F@H - what a bunch of crap

[BCinSC]

Turned in 6 WUs today for a whopping 69 points. First 5 were only worth 28; last was 41. This is PATHETIC!

 

[ProviaFan]

What machines, what kind of WUs, what core, etc.?

 

[BCinSC]

Heck, I don't know which of the heap turned in today, but at least a P3. Probably 78 core.

 

[BCinSC]

Got a bunch that will turn in with next couple hours. I hope the point scale is more kind or SETI will be getting a boost instead.

 

[GLeeM]

This from:
http://forum.folding-community.org/viewtopic.php?t=7954


Posted: Wed May 12, 2004 10:28 pm Post subject: Recent stats problem reports

--------------------------------------------------------------------------------

Hi Folders,

We've noticed a number of reports that some credits were not properly given for WU's returned to server .143 (p6XX projects). We are currently looking into this matter, and hope to have it resolved soon. Feel free to post missing WU info (proj,run,clone,gen) here and we will add it to our list. Thanks for your patience.

 

[GLeeM]

Also I noticed a new version 78 core on 5-10

 

[BCinSC]

Stopped client using 78 core v1.56 2/2/04, deleted 78 core, and restarted. Downloaded v1.64 4/29/04. Do you have something even newer? Isn't client suppose to download newer cores automatically?

 

[BCinSC]

Well, isn't that special. Just noticed that I lost all work in progress for that client I stopped and downloaded new core. Growing more annoyed with the project.

 

[GLeeM]

version I noticed client automatically downloading on 5-10 is 1.64

They hope to have the stats problem resolved soon...

 

[BCinSC]

Does anyone know what the changes in the interim 8 steppings of the core?

 

[ProviaFan]

(I'm not in a good mood, so beware)

Look, somebody posted something to try to help you... If all you can do is whine about F@H and how their stats are such a detestable and filthy codpatch, then please do us a favor and go over to SETI or somewhere where you'll be happy. It's best to do F@H for the science, not the stats. If all you care about are the latter, then you're bound to be disappointed sooner or later. Most projects get their stats screwed up at one point or another... Except, of course, that's never happened to SETI, because they're perfect. So you can take advantage of the people's offer to help in the thread the GLeeM posted, grab a :beer:, relax and do something not relating to computers for a day or so while things get sorted out, or you can get your panties in a wad because they won't (never mind can't) fix your stats THIS FREAKING SECOND and run off in a huff.

Basically, I'm sick and tired of people who are impatient and think that the world (and everybody) owes them something and everything, in perfect working order, RIGHT NOW. These guys run the project for the benefit of mankind, and have other things to worry about than stats. But even then, they seemed to catch notice of the problem quickly, and are trying to rectify it. If that's not enough for you, then fine, find something more to your taste, with PERFECT stats that update every microsecond and NEVER have issues with lost work, or a multitude of other things (like cheaters who inflate their score and make your rank go down).

Sorry, I've gotten lots of complaints from utterly impatient people today, and don't feel like dealing with more of it around here.

Have a good day, and good luck finding a project that will satisfy you.

Edit: Or, you can run without -advmethods to increase your chances of getting TINKERS, or run with GAH work unit preference, to get guaranteed TINKERS... but no, you'll loose 0.5 points per day doing that...

 

[MDE]

^ What Jliechty said. Run the project for the research first, stats second. Sometimes we get caught up in stats races and whatnot, but in the end I hope most of us are doing this for its potential benefit to mankind. It's a drag when the stats aren't updated or you get "cheated" on points but you win some and you lose some, this is something that falls into the latter category. I hope you stick around, but please realize that Stanford is running the F@H program to get help with research, not hand out points.

 

[BCinSC]

Sorry, Dude, for being such a whinny butt. I did kick back and have a beer. Just cranky 'cause one Barton is down and the other 2 won't OC worth beans. Anyway, I did take GLeeM's post to heart and will review logs to see if that is what happened of if I lost some work with unit crashes. I do F@H, FaD, and D2OL for the science, but I have become a stats whore. All projects have their issues and I'll just shut up now.

But I do ask whether latest release of Gromacs Core resolves 3DNow issues or should I continue to run with -forcesse.

 

[ProviaFan]

Originally posted by: BCinSC
But I do ask whether latest release of Gromacs Core resolves 3DNow issues or should I continue to run with -forcesse.

Sorry for being a bit rough in my post... hopefully you caught it after the edit, not before.

Anyway, I use -forcesse on my AXP because I thought that it offered more performance than 3DNow. At least that's what someone told me a while ago. And, I thought that it was that people used 3DNow because SSE caused crashes, though SSE was supposedly faster, and now that the SSE crashes have been fixed, that SSE was to be preferred... Shows how much I know.

 

[BCinSC]

Oh Goody. Just took down another Barton 3200+. I hate Windows Evals. If your CMOS gets screwy with oc'ing and reset to prior to install date, Windows is forever toast. Going to bed. Can't deal with reload now.