As I already mentioned in some other posts, I am working on an implementation of the MW application on ATI graphics hardware (RV670 and up, that means HD38x0, HD4670?? and HD48x0). It is not finished yet (still some scheduling problems as BOINC only supports CUDA atm), but it calculates some results that are in line with the requirements (at least when running offline, have to ask Travis for this ones).

But I guess you want to know how fast it is! So without further fuss here is a screenshot for you (done on a PhenomII 940BE with a Radeon HD4870):



That's something I would call fast!

Holy mother of moo moo's... 8seconds per WU!!

Great to see some people can develop such an application on a GPU (not the easiest of things!) Just a shame that the BOINC dev's have decided to cast ATI to one side and adopt only CUDA is all i can say.

Well done, keep it up :)

I agree, excellent work!!

I know nvidia cards aren't as fast as ATI for this kind of work but when you have completed the ATI app would you consider a CUDA app?

I've just had a go at knocking out my first GPU WU which recently completed giving me nearly 4,000 credits.

I was very impressed, but much more so with the fact that all the while my laptop was churning out 2 MW WUs every 8 minutes or so, at the same time.

I knew VM would eventually let me squeeze out more.

But this is all nothing compared to how I now have an optimized signature with 6 images and 5 links. Who can beat that ? :P


____________

I've just had a go at knocking out my first GPU WU which recently completed giving me nearly 4,000 credits.

I was very impressed, but much more so with the fact that all the while my laptop was churning out 2 MW WUs every 8 minutes or so, at the same time.

I knew VM would eventually let me squeeze out more.

But this is all nothing compared to how I now have an optimized signature with 6 images and 5 links. Who can beat that ? :P

Now there is the making of a conundrum ... which card to get ... ATI so you can do Milky Way or Nvida so you can do SaH and / or GPU Grid?

Or different cards for different systems ... :)

Just a side note, it is possible that GPU Grid got the CPU load issue under control, at least at first blush ... I have one task in flight right now with 1% tops CPU load ... over 10% done in two hours and just over a minute of CPU time ...
____________

w00t, sign me up! Just been waiting to get my 4870 1GB rolling. I was waiting on AI because they're working on an official app last I heard, but Milkyway will work too! :-D
____________

That's really crazy. Now I wish I had an ATI card...
But:
Does this break the credit limit barrier of 108? Wonder if.
Otherwise it's only another method to crunch more WUs without getting more credits from it.
____________
Member of BOINC@Heidelberg and ATA!

My BOINCstats

Otherwise it's only another method to crunch more WUs without getting more credits from it.

Otherwise it's only another method to crunch more WUs without getting more credits from it.

You mean, like work without pay?

Not without pay, but close to it!

Not without pay, but close to it!

Gulp! I can't image anyone providing computer resource to a BOINC project like this and getting almost negligible credit results in return.

If someone has thought, the screenshot may be a fake, there is a video of the running app.

I've implemented a solution to the scheduling problem (at least for Windows) that seems to be working on that specific machine. It is able to calculate several WUs at once, but that doesn't raise the throughput of the app, so calculating 4 WUs at the same time on a quadcore isn't faster than one after another. It uses up one core independent on the number of concurrent WUs.
In principle it works like a small scheduler that switches between the WUs 20 times a second (HD4870, 6 times a second on a HD3850). The video shows that GPU-Z is somehow incompatible with this approach as one experiences severe lags or short freezes of the system after it is started.

Thanks Emploi for that video!

If someone has thought, the screenshot may be a fake, there is a video of the running app.

I've implemented a solution to the scheduling problem (at least for Windows) that seems to be working on that specific machine. It is able to calculate several WUs at once, but that doesn't raise the throughput of the app, so calculating 4 WUs at the same time on a quadcore isn't faster than one after another. It uses up one core independent on the number of concurrent WUs.
In principle it works like a small scheduler that switches between the WUs 20 times a second (HD4870, 6 times a second on a HD3850). The video shows that GPU-Z is somehow incompatible with this approach as one experiences severe lags or short freezes of the system after it is started.

Thanks Emploi for that video!

Jeeeeeezzzz... And i thought you mean't 1 result every 10 or so seconds! But 4!! Wohay, why the hell did i get Nvidia! :P

Great app!

w00t, sign me up! Just been waiting to get my 4870 1GB rolling. I was waiting on AI because they're working on an official app last I heard, but Milkyway will work too! :-D

Two questions.

1) "...that means HD38x0, HD4670?? and HD48x0" What the heck does that mean? Which cards/chips are we talking about?

2) Will this app work on Macs?

[quote]

I have an ATI Radeon X1900 in my Mac. Two questions.

1) "...that means HD38x0, HD4670?? and HD48x0" What the heck does that mean? Which cards/chips are we talking about?

2) Will this app work on Macs?

8 tasks on 1 gpu ;)

http://de.youtube.com/watch?v=nnsW-zB95Is (new video)

Is there any way to get the application now, or is it just open to Planet 3DNow members?
____________

No, it's here: http://www.planet3dnow.de/vbulletin/showpost.php?p=3850506&postcount=762 click: "die neue Version (V3)"

And report (Feedback) in this thread http://www.planet3dnow.de/vbulletin/showthread.php?t=353616 please ;) (also in English)

No, it's here: http://www.planet3dnow.de/vbulletin/showpost.php?p=3850506&postcount=762 click: "die neue Version (V3)"

And report in this thread http://www.planet3dnow.de/vbulletin/showthread.php?t=353616 please ;) (also in English)

...
Heh, that was quick. ..

Is there any way to get the application now, or is it just open to Planet 3DNow members?

Is there any way to get the application now, or is it just open to Planet 3DNow members?

If you browse our forum, you will find it. It's not in a somehow closed forum where you have to login or even have to be member of the team. You will find it also as a guest and now even easier as Emploi posted the link ;)


But it would be helpful if you could read some German, as there are instructions in that thread (you may need to edit the included app_info.xml). It is still a test version (think of it as an alpha version or technology demonstrator, it not a release candidate yet) and some things are simply not working in the moment. I'm in contact with Travis about it as well.
My plan is to give the app including the code to the project when it is working as desired and can be distributed as the stock GPU application of Milkyway.

And if you are interested in your credit standings, you better not run this app with the current credit and WU limits. A quad core with a HD4870 gets only about 330 credits a day, a dual core only 165 credits or so (as the number of WUs you get scales with the number of your cores, even if the calculation speed does not depend on them). It is just the pure enthusiasm that let one lone guy run that app, not the credits. Furthermore we simply need to test it, so I'm really thankful for that (as I don't have a compatible card).

Understood. I know it's just for testing, but it also helps to heat up the room here. :-P The only thing is I don't speak any German, but I've tweaked plenty of app_info.xml files before, so I can get it running. Is it fine if I were to just report anything I find here, or should I still get an account on Planet 3DNow's forums?

Understood. I know it's just for testing, but it also helps to heat up the room here. :-P The only thing is I don't speak any German, but I've tweaked plenty of app_info.xml files before, so I can get it running. Is it fine if I were to just report anything I find here, or should I still get an account on Planet 3DNow's forums?

I guess it does not matter too much, you can also ask here.

For starters, the content of the app_info.xml does not only decide if it runs or not, but also how it runs. With the values avg_ncpus and max_ncpus and the ncpu value of your cc_config.xml you can control (actually together with the resource share, if you run other projects at the same time) how much MW-WUs run concurrently. It will use one core, no matter how many WUs are calculated in parallel. Running two at once could give a slightly better throughput (think about saving 0.2 seconds per WU or so), running more could give diminishing returns. Especially if you set the avg_ncpus value very low, it may become slower. I would suggest setting the ncpu value of your cc_config.xml to your actual numbers of cores +1, set avg_ncpu and max_ncpu both to one and then chose the resource share of milkyway that way, that 2 WUs run concurrently.

Example: Quadcore, ncpu in cc_config.xml set to 5 (calculates then 5 WUs in total parallel) and MW resource share to 2/5 = 40%. Milkyway will use then effectively one core and three are left for other things and it calculates 2 MW-WUs in parallel.

You can also test a very low resource share for MW and setting the avg_ncpus in the app_info.xml to a low value (like 0.25). The slower the card, the lower you can set this before loosing too much performance. 0.2 should be the absolute minimum for a HD4870, a HD3850 may run also with 0.1 quite well. These values get really interesting if/when I release the CPU during the computation.

Tested GPU app on HD3870.
Performance is far superior to any CPU, no doubt.

I've notice that some WUs are "not compatible, falling back to a somewhat slow CPU code."

For example nm_s79, nm_s86 are run on GPU but nm_s20 or nm_s21 are run on CPU.
It is a known issue or some types of WUs can't be completed on a GPU at all?
____________
BOINC Project specifications and hardware requirements

Tested GPU app on HD3870.
Performance is far superior to any CPU, no doubt.

I've notice that some WUs are "not compatible, falling back to a somewhat slow CPU code."

For example nm_s79, nm_s86 are run on GPU but nm_s20 or nm_s21 are run on CPU.
It is a known issue or some types of WUs can't be completed on a GPU at all?

Have you played around a bit with the settings in the app_info.xml, cc_config.xml and the resource share?

I see you have updated the client from 5.10.30 to 6.6.3 before running the GPU app. This should have been unnecessary as the GPU stuff in there is only for nvidia cards and is not used at all.

Thank you for all your work on this. I must say it's good to hear AMD/ATI's cards are performing so well now, considering they've been playing catchup for a while (from what I can gather, anyway). Is the code very different from the CUDA equivalent? And would a CUDA conversion be worth it? (I heard the GTX280 is much better at double precision operations than earlier cards, but it sounds like even it is outmatched by current offerings by AMD)

Thank you for all your work on this. I must say it's good to hear AMD/ATI's cards are performing so well now, considering they've been playing catchup for a while (from what I can gather, anyway). Is the code very different from the CUDA equivalent? And would a CUDA conversion be worth it? (I heard the GTX280 is much better at double precision operations than earlier cards, but it sounds like even it is outmatched by current offerings by AMD)

Is the code very different from the CUDA equivalent? And would a CUDA conversion be worth it? (I heard the GTX280 is much better at double precision operations than earlier cards, but it sounds like even it is outmatched by current offerings by AMD)

Got some errors with the workunits. I made the mistake of setting the app in, then putting Milkyway wide open. About 7 tasks quit with an error before I suspended the rest. Looking in one of the reported WUs, the exit code is 0xc0000135. Googling for that returns many results for BOINC, most of which seem to say it's a missing DLL. I'm presuming this to be brook.dll. So maybe the app_info has to be tweaked so it's also copied to the slots directory? I'm not sure how to verify what files are in there to check, as BOINC deletes them too quickly for me.
____________

Quick notes:

* ~20s GPU crunch times on i920@3.5 + 3850/256.
* You can feel the whole machine lagging when the GPU fires up lol. (think the problem is noted a few messages ago)
* ~450s CPU runtime? I hope there's a way of disabling the GPU part so I can run this on my other x64 box with Nvidia card...

Got some errors with the workunits. I made the mistake of setting the app in, then putting Milkyway wide open. About 7 tasks quit with an error before I suspended the rest. Looking in one of the reported WUs, the exit code is 0xc0000135. Googling for that returns many results for BOINC, most of which seem to say it's a missing DLL. I'm presuming this to be brook.dll. So maybe the app_info has to be tweaked so it's also copied to the slots directory? I'm not sure how to verify what files are in there to check, as BOINC deletes them too quickly for me.

Quick notes:

* ~20s GPU crunch times on i920@3.5 + 3850/256.
* You can feel the whole machine lagging when the GPU fires up lol. (think the problem is noted a few messages ago)
* ~450s CPU runtime? I hope there's a way of disabling the GPU part so I can run this on my other x64 box with Nvidia card...

So where is the link to the zip file?? Sorry in reply to Message 9513
____________

So where is the link to the zip file?? Sorry in reply to Message 9513

So where is the link to the zip file?? Sorry in reply to Message 9513

@ the guys, where it runs, had you similiar problems?

I installed it this morning on a vista 64 i7 with ati hd 4800, and they've been cranking out fine since. the short ones are running 5 - 9 seconds; the longer apps are running just shy of 10 minutes each.

Got some errors with the workunits. I made the mistake of setting the app in, then putting Milkyway wide open. About 7 tasks quit with an error before I suspended the rest. Looking in one of the reported WUs, the exit code is 0xc0000135. Googling for that returns many results for BOINC, most of which seem to say it's a missing DLL. I'm presuming this to be brook.dll. So maybe the app_info has to be tweaked so it's also copied to the slots directory? I'm not sure how to verify what files are in there to check, as BOINC deletes them too quickly for me.

Hmm, what have you downloaded?
The brook.dll is supplied in the zip file and also correctly set up in the also supplied app_info.xml. Just copy all 3 files to your Milkyway folder (and completely quit BOINC before). But maybe you need to download new WUs, as for the ones you already have there could be some issues with the data in the client_state.xml.

@ the guys, where it runs, had you similiar problems?

* ~20s GPU crunch times on i920@3.5 + 3850/256.

The files are there, and the only thing I've tweaked in the app_info is the avg_ncpus to 1. However, could it be caused by an older version of BOINC? I'm running 6.2.19.

The files are there, and the only thing I've tweaked in the app_info is the avg_ncpus to 1. However, could it be caused by an older version of BOINC? I'm running 6.2.19.

The files are there, and the only thing I've tweaked in the app_info is the avg_ncpus to 1. However, could it be caused by an older version of BOINC? I'm running 6.2.19.

You may try to delete the line with <api_version> in the app_info.xml, but actually it should even run with the old 5.x.x clients. Remember that you have to stop BOINC also for editing the app_info.xml.

The files are there, and the only thing I've tweaked in the app_info is the avg_ncpus to 1. However, could it be caused by an older version of BOINC? I'm running 6.2.19.

You may try to delete the line with <api_version> in the app_info.xml, but actually it should even run with the old 5.x.x clients. Remember that you have to stop BOINC also for editing the app_info.xml.

Yes, I know I have to stop BOINC. I'll try deleting the API version line, and try again.

The files are there, and the only thing I've tweaked in the app_info is the avg_ncpus to 1. However, could it be caused by an older version of BOINC? I'm running 6.2.19.

You may try to delete the line with <api_version> in the app_info.xml, but actually it should even run with the old 5.x.x clients. Remember that you have to stop BOINC also for editing the app_info.xml.

Yes, I know I have to stop BOINC. I'll try deleting the API version line, and try again.

I googled a bit and it appears you may need the .NET framework installed. Thought I've unchecked that option in the compiler (it's not used either way), but who knows.

Edit: Hmm, but normally one needs the .NET framework 2.0 either way for the Catalyst driver, isn't it?

Edit2: Has anyone tried the new Cat 9.1? It shouldn't make a difference, as the CAL compiler (part of the driver) still generates the same binary code to run on the GPU.

I don't know about .NET for Catalyst, but I'm running Windows 7 beta x64. It should have .NET already installed, same version as or later than Vista. As for drivers, no I haven't tried 9.1. 8.12 is all that's available for Win7 for now.

...I have not thought of the 256MB variants and 512MB are enough for any sensible amount of concurrent WUs, but it would be interesting to see what happens, just in case.

As stated 4670 does not seem to be working this is an MSI 4670 512 MB

Seems to get compute error instantly


on this host

http://milkyway.cs.rpi.edu/milkyway/show_host_detail.php?hostid=25888

...I have not thought of the 256MB variants and 512MB are enough for any sensible amount of concurrent WUs, but it would be interesting to see what happens, just in case.

On my system the WUs "stalls" when more than 6 WU are run concurrently, ie the CPU and GPU usage goes to 0%. Restarting Boinc manager solves the problem for a while then it just stalls again.

As stated 4670 does not seem to be working this is an MSI 4670 512 MB

Seems to get compute error instantly


on this host

http://milkyway.cs.rpi.edu/milkyway/show_host_detail.php?hostid=25888

Thank you kindly for the info and the link... Lets hope it works out maybe someone wil do an app for w32 and xseries ATI cards..
Glenn.
____________

Any more suggestions on what I should do to test the GPU app? I have a Vista x64 installation I can try with to, if that might help.
____________

Works perfect here on a Q6600 with HD4870 running on Vista X64. Should get the label "release candidate" imho.

Works perfect here on a Q6600 with HD4870 running on Vista X64. Should get the label "release candidate" imho.

I have the next alpha version ready, still limited to Win64 and running on ATI HD38x0 and HD48x0 cards with Catalyst driver 8.12 or 9.1. But this time it should be able to handle again all WUs flying around, also those with more than one stream. They should take about 50% longer (there is one stream more to calculate). I have squeezed the code a bit, it should need now about 0.1 seconds less for the single stream WUs ;) But there is not much left what one could get out of it performance wise. Don't expect major speed gains for the future!

Let me know how it is running for you!

Link to the latest alpha-version

Next thing to do will be more diagnostic output (GPU detection) as preparation for a major overhaul of the WU scheduling, which will hopefully support multiple GPUs afterwards and maybe also the release of the CPU during the GPU tasks. After that, I will have a look to Linux.

Running ATI Tray Tool and setting fan speed to 40-50% helped was good solution.
I'm able to monitor GPU load, temps and fan speed with ATT without problems.
In general, app is working just fine (with comments in Message 9432).

So triple 4870's are out of the question for now I see :)
I have an linux box with a 3870 and Ubuntu 64bit with Cat 9.1 set up when you get around to a linux app.

So triple 4870's are out of the question for now I see :)
I have an linux box with a 3870 and Ubuntu 64bit with Cat 9.1 set up when you get around to a linux app.

I have the next alpha version ready, still limited to Win64 and running on ATI HD38x0 and HD48x0 cards with Catalyst driver 8.12 or 9.1. But this time it should be able to handle again all WUs flying around, also those with more than one stream. They should take about 50% longer (there is one stream more to calculate). I have squeezed the code a bit, it should need now about 0.1 seconds less for the single stream WUs ;) But there is not much left what one could get out of it performance wise. Don't expect major speed gains for the future!

I have the next alpha version ready, still limited to Win64 and running on ATI HD38x0 and HD48x0 cards with Catalyst driver 8.12 or 9.1. But this time it should be able to handle again all WUs flying around, also those with more than one stream. They should take about 50% longer (there is one stream more to calculate). I have squeezed the code a bit, it should need now about 0.1 seconds less for the single stream WUs ;) But there is not much left what one could get out of it performance wise. Don't expect major speed gains for the future!

Just a quick note that it still crashes on my machine. I'm going to try and clear out at least the Milkyway tasks I have left, then see if it works on my Vista x64 installation. It's possible ATI's 8.12 drivers for Win7 don't have Stream support yet, which is my suspicion at this point.

I have the next alpha version ready, still limited to Win64 and running on ATI HD38x0 and HD48x0 cards with Catalyst driver 8.12 or 9.1. But this time it should be able to handle again all WUs flying around, also those with more than one stream. They should take about 50% longer (there is one stream more to calculate). I have squeezed the code a bit, it should need now about 0.1 seconds less for the single stream WUs ;) But there is not much left what one could get out of it performance wise. Don't expect major speed gains for the future!

Just a quick note that it still crashes on my machine. I'm going to try and clear out at least the Milkyway tasks I have left, then see if it works on my Vista x64 installation. It's possible ATI's 8.12 drivers for Win7 don't have Stream support yet, which is my suspicion at this point.

Well, Vista x64 gives a better error: "No compatible GPU found!" I know I have a Radeon HD4870 1GB though, so I'm not sure what's causing this. Right now, I'm running with Cat 8.12. I'll update to Cat 9.1 and try again.

EDIT: Same result with Cat 9.1. Are you checking for a compatible GPU, and if so, how?

Well, Vista x64 gives a better error: "No compatible GPU found!" I know I have a Radeon HD4870 1GB though, so I'm not sure what's causing this. Right now, I'm running with Cat 8.12. I'll update to Cat 9.1 and try again.

EDIT: Same result with Cat 9.1. Are you checking for a compatible GPU, and if so, how?

Could you post the content of the stderr.txt (shown under task details) please? I have not found it in your host list (already purged?).

If "No compatible GPU found!" is the only error there, the app does not find the CAL (Compute Abstraction Layer) libraries, which are installed as part of the driver since 8.12. If the libs would be there and one has an unsupported card, it would tell you something about double precision not supported.
The error you are seeing hints at an incorrectly installed or too old driver. The alternative would be a missing brook.dll, but I guess you checked that already.

Or are you trying to run it when accessing the machine via a remote desktop connection? That is also not working, because Win uses some kind of a virtual framebuffer driver for that (you don't have the Catalyst as display driver in a remote session). You could use VNC and check "disable mirror driver".

Is there any way I can find out manually if I have the CAL libraries installed?

Is there any way I can find out manually if I have the CAL libraries installed?

Look into the Windows\system32 folder. There should be the files "amdcalcl64.dll", "amdcaldd64.dll" and "amdcalrt64.dll".

I've been using the gpu app for almost 2 weeks now and just wanted to give a little feedback. I've rolled back my driver to the 8.12 version from 9.1 yesterday. I was running 8.12 fine but wanted to get the latest driver installed when it came out. After the upgrade i had a lot of pc responsiveness issues and a bsod every few days. Since then my computer has ran much smoother. I'm not sure why the newer driver just didn't seem to perform as well as 8.12. I should mention i'm running gpu version 0.17.

I have the next alpha version ready, still limited to Win64 and running on ATI HD38x0 and HD48x0 cards with Catalyst driver 8.12 or 9.1. But this time it should be able to handle again all WUs flying around, also those with more than one stream. They should take about 50% longer (there is one stream more to calculate). I have squeezed the code a bit, it should need now about 0.1 seconds less for the single stream WUs ;) But there is not much left what one could get out of it performance wise. Don't expect major speed gains for the future!

Just a quick note that it still crashes on my machine. I'm going to try and clear out at least the Milkyway tasks I have left, then see if it works on my Vista x64 installation. It's possible ATI's 8.12 drivers for Win7 don't have Stream support yet, which is my suspicion at this point.

Well, Vista x64 gives a better error: "No compatible GPU found!" I know I have a Radeon HD4870 1GB though, so I'm not sure what's causing this. Right now, I'm running with Cat 8.12. I'll update to Cat 9.1 and try again.

EDIT: Same result with Cat 9.1. Are you checking for a compatible GPU, and if so, how?

Just a quick note that it still crashes on my machine. I'm going to try and clear out at least the Milkyway tasks I have left, then see if it works on my Vista x64 installation. It's possible ATI's 8.12 drivers for Win7 don't have Stream support yet, which is my suspicion at this point.

Well, Vista x64 gives a better error: "No compatible GPU found!" I know I have a Radeon HD4870 1GB though, so I'm not sure what's causing this. Right now, I'm running with Cat 8.12. I'll update to Cat 9.1 and try again.

EDIT: Same result with Cat 9.1. Are you checking for a compatible GPU, and if so, how?

Hi,
Same problem for me with Vista64 + Catalyst 9.1

<core_client_version>6.2.19</core_client_version>
<![CDATA[
<message>
Fonction incorrecte. (0x1) - exit code 1 (0x1)
</message>
<stderr_txt>
No compatible GPU found!

</stderr_txt>
]]>

Works fine on my E7200 2.53 GHz & HD4850. I get about 300 credits per hour, with four workunits at the same time (Core2Duo).

Works fine on my E7200 2.53 GHz & HD4850. I get about 300 credits per hour, with four workunits at the same time (Core2Duo).

Works fine on my E7200 2.53 GHz & HD4850. I get about 300 credits per hour, with four workunits at the same time (Core2Duo).

Thanks for the feedback on the Vista64 issue.
The credits you get are less than a Quadore with an optimized CPU app gets (432credits/hour). Furthrmore you can't let it run for 24h a day. After about 6 hours you have used up your WU limit of 1000 WUs per day and core and that will earn you less than 2000 credits a day.
So you should definitely run another project at the same time. The MW GPU app uses only one core no matter how many WUs run concurrently. The work done for MW won't change, if you run it only with 50% resource share (2 WUs at the same time). Just try it!

Just a quick note that it still crashes on my machine. I'm going to try and clear out at least the Milkyway tasks I have left, then see if it works on my Vista x64 installation. It's possible ATI's 8.12 drivers for Win7 don't have Stream support yet, which is my suspicion at this point.

Well, Vista x64 gives a better error: "No compatible GPU found!" I know I have a Radeon HD4870 1GB though, so I'm not sure what's causing this. Right now, I'm running with Cat 8.12. I'll update to Cat 9.1 and try again.

EDIT: Same result with Cat 9.1. Are you checking for a compatible GPU, and if so, how?

Hi,
Same problem for me with Vista64 + Catalyst 9.1

<core_client_version>6.2.19</core_client_version>
<![CDATA[
<message>
Fonction incorrecte. (0x1) - exit code 1 (0x1)
</message>
<stderr_txt>
No compatible GPU found!

</stderr_txt>
]]>

The screenshot and two videos posted in this thread are done on Vista64. There are quite some people where it runs. So please check again the points already mentioned here in the forum:
- driver correctly installed (three amdcal??64.dll files present in Windows/system32 folder)
- you are not accessing the machine over a remote desktop connection (use VNC and disable mirror driver)
- the card should be active (monitor connected to it)

How have you guys installed the BOINC client?
In the protected mode under Vista the application may not have the necessary access rights. CUDA apps also fail when tried to run in such a scenario. On XP64 and Server2003 it should also run in the protected mode.

If also the people where it runs could report on this?

How have you guys installed the BOINC client?
In the protected mode under Vista the application may not have the necessary access rights. CUDA apps also fail when tried to run in such a scenario. On XP64 and Server2003 it should also run in the protected mode.

If also the people where it runs could report on this?

How have you guys installed the BOINC client?
In the protected mode under Vista the application may not have the necessary access rights. CUDA apps also fail when tried to run in such a scenario. On XP64 and Server2003 it should also run in the protected mode.

If also the people where it runs could report on this?

How have you guys installed the BOINC client?
In the protected mode under Vista the application may not have the necessary access rights. CUDA apps also fail when tried to run in such a scenario. On XP64 and Server2003 it should also run in the protected mode.

If also the people where it runs could report on this?

I reinstalled Boinc this morning with default value, and now all seems to work, some freeze on the screen but some tuning needed, i suppose. Thanks for your support

Probably silly question, but, Why can these certain graphics cards do their own work for projects and not older cards?
____________
Doesn't expecting the unexpected make the unexpected the expected?
If it makes sense, DON'T do it.

Probably silly question, but, Why can these certain graphics cards do their own work for projects and not older cards?

I reinstalled Boinc this morning with default value, and now all seems to work, some freeze on the screen but some tuning needed, i suppose. Thanks for your support

Isn't the default the protected mode?...

I reinstalled Boinc this morning with default value, and now all seems to work, some freeze on the screen but some tuning needed, i suppose. Thanks for your support

Isn't the default the protected mode?...

No, per default the protection mode is off, you have to choose the "Protected application execution" option separately.

I reinstalled Boinc this morning with default value, and now all seems to work, some freeze on the screen but some tuning needed, i suppose. Thanks for your support

Isn't the default the protected mode?...

No, per default the protection mode is off, you have to choose the "Protected application execution" option separately.

Oh, I guess they have changed it because of the CUDA apps for SETI *lol*

Sorry to sound stupid but is a 9700 all in one wonder to old to use this application?

Thanks in advance.

According to GPU-Z the GPU is an R300

Sorry to sound stupid but is a 9700 all in one wonder to old to use this application?

Thanks in advance.

According to GPU-Z the GPU is an R300

Probably silly question, but, Why can these certain graphics cards do their own work for projects and not older cards?

Because the older cards can only do single precision calculations (32bit) and the HD38x0 and HD48x0 are the only ones (besides nvidias GTX2xx series) that can handle double precision (64bit).

Sorry to sound stupid but is a 9700 all in one wonder to old to use this application?

Thanks in advance.

According to GPU-Z the GPU is an R300

Sorry, but some years too old.

Probably silly question, but, Why can these certain graphics cards do their own work for projects and not older cards?

Because the older cards can only do single precision calculations (32bit) and the HD38x0 and HD48x0 are the only ones (besides nvidias GTX2xx series) that can handle double precision (64bit).

Isn't that what pc's do? But older cards could do single calc's then. Why not add support for those? There would be plenty, since these new cards aren't that old.

This is about the floating point precision. Travis has set quite strict limits for the reslts of the test WUs. You can't reach them with single precision calculations on older cards (the really old ones support only 16 or 24bit FP). It may be possible to get there with some kind of software emulation, but this would be a lot of effort which would be lost in the future as more and more cards get double precision support. Furthermore it would be most likely as slow ;) or even slower than to do the computations on a CPU.

Ok. I guess if it was worth it, other projects would be using it for the 'older' cards.

Ok. I guess if it was worth it, other projects would be using it for the 'older' cards.

Don't forget for most other projects (like SETI or GPUGrid) single precision is enough. That is the reason you can also use slightly older cards there. MW is just more demanding in this specific area.

Running pretty well on my HD4830, but I would like it to be running 2 wu's instead of the 8 or so it's running currently. I have played with the app_info file and so far have had no success. Right now it's set on avg_ncpus to 0.2, and max_ncpus to 5 (quad core intel). Resource share on MW is set to 10% and it still continues to run 10 wu's in parallel. Any suggestions?

Running pretty well on my HD4830, but I would like it to be running 2 wu's instead of the 8 or so it's running currently. I have played with the app_info file and so far have had no success. Right now it's set on avg_ncpus to 0.2, and max_ncpus to 5 (quad core intel). Resource share on MW is set to 10% and it still continues to run 10 wu's in parallel. Any suggestions?

Running pretty well on my HD4830, but I would like it to be running 2 wu's instead of the 8 or so it's running currently. I have played with the app_info file and so far have had no success. Right now it's set on avg_ncpus to 0.2, and max_ncpus to 5 (quad core intel). Resource share on MW is set to 10% and it still continues to run 10 wu's in parallel. Any suggestions?

I am running another project, and I switched the resource share to 40% here at MW, bumped the avg_ncpus up to 0.5. I closed CCC and it seemed to help a bit, more towards the acceptable level.

Also, is there going to be a new version for .18?

avg_ncpus set to 0.1

max_ncpus set to 3 (core2duo)

Seems to work very fine, the Milkyway units are still completed in 8 or 10 seconds. It's the same time when I used 0.50 core, is it "normal" ?

Too bad we have the 1000 workunit-per-cpu limit, is it possible to send a request somewhere to remove this limit when using an optimized app ? This limit is now obsolete when you can calculate 3 or 4 times more units in a day with an optimized app. The only reason I'm moving a computer on milkyway is to help you for your excellent work on ATI graphic cards, but the credits are not very interesting ^^

THank you for this app, anyway ^^
____________
Star Wars BOINC Team

Also, I can raise the workunit-per-cpu limit. What would be a good value?

Sounds like a good idea to me.

I think one of the recent updates to the BOINC server code allows for a separate daily WU queue for GPUs.

avg_ncpus set to 0.1

max_ncpus set to 3 (core2duo)

Seems to work very fine, the Milkyway units are still completed in 8 or 10 seconds. It's the same time when I used 0.50 core, is it "normal" ?

Too bad we have the 1000 workunit-per-cpu limit, is it possible to send a request somewhere to remove this limit when using an optimized app ? This limit is now obsolete when you can calculate 3 or 4 times more units in a day with an optimized app. The only reason I'm moving a computer on milkyway is to help you for your excellent work on ATI graphic cards, but the credits are not very interesting ^^

THank you for this app, anyway ^^

I think one of the recent updates to the BOINC server code allows for a separate daily WU queue for GPUs. I'll do a little looking into it and if thats the case then we can give the GPUs a 10k daily limit without touching the other one.

max_ncpus back to "1".

avg_ncpus set to 0.25. If I keep the default value of 0.50, I have only three WU running at the same time : two optimized MW and one World Community Grid. With avg_ncpus to 0.25 I have four workunits (two for each project)

Travis > thank you very much for your answer about the credits, it's a very good news. For maximum number of units per core, I can say that : with a Core2Duo and a HD4850 512Mo, I crunched 2000 workunits in 6 hours (with no other projet running at the same time, so the CPU wasn't at full use). So 4.000 workunits per core and day will be enough for this configuration (my computer is working "only" 12 hours per day or so, so 4.000 WU per core and day is the very, very maximum). I don't know how many workunits can be crunched with another model of ATI.

Cluster > If I understand correctly, your optimization uses "only" one core at a time, is that right ? Is it possible to use more core, so we can use only Milkyway on one computer with more than one core ?

Do you need something special to help you for your tests ?

PS : during this writing I reached my 2.000 workunits limit - later than before, probably because I tried avg_ncpus to 0.10 ...
____________
Star Wars BOINC Team

max_ncpus back to "1".

avg_ncpus set to 0.25. If I keep the default value of 0.50, I have only three WU running at the same time : two optimized MW and one World Community Grid. With avg_ncpus to 0.25 I have four workunits (two for each project)

Travis > thank you very much for your answer about the credits, it's a very good news. For maximum number of units per core, I can say that : with a Core2Duo and a HD4850 512Mo, I crunched 2000 workunits in 6 hours (with no other projet running at the same time, so the CPU wasn't at full use). So 4.000 workunits per core and day will be enough for this configuration (my computer is working "only" 12 hours per day or so, so 4.000 WU per core and day is the very, very maximum). I don't know how many workunits can be crunched with another model of ATI.

Cluster > If I understand correctly, your optimization uses "only" one core at a time, is that right ? Is it possible to use more core, so we can use only Milkyway on one computer with more than one core ?

Do you need something special to help you for your tests ?

PS : during this writing I reached my 2.000 workunits limit - later than before, probably because I tried avg_ncpus to 0.10 ...

Cluster > If I understand correctly, your optimization uses "only" one core at a time, is that right ? Is it possible to use more core, so we can use only Milkyway on one computer with more than one core ?

With the gpu, I can crunch 16 @ a time with the i7, so 10k limit per core, or 80,000 in my case, would be a more realistic than 4000.

Cluster > If I understand correctly, your optimization uses "only" one core at a time, is that right ? Is it possible to use more core, so we can use only Milkyway on one computer with more than one core ?

The goal is actually to use not a full core (or even more), but maybe only 10% of a core or so. This way your CPU would be free to crunch something else.
If it is really wanted I could put in support for simultaneous crunching of MW on GPU and CPU. But this would have a low priority on my list.

Cluster > If I understand correctly, your optimization uses "only" one core at a time, is that right ? Is it possible to use more core, so we can use only Milkyway on one computer with more than one core ?

The goal is actually to use not a full core (or even more), but maybe only 10% of a core or so. This way your CPU would be free to crunch something else.
If it is really wanted I could put in support for simultaneous crunching of MW on GPU and CPU. But this would have a low priority on my list.

What I would like to see is the ability to use my Nvidia in MW. I know you are using an ATI since it is faster, but I only have the Nvidia which I'd be interested in transferring from crunching on GPUGRID to MW.

What I would like to see is the ability to use my Nvidia in MW. I know you are using an ATI since it is faster, but I only have the Nvidia which I'd be interested in transferring from crunching on GPUGRID to MW.

Afaik, there is already a student starting to work on a CUDA app. As this is easier to work with, I guess we could see some results soon ;)
But don't expect times much below 25s per WU for nvidias GTX line. And older ones won't work at all (lack of double precision units).

25 secs? Blimey, hurry up student :)

25 secs? Blimey, hurry up student :)

But that is soooo slooooow compared to the less than 10 seconds on ATIs HD4870 ;)

25 secs? Blimey, hurry up student :)

But that is soooo slooooow compared to the less than 10 seconds on ATIs HD4870 ;)

25 sec is good enough for me, hurry up student :-)

With the gpu, I can crunch 16 @ a time with the i7, so 10k limit per core, or 80,000 in my case, would be a more realistic than 4000.

But the throughput is still be one WU every 9 seconds or so with a HD4870. It is not getting faster with more concurrent WUs. So with a HD4870 a limit of 10,000 WUs a day would be enough as long there is no multi GPU support implemented (or massive overclocking involved).

I would say 10,000 WUs per host and day are needed now. When multiple cards are working and/or newer GPUs are available, this needs to be raised again.

With the gpu, I can crunch 16 @ a time with the i7, so 10k limit per core, or 80,000 in my case, would be a more realistic than 4000.

But the throughput is still be one WU every 9 seconds or so with a HD4870. It is not getting faster with more concurrent WUs. So with a HD4870 a limit of 10,000 WUs a day would be enough as long there is no multi GPU support implemented (or massive overclocking involved).

I would say 10,000 WUs per host and day are needed now. When multiple cards are working and/or newer GPUs are available, this needs to be raised again.

oh, i agree with that too. the turn around time for the gpu app was about 2 1/2 minutes. i've been running the op app this week and it's crunching 8 wu's in 6 minutes since the .19's came out, so that limit needs to go much higher as well.

Yeah hopefully within the next week or two we'll have an alpha CUDA application for you guys to crash :D

Yeah hopefully within the next week or two we'll have an alpha CUDA application for you guys to crash :D

Well, I have one GTX 280 and 2 GTX 295s ... start your engines ...

Of course we will need a setting on the site to get only CPU work, only CUDA work ... or both ...

My HD4830 times have been going up steadily in the last day or so. Running almost 30 seconds now.

My HD4830 times have been going up steadily in the last day or so. Running almost 30 seconds now.

Awesome, thank you for all your help.

Earlier you mentioned leaving the max_ncpus at 1, is there any harm in having it higher? I set it back to 5 and the times are immediately back to 2-3 seconds. Just bad timing I suppose, or could it be bottlenecked and needing some extra CPU power?

Awesome, thank you for all your help.

Earlier you mentioned leaving the max_ncpus at 1, is there any harm in having it higher? I set it back to 5 and the times are immediately back to 2-3 seconds. Just bad timing I suppose, or could it be bottlenecked and needing some extra CPU power?

How big of an impact does amount of memory on the card (ie 512MB vs 1GB 4870) have on crunch times?

How big of an impact does amount of memory on the card (ie 512MB vs 1GB 4870) have on crunch times?

How big of an impact does amount of memory on the card (ie 512MB vs 1GB 4870) have on crunch times?

None.
Also the speed of the memory does not have a pronounced effect (cache hit rates are quite high). The advantage of a HD4870 compared to a HD4850 stems only from the higher clockrate (+20%). The almost doubled memory bandwidth of the HD4870 (GDDR5 vs GDDR3) does not help in case of MW. The algorithm is dominated by computations, not memory accesses. One has only one memory access every eleven instructions or so. And as most of these accesses can be cached quite well it doesn't need much memory bandwidth. Actually it is the same behaviour as one sees with the CPU version.

More memory on the card would only enable the possibility to crunch more WUs concurrently (think of 30+ on a 1 GB card). But as more WUs in flight does not help the througput, it is only an esoterical advantage.

32 bit XP version in the works?? ;)

Regarding Multi-GPU: does that mean that at the moment you can run only GPU or only one card per PC ? What should one look for: an 4870X2 or an 4870 ?

Regarding Multi-GPU: does that mean that at the moment you can run only GPU or only one card per PC ? What should one look for: an 4870X2 or an 4870 ?

Thanx, that did save me some money... :)

I noticed something strange of the website (zslip.com) : the link for the ATI GPU app is labeled "Windows optimized apps (0.17 - latest, cuda)".

Cuda ?
____________
Star Wars BOINC Team

I noticed something strange of the website (zslip.com) : the link for the ATI GPU app is labeled "Windows optimized apps (0.17 - latest, cuda)".

Cuda ?

So is there currently a WU/credit limit on the GPU app, and if so, what is it?

I noticed something strange of the website (zslip.com) : the link for the ATI GPU app is labeled "Windows optimized apps (0.17 - latest, cuda)".

Cuda ?

I guess nvidias marketing has done its job ;)

lol. They probably want to change that :)

I noticed something strange of the website (zslip.com) : the link for the ATI GPU app is labeled "Windows optimized apps (0.17 - latest, cuda)".

Cuda ?

I guess nvidias marketing has done its job ;)

lol. They probably want to change that :)

Haha, sorry about that. I'll get it changed later today when I can get to the site(s). I meant to put 'gpu' not 'cuda'.

Might want to put ATI because I think it will only work for certain ATI cards. You might want to mention which ones as well.

It's my personal opinion, but I think that a full list of compatible GPUs might be great, because I can read here and there some questions about the compatibility of cards. Maybe the label "HD38x0 or HD48x0" isn't clear for all people. Maybe something like this :

Compatible Cards :

- ATI Radeon HD 3840
- ATI Radeon HD 4850
- ATI Radeon HD 4870

Cluster > do not be offensed by my next question : do you want a simple html/css code to present your different pages ? With some nice tables, colors, etc. ?
____________
Star Wars BOINC Team

It's my personal opinion, but I think that a full list of compatible GPUs might be great, because I can read here and there some questions about the compatibility of cards. Maybe the label "HD38x0 or HD48x0" isn't clear for all people. Maybe something like this :

Compatible Cards :

- ATI Radeon HD 3840
- ATI Radeon HD 4850
- ATI Radeon HD 4870

Cluster > do not be offensed by my next question : do you want a simple html/css code to present your different pages ? With some nice tables, colors, etc. ?

You can take a look (it's an alpha version ^^) :

http://www.starwars-holonet.com/boinc/mw.html

Just to get an idea. You can take the code of course ^^
____________
Star Wars BOINC Team

You can take a look (it's an alpha version ^^) :

http://www.starwars-holonet.com/boinc/mw.html

Just to get an idea. You can take the code of course ^^

http://www.starwars-holonet.com/boinc/mw.html

I'm always pleased to take advantages of improvements, so if you're serious in that I can use the code for your table, I might just do that. Thanks ;)

I like that :)

Take everything you want.

What can I have? ;P

I'm always pleased to take advantages of improvements, so if you're serious in that I can use the code for your table, I might just do that. Thanks ;)

I'm serious. It took me 5 or 10 minutes to do that, so it's not a big work ^^

Take everything you want. If you want me to do something else, do not hesitate ^^

(exarkun_AT_free.fr if you want to talk more about it)

I like that :)

Thank you ^^

Many thanks for that Exar Kun. I have imported your code into zslip.com and tweaked it just a little bit.

You should mention cat /proc/cpuinfo for the linux CPU capabilities.

@ Cluster Physik:

Does CPU speed matter in re: to GPU application? or does GPU app not have any significant dependency on CPU? Would one expect a major difference running Pent D or AMD X2 for example vs. a Core2 CPU?

Thx.

Does CPU speed matter in re: to GPU application? or does GPU app not have any significant dependency on CPU? Would one expect a major difference running Pent D or AMD X2 for example vs. a Core2 CPU?

Does CPU speed matter in re: to GPU application? or does GPU app not have any significant dependency on CPU? Would one expect a major difference running Pent D or AMD X2 for example vs. a Core2 CPU?

As not the whole computation is done on the GPU (just 99.5% or so) it may have a small effect. But I would say it is less than one second difference per WU between a P4 and a Core2 for instance.
Quite a bit of this CPU computation time can be hidden when running 2 WUs (or more) concurrently. When one WU waits some milliseconds for the GPU to complete the current task, another WU can calculate its CPU part.
That's actually the reason I've integrated the possibility to run several WUs concurrently. It slightly increases the efficency. For slower CPUs it just pays off more.

When can we expect a 32-bit version? I'm so excited! :D

EDIT: I found both versions here.
http://www.file-upload.net/download-1465057/Milkyway_0.19_ATI.zip.html

EDIT2: This is what I'm getting from the 32-bit app on a 4850.
2/18/2009 10:44:59 PM|Milkyway@home|Starting ps_s86_9_219475_1235014948_0
2/18/2009 10:44:59 PM|Milkyway@home|Starting task ps_s86_9_219475_1235014948_0 using milkyway version 19
2/18/2009 10:45:00 PM|Milkyway@home|Computation for task ps_s86_9_219475_1235014948_0 finished
2/18/2009 10:45:00 PM|Milkyway@home|Output file ps_s86_9_219475_1235014948_0_0 for task ps_s86_9_219475_1235014948_0 absent

Aha, I fixed my Issue, turns out my CCC was slightly corrupted. A reinstallation fixed it.

Edit: DAAAAAAAAAAYUM!!! Nearly a credit/sec of CPU time. Amazing!!

I just saw the new 32 bit version for ATI GPU. It's a good news, but is there any performance difference between 32 and 64 bits ? If the calculation times are the same, I will come back to my Xp 32 bits ^^
____________
Star Wars BOINC Team

Another question (sorry if it's stupid ^^) : why the optimized application, specially the GPU app, are not available on the MW website with a big-red-flashing font ? :) I found this thread only by (extreme) luck, and I think that a lot of people don't know they can optimize their participation. But maybe it will be a charge to heavy for the server ?
____________
Star Wars BOINC Team

Another question (sorry if it's stupid ^^) : why the optimized application, specially the GPU app, are not available on the MW website with a big-red-flashing font ? :) I found this thread only by (extreme) luck, and I think that a lot of people don't know they can optimize their participation. But maybe it will be a charge to heavy for the server ?

I just saw the new 32 bit version for ATI GPU. It's a good news, but is there any performance difference between 32 and 64 bits ?

It's my personal opinion, but I think that a full list of compatible GPUs might be great, because I can read here and there some questions about the compatibility of cards. Maybe the label "HD38x0 or HD48x0" isn't clear for all people. Maybe something like this :

Compatible Cards :

- ATI Radeon HD 3840
- ATI Radeon HD 4850
- ATI Radeon HD 4870

Cluster > do not be offensed by my next question : do you want a simple html/css code to present your different pages ? With some nice tables, colors, etc. ?

If you do that, and it's in these MW forums, I'd be happy to link zslip to reference it.

I am doing a quick inventory on my hosts and to my surprise I have found several with ATI cards.. Not sure if they are compatible...

• HD3850
  
• HD3870
  
• HD4830
  
• HD4850
  
• HD4870
  

are the comaptible cards. The upcoming HD4890 will be running too.

Okay, so far, a speed report.

ATI Radeon HD 4850 512MB. Running Cat 8.12.
Intel Core2Duo E8400, running stock at 3.0Ghz.
Crunched through daily quota of 2000 tasks in around 10 hours.
Credits - 23,000. This equals around 46,000 credits per day without a WU limit, or 2,300 credits/hour.
I can also run GPU-Z while crunching without a problem. Had to use RivaTuner to increase my fan speeds, it was running at 85C.
This is amazing work!

I am doing a quick inventory on my hosts and to my surprise I have found several with ATI cards.. Not sure if they are compatible...

• HD3850
  
• HD3870
  
• HD4830
  
• HD4850
  
• HD4870
  

are the comaptible cards. The upcoming HD4890 will be running too.

I am doing a quick inventory on my hosts and to my surprise I have found several with ATI cards.. Not sure if they are compatible...

• HD3850
  
• HD3870
  
• HD4830
  
• HD4850
  
• HD4870
  

are the comaptible cards. The upcoming HD4890 will be running too.

Will a 4550 work ? I have a couple of those.

Okay, so far, a speed report.

ATI Radeon HD 4850 512MB. Running Cat 8.12.
Intel Core2Duo E8400, running stock at 3.0Ghz.
Crunched through daily quota of 2000 tasks in around 10 hours.
Credits - 23,000. This equals around 46,000 credits per day without a WU limit, or 2,300 credits/hour.
I can also run GPU-Z while crunching without a problem. Had to use RivaTuner to increase my fan speeds, it was running at 85C.
This is amazing work!

Okay, so far, a speed report.

ATI Radeon HD 4850 512MB. Running Cat 8.12.
Intel Core2Duo E8400, running stock at 3.0Ghz.
Crunched through daily quota of 2000 tasks in around 10 hours.
Credits - 23,000. This equals around 46,000 credits per day without a WU limit, or 2,300 credits/hour.
I can also run GPU-Z while crunching without a problem. Had to use RivaTuner to increase my fan speeds, it was running at 85C.
This is amazing work!

Is that one task every third of a minute?

One other newbie ATI /GPU cruncher..

I have done some GPU crunching with Nvidia - so I understand this.


1: Boinc 6.4.5 does not "find" CUDA after upgrading.. I am assuming this is OK?

2: Is there a difference between CAT 8.12 and 9.1 ? I can only find 9.1 on the ATI web site.

3: Other than loading the catalyst driver and copying in the MW opti app/xml and brook.dll is there any other special things I need to be made aware of.

4: I am assuming that the brook.dll is copied into the milkyway projects directory as well.




____________
.

I am doing a quick inventory on my hosts and to my surprise I have found several with ATI cards.. Not sure if they are compatible...

• HD3850
  
• HD3870
  
• HD4830
  
• HD4850
  
• HD4870
  

are the comaptible cards. The upcoming HD4890 will be running too.

One other newbie ATI /GPU cruncher..

I have done some GPU crunching with Nvidia - so I understand this.


1: Boinc 6.4.5 does not "find" CUDA after upgrading.. I am assuming this is OK?

2: Is there a difference between CAT 8.12 and 9.1 ? I can only find 9.1 on the ATI web site.

3: Other than loading the catalyst driver and copying in the MW opti app/xml and brook.dll is there any other special things I need to be made aware of.

4: I am assuming that the brook.dll is copied into the milkyway projects directory as well.

Screenshot, I couldn't resist. I still can't believe what amazing work this is!

Screenshot

Screenshot, I couldn't resist. I still can't believe what amazing work this is!

Screenshot

100% in 6 to 9 seconds. Amazing. I've been looking at the rumours about the ATI Raedon HD 4890 coming out in April with 960 stream processors and 48 texture units. Apparently there will be an overclocked version which will beat the HD 4870 with 25-30 percent performance boost with 1GB GDDR5 memory.

I'm saving my pennies already ;)

Screenshot, I couldn't resist. I still can't believe what amazing work this is!

Screenshot

100% in 6 to 9 seconds. Amazing. I've been looking at the rumours about the ATI Raedon HD 4890 coming out in April with 960 stream processors and 48 texture units. Apparently there will be an overclocked version which will beat the HD 4870 with 25-30 percent performance boost with 1GB GDDR5 memory.

I'm saving my pennies already ;)

Can I have your old one? I just have a Radeon 9600. ;P

I think I have a little problem :

20/02/2009 09:21:02 Milkyway@home Sending scheduler request: Requested by user.
20/02/2009 09:21:02 Milkyway@home Not reporting or requesting tasks
20/02/2009 09:21:07 Milkyway@home Scheduler request completed: got 0 new tasks

For the past hour I tried a few times, no workunit for me :(

I will try the 32 bits version, with an overclocking of my HD4850 (and CPU, too ^^)
____________
Star Wars BOINC Team

It works now, sorry for the previous message ^^

Everything is ok, with the WU limit removal, I try now to keep only the MW project. I have 8 workunits running on my Core2Duo, less than 10 seconds by WU...
____________
Star Wars BOINC Team