yep, make that bolded in first post too.
At least Brent's case is closed.
He has 32-bit XP host.

Anybody got an incompatibility wizard I can use to detect a 32-bit operating system from a 32-bit application (the installer itself), to say "Hi. I'm a 64-bit application really"?

Don't use the sleep command with SoG/Lunatics 0.45, at present it causes wu's to slow down and report estimated completion times in the thousands of days, then the status will say 'Waiting to run', while other wu's get worked on, I've had nothing but trouble with this, it doesn't matter if the wu is a guppi or not, I've aborted at least 5 wu's that would just sit there and do nothing.

. . May I ask how much memory you have?

. . I am running 0.45 Beta with -use_sleep turned on on this rig. I have not experienced any such problems but it does cause the GPU utilisation to become sporadic and drop quite low. My average was about 30%. For non Guppi tasks this was overcome by running multiple instances which behaved nicely and well mannered. With three WUs running the GPU is now achieving almost 100% usage and runtimes are only about 75% greater (20 mins running singly, 36 running triples).

. . Sadly Guppis are as cantankerous as ever, they take about 50 to 52 mins with sleep on, but instead of working concurrently as triples they seem to want to run as singles playing tag with the GPU and work load did not budge, staying at about 30%, and runtimes went from 52 mins to 3 hours and 20 mins. Killed my hopes and expectations that they would take about 1 hour 30 mins or less. It seems there is no way to run Guppis on an Nvidia card without wasting large amounts of resources.

Then I think I'd rather not do Guppi wu's on My gpu, I'd rather abort any that I find on My 580 gpu. I think I might go back to Lunatics 0.44.

. . Here, here!

. . Guppis are fine as CPU tasks but on Nvidia cards they are total brats.

I'd rather abort any that I find on My 580 gpu.

Better not to crunch at all than to abort.

. . So what is it about aborting WUs that gets so many people upset. Does it release the Kraken or something?

. . Better to abort a WU than to have it stuff up a machine and prevent it from working on other tasks.

I'd rather abort any that I find on My 580 gpu.

Better not to crunch at all than to abort.

Well I aborted all the cuda guppi's/vlar's, it's done...

So sue Me.

The only effect of the Guppies on my systems is they take forever to finish & don't pay much credit.

So it would be better just to block all those that abort from using the project. They have the option of not using the GPU in their settings.


EDIT-
Just in case you didn't get around to reading the initial post in this thread, i'll repost the first couple of lines.

"As the title says. Beta test only, try at your own risk.

I think we should dig out the old Advanced Users only tag for this one"

. . Sorry Grant,

. . Unlike yourself that is exactly what frustrates and annoys many other contributors. They do take forever to complete and so prevent the host machine from doing more productive tasks.

. . As you fully know disabling GPU processing on a host completely would remove a valuable resource from the project, and there is NO option to allow GPU crunching on other tasks while excluding Guppis.

. . While you are correct about the caveat on the Lunatics Beta release the problem with Guppis is not a Lunatics issue in any way, the installer works fine, it is an issue probably more suited to the SoG thread.

The only effect of the Guppies on my systems is they take forever to finish & don't pay much credit.

So it would be better just to block all those that abort from using the project. They have the option of not using the GPU in their settings.

Fine then more people will leave, vlars it was agreed would not be on the gpu again We were told, promised really, sounds like that was an empty(hollow) promise...

. . I don't think people need to leave. But it would be nice if we could find a Guppi tamer to get these critters to behave and play well with Nvidia cards.

I don't suffer from lag here, I have an i7 3820 and the coolest 580(temperature wise) gpu, a PNY LC 580.

I'm not a programmer, I'm a hardware guy, but it would be nice if running vlars could be optimized.

I'm also going to apologize for what I said when I was having hardware problems, I have severe anxiety problems, the pc hardware I could replace and I've done so, it just required time and money, the severe anxiety problems I can do nothing about.

. . Exactly, it just requires time and money, resources that contributors donate to the project and which should be respected.

Anybody got an incompatibility wizard I can use to detect a 32-bit operating system from a 32-bit application (the installer itself), to say "Hi. I'm a 64-bit application really"?

Something simpler than https://msdn.microsoft.com/en-us/library/windows/desktop/ms684139(v=vs.85).aspx?

Fine then more people will leave, vlars it was agreed would not be on the gpu again We were told, promised really, sounds like that was an empty(hollow) promise...

That was MB data & that is still the case. There are still no MB VLARs going to GPUs.

. . It seems strange to me that such a commitment is being honoured when Guppis which are at the very least as demanding as MB VLARs are being distributed as GPU tasks. Anyone else see a pradox there?

I don't suffer from lag here, I have an i7 3820 and the coolest 580(temperature wise) gpu, a PNY LC 580.

I'm not a programmer, I'm a hardware guy, but it would be nice if running vlars could be optimized.

If you run the stock application then it will (eventually) select the application that gives the best throughput.


If you choose to run something that is only at the Beta stage of release- not yet close to general release- then it's up to you to select the application & then tweak it to give the best results. The information you provide back will allow those developing the applications to make the release version better. Unfortunately the documentation on what each variable is & does & what's best & not best doesn't really exist- this is still a Beta release remember.
But there are people around that can help.

. . Sadly that is true of most of the documentation available.

I run Lunatics 0.44...

I tried 0.45, opencl and Me aren't compatible, since most of opencl seems to be cpu work, I opencl you'd think would work on the gpu like cuda does.

. . While I share your allergy to Guppis, unless you are running a very, very old graphics card it should support OpenCL language. I am probably not the first to respond to this but most Nvidia cards less than 5 years old support OpenCL 1.2, and the Intel integrated GPUs support OpenCL 2.0 (at least my HD530 does).

I run Lunatics 0.44...

I tried 0.45, opencl and Me aren't compatible, since most of opencl seems to be cpu work, I opencl you'd think would work on the gpu like cuda does.

. . While I share your allergy to Guppis, unless you are running a very very old graphics card it should support OpenCL language. I am probably not the first to respond to this but most Nvidia cards less than 5 years old support OpenCL 1.2, and the Intel integrated GPUs support OpenCL 2.0 (at least my HD530 does).

The PNY LC GTX 580 supports OpenCL 1.1, runs like a slightly run away freight train, yet very cool in temps.

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

Anybody got an incompatibility wizard I can use to detect a 32-bit operating system from a 32-bit application (the installer itself), to say "Hi. I'm a 64-bit application really"?

I'm glad we have the experts around here to identify the error code ID10T :)

Thanks for looking into it everyone.

There are a few things mixed up here.
Strong opinions about what guppi runs well on or not etc.

Maybe some facts should be collected first, instead of blaming one thing or the other or demanding one thing or the other.

I run Lunatics 0.44...

I tried 0.45, opencl and Me aren't compatible, since most of opencl seems to be cpu work, I opencl you'd think would work on the gpu like cuda does.

. . While I share your allergy to Guppis, unless you are running a very very old graphics card it should support OpenCL language. I am probably not the first to respond to this but most Nvidia cards less than 5 years old support OpenCL 1.2, and the Intel integrated GPUs support OpenCL 2.0 (at least my HD530 does).

The PNY LC GTX 580 supports OpenCL 1.1, runs like a slightly run away freight train, yet very cool in temps.

. . :)

Fine then more people will leave, vlars it was agreed would not be on the gpu again We were told, promised really, sounds like that was an empty(hollow) promise...

That was MB data & that is still the case. There are still no MB VLARs going to GPUs.

. . It seems strange to me that such a commitment is being honoured when Guppis which are at the very least as demanding as MB VLARs are being distributed as GPU tasks. Anyone else see a pradox there?

No.
MB VLARs caused significant system usability issues. Guppie VLARs don't. Since they don't, why wouldn't they be run on GPUs?

---

Grant
Darwin NT

There are a few things mixed up here.
Strong opinions about what guppi runs well on or not etc.

Maybe some facts should be collected first, instead of blaming one thing or the other or demanding one thing or the other.

. . Well Guppis definitely run well on CPUs, on my core i5-6400 they actually take slightly less time then normal MB tasks.

. . And they definitely have a hate/hate relationship with Nvidia GPUs. In fact they are total brats.

. . Anything else? :)

Fine then more people will leave, vlars it was agreed would not be on the gpu again We were told, promised really, sounds like that was an empty(hollow) promise...

That was MB data & that is still the case. There are still no MB VLARs going to GPUs.

. . It seems strange to me that such a commitment is being honoured when Guppis which are at the very least as demanding as MB VLARs are being distributed as GPU tasks. Anyone else see a pradox there?

No.
MB VLARs caused significant system usability issues. Guppie VLARs don't. Since they don't, why wouldn't they be run on GPUs?

. . Now that is a scary thought, so MB VLARs on GPUs actually cause more system disruption than Guppis do?

. . That could cause me nightmares.

. . Now that is a scary thought, so MB VLARs actually cause more system disruption than Guppis do?

What system disruption do the Guppies cause on your system?

The only issue everyone else appears to be having is they take considerably longer than their estimated time to crunch.

---

Grant
Darwin NT