And I'll repeat the last line of my answer to Allen:

If I see anyone else damaging the main science database by running v8 tasks using a v7 application, I will have no hesitation in asking Eric to prevent their computer downloading any new work.

peanut wrote:

I came here from the Q&A on Apple PCs so hopefully this is the correct spot to pose my question/issue. I have a somewhat out-of-date MacPro that I have not upgraded the OS since Tiger (10.4). Currently, my messages seem to be saying something about powerPC which seems odd because my MacPro is an intel machine.


This is the initial part of my messages (i686 seems normal). Then I get errors about powerPC ???

Wed Jan 6 19:05:09 2016||Starting BOINC client version 6.2.18 for i686-apple-darwin

Then comes this which is repeating right now.


Wed Jan 6 21:41:51 2016|SETI@home|Scheduler request succeeded: got 4 new tasks
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802

The Mac apps were updated to v8.02 today. This looks like some server confusion in that respect.

Edit: I've just re-enabled work fetch on my PPC machine and, after getting a similar message, have received an app with the version number 8.01.

---

NG

Just a little warning to everybody who wants to use the latest Nvidia app. It was running fine for me until I added the recommended commandline (+ hp + use_sleep) for mid tier cards from the read me file. When restarting the project my cruncher immediatly froze and crashed. Couldn't even restart my cruncher after that so I have to check this evening what the hell happened there.

So, at least for Gtx 750, DO NOT use this commandline!

---

And I'll repeat the last line of my answer to Allen:

If I see anyone else damaging the main science database by running v8 tasks using a v7 application, I will have no hesitation in asking Eric to prevent their computer downloading any new work.

Richard,

I assure you, I had no idea that this was a bad thing to do. I will shut it down as soon as I return home tonight.

Very sorry!

Allen

---

Thanks - no major harm done, you'll just have diddled yourself out of some points by running tasks which turn out to be invalid.

Sorry for using you as an example to warn the others.

Workunit 2027793629 computer ID 7836483 using GPU v8 app from beta?

---

Workunit 2027793629 computer ID 7836483 using GPU v8 app from beta?

clickable links. investigating

from stderr:

SETI8 update by Raistmer

OpenCL version by Raistmer, r3314

yes, app from beta or from the published dl link. not sure if latest version. the svn stands at r3321, that build may not be public yet.

I hope he keeps an eye on the releases, we're not quite there yet!

---

A person who won't read has no advantage over one who can't read. (Mark Twain)

peanut wrote:

I came here from the Q&A on Apple PCs so hopefully this is the correct spot to pose my question/issue. I have a somewhat out-of-date MacPro that I have not upgraded the OS since Tiger (10.4). Currently, my messages seem to be saying something about powerPC which seems odd because my MacPro is an intel machine.


This is the initial part of my messages (i686 seems normal). Then I get errors about powerPC ???

Wed Jan 6 19:05:09 2016||Starting BOINC client version 6.2.18 for i686-apple-darwin

Then comes this which is repeating right now.


Wed Jan 6 21:41:51 2016|SETI@home|Scheduler request succeeded: got 4 new tasks
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802

The Mac apps were updated to v8.02 today. This looks like some server confusion in that respect.

Edit: I've just re-enabled work fetch on my PPC machine and, after getting a similar message, have received an app with the version number 8.01.

Hi,
I got same error messages on my Mac.
It looks like there is no app version available on SETI servers...

---

workunit 2027793629 computer ID 7836483 using GPU v8 app from beta?

clickable links. investigating

from stderr:

SETI8 update by Raistmer

OpenCL version by Raistmer, r3314

yes, app from beta or from the published dl link. not sure if latest version. the svn stands at r3321, that build may not be public yet.

I hope he keeps an eye on the releases, we're not quite there yet!

Judging by the number of restarts, followed by a false overflow on spikes, I'd say he's got a way to go.

I don't recognise the name as an active tester. He hasn't run any tasks at Beta since May 2010 (and he's got a new laptop since then).

peanut wrote:

[snip]
Then comes this which is repeating right now.


Wed Jan 6 21:41:51 2016|SETI@home|Scheduler request succeeded: got 4 new tasks
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802
Wed Jan 6 21:41:51 2016|SETI@home|[error] No app version for result: i686-apple-darwin 802

The Mac apps were updated to v8.02 today. This looks like some server confusion in that respect.

Edit: I've just re-enabled work fetch on my PPC machine and, after getting a similar message, have received an app with the version number 8.01.

Hi,
I got same error messages on my Mac.
It looks like there is no app version available on SETI servers...

My Tasks page shows that my PPC is currently processing tasks with v8.02, but in fact the app it received was 8.01 and those tasks were "discarded" immediately upon receipt.

---

NG

I'm getting the same errors as several of the other Mac users. I say it is best to just wait it out. It will probably be fixed soon. I'm running OS X "Tiger" on a Core Duo (not Core2Duo) and all has been fine until now..

---

Plus SETI Classic = 21,082 WUs

I've relayed the error reports about mac apps to Eric at beta. He doesn't usually read here.
[there are enough messenger pidgeons around to draw his attention to any problems though, but it would be good if you made sure that one of the usual crew has ack'ed the report and it didn;t get lost in the noise]

---

A person who won't read has no advantage over one who can't read. (Mark Twain)

On what host you testing it?
Your single active beta host has NVIDIA NVS 5200M and did not process (and even recived) single GPU task.

EDIT: and your host on main with low-end 430 has
"Number of period iterations for PulseFind set to:8" manually. So shooted yourself in the leg and blaming gun?
If you use non-default settings attempt to use them properly. If you really want to test new app - test it on beta.

I did not shoot me in the leg, instead i set the iterations to 8 on purpose. Regarding the accompanied readme file it should reduce screen lags. From the readme this value also should default to 20 where in reality it defaults to 50 - so even setting it to 8 causes driver restarts. So i decided to kick it from my system.

[edit/PS:]
Or should this value be INCREASED for reduced screen lag/driver restart?!
Then the documentation should reflect that a lot more clearly! :?

---

Aloha, Uli

Or should this value be INCREASED for reduced screen lag/driver restart?!

Yep.

Splits single PulseFind kernel call to N calls for longest PulseFind calls.

But, perhaps it's not enough.
But if I'll start to write "MultiBeam for Dummies" book seria I'll require adequate payment :P ;)

On what host you testing it?
Your single active beta host has NVIDIA NVS 5200M and did not process (and even recived) single GPU task.

EDIT: and your host on main with low-end 430 has
"Number of period iterations for PulseFind set to:8" manually. So shooted yourself in the leg and blaming gun?
If you use non-default settings attempt to use them properly. If you really want to test new app - test it on beta.

I did not shoot me in the leg, instead i set the iterations to 8 on purpose. Regarding the accompanied readme file it should reduce screen lags. From the readme this value also should default to 20 where in reality it defaults to 50 - so even setting it to 8 causes driver restarts. So i decided to kick it from my system.

[edit/PS:]
Or should this value be INCREASED for reduced screen lag/driver restart?!
Then the documentation should reflect that a lot more clearly! :?

Yes, you should increase it.
I have seen GPU which needs 150.
We can`t test each hardware but i will rework the read me`s before next installer release.

---

With each crime and every kindness we birth our future.

Ok, mea culpa! X)

Thank you both, i'll reinstall and change the settings when i'm back from Thursdays regulars' table. ;)

:beer: Cheers!

---

Aloha, Uli

Thanks - no major harm done, you'll just have diddled yourself out of some points by running tasks which turn out to be invalid.

Sorry for using you as an example to warn the others.

No problem Richard, sorry to have been a pain in the neck, I really didn't understand what I was doing, but I have a much better sense of what is going on now. Seems like the app_info file is much like a .bat file in DOS....didn't realize that.

I've removed everything that has to do with graphics and am in the process of trying to get my cpu cores back to work now on SETI. Hope I don't cause you and the guys any more trouble in the future.

I may try to get my GPU back to work, but on another project. *:)

Thanks!

---

MB8_win_x86_SSE2_OpenCL_ATi_HD5_r3299.exe still running fine. No errors as off yet.

SETI@home v8 (anonymous platform, ATI GPU)
Number of tasks completed	14
Max tasks per day	50
Number of tasks today	22
Consecutive valid tasks	17
Average processing rate	198.64 GFLOPS
Average turnaround time	3.18 days

---

There were no real issues with that build besides cosmetic one for those who use tuning. But currently we on 3316 revision on beta

I may try to get my GPU back to work, but on another project. *:)

If you're that enthusiastic to keep your GPUs busy and crunch v8 early (can't blame you for that) and want to help the project, why not join us in Beta? Just connect BOINC to http://setiweb.ssl.berkeley.edu/beta/

It's recommended to read through the News thread to know the caveats (any important updates will also appear in your BOINC client), check your Tasks to see if there are excessive Invalids/Errors/Inconclusives and post if you find anything wrong that isn't covered.

(edit: and no tinkering... lol)

By helping there, you'll help get the v8 GPU clients released sooner here...

---