|
|
| Re: agSetClip: can not nest agSetClip() in win32 ..? [message #105143 is a reply to message #105142] |
Fri, 02 April 2010 10:10   |
|
|
Happy Easter to you too!
Thanks for the heads-up - as it happens that error message is one that I've seen within the last few months myself.
I can't recall why it cropped up, or what I did to beat it - but I know I don't get it any more. It was flushed into the light by something we did over the last year, so reducing variables will help us nail it.
If you're using a "skinned" PARIS app, eliminate that first by booting your original "unskinned" EXE instead. Then revert to a plain "vanilla" version of the PARIS.dat. If that doesn't clear things up we'll run it by Mike to check whether a bug was introduced somewhere in the revisions to the drivers (or alternately whether the new drivers now cause PARIS to give a new error message for an old bug).
"... being bitter is like swallowing poison and waiting for the other guy to die..." - anon
|
|
|
|
| Re: agSetClip: can not nest agSetClip() in win32 ..? [message #105147 is a reply to message #105143] |
Sun, 04 April 2010 07:46 |
mani1147
Messages: 130
Registered: May 2009
Location: NB Canada
|
Senior Member |
|
|
Hi Kerry, so you have seen this error as well. Its a weird one, it only starts happening when I start using automation. I lay down automation across my 3 card mixs using 3 c16s, then go back and start editing fader moves via mouse, typical tweeks, and bam..sure as hell, it starts happening. I have noticed that if I reboot the computer rather than just rebooting Paris, I seem to be able to run longer. Paris is not skinned, I haven't ventured down that path YET but I must say, it looks pretty sweet.
Things I've tried so far with no luck:
-Pulling out 1 stick of Ram (ddr3 2 gig stick) and just running on one stick.
-Un-pluging my 3rd c16
-disable virtual desktop
I wonder if a bad VST plugin could do this?
Thanks
Robert
|
|
|
|
| Re: agSetClip: can not nest agSetClip() in win32 ..? [message #105149 is a reply to message #105147] |
Sun, 04 April 2010 15:46 |
|
|
I got Mike's OK to talk about this one in a bit more detail. We know what this is, and it's being caused by an issue inside the application itself where we can't get at it, so it's not fixable.
So of course Mike's gone ahead and come up with a fix nonetheless.
Background from my layman's perspective: there's always been a window of time in the app where a Windows screen re-draw can "collide" with PARIS automation data and get "smooshed", resulting in this "agSetClip" error. Even old single-CPU computers were theoretically vulnerable to it, there's always been a miniscule chance of getting this error whenever you used automation, no matter what kind of processor you used. But on single core CPUs the window of time in which it could happen was so miniscule and the chance of hitting it by chance sufficiently low that we don't seem to have any reports of it having been accidentally stumbled upon before.
Then we moved to multiple processors, and the chance of encountering this "window" increased - not just by a simple multiple of probability, but by an enormously greater factor. So now a heavy automation session has a better chance of triggering it than not, and you're actually the third user who's reported it now. In fact Mike can replicate it there at his place with a heavy automation session.
Mike can't *cure* this without the app's source code - so instead he's worked around it with a soon-to-be-released update of the PSCL that drops your chance of encountering it back down to the original numbers (ie very unlikely), as well as containing a number of other tweaks, enhancements and updates. It's essentially finished and ready to go now, and Mike's just figuring out the best way to roll this update out to existing users via email (a lot of peoples' ISPs filter out those sort of attachments even if zipped) - contact him directly in the mean time and he'll send you a copy.
"... being bitter is like swallowing poison and waiting for the other guy to die..." - anon
[Updated on: Sun, 04 April 2010 22:01] Report message to a moderator |
|
|
|
|
|
|
|
| Re: agSetClip: can not nest agSetClip() in win32 ..? [message #105156 is a reply to message #105153] |
Mon, 05 April 2010 07:27 |
drfrankencopter
Messages: 137
Registered: July 2009
|
Senior Member |
|
|
Can we rally together to push ID to release the code? Would this take a petition, or some $, or both?
Maybe not make it open-source (I can see ID's resistance to that), but subject to a non-disclosure agreement, and available only to a few people who could ultimately tweak it up, and release new executables, or patches.
I really think that the VST implementation in Paris is broken, and could probably be fixed with a re-write of the Paris_fxVst.dll file. But, it's hopeless without the source.
I've emailed ID about obtaining the source, and further information regarding some the 'secret' config file parameters, but it's like talking to a wall and I've received no signs of life from their end.
Perhaps with a united effort we'd have more 'pull' with them.
Cheers
Kris
|
|
|
|
|
|
| Re: agSetClip: can not nest agSetClip() in win32 ..? [message #105159 is a reply to message #105158] |
Mon, 05 April 2010 19:12 |
mani1147
Messages: 130
Registered: May 2009
Location: NB Canada
|
Senior Member |
|
|
All I can say is WOW...!!! My 3 card Paris rig is an entirely new beast with this updated code from Mike. I mixed all afternoon, not one error where 2 days ago, I would have rebooted at least 10 times. Even the dreaded loop error didn't show up and I was pushing my luck, driving paris hard and it stayed rock solid....amazing work.
Paris + Mike = Kickass
Thanks again Mike and Kerry
Rob_A
[Updated on: Mon, 05 April 2010 19:18] Report message to a moderator |
|
|
|
The PARIS Forums
Search
Help
Members
Register
Login
Home
