoother. This config seemed best out of all the changes I made. I did try others.

After other forum searches, I lowered the Host processors in PT from 8 to 7 and upped the CPU usage to 95%. Bingo, I managed to load the UAD card to 93% and load other RTAS as well. Played at 128 samples but not stable. At 256 it worked fine. Not great but certainly usable. I'm sure at higher Buffer sizes it would be sweet.

I haven't revisited the Vista64 install yet but will out of interest.

I'll get a couple of session under my belt before I claim full success. I'll keep you posted.

Cheers for your help.

[LEX]

Ok, for you gluttons who want to navigate this minefield here are some results of a days testing and changes.

I narrowed this down to the UAD card. Without it the XP install ran flawless and I could load up almost limitless RTAS plugs. Insert the UAD2 Quad and load a couple of Wrapped UAD plugs and boom, 6031 error. This was with the UAD in the no# 2 16x slot. Sharing IRQ with the HD card in Magma on PCIx8-1 slot. This on the Gigabyte EX-58 UD5. Video card on 16x slot 1 is ATI Radeon HD4670.

Changed the UAD to PCIx4_1 slot and I gained a heap of stability and and allowed more UAD instances. Still not great. Still sharing IRQ with HD but much smoother. This config seemed best out of all the changes I made. I did try others.

After other forum searches, I lowered the Host processors in PT from 8 to 7 and upped the CPU usage to 95%. Bingo, I managed to load the UAD card to 93% and load other RTAS as well. Played at 128 samples but not stable. At 256 it worked fine. Not great but certainly usable. I'm sure at higher Buffer sizes it would be sweet.

I haven't revisited the Vista64 install yet but will out of interest.

I'll get a couple of session under my belt before I claim full success. I'll keep you posted.

Cheers for your help.

Sounds like a wrapper issue. Have you contacted UAD?

At least you are able to use over 90 percent of your available DSP.

LEX

[nikki-k]

Fun, huh? Ah, the old stomping grounds..

From what I have read, UAD cards can be.. interesting. Granted, many seem to be happily, erm, plugging along ;) . I would venture a (unfortunate) guess that it may come down to choosing to have the UAD or stability. What happens if you place the UAD in the chassis?

PT and the almighty RTAS. Wish I could say (so much) more. Only hint I can state is... go back to that interesting era of PT5, and then jump to, say, 6.3 I think? There are some interesting things you can try that will demonstrate an impending doom. HTDM pretty much disappeared quickly, swept not under the rug but out of the country completely. But, a little digging can reveal interesting caskets of goodies. Pro Tools was built on a VERY interesting foundation, and, unfortunately, repouring a new foundation of your own ingredients for the cement is financial suicide. And then some. So, reworking framework is the next choice, albeit still expensive. Hence, RTAS today is buggered, mostly due to the capabilities of natives systems, both in hardware and software. Moore's Law has simply outpaced Digi's capabilities (IMO! And possibly Kurzweil's Law stepping in as well..lol). Well, that outpacing truly first showed teeth about 7 years ago.. maybe more.

Sorry for the length, but point is... RTAS is buggered. A TDM system is (currently) best left TDM. I stated a minimum of HD3 for simple use, HD4 for 24/96k, and greater than HD4 for 24/06k 5.1+. If RTAS must be used (likely improbable to avoid with HD1 or HD2), great care must be taken, almost more so than with *purely* native solutions. Processor usage is first.. and currently, RTAS will require at least one core of a more than 2 core system to "absorb" spiking. With more than 4 true cores (ie, dual quad core cpu; not a "virtual" core set, such as is with the i7), the likelihood of sacrificing further cores for this becomes very probable. I have spoken with and seen users complain of having to set a dual quad core system to 6 or even 5(!!!) core usage in PT when RTAS use gets up there, especially with demanding VI's.

I would try setting to 5 cores and see how it behaves. If a 4 core setting will not allow for stability with the UAD, then it returns to my second sentence: stability, or UAD. (for now?) Both UAD and TDM seem to be... picky. ;)

[tunetown]

You'd think so. I did some searches on their forum and didn't find and related issues. Strangely though, when reducing the number of RTAS processors this problem improves significantly. I feel there is more a multiprocessor issue than a Wrapper one. Turning hyperthreading off i bios made no difference.

There's something going on here that's out of my control. When I've got some more time I will load up PT7.4 and see if the problem is still there. There are many that say these errors became apparent after 8.01cs-3. The initial release seemed stable.

I'll keep testing and post my findings. I will shoot UAD an email to ask if they are aware of my issues. I will also post my findings at the DUC once I've completed more tests.

Cheers

[nikki-k]

The UAD stuff: is it RTAS native now, or wrapped? I know users had to use the FXpansion wrapper before, but I could have sworn UAD states RTAS as a plug-in type.. no? Or are they simply wrapping it themselves so the user does not have to actually buy the wrapper?

[tunetown]

Hi Nikki, Yeh, UAD are wrapping it for us. It looks like they bought the rights to use the fxpansion wrapper inside their plugin suite. I'm happy about that.

RTAS does seem like a pig. I have not tried the UAD in the chassis. Maybe I will when time allows. I wanted to get them as far apart as possible although they still are sharing irq's. The setup I have now seems stable. I can fully load the UAD Quad without errors. I really need to do a few sessions with VSTi's and plugs to see how they all work together. I suspect this issue is far from resolved.


Cheers