The new win-cmd 3.08 seti-client is out.

Toggle sidebar Toggle sidebar
Assimilator1

Assimilator1

Elite Member
Nov 4, 1999
24,125
508
126
Didn't think I'd see another version of S@H1! ,I guess that's one of the things thats delayed BOINC?

Btw I take it this isn't an enforced upgrade? (seeing as its only a minor security risk......apparently)
I wonder if they've sneaked anything else in whilst they were at it!
rolleye.gif


I'm going to try it out on my rigs
 
R

Robor

Elite Member
Oct 9, 1999
16,979
0
76
I'm interested in seeing some benchmarks on 3.03 vs. 3.08. If there's no performance increase and this upgrade isn't forced I'll be leaving my fleet alone. However, if there's a decent performance bump I'll have to look at upgrading my faster rigs. :)
 
S

Spacehead

Lifer
Jun 2, 2002
13,067
9,858
136
Well, i just downloaded the new CLI & have the Ars benchmark WU running. I'll let you know the processing time when it finishes, hopefullly in about 3.5 hours :)

v3.03 time was 3:30:23
 
lane42

lane42

Diamond Member
Sep 3, 2000
5,721
624
126
Is this what the "New Client" Button is for in Seti Driver ?
 
S

Spacehead

Lifer
Jun 2, 2002
13,067
9,858
136
SETI Driver/CLI v3.03 - 3:30:23 :)
SETI Driver/CLI v3.08 - 3:51:43 :frown:

GUI v3.07 - 4:41:43
rolleye.gif


All WU's are the Ars BM(0.417) run on my XP Home, XP1800, PC2700 RAM computer.
 
Assimilator1

Assimilator1

Elite Member
Nov 4, 1999
24,125
508
126
Lane42
Yep!

Space head
I get almost exactly the same results as you:(
Though I haven't used the TLC WUs I've been comparing 0.41x WU times on my XP1800 @1.88GHz (SDR ,150MHz FSB/RAM).

Here's what I get with v3.03

2003 Apr 02 17:25 22.109,17.21 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Apr 02 13:50 14.611,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:36
2003 Apr 02 10:14 8.842,17.23 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Apr 02 06:39 16.651,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Apr 02 03:04 18.617,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:34
2003 Mar 31 01:37 19.822,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:33
2003 Mar 30 22:04 15.791,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:33
2003 Mar 30 18:30 26.213,18.01 0.417 Assimilator1 Ath XP1800 @1.88GHz 3:34
2003 Mar 30 14:56 24.957,18.01 0.417 Assimilator1 Ath XP1800 @1.88GHz 3:34
2003 Mar 30 11:22 7.052,13.43 0.427 Assimilator1 Ath XP1800 @1.88GHz 3:33
2003 Mar 30 07:49 9.852,17.23 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:33
2003 Mar 30 04:16 19.298,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:33
2003 Mar 29 22:15 20.884,16.01 0.422 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Mar 29 12:14 9.87,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Mar 29 08:39 24.89,18.01 0.417 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Mar 29 05:03 24.799,18.01 0.417 Assimilator1 Ath XP1800 @1.88GHz 3:36
2003 Mar 29 01:27 1.419,21.84 0.484 Assimilator1 Ath XP1800 @1.88GHz 3:34
2003 Mar 28 21:53 8.422,22.75 0.469 Assimilator1 Ath XP1800 @1.88GHz 3:46

So thats about 3hrs 35mins for 0.417 WUs

Here's what I've got so far from v3.08 CLi

2003 Apr 08 15:24 21.868,16.18 0.421 Assimilator1 Ath XP1800 @1.88GHz v3.08 3:59
2003 Apr 08 11:26 0.945,18.02 0.417 Assimilator1 Ath XP1800 @1.88GHz v3.08 3:59
2003 Apr 08 07:27 19.812,16.18 0.421 HOMER ....................................................... 3:58

Looks like its adding nearly 1/2hr for average WUs :|:disgust: ,I haven't heard anything about any added to science to the client so why this damn performance hit!?
rolleye.gif

The current WU this rig is on is about 1.5 AR WU ,its about 65% done & SETIspy is predicting about 3hrs 13mins for the WU ,if true then the VHAR WUs don't seem to be affected but the average ones certainly seem to be!.
I'm going to leave it running v3.08 for a while longer just incase some odd 'effect' occured.

But so far I'd say stay away from v3.08 if you value best performance the most

[update]
SETISpy held up true

2003 Apr 08 18:44 14.457,9.32 1.495 Assimilator1 Ath XP1800 @1.88GHz v3.08 3:15
 
R

Rattledagger

Elite Member
Feb 5, 2001
2,989
18
81
As for speed, I haven't much data yet, but for normal wu the new client seems to use 5-10% longer time.
For high angle-range, it seems to be like before, or a little bit faster...
For VLAR, I've only crunched 1 wu, but this was much faster than before, using the same time as a normal wu with v3.03. :)

I'm sticking with v3.08. I've used worse clients than this... like v2.66. ;)
 
C

CraigRT

Lifer
Jun 16, 2000
31,440
5
0
God, I'd reaaaally like to leave my fleet alone... not sure how the service installs will take to changing the client on them.... :eek:

hopefully I can still use 3.03
 
Assimilator1

Assimilator1

Elite Member
Nov 4, 1999
24,125
508
126
Well I'm switching back to v3.03 on my 2nd rig.
I got another result from it on v3.08

2003 Apr 08 22:45 13.551,8.1 0.434 Assimilator1 Ath XP1800 @1.88GHz v3.08 3:59

Still just under 1/2hr slower ,I'm going to use SETIDrivers 'new client' to switch back to v3.03 so they'll be 1 more result on v3.08 & then its back to v3.03.
That makes v3.08 about 12% slower than V3.03 for average AR WUs

VLAR WUs aren't a problem to me as I have 'VLAR routing' enabled on my home Q & it runs Win2k where VLAR WUs are hardly any slower.

RD
I'd be interested to see more VLAR WU times from you on v3.08:)
 
Confused

Confused

Elite Member
Nov 13, 2000
14,166
0
0
Posted the stuff below a lot of times over in the S@H message boards regaring 3.08 :) Also asked in here whether there is any new science! ;) Will wait for answers!


We have done some tests on Team Anandtech, and have found 3.08 CLI to be much slower than the 3.03 CLI.

Here are some examples:

<Quote>
SETI Driver/CLI v3.03 - 3:30:23
SETI Driver/CLI v3.08 - 3:51:43

GUI v3.07 - 4:41:43

All WU's are the Ars BM(0.417) run on my XP Home, XP1800, PC2700 RAM computer.
</Quote>

That's one person who noticed an increase in WU times, that's using the Ars Technica 0.417 WU.

<Quote>
Here's what I get with v3.03

2003 Apr 02 17:25 22.109,17.21 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Apr 02 13:50 14.611,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:36
2003 Apr 02 10:14 8.842,17.23 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Apr 02 06:39 16.651,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Apr 02 03:04 18.617,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:34
2003 Mar 31 01:37 19.822,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:33
2003 Mar 30 22:04 15.791,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:33
2003 Mar 30 18:30 26.213,18.01 0.417 Assimilator1 Ath XP1800 @1.88GHz 3:34
2003 Mar 30 14:56 24.957,18.01 0.417 Assimilator1 Ath XP1800 @1.88GHz 3:34
2003 Mar 30 11:22 7.052,13.43 0.427 Assimilator1 Ath XP1800 @1.88GHz 3:33
2003 Mar 30 07:49 9.852,17.23 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:33
2003 Mar 30 04:16 19.298,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:33
2003 Mar 29 22:15 20.884,16.01 0.422 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Mar 29 12:14 9.87,17.15 0.419 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Mar 29 08:39 24.89,18.01 0.417 Assimilator1 Ath XP1800 @1.88GHz 3:35
2003 Mar 29 05:03 24.799,18.01 0.417 Assimilator1 Ath XP1800 @1.88GHz 3:36
2003 Mar 29 01:27 1.419,21.84 0.484 Assimilator1 Ath XP1800 @1.88GHz 3:34
2003 Mar 28 21:53 8.422,22.75 0.469 Assimilator1 Ath XP1800 @1.88GHz 3:46

So thats about 3hrs 35mins for 0.417 WUs

Here's what I've got so far from v3.08 CLi

2003 Apr 08 15:24 21.868,16.18 0.421 Assimilator1 Ath XP1800 @1.88GHz v3.08 3:59
2003 Apr 08 11:26 0.945,18.02 0.417 Assimilator1 Ath XP1800 @1.88GHz v3.08 3:59
2003 Apr 08 07:27 19.812,16.18 0.421 HOMER....................................................... 3:58
</Quote>

That's about 1/2hr added to an average WU! So, why the increased WU time, if there's no new science added, and it's only a security update?

This isn't good, and these numbers prove it.


There is no way i'm changing from 3.03 to 3.08 until we get a decent explanation as to why, for no known added science, it's adding so much time.


Confused
Team Anandtech
Click to expand...


Garry
 
lane42

lane42

Diamond Member
Sep 3, 2000
5,721
624
126
Most all my machines just run Seti only. What's the worst that can happen, ill just format and reload windows and Seti again. Come get me Hackers :p
 
C

CraigRT

Lifer
Jun 16, 2000
31,440
5
0
Originally posted by: lane42
Most all my machines just run Seti only. What's the worst that can happen, ill just format and reload windows and Seti again. Come get me Hackers :p
Click to expand...

hahaha yeah same here!
 
Assimilator1

Assimilator1

Elite Member
Nov 4, 1999
24,125
508
126
Well I've switched my other rig back to v3.03 & its back down to ~3.5hrs/WU:)

One last v3.08 WU time

2003 Apr 09 07:24 22.435,21.32 0.409 Assimilator1 Ath XP1800 @1.88GHz v3.08 4:07 :disgust:

I'd still be interested to see WU times on VLAR WUs though.

Nice 1 Garry ,though I won't hold my breath for an answer;)
 
R

Rattledagger

Elite Member
Feb 5, 2001
2,989
18
81
Well, here is the rest of my VLAR's, and some old ones...:
box1 - 0,078 - 3h37m - v3.08
box1 - 0,08 -- 4h15m - v3.03
box1 - 0,057 - 4h20m - v3.03
box1 - 0,075 - 4h22m - v3.03
box1 - 0,069 - 4h26m - v3.03
box2 - 0,053 - 4h26m - v3.08
box2 - 0,056 - 4h27m - v3.08
box2 - 0,09 --- 4h31m - v3.08
box2 - 0,073 - 5h03m - v3.03
box2 - 0,06 -- 5h06m - v3.03

And some averages...
box1 - 288 wu -- 3h39m - v3.03
box1 -- 21 wu -- 3h57m - v3.08
box2 - 562 wu -- 4h16m - v3.03
box2 -- 19 wu -- 4h34m - v3.08
box3 - 223 wu -- 9h58m - v3.03
box3 --- 7 wu - 10h24m - v3.08

OS: win2k

Atleast for VLAR's under 0,1 the new client is much faster. :)
 
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom