I've just built a PC and used an Antec Basiq Plus 550W PSU. It is expensive at £70 but it is a real quality piece of kit, weighs 1/2 ton, and comes with so many types of leads its amazing. It also has 22amps output at 12 volts. It is the output amps that matter not the notional wattage.

I've seen 750 Watt PSU's at half the price but they only have 16A output, this one runs at 80% efficiency. Top of the range graphics cards need that sort of amperage to run at 100% full power.

I'm no expert, but those that are have advised me of the above. If people are running cards on lower output PSU's then they are very fortunate, and I wouldn't knock it. Most people can't.

---

Don't drink water, that's the stuff that rusts pipes

I'm running a Q6600 (95W) overclocked to 3.21ghz with a 9800GT card and 1 SATA2 Drive, using a Corsair 520W PSU and system is stable.

Again here you're talking about a piece of high quality kit. The specs say 40Amps across triple 12V rails, is that 13amps on each? I'm not sure. Whatever, at £70 or so you wouldn't go far wrong with that!

PSU link

Whatever. I'm going to continue to run my Sapphire HD4850, SATA HDD and DVD reader/writer etc. in a PC with a 250W PSU.

---

Whatever. I'm going to continue to run my Sapphire HD4850, SATA HDD and DVD reader/writer etc. in a PC with a 250W PSU.

As long as the system is stable (as you say), clearly crunches well and nothing overt shows over time (been running for months) then one must carry on.

Clearly, the manufacturers of high end graphics cards are covering themselves (for legal retail purposes) by recommending lots of spare capacity within the PSUs.

---

Go away, I was asleep

Whatever. I'm going to continue to run my Sapphire HD4850, SATA HDD and DVD reader/writer etc. in a PC with a 250W PSU.

As long as the system is stable (as you say), clearly crunches well and nothing overt shows over time (been running for months) then one must carry on.

Clearly, the manufacturers of high end graphics cards are covering themselves (for legal retail purposes) by recommending lots of spare capacity within the PSUs.

Sounds about right.

I have my HD4870 + Q6600 (95W) OC'd to 3.26GHz + 4x DDR2-1066 + 2 SATA + 2 IDE on a 750W PC Power and Cooling Silencer unit. It's plenty overkill, as this machine draws about 420W on full load, but that's fine with me.

---

Whatever. I'm going to continue to run my Sapphire HD4850, SATA HDD and DVD reader/writer etc. in a PC with a 250W PSU.

As long as the system is stable (as you say), clearly crunches well and nothing overt shows over time (been running for months) then one must carry on.

Clearly, the manufacturers of high end graphics cards are covering themselves (for legal retail purposes) by recommending lots of spare capacity within the PSUs.

Sounds about right.

I have my HD4870 + Q6600 (95W) OC'd to 3.26GHz + 4x DDR2-1066 + 2 SATA + 2 IDE on a 750W PC Power and Cooling Silencer unit. It's plenty overkill, as this machine draws about 420W on full load, but that's fine with me.

That PSU gives you the capability of placing one or two nVidia GTX28x CUDA cards in the PCI-e slots of the MoBo, with the additional power leads from the PSU, so the MW CUDA project can crunch as well as MW normal?

---

Go away, I was asleep

I have my HD4870 + Q6600 (95W) OC'd to 3.26GHz + 4x DDR2-1066 + 2 SATA + 2 IDE on a 750W PC Power and Cooling Silencer unit. It's plenty overkill, as this machine draws about 420W on full load, but that's fine with me.

That PSU gives you the capability of placing one or two nVidia GTX28x CUDA cards in the PCI-e slots of the MoBo, with the additional power leads from the PSU, so the MW CUDA project can crunch as well as MW normal?

Yes, it has 2x 6-pin and 2x 6/8-pin PCIe connectors on it.

---

Whatever. I'm going to continue to run my Sapphire HD4850, SATA HDD and DVD reader/writer etc. in a PC with a 250W PSU.

I think you should get ready for a bang in the night....

Whatever. I'm going to continue to run my Sapphire HD4850, SATA HDD and DVD reader/writer etc. in a PC with a 250W PSU.

I think you should get ready for a bang in the night....

As I said, it's the PC I put my first ATI in 8 million credits ago, many weeks ago, running 24/7 since then, and no bangs yet and why should there be?

What happened to your ATI then? Bang!

Oh mine must have gone BANG! My RAC has risen from 75,000 to 76,700 today. Bang! Wonder how much it will Bang! tomorrow night.

---

I have managed to get work on all 3 computers so far today, the iMac and the Dell take the longest of course since they only run them on the CPU, they also managed to get 12 each.

The quad with the 4830 usually gets 3 at a time and zips them away in 2 minutes.

---

The team search throws the following messages.

Notice: Undefined variable: params in /export/share0/www/boinc/milkyway_gpu/html/user/team_search.php on line 167

Notice: Trying to get property of non-object in /export/share0/www/boinc/milkyway_gpu/html/inc/team.inc on line 35

Notice: Trying to get property of non-object in /export/share0/www/boinc/milkyway_gpu/html/inc/team.inc on line 42

Notice: Trying to get property of non-object in /export/share0/www/boinc/milkyway_gpu/html/inc/team.inc on line 43



While creating a team those messages show up.

Notice: Trying to get property of non-object in /export/share0/www/boinc/milkyway_gpu/html/inc/team.inc on line 409

Privacy note: if you create a team, your project preferences (resource share, graphics preferences) will be visible to the public.


Notice: Trying to get property of non-object in /export/share0/www/boinc/milkyway_gpu/html/inc/team.inc on line 426

Notice: Trying to get property of non-object in /export/share0/www/boinc/milkyway_gpu/html/inc/team.inc on line 432

Notice: Trying to get property of non-object in /export/share0/www/boinc/milkyway_gpu/html/inc/team.inc on line 436

Notice: Trying to get property of non-object in /export/share0/www/boinc/milkyway_gpu/html/inc/team.inc on line 442

Notice: Trying to get property of non-object in /export/share0/www/boinc/milkyway_gpu/html/inc/team.inc on line 445

Notice: Trying to get property of non-object in /export/share0/www/boinc/milkyway_gpu/html/inc/team.inc on line 453

Yes. See this post.

As MilkyWay_GPU is not listed as a project by BOINC, I can't sign up via my normal method "Signup for projects" from my BAM account page.

I could attach to project directly instead of using the account manager, but will that cause problems later when I try to manage it with BAM?
I could also create an account directly from the milkway_gpu page, but I still would have to attach directly to the project.

So what is the best way to get signed up to get cuda gpu wus when they are ready?

So what is the best way to get signed up to get cuda gpu wus when they are ready?

The best way is to wait for the project team to state that the GPU project is ready and for them to provide any different instructions on how to attach to the GPU project, if it is different from other normal projects. To forge ahead without the project truly being ready could end up with you just having to do it all over again.

While agreeing with Brian, if you really want to sign in to create an account and then attach to the project via this -

http://milkyway.cs.rpi.edu/milkyway_gpu/

- You can create an account from the Home page, or a new one at the Milkyway_GPU account via the BOINC Manager attach to a project - new account). Just paste in the above link. That way you will be ready when it launches. But, as Brian indicates, you may have to do it again?

---

Go away, I was asleep

But, as Brian indicates, you may have to do it again?

IIRC, there was a post by Travis or Dave that stated that they wanted people to hold off on doing things there, but that may have just been profiles. Not sure, but anyway... I personally don't see the need in what equates to "pre-registration". You might get a few of the very first tasks, which could be a status thing, but if they run out quickly like over here, whether you signed up first or last won't really matter as everybody will be competing for work like it is here right now... So, my view, even if I had a card capable of doing either CUDA or the ATI/CAL, is that I'd wait until the project was established first...

I wish the GPU team was a little more forecomig with the updates ....
I just would like to know whats going on ... the last update was on the 16th April ...... :-(

1 x i7-920 (Overclocked to 3.6) - Win 32 XP 3Gb DDR3 1600.. GTX9800
1 x Q6600 2.4Ghz - Win 32 XP 2Gb DDR2 800
1 x Q9650 3.0Ghz - Win 32 XP 2Gb DDR2 800 ... 1 x HIS4830 (ATI)
1 X Pentium D 3000..Win 32 XP 1Gb DDR2 533... 1 x HIS4830 (ATI)
+++++++++++++++ ( Total = 233 GFlops ) +++++++++++++++

---

I wish the GPU team was a little more forecomig with the updates ....
I just would like to know whats going on ... the last update was on the 16th April ...... :-(

1 x i7-920 (Overclocked to 3.6) - Win 32 XP 3Gb DDR3 1600.. GTX9800
1 x Q6600 2.4Ghz - Win 32 XP 2Gb DDR2 800
1 x Q9650 3.0Ghz - Win 32 XP 2Gb DDR2 800 ... 1 x HIS4830 (ATI)
1 X Pentium D 3000..Win 32 XP 1Gb DDR2 533... 1 x HIS4830 (ATI)
+++++++++++++++ ( Total = 233 GFlops ) +++++++++++++++

The History of MW GPU
by Slicker

Step 1. Download latest nVidia drivers.

Step 2. Download CUDA toolkit and SDK.

Step 3. Install VS C++ (hope the academic version works ok....)

Step 4. Compile and run "Hello World" successfully. (yep it does.)

Step 5. Announce you are almost ready to release the CUDA app.

Step 6. Discover it takes longer for the CUDA version to run than the CPU version. Oops. Looks like some optimization is required.

Step 7. RTFM. (Then again. And again. And again, but this time only the text since you finally figured out the diagrams are all marketing hype anyway written by some high priced consultant who is a master at powerpoint.)

Step 8. Rewrite the code to use streams, async calls, adjust the thread and block sizes, decide when to use shared memory.

Step 9. Recompile. App runs really fast now but gives wrong results. The code is now 10 times more complicated with all the optimization in place and just a bugger to debug. Plus, you can't afford to puchase an nVidia card since you spent all your money to buy kleenex to wipe off your brown nose.

Step 10. You chew off someones head one day since you were up until 3 am the night before figuring out why the results when using multidimensional streams are so screwed up until you realize that the gpu multi-dimensional arrays don't necessarily use J[x][y] but rather J[y][x] which really starts to matter when x and y are not equal.

Step 11. Frustrated, you give up for now. Determine whether you have enough in the budget to hire Gipsel, Crunch3r, etc. to fix it and then optimize it for you -- but you don't.

Step 12. After finally getting the app working, you find out that the boinc server code is flawed but rather than just fix it, you ask for permission first. Hindsight being 20/20, you now realize that you have now learned that it is better to beg for forgiveness later than to ask for permissions first.

Step 13. You decide to get a Ph.D. in communications instead of continuing your scientific endeavors and then just drop off the face of the earth.

Step 14. People start making up stories with "16 steps" in one of several GPU announcement threads all of which are now totally off topic.

Step 15. The undergrad developers who have really been doing all the work for minimum wage get the code working. They release it. It works really well. It gives 1000 credits per second.

Step 16. DA shuts it down.

The end.

Step 16. DA shuts it down.

I think that is a bit unfair. He generally gets a bad press in Boinc which is not deserved. OK, he could be better at the PR thing, but scientists are what they are.

---

Don't drink water, that's the stuff that rusts pipes