• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

SETI 3.03 is slower, also CLI available..

Yes, it is way slower. 🙁
But we get much more science for the bandwidth dollars spent at Berkeley. 🙂
I think it also increases the odds of finding ET. 😀
 
Slower but more science? Oh well, that's the point. I don't mind the client being slower as long as we're all using the same client.

Rob
 
Same here ,as long as we're all on a level playing field.
Mind you I do feel sorry for Pentium & other lower clocked S7 owners
 


<< Mind you I do feel sorry for Pentium &amp; other lower clocked S7 owners >>


Forget them! 🙁 I'm runing the TLC bench using the 3.03 CLI (via WINE on Linux) on my P3 600@800 and am kissing my 3:41 hr WUs goodbye... 🙁

Looks like it might complete in 6 hours or more! :|
 
My 233MMX is at 51% after 25 Hrs.
It used to do a WU in about 23-24 Hrs.

My Athlon 700 took about 10 Hrs.
It used to do a WU in about 7-8 hrs.
EDIT:The current unit looks like it will take about 12 Hrs.
 
That's a good point Curt, we've still got the same amount of WU's to make up to catch the teams ahead of us but it's taking us longer to do WU's.

Rob
 
I ran the Ars benchmark (AR = ~6.7) using the 3.03 CLI. System is the one in my sig running WINE w/nt40 file system option. Time:

6:14:06 hrs :Q

Previous time obtained for same benchmark unit on same system using WINE w/win95 file system (which is just a little bit slower than the nt40 one by a few minutes generally):

3:42:11 hrs

So I have kissed my sub-4 hr WU times goodbye... 🙁

😉

 
Here's a big suggestion for the SETI Client programmers:

MAKE THE DARN THING SMP COMPLIANT!!!!!!!!! :|

Arghh.. that is the single biggest gripe I have about the client, and the single biggest thing going for the RC5/OGR client.

Sheeesh! 😕
 
Poof
Ouch! thats a big hit to WU times.

Curt Oien

Damn it! ,hadn't thought of that 🙁.Don't hold your breathe on 2WU's=1 😉
 

is the 3.03 slower than the 3.0 client ?

i knew that it could be slower than the 2.x version but is it
30% slower thant the 3.0 client?
 
Nooooooooooooo! Well, I'm not installing 3.03 CLI until they start rejecting my CLI 3.0 results. Plus, it's a major pain in the a$$ for me to go around updating my herd with new clients. Hell, I just got done putting the v3.0 CLI on!!

networkman - I hear ya bro. Not only SMP, they need to take advantage of SSE and 3DNow. What a waste. I'm sure there is a few thousand programmers out there willing to optimize the clients at no charge. Look how fast Intel and AMD modified that FlasK program for Tom's Hardware.
 
Cory - my 3.x benchmarks were lost in my earlier post... 😉 But using the TLC bench work unit, I got:

3:42:11 hrs for the old 3.00 CLI

6:14:06 hrs for the new 3.03 CLI

This is on an overclocked PIII (600@800).

Assimilator1 - yeah... it's a MAJOR hit. I think this client will about obsolete some of my older machines that already take 20+ hrs/WU. I also have a few Hall of Shamers that normally take in excess of 35-40hrs/WU with the 3.00 clients and would probably double the times with 3.03. I'm afraid at what they'll do to my alpha, since I've yet to see a 3.x client for linux on alpha...

Only way I'll see a sub 4hr high angle range WU result again would be to get to 1Ghz somehow... :Q

Hmmmm.... OC'd Duron 700 anyone? 😉


 
At least the hit is due to increased science, which after all is what seti is all about in the first place.🙂
 
'increase science'... if it was THAT important, they wouldn't have left it out in v3.0. personally, i would be happier with larger packets.
 
With all due respect TuffGuy, what good would larger packets do? I understand this would decrease the number of times the seti server would get hit but the total amount of bandwith needed would remain the same.

Seti already has more than enough computing power behind it to crunch all the data they can give us and then some. So wouldn't the most efficient thing for them to do be to increase the calculations the client is doing? It will take us longer to do a wu but they'll be getting more data out of each wu and their servers won't be hit as often and alot of bandwith will be freed up. To me their actions seem logical enough. They are after all a nonprofit organization with very limited funding &amp; staffing.

Not to mention their being forced to throttle back their bandwith by the university. I'm too lazy to look for the link so maybe someone else can post it, but anyway the university is tired of them using so much bandwith (somewhere around 30% of all Berkley's traffic is due to seti).
 
i was being facetious, but:

i just think that it's better to do a &quot;rough&quot; analysis and then if anything suspicious turns up, do a fine analysis. i'm not sure how accurate v3.0 is compared to v3.03. my theory is that if packets are larger, and people connect less often, the amount of simultaneous hits will decrease. they only started experiencing bandwidth problems when the new clients came out. it did not occur with v2.04. so if you increase the packets to where the times to process is similar to that of the old clients, the stress on the servers should also decrease.

increasing the amount of calculations is more efficient though since they only need to modify the new client and not the whole packet structure.

 
Blade 47

Dang! ,I agree with you again! 😉

Tuffguy

The only trouble with doing a rough analysis is that it might miss THE signal altogether ,so it would never be passed on for finer analysis.
Ref making the WU's to about the same hrs/Wu as v2.04 ,don't forget that there's a whole load more computors doing SETI now since it was 1st released ,not to mention people upgrading.

We'll just have to live with the fact that they have limited resources ,so they need to slow everyone down to compensate 🙁 .but at least the search is better 🙂.

BTW what did you change your SETI name to?
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top