You could try MilkyWay@home, but if you don't have a beefy enough GPU (>512 MB), and aren't running BOINC for at least 4-5 hours per day, I wouldn't recommend it due to the size of the work units and the relatively short deadlines (when crunching with the CPU alone, work units may take 30 hours or more, and the deadline is 8 days).

I'm using optimized aps for MilkyWay, and my dual core CPU with 1G of RAM takes about 5 hours for a WU. Deadlines are usually 7 to 10 days.

---

I´ve got a few einstein tasks to keep my Phenom busy.
Added Collatz to test my GPU (HD 5770)
Thinking about Milkyway too.

Congratulations on getting your Phenon running with Einstein, Mike. The latter came back on the servers this afternoon.

Einstein on my Penryn@3.85GHz take 7.5 hours.

Your AMD HD5770 should give you an RAC of about 140K to 150K on Collatz. But, unfortunately, the GPU does not support double precision. which means it will not work with Milkyway. The ATI/AMD GPUs that do work on Milkyway are - a 3850/70, or a 4850/70 4970 or a 5850/70 or 5970.

The other GPU project, which should see an RAC of 160K on the HD5770 is DNETC.

Happy crunching until S@H is back up.

---

It's good to be back amongst friends and colleagues

It doesn't run under BOINC so no RAC or BOINC points but

http://folding.stanford.edu/

Folding@home will run on cpus and the proper ATI and Nvidia gpus, Linux, Windows, etc.

It's medical research.

The ATI/AMD GPUs that do work on Milkyway are - a 3850/70, or a 4850/70 4970 or a 5850/70 or 5970.

HD 5830 works, too, I'm using a couple of them.
[edit] And HD 4770, was using a couple of those for a while, too...

-Dave

---

I'm using optimized aps for MilkyWay, and my dual core CPU with 1G of RAM takes about 5 hours for a WU. Deadlines are usually 7 to 10 days.

?

AFAIK, the old opt. CPU app don't work with the new WUs..
At least not on my E7600.

http://milkyway.cs.rpi.edu/milkyway/apps.php
MW@h have since 07. Oct. a new CPU app (V0.4).
Also additional a new CPU(-only) app, nbody V0.21 .


EDIT: AFAIK, stock (CPU and) GPU apps are (currently) the fastest.

---

Maybe Milkyway downloaded a new executable without me noticing. At any rate, my CPU today still takes 5 hours or less per WU.

And I still get this message when I start up BOINC:

31/10/2010 2:16:09 PM Milkyway@home Found app_info.xml; using anonymous platform

---

milkyway@home
einstein@home
cosmology@home
orbit@home

Something odd over at cosmolgy@home. You might want to review the message boards first.

No work at the moment at orbit@home. They also are undergoing a hardware upgrade.

Yeah I'm not really sure what's up with cosmology@home tbh. I don't really feel like calling the guy to find out.

From what I understand in This Post He's working away from the project and theres nobody looking after it.

IOW, it's a zombie. How apropos for Halloween.

A similar list here -- for CPU crunching -- plus Climate, Spinhenge and Aqua.

For GPU crunching -- Collatz, Dnetc, Milkyway.

If you want heavy-duty, long-running wu's - try http://climateprediction.net/

They trundle along for long hours, with credit assigned along the way.

Others are Einstein, Docking, Rosetta, Poem - see the BOINC list. These seem to play nice. I stopped PrimeGrid because it kept assigning wu's with short turnaround so they were always running as high priority. Didn't seem like they played nice with others.

(Edited to fix late-night grammar.)

---

I'm using optimized aps for MilkyWay, and my dual core CPU with 1G of RAM takes about 5 hours for a WU. Deadlines are usually 7 to 10 days.

?

AFAIK, the old opt. CPU app don't work with the new WUs..
At least not on my E7600.

http://milkyway.cs.rpi.edu/milkyway/apps.php
MW@h have since 07. Oct. a new CPU app (V0.4).
Also additional a new CPU(-only) app, nbody V0.21 .


EDIT: AFAIK, stock (CPU and) GPU apps are (currently) the fastest.

Yep, have noticed that too.

---

With each crime and every kindness we birth our future.

Maybe Milkyway downloaded a new executable without me noticing. At any rate, my CPU today still takes 5 hours or less per WU.

And I still get this message when I start up BOINC:

31/10/2010 2:16:09 PM Milkyway@home Found app_info.xml; using anonymous platform

Hmm.. strange..

I was curious and looked at your PC at MW@h.
In the tasks overview I saw:

<stderr_txt>
Running Milkyway@home version 0.19 by Gipsel
relabeled by input argument in app_info.xml to version -np
CPU: Intel(R) Core(TM)2 Duo CPU T5750 @ 2.00GHz (2 cores/threads) 1.99522 GHz (2204ms)

WU completed. It took 15665.6 seconds CPU time and 20064.6 seconds wall clock time @ 1.78869 GHz.

</stderr_txt>

Strange, which extension app is this? The SSE2?
It's an app from here: http://www.arkayn.us/milkyway ?
In app_info.xml file, which <version_num> you use, 19, 21, 40?

I used the SSE3 app and didn't worked.
The WU 'calculated' ~ 20 mins and the % progress didn't changed.

---

I'm using optimized aps for MilkyWay, and my dual core CPU with 1G of RAM takes about 5 hours for a WU. Deadlines are usually 7 to 10 days.

?

AFAIK, the old opt. CPU app don't work with the new WUs..
At least not on my E7600.

http://milkyway.cs.rpi.edu/milkyway/apps.php
MW@h have since 07. Oct. a new CPU app (V0.4).
Also additional a new CPU(-only) app, nbody V0.21 .


EDIT: AFAIK, stock (CPU and) GPU apps are (currently) the fastest.

Yep, have noticed that too.

From your projects overview, you don't let run MW@h.
Why/from where you know this? ;-) (if you meant that the opt. app don't work)

---

Sutaru;

Yes, I downloaded optimized aps from Arkyans's site many moons ago. Haven't really watched closely since then. Here is my MilkyWay ap_info file:

<app_info>
<app>
<name>milkyway</name>
</app>
<file_info>
<name>astronomy_0.19_SSSE3.exe</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway</app_name>
<version_num>16</version_num>
<file_ref>
<file_name>astronomy_0.19_SSSE3.exe</file_name>
<main_program/>
</file_ref>
</app_version>
<app_version>
<app_name>milkyway</app_name>
<version_num>19</version_num>
<file_ref>
<file_name>astronomy_0.19_SSSE3.exe</file_name>
<main_program/>
</file_ref>
</app_version>
</app_info>

---

Thanks!

I test again the opt. SSE3 CPU app as milkyway 0.19 .
BOINC got WUs, but normally he shouldn't, because 0.19 isn't current.
The calculation time increase (now ~ 30 mins, in past ~ 2 hours/WU), estimate time decrease, but the % progress stay at 0. The CPU temp increase, so something is calculating.. ;-) ..or this are only 'calculation loops'.

Funny, from where you have the SSSE3 app? This was the x87 app? (But this was in the 32bit area)
I DLed the SSE3 app in past.

---

Strange, your app is named: astronomy_0.19_SSSE3.exe .
My app: astronomy_0.21_SSE3.exe .

Maybe your app is a previously version?

I have my/this app in app_info.xml nevertheless as 0.19 named (because 0.19 was in past the currently stock version, before 0.40).

---

I think it is an old optimized ap. My copy of the executable is dated July 2009. I get WUs steadily, they are all 4.5 to 5 hours to complete, and I think I had maybe one WU hang up in the last year. I get slightly more RAC per CPU hour with this than with optimized S@H.

From posts over at the MW Forums, it looks like the newer stock aps are sometimes fast, but are causing problems on some operating systems. I will stick with this until I see a reason to change.

Note to mods: I know this is all way off topic, but I hope you will let us sort out how to run other projects until the REAL project comes back.

---

GPUGrid.net, to keep CUDA GPU hot.
Einstein@Home for CPU-s, to find new pulsars and gravitational waves.
Keep crunchin'!

---

I like Primegrid. there are several subprojects that have very short TAT and if you prefer longer WU's well they have those to choose from as well.

For the GPU I use Milkyway and Collatz as the backup.

Other projects that I run are World Community Grid - I like their work on Muscular Dystrophy.

I've even ran a bit of work on EON2 though I think this project is still very alpha.

I've ran Yoyo@home but it requires a large amount of RAM

---

In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

I'm using optimized aps for MilkyWay, and my dual core CPU with 1G of RAM takes about 5 hours for a WU. Deadlines are usually 7 to 10 days.

?

AFAIK, the old opt. CPU app don't work with the new WUs..
At least not on my E7600.

http://milkyway.cs.rpi.edu/milkyway/apps.php
MW@h have since 07. Oct. a new CPU app (V0.4).
Also additional a new CPU(-only) app, nbody V0.21 .


EDIT: AFAIK, stock (CPU and) GPU apps are (currently) the fastest.

Yep, have noticed that too.

From your projects overview, you don't let run MW@h.
Why/from where you know this? ;-) (if you meant that the opt. app don't work)

I have no idea why MW dont show up on this page.
Already finnished 6 units so it should.
I also added Collatz

---

With each crime and every kindness we birth our future.

I think it is an old optimized ap. My copy of the executable is dated July 2009. I get WUs steadily, they are all 4.5 to 5 hours to complete, and I think I had maybe one WU hang up in the last year. I get slightly more RAC per CPU hour with this than with optimized S@H.

From posts over at the MW Forums, it looks like the newer stock aps are sometimes fast, but are causing problems on some operating systems. I will stick with this until I see a reason to change.

Note to mods: I know this is all way off topic, but I hope you will let us sort out how to run other projects until the REAL project comes back.

My app is from 03/2010.
Hmm.. strange that your old opt. (0.19) app is working with the new WUs.

In the meantime, MW@h released the 0.45 app. (after 0.4 and 0.44 was/is buggy)

Maybe arkayn could reactivate the DL for the old opt. 0.19 apps on his site?
If they work well with the new WUs.. to the time, maybe Gipsel looked again to the code and release again new opt. apps..

It's not off topic, this thread is about alternate projects.. :-)

---