Ive created new apps for x86 and PPC for the Mac which should be more optimized. However, i'm not sure if they run without error so if you encounter any errors,please post here, and i will put the old version back up until i resolve them.

---

Dave Przybylo
MilkyWay@home Developer
Department of Computer Science
Rensselaer Polytechnic Institute

Awesome. Awesome to the max. -- '80s Guy. Fututama

I'll let you know if there are any issues as soon as I burn through this SIMAP queue.

---

Runs fine on a G5. Crunch time reduced to 71% of previous app. Thanks Dave. Whatever you did, keep doing it!

Nice work!

Early results show my G4s running v1.22 ten to twenty percent faster than previously but, like Harry’s, my G5 appears to have gained considerably more, nearly thirty percent.

---

Nice work!

Early results show my G4s running v1.22 ten to twenty percent faster than previously but, like Harry’s, my G5 appears to have gained considerably more, nearly thirty percent.

Version 1.22 is optimized to take full advantage of G5 extensions that are not present in the G4. :)

---

Dave Przybylo
MilkyWay@home Developer
Department of Computer Science
Rensselaer Polytechnic Institute

Hi,

No errors here.
- 1 Macbook pro Core duo 2Ghz, 2GB ram, Os X 10.4.11.
- 1 iMac Core 2 duo 2.16Ghz, 1GB ram, Os X 10.5.2.

Speed increased but not much compared to PPCs.
The mbp is 5 seconds faster than before on a WU.
The iMac is 15 secondes faster than before on a WU.

But it's still strange. The iMac should be faster than the mbp, and is not ! (13mn40s for mbp, 14mn for the iMac).

Regards,

Hurrican
Member of L'Alliance Francophone.

(13mn40s for mbp, 14mn for the iMac).

Hmm, both of those times seem very long for those types of processor. Have you tested either of them on Windows with bootcamp?

Sorry for my ugly english :
I'm from the Alliance Francophone !

Thanks a lot for trying improve the app on Macs.
It seems that the new one is perfect for PPC, but i've notived no improvement on Mac Intel.

At the moment i'm running boinc on Ubuntu 64bits in a virtual machine : 5min/WU
But if i try running boinc on OS X (leopard) with the new app : 12min/WU

I suppose that the app for OS X is not 64bits but why? My Core 2 Duo 2,4Ghz is 64bits and leopard too ... (i think that leopard is both 32 et 64bits depending of the CPU)...

By the way thanks a lot for what you does for mac! Keep on :p
And next time try doing something for mac intels et why not trying to do it in 64bits :) It'll be perfect :p

See you soon :)

(13mn40s for mbp, 14mn for the iMac).

Hmm, both of those times seem very long for those types of processor. Have you tested either of them on Windows with bootcamp?

Not done, but I'll do it as soon as possible ! ;)

Hi fellow crunchers of Macgé,

You will probably see that under Windows or Linux your Mac crunches 30% faster.

That's what I'm noticing when I compare my Dell Latitude with my Mac Mini (very same CPU).

http://forum.macbidouille.com/index.php?s=8987193da7a5a89a366e61a99d549206&showtopic=248789&st=270

Read you on MacG one of these times....

;-)

Ok.. Did some tests on my mbp (CD 2Ghz, 2GB ram, 90% cpu limit, one task per core).
On each result I indicate the time for one core to calculate one WU.
As the processor has two cores, of course the score of the processor is twice faster (two WU calculated at the same time).

On Os X 10.4.1 : about 13mn30s for a WU.
On XP SP2 via BootCamp : about 10mn20s for a WU.
On W2000proSP4 via Parallels Virtual Machine : about 11mn for a WU.

As you can see, the OS X client is not enough optimized. It's obvious, as running BOINC through Parallels 3.0 (Windows 2000 on MacOs X virtualization), is faster than Os X alone ! :)

Yep,
But i am sure that one day, milkyway'll be perfect for mac users, isn't it ? :)