I want to go faster (in Seti!)

[agallag]

So, how do you figure out your time per work unit? I can't see anywhere that it lists your time for each individual wu, just your overall average.

On a side note, why is the text in Netscape 4.72 for linux so frickin small, and how do you make it bigger?

Andrew

[Ees]

Rags,

What can I say? This is silly. Please come back to seti@murc.

The way I see it, there are two possibilities:

- You were using a patched seti-client. Just don't do that, get back to murc and show that you're still in the top-20 fastest (I hardly doubt that you are, considering your skills).
- You weren't; it's simply that your setup is godawful fast. In that case, just say so, and I offer you my humblest apologies. I'd still love to know how you did that though.

If my suspicion was wrong, I'll be the first one to eat dust, but just bailing out seems like a pretty poor solution to me.

Martin

[Ees]

Mr D. Ackerot: I see you listed with 131 WUs. Where do you see the 18?

Agallag: You're correct, it only shows average times. I'm able to provide the table above (with "last weeks" fastest times) because I keep a database with all seti@murc info. It's still an average though. Some seti helper-programs will be able to show you individual WU-times.

About the fonts: check out the Linux Font HOWTO. Short summary: get, install, and use some better fonts than the ones that come with Linux.

Martin

[Rags]

No, it isn't silly. Before MURC had a SETI team, I was on team anandtech, and I was never flooded with emails and posts accusing me of cheating. Now, that you guys have brought up the issue of a patch, I have had a stream of emails asking me for the patch. Well, if you want your WU's to be quicker use a better OS, fast/good Hardware, set it up to it's fastest and let it rip. My only comments in this thread were offers of assurance that with the original GUI version of the SETI program, a P3 550E@550 is, indeed, faster than my Katmai 450@560, by about 45 min. at that. I could care less if you guys believe me or not. This doesn't matter to me so much that I will make a deal of it. I am just doing the best thing. Moving to another team/or quit using SETI all together, and being left alone. If there was anything I could do to help you get your WU's faster, I would post some tips for you. But, my config. is quite a bit different than all of yours. End of story.

Rags

[jsb]

Okej Rags, if you are not using any illegal patches, what is the problem? Just say so, but if you are stop and continue to run a normal seti, same as everybody else.
Leaving because someone asked what config you have is a bit strange to me.
Stefan

[Rags]

Rags
Until you tell us how you cheated, I can hardly take it seriously that you complete your WUs in about 1.5 hours on PIII 550 machines.

I'd hardly expect to get WU times below two hours from anything short of an Alpha EV6...

Rags,
there is no way that you could achieve those times with a P3. The Compaq Washington benchmark centre are averaging 1 hour 39 minutes, and they probably have some pretty funky machines.

But if I read between the lines correctly, Rags' times are less than official... I sure hope that we can keep Team Matrox Users patch-free.

Sort of cheating, Isn’t it

Hmmmm...sounds pretty accusing doesn't it. Well you should see the emails I have gotten in the past few days. I never thought that people would be so upset that a P3 could be so fast in SETI, I guess I was wrong. I will not go for it. I am sorry. There was another thread where I was prodded a bit about my system, and I explained there vaguely what my system is like, and I was left alone. There are just some things I won't divulge in, and the setup my SETI WU's come from is one of them. I can tell you that I am using Win2k as my OS, and am running the original GUI version. The only problem with this is that I have to manually connect myself, but my output is still greater. I have considered (and it can be done ) fixing the GUI version so it will use DUN under win2k properly, but have chosen to keep the seticlient un-touched, the program itself is untouched. Why can't people leave it at that? Maybe the remaining WU's on the original SETI client are just easier to crunch, I dunno.

Rags

[jms]

Rags,
I apologize if I've sounded accusing. That was NOT my intention. I was just getting curious, because you are obviously churning out WUs like there is no tomorrow.
Please reconsider, and come back.
Jan M.


[This message has been edited by jms (edited 12 May 2000).]

[Ees]

Rags,

I still think you're overreacting.

What did you expect to happen when you do your WUs on a 550e in less than 1.5 hours? People are going to be very, very curious.
I don't give a rat's ass that a 550e is faster than an overclocked Katmai; I'll take it for fact anyday. I was just stupified by the fact that you get it to churn out a unit in 1h 16m.

Your vagueness about your setup didn't help too much either. The statement that you're running the "original GUI version" is new. You could have stated something similar anywhere in this thread, but you didn't. And the way I read it, you *purposely* didn't. That's where I got the idea that you might be running a patched client.

So you're achieving those results entirely official. I apologize for mentioning a patch. Please note however that the full quote should have been:

I apologize up front if I misunderstand, but I sure hope that we can keep Team Matrox Users patch-free.

(Emphasis added.)

Best of luck, and happy seti-ing (anywhere, but hopefully for seti@murc),

Martin

[Jorden]

I've done some testing over the past weekend, since I couldn't connect to the internet anyway. Crunched one and the same WU 4 times (btw, I only sent it in once, before I get comments on that ), and here's my outcome (from memory, as I'm at work now, will edit it if I'm wrong when I'm home later today):

P3-450, non OC, 256Mb RAM, P2B
HDD's used: 10.8G Fujitsu 5400rpm, 9.1G Seagate 7200rpm.

I've tested with the APM on and off.
Fujitsu with APM on: approximately 6 hours 20 minutes.
Fujitsu with APM off: approximately 7 hours 15 minutes.
Seagate with APM on: approximately 6 hours 0 minutes.
Seagate with APM off: approximately 6 hours 20 minutes.

Leaving my system idle, with GUI 1.06 in the system tray only, and enabling my HDD to power down after 1 minute, speeds up Seti for some unknown reason.

I agree with Matt in believing the original GUI works faster. The new one(s) crunch other data as well, which could explain the difference between the GUI's. And as long as Seti still enables me to receive data using the 1.06 GUI, I'll be using that one.

The last witchhunt so to say costs us dearly when people wanted to know who Charlotorn (or something alike) was. I'd say we stop witchhunting, before Manex for example is the next one on your hunt-list.

Matt, you know what I wrote in my email, and I meant it Just call me up to go grab a beer

Jord.

[edit]Checked the times and they were correct, just the APS should've been APM )[/edit]

[This message has been edited by Jorden (edited 12 May 2000).]

[Guru]

Rags I have just tried the old setiathome_win_1_06 client under wk2 and it takes over 8h to complete a unit for me (the setiathome-2.4.i386-winnt-cmdline takes 6h 30min!) what am i doing wrong? Is this the same ver. as you are using? Oh by the way if I remember your system specs right the are (correct me if I am wrong) :
Intel P3 550e @ 733 825MHz using a 150MHz FSB right that would make it a fast seti cruncher, 384MB Cas 2 PC133 Ram!

------------------
PIII450@558, ABIT BX6-2, 256RAM, G400MAX, SBLIVE, HOTROD-UDMA66
Join the MURC SETI team @ <A HREF="http://setiathome.ssl.berkeley.edu/cgi-bin/cgi?cmd=team_join_form&id=25678


" TARGET=_blank>http://setiathome.ssl.berkeley.edu/cgi-bin/cgi?cmd=team_join_form&id=25678


</A>

[Assimilator1]

I've seen some conflicting reports regarding GUI v1.06 vs GUI v2.0x as well

On my Celeron 550 PC the v2. cut my times by about 2 hrs on average!!! :Q
But a fellow team member who had a PII @ 504MHz found it made virtually no difference!

Strange! ,if I remember rightly most of our team found v2. is a bit quicker.
I guess the best thing to do is to try both on your own systems to find out which is quicker.
Maybe the differences are to do with cache size & speed?

Further thoughts anyone?

[Jorden]

up

[Pace]

.luaP

?crum detartlifni slairtserret-artxe evaH

[CHHAS]

.ecaP

!noitcnuflam rojam a sah draobyek ruoy kniht I

oot enim ,yeH

------------------
Katmai 500@560, 196 MB ram, G400 16SH , Maxtor DM 40+ 30GB, Maxtor 4320 13 GB

[Pace]

?draobyek ym htiw gnorw si tahw dnA ?taht gniod uoy era woH !sdrawkcab gnipyt era uoy ,yeH There's gnihton wrong htiw my draobyek!

.luaP