stephenbrooks.orgForumMuon1GeneralThe stats for version 4.x are knackered
Username: Password:
Search site:
Subscribe to thread via RSS
Stephen Brooks
2003-04-11 03:28:02
Yes, indeed they are.

I'll be working on getting stats up this weekend - both for the new versions v4.3+ and also for v4.2x. Once that's done I'll get the stats info for v4.0x and 4.1x from one of the people with timeseries databases and put those figures back in.

Today's weather in %region is Sunny/(null), max.  temperature -99999°C
Jetta
2003-04-11 11:27:42
great to hear it!  Smile thanks!  Big Grin
DanC
2003-04-11 17:44:02
Stephen,

You've done a dandy job so far!  We'll get over not having stats for a couple of days (I think) I WANT MY STATS!!!!  - ok, I'm better now.  Wink

Question: will it be possible, (or is it now) to replay the simulations of spectacularly efficient muon transfer rates for the edification of those that wish to learn more about the project and study the results?

I found the simulations to be fascinating - although having a better results.dat for the 4.3 version would sure help my efficiency.
TheFinalLoser
2003-04-14 01:58:07
just one stupid question: will the stats be reset for v4.3?  or will they get added up with the 4.x statistics?

"So the last will be first, and the first will be last."
Mt.20:16
Stephen Brooks
2003-04-14 02:42:19
quote:
Originally posted by DanC:
Question: will it be possible, (or is it now) to replay the simulations of spectacularly efficient muon transfer rates for the edification of those that wish to learn more about the project and study the results?


Try cut and pasting one result (that's a long line of parameters followed by a shorter line of other info) into queue.txt.  Muon1 ought to then re-play it, although if the result already had a "#runs=005;" in its first line, it might get added to results.txt/dat with a lower score than it really had because Muon would think the score shown was a sum of 5 runs rather than just one.  This isn't a major problem because low scores just get ignored, and if that result scored high previously it would still be often used as the basis for other designs.  Of course if you manually edit the result in any way (e.g. remove the #runs thing), the checksum will be wrong and Muon will ignore the result.

quote:
Originally posted by TheFinalLoser:
just one stupid question: will the stats be reset for v4.3?  or will they get added up with the 4.x statistics?


They can't be added up directly because they use a different scoring system.  v4.0-4.2x stats will be preserved in a table somewhere else, showing people's final positions for those versions in total (although I'll still accept v4.2x results for a while yet).  Then the v4.3+ stats will be on a new page, where you'll have totals for that.

quote:
Originally posted by DanC:
I WANT MY STATS!!!!


I actually took the stats generator program to work today (it usually only lives on my home machine where the Muon database is), so maybe with fewer distractions I'll get it working some time soon.

Today's weather in %region is Sunny/(null), max.  temperature -99999°C
DanC
2003-04-14 09:05:24
Stephen,

Of course I was just throwing a "tantrum" for effect.  (note the smiley)

It's understandable that you've undertaken a huge effort here.
Thanks very much for your reply!  We're already building some much better results, approaching the 3.xxx% transfer rate!  Big Grin
Stephen Brooks
2003-04-17 14:06:04
quote:
Originally posted by DanC:
Of course I was just throwing a "tantrum" for effect.  (note the smiley)


Oh yeah, I noticed that, I just didn't seem to put any inflection into my reply to indicate that I noticed it (?  I think I might have been tired when I wrote it!) Actually I suspected you also genuinely wanted to know how the stats were going so I told you.

quote:
We're already building some much better results, approaching the 3.xxx% transfer rate!  Big Grin


OK, have a look at this v4.3+ stats beta page... someone from DPC (three, in fact), appear to have reached 13% transfer.  I'm really worried there's a bug in my program, but it'll be even weirder if there isn't!  I'll check it some time soon.

Today's weather in %region is Sunny/(null), max.  temperature -99999°C
DanC
2003-04-17 17:03:53
Stephen,

Actually - that is a pretty old comment.  We're doing over 5.xxx routinely, and popped a 12.92 the other day!  Big Grin

Anyway - we appreciate you keeping us up-to-date on the progress.  Can't wait to see the 4.3 stats!

dc
DanC
2003-04-17 20:04:21
Hey Stephen,

Just noticed something odd.  I'm not sure if the other versions were doing this, but the command-line version of 4.3a reads

"Reading genome scores"
"Making new genome"

Didn't know if you were aware of that... Wink

I'm building particle accelerators... heheh.
[DPC]Stephan202
2003-04-18 00:59:10
I send my results like two days ago, and they are still not in the stats.
20030416-174429 200 results sent to cgmuonftp.homelinux.org

Did you empty that server too?

---
Dutch Power Cow.
MOOH!
wirthi [Free-DC]
2003-04-18 01:35:36
The stats for version 4.3x are obviously not working.  Nobody can see his/her stats Smile AFAIK Stephen is working on that issue.
Henri Valta
2003-04-18 02:02:34
quote:
Originally posted by [DPC]Stephan202:
I send my results like two days ago, and they are still not in the stats.
20030416-174429 200 results sent to cgmuonftp.homelinux.org

Did you empty that server too?



Looking at my server log shows your results, and that they have been transferred to Stephen's computer.
I have no idea why they do not show up in the stats.
[16/Apr/2003:20:44:48 +0300] "STOR 20030416-174429-dYCFz-[DPC]Stephan202.txt" 226 205415
[17/Apr/2003:02:16:03 +0300] "RETR v4.31/20030416-174429-dYCFz-[DPC]Stephan202.txt" 226 205415
[17/Apr/2003:02:16:03 +0300] "DELE v4.31/20030416-174429-dYCFz-[DPC]Stephan202.txt" 250 -
[DPC]Stephan202
2003-04-19 07:29:06
Stephen, what about this?  Should I resend my results, or be patient?

---
Dutch Power Cow.
MOOH!
Stephen Brooks
2003-04-19 13:11:29
Stephan202, you're using v4.31, which was later found to be very buggy.  Use v4.31b instead.  I have your results here but they'll probably get deleted anyway, as v4.31 used the wrong checksums.  I actually fixed this error something like 6 hours after uploading v4.31 and produced v4.31a, and now there's v4.31b. You were just unfortunate to get the dud version.

The thing about "genome scores" is fairly normal and it's been there since around version 3.

The v4.3x stats _do_ show up on that beta page I linked to there.  When I fix a few glitches with that list I'll make it the default page instead of the older v4.0-4.2x one.

Today's weather in %region is Sunny/(null), max.  temperature -99999°C
[DPC]Stephan202
2003-04-20 00:57:51
Aargh!!!
Auch that hurts.  Frown

But besides the checksum, are the configurations legal?  If so, can you please add them to my stats?  (if ít's not too much trouble).
Oh, and there's one results of v4.31 at the end of the results.txt, so that one should at least be counted.  Smile

---
Dutch Power Cow.
MOOH!
Xanathorn
2003-04-20 02:11:07
Hmm I just uploaded about 50 results.  The number of megapoints do show up however not the number of results nor the highest muon percentage.  Alot of those were v4.31b results, and I noticed alot of people who uploaded in the last few hours also got no highest muon percentage nor results for 4.31b. Are those statfields running behind or is it a (new) glitch?

Stephan too bad for ur results Frown, now assimilate a new pc to make up for the lost results Wink.

___________
Proud member of DPC.
[DPC]WAU-Spons
2003-04-20 02:42:06
Stephen, what about the results gathered with v 4.3 (e.g. the linux port...) do these have legal checksums?  will the be counted?
px3
2003-04-20 09:58:00
quote:
Stephen, what about the results gathered with v 4.3 (e.g. the linux port...) do these have legal checksums?  will the be counted? 


I'm not Stephen, but the Linux/Unix port results are counted and have valid checksums.

PX3
DanC
2003-04-20 17:39:58
Stephen,

Does that all mean our 12.x and 13.x results are no good?

I thought that might be just a tad bit too good to be true.  Frown
Ahh well... Que Sara, sara.  :Q
[DPC]fenrir
2003-04-21 00:58:36
Stephen could you just make an 4.3x column in the stats instead of a 4.3 and a 4.31b ??  In principle are the results the same; version 4.3 and 4.31b calculate the same simulation.  And now because of your mistake the calculated 4.31b results which are send in manuallly are seen as 4.3 results, which results in a 0.0000 % in the best muon column.  I don't think it's fair to the users this way and in my opinion there is no difference between the 2 columns. 
After all there is for version 4.2x only 1 column in the stats while there were different 4.2 versions .

[DPC]Clan of ToeT
Opera user
Jwb52z
2003-04-21 12:19:01
DPC, Stephen will probably go back through the results manually to check the checksums to sort out the version problem with the results manualsend after he fixes the version signification in the manualsend itself.
[DPC] Jiriki
2003-04-22 03:52:18
I'm not sure if this is the same issue, but even after uploading the latest results with the autosend feature, my top results in stats stay on 0.00000. Frown

The number of results has been updated after sending.

EDIT: Oke, now has been updated, after a manualsend.  Guess it was temporary...

[This message was edited by [DPC] Jiriki on 2003-Apr-22 at 14:39.]
Flumsi
2003-04-22 05:48:11
dont use corrupt results.dat

[img]
http://www.rechenkraft.info/muoncounter.php?name=Flumsi&team=set[/img]
[DPC] Jiriki
2003-04-24 14:19:42
You're still looking for the DPC best results?  Looks like you're doing fine on your own Smile

Nothing corrupt about my results.dat as far as I can see...

Oh, and when can we welcome [DPC]Flumsi on our team?  Big Grin
tron
2003-04-26 02:37:56
just a question: is it normal that the stats show, that i havent sent results yet and that my best muon-transfer-score is 0,00000%?
px3
2003-04-26 02:59:58
quote:
Originally posted by tron:
just a question: is it normal that the stats show, that i havent sent results yet and that my best muon-transfer-score is 0,00000%?


hmmm... :
50. tron 0 0 0 2769 102398.3 2.406841 20

PX3
Flumsi
2003-04-26 04:40:51
quote:
Originally posted by [DPC] Jiriki:
Oh, and when can we welcome [DPC]Flumsi on our team?  Big Grin


I say Hello for 10 results Big Grin

Please open your database.

greets
Flumsi [SwissTeam.NET]

[img]
http://www.rechenkraft.info/muoncounter.php?name=Flumsi&team=set[/img]
[DPC]fenrir
2003-04-27 13:21:34
quote:
Originally posted by Flumsi:

I say Hello for 10 results Big Grin

Please open your database.

greets
Flumsi [SwissTeam.NET]




You can't be a member of 2 teams

[DPC]Clan of ToeT
Opera user
vaughan
2003-05-06 01:23:17
Where are the Teams Stats for 4.31?

Crunching for
TheFinalLoser
2003-05-07 12:53:54
well there is a unoffical team and overall statistic page from DPC:
http://muon.qik.nl/
well, just to say it, NICE WORK, I like that page.

"So the last will be first, and the first will be last."
Mt.20:16
vaughan
2003-05-13 06:45:15
Thanks for the link.  I have added it in our team forum at www.amerak.com/amdusers

Crunching for
: contact : - - -
E-mail: sbstrudel characterstephenbrooks.org
Yahoo: scrutney_mallard
Jabber: stephenbrooksstrudel characterjabber.org
Twitter: stephenjbrooks

Site has had accesses.