Weird folding stats problem?

[mj_1903]

I am having a weird stats problem it seems.

Today I submitted a 235 point tinker and a 60 point gromacs but neither have been reported in my stats (on macnn or overclocking). They do show up on folding.stanford though.

The tinker came from a machine that has not submitted any work units before. Is it possible that I am getting interference between the machines because they are behind the same firewall?

I also have a G5 almost finished with ~260 odd points of work units, plus another 235 point tinker and I would rather it not lose those points as well.

All the machine ID's (powerbook, powermac and powermac g5) are showing up as 1, could this be the problem?

Update: Well the tinker work unit showed up the next day (when it was submitted the day before). The PowerBook's workunit is still MIA though. Is this normal behaviour for a new machine?

Update 2: Well all my units appeared. I assume the servers can get backlogged?

 

[bousozoku]

You're not losing points if they show up at Stanford Univ.'s servers.

The various statistical sites are run by other teams and are often prone to problems.

 

[mj_1903]

You're not losing points if they show up at Stanford Univ.'s servers.

The various statistical sites are run by other teams and are often prone to problems.

Yeah, it seems the stats sites were throwing up garbage for a time there this morning. All ~1100 odd points made it though for today which I am glad about.

 

[Dreadnought]

The machine id is correct. that is (what I have heard) the processor of the machine. If you have only single proc machines the 1 is correct. If you got a dual and a dual foldingscript on it, then it will show up as machine id 1 and 2.

 

[zimv20]

a couple hours ago, i finished a p638_L939_K12M_ext. according to the summary page, it's worth 239 points.

but i've been given only 128, as evidenced here.

i've only got 4 cpus going, the other 3 are nowhere near finished, so i know i'm looking at the right results. anyone else see anything like this?

 

[bousozoku]

a couple hours ago, i finished a p638_L939_K12M_ext. according to the summary page, it's worth 239 points.

but i've been given only 128, as evidenced here.

i've only got 4 cpus going, the other 3 are nowhere near finished, so i know i'm looking at the right results. anyone else see anything like this?

That's the old point value. It must be an older WU, whether you got it later or not.

 

[zimv20]

That's the old point value. It must be an older WU, whether you got it later or not.

here's a dump of unitinfo.txt from this afternoon:

Name: p638_L939_K12M_ext
Download time: June 14 11:53:08
Due time: July 21 11:53:08
Progress: 99% [|||||||||_]

it's changed since the 14th, then, eh?

 

[bousozoku]

here's a dump of unitinfo.txt from this afternoon:

it's changed since the 14th, then, eh?

The point values were changed in the summary but it may be that the work unit you have has been recycled since before that time and retains the original point value.

 

[zimv20]

The point values were changed in the summary but it may be that the work unit you have has been recycled since before that time and retains the original point value.

first i've seen this. thanks for the info.

 

[bousozoku]

first i've seen this. thanks for the info.

Hey, I know it's a pain. I recently had one worth 128 points and I never received anything for it. I know how it feels.

 

[Don't panic]

The point values were changed in the summary but it may be that the work unit you have has been recycled since before that time and retains the original point value.

I don't think that's correct. It depends on when you received the unit.

p638 changed value a couple of times:
239.00 points current
128.00 points until Thu Jun 17 2004
70.90 points until Mon Apr 19 2004

if you got it on the 14th you were just out of luck.
I read somewhere of someone getting it a couple hours before the change!

it shouldn't happen again

fold on

 

[zimv20]

if you got it on the 14th you were just out of luck.

is it the case that:
1) the point value does reflect the amount of work done, or
2) stanford realized they were undervaluing it?