Is it just me........

[BurnUnit]

I've been using both the win-nt command line ver. 3.03 and 3.08 back and forth recently. Is it just me or has anybody else noticed that the 3.08 version uses on average from 10-15% more processing time per work unit?

Maybe I ran too short a sampling with 3.08 (around 50 wu's), but I immediately noticed higher numbers on a substantial number of wu's that I would rarely see in 3.03. I guess I could run identical wu's on each version under identical conditions for the definitive answer, but I'm hoping somebody out there may already have done the leg work that I'm not aware of yet. Hence, this post.

Needless to say, at this point I'm staying with 3.03 until I know for sure, it just FEELS better. Does anybody else have any comments or observations on this?

Thanks all!
Burn

[TABwebmaster]

Yes, indeed. I didn't do any more legwork than to download 3.08 and try it. I noticed significantly more time to crunch a wu. I did about 20 on 3.08 and I wasn't satisfied. I'm sticking with 3.03 for the duration of SAH. When BOINC is released I think everyone will be pleased. It looks like it's going to be a great backend to the clients: Astropulse and SAH2 to start.



Mark

[Terry Groff]

Someone in one of the seti NG's posted that it seemed like Ver 3.08 was calling for the date every second. This would significally increase process time.

I went back to 3.03 almost immediately.

As for BOINC, I'm sticking with SAH until they fix the Large file upload problem. It just wastes CPU time to continually try uploading the same WU over and over.

[SuperDave]

is 3.03 still available for download? Got a link?

Thanks.

[TABwebmaster]

Dave,

What kind of friend would I be if I didn't host 3.03 on our server?

Click the following link to get the cmdline 3.03:

http://www.seti-teamartbell.com/setiath ... mdline.exe

Mark