I normally like to try new versions reasonably quickly, but in the case of 0.20b I decided to wait until it was complete or help was requested in testing it.

As far as I am aware version 0.20 remains the recommended version and the last message regarding the use of 0.20b stated the following:

....."But before you try to download it, it's not complete and not correctly set up yet. So it will probably not work right now. Be a little bit more patient"...... from Message 32135

I must admit I wasn't sure about the status of 0.25b, so I just went ahead and installed it on all my PCs with ATIs, and then proceded to find every one of them failing all WUs with compute errors. I usually test all win_32 op apps before putting them up on www.brilliantsite.com, which I did in this case and 0.20b won't be going up there just yet.

maybe i misinterpreted this

v0.20b is working fine here and for many others, most likely it isn't set up correctly.

maybe i misinterpreted this

No that was due to me making a mistake and misunderstanding Cluster Physik's post about 0.20b. I incorrectly thought that we really weren't supposed to be using 0.20b yet and when I saw some were getting errors I mistakenly thought that was the reason.

I just installed 0.20b myself after reading all of the readme file again to make certain I was using the correct version to suit my card's Catalyst 9.4. It works very well for me and is 1.5-2 seconds faster with no parameters used with my usual 7 cores used for CPU projects. Runs a bit hotter so I'm back to one task at a time during the day, but should be able to run 2 concurrently at night.

What's really excellent is that I experimented with using the full 8 cores for my CPU project with the b-1 parameter and I still get the same improved MilkyWay runtime. Previously I always had to leave a CPU core free to get a faster MilkyWay performance. So I don't just gain a speed improvement in MilkyWay but an extra core for CPU projects as well. That's very pleasing. It's not all about the credit either because the Ibercivis neurosim I'm currently doing on the CPU grants about 12 credits per hour on my i7, which is so low it may as well be non-existent, but an extra core available gets a bit more science done. (Don't know what's up with those Spanish project admins, they must prefer a low number of active contributors.)

Many thanks to Cluster Physik for this new version, it's terrific.

Weird thing for me is I saw a 25 second speed-up on my 4830.

I was running around 105 seconds per unit and am now running around 80 seconds. This is with running 2 units at a time and using Catalyst 9.2 without the CCC on Win 7-64.

---

One hour ago i tried the new 0.20b application. Sorry, not even one wu created results, all of them ended up with errors, right after starting them. Application is still reported as milkyway 0.20, without b.

If it doesn't report being 0.20b it simply is not 0.20b. Are you sure you downloaded the correct version? Could you post the task details of a failed WU?

Hi,
i tried to reinstall 0.20b The result can be seen here: http://festplatte.aon.at/a/filemanager.get/11746831/0.20b.JPG
I switched back to 0.20, that works. But if anyone has an idea how to fix it, i would prefer to use the newer app.

But anyway, other posts say, this application is faulty and it has been removed from the brilliantsite.com.
Let's wait for the next version 0.25 ...

No, it's not faulty, actually some issues where fixed. I'm quite sure you simply used the wrong version ;)

But nobody is forcing you to use 0.20b of course. Maybe I should wait someone builds v0.25 ;)

Hihih how about you ;)

But seriously when something goes wrong with the app i found with patients and reading CP his instructions ( i know they are complex ;) )
All should work as planned and since app_info does work fine, i would not recommend people to do without.
Simply the extra parameters possible and so on which you can't when not using app_info makes it worth while
The funny thing is after testing almost all 10.XX versions i went back to the good old 6.XX versions since they run smooth and without problems for me.

---

Its new, its relative fast... my new bicycle

Weird thing for me is I saw a 25 second speed-up on my 4830.

I was running around 105 seconds per unit and am now running around 80 seconds. This is with running 2 units at a time and using Catalyst 9.2 without the CCC on Win 7-64.

I see a similar increase on my 4770 it simply runs better now, although i changed it to 3 units to see if this gains overall performance.
After i set this card to 3 its constant under load at 98% while with 2 units it dropped at constant intervals to 45-50%
So i hear now the constant whirring of the cooler :D, time will tell if it will be better for the outcome.
As previously stated some applications seem to make the ati driver crash, so its not the fault of CP his program but my guess is some issue in the driver
(w7 cat 9.9 win32) the w7 x64 version here seems to run much more stable.
As a sidenote i have no problems running CCC nor with rivatuner or can even game with my X64 machine while running MW albeit slow :D

---

Its new, its relative fast... my new bicycle

i tried to reinstall 0.20b The result can be seen here: http://festplatte.aon.at/a/filemanager.get/11746831/0.20b.JPG
I switched back to 0.20, that works. But if anyone has an idea how to fix it, i would prefer to use the newer app.

you've installed the 0.20b_ATI version of the app, do you have the ati version of the dlls??

"Cluster Physik" wrote:

You are running WindowsXP, therefore Catalyst 8.12 is recommended. That also means you have to use the Win32_amd or Win32_SSE2_amd version (not the ones with "_ati" at the end). That is all said in the readme by the way.

I suspect some of the people having trouble with v0.20b may be running the wrong app for their catalyst version, so I thought this could do with highlighting :)

i tried to reinstall 0.20b The result can be seen here: http://festplatte.aon.at/a/filemanager.get/11746831/0.20b.JPG
I switched back to 0.20, that works. But if anyone has an idea how to fix it, i would prefer to use the newer app.

you've installed the 0.20b_ATI version of the app, do you have the ati version of the dlls??

Hmm, i'm running astronomy_0.20_ATI_SSE2, it works fine. Do i need different dll's?
System requirements:

- x86 CPU running Win32 or Win64 (Windows7 may require a workaround, see below)
Q9300, XPpro SP3
- ATI Radeon HD3800, HD4700/HD4800 and HD5800 series or
ATI Firestream 9100/9200 series card
HD4850
- correctly installed Catalyst 8.12 or up (see below which application version you need)
8.12
- BOINC client 6.10.x or higher for ATI support, but works also with older versions,
Boinc 6.10.11

Usage:

- Read this file carefully! maybe i need more practice in english ..
- If problems arise, read it again ;)
- installed like any optimized app using the anonymous platform mechanism i did it like 0.20 app - stop BOINC completely (stop the service if you are running a protected mode installation) i did that
- copy the content of the appropriate folder of the zip file (3 files) to your Milkyway
folder (location may vary) i did that too
- restart BOINC
- enjoy the incredible crunching speed here was something wrong.

Actual settings can be seen her: http://festplatte.aon.at/a/filemanager.get/11784532/0.20.JPG

Did i miss something?

Yes I think you missed something. "(see below which application version you need)"

You need to read all of the Readme file. Near the end ("below") you will find a section "Note on Catalyst driver versions:"
Although I suggest you read all of it yourself or read it again to improve your understanding I will mention it contains: "For Catalyst 8.12 and 9.1 use the application variants with "_amd.exe" at the end"

"Cluster Physik" wrote:

You are running WindowsXP, therefore Catalyst 8.12 is recommended. That also means you have to use the Win32_amd or Win32_SSE2_amd version (not the ones with "_ati" at the end). That is all said in the readme by the way.

I suspect some of the people having trouble with v0.20b may be running the wrong app for their catalyst version, so I thought this could do with highlighting :)

Duhhhh, this is all becoming double Dutch. I don't have any AMD processors, but I >do< have ATI cards, so I'm supposed to realize that I should pick the AMD version?

I started up zslip/brilliantsite.com to try and make it easier for people to find and decide which one to use. But you got me here CP ;)

---

i tried to reinstall 0.20b The result can be seen here: http://festplatte.aon.at/a/filemanager.get/11746831/0.20b.JPG
I switched back to 0.20, that works. But if anyone has an idea how to fix it, i would prefer to use the newer app.

you've installed the 0.20b_ATI version of the app, do you have the ati version of the dlls??

Hmm, i'm running astronomy_0.20_ATI_SSE2, it works fine. Do i need different dll's?
Actual settings can be seen her: http://festplatte.aon.at/a/filemanager.get/11784532/0.20.JPG
Did i miss something?

You haven't missed anything for how you're running now with v0.20, all looks to be working fine :)

But before, when you tried version 0.20b, you installed the ati version of the app. In this case ati means you have 3 ati*.dll in the system32 directory and i think that's where you went wrong.
I suspect you have 3 amd*.dll files instead, which would mean you should install the amd version of the app

Yes I think you missed something. "(see below which application version you need)"

You need to read all of the Readme file. Near the end ("below") you will find a section "Note on Catalyst driver versions:"
Although I suggest you read all of it yourself or read it again to improve your understanding I will mention it contains: "For Catalyst 8.12 and 9.1 use the application variants with "_amd.exe" at the end"

Aha. I kept in mind, that this needs to be done with catalyst 9.2 and above, i have done this a lot of times.
But after downgrading it worked fine and i did no longer thik about that. But anyway, i did not read the readme till the end.

Shame on me

Its working now. Thanks to you all!