Hey Zanth,
will be great to compare if when you post the results, you provide one per one the Hw description and OS Sw from each machine, and the number of BOINC version running in each one.

Will be great too, if you could share a little beat more information with us, based on event logs from each WU in each machine, for example,
and if something distinct happened during crunching time.

In my opinion, appear that you have an excellent sample of WCG kind of distribution, and your comments and feedbacks are very welcome.

Thank you in advance. Best Regards.

I have 6 other ones waiting to start. It seems that I will be OK for the bronze Beta badge this time : the E000370 are currently on a 14 hours basis while the E000007 has completed in 3 hours and 34 minutes!

Something else which might interest members with restrictions on their traffic: the E000007 has uploaded 5 MB to the servers. But maybe they return more data during the beta test than they will once in production.

Cheers. Jean.

Goeieavond!

I also have a beta WU:

BETA_ E000370_ 089A_ 001e8z151_ 0-- In Progress 12/2/08 22:59:46 12/4/08 22:59:46 0.00 0.0 / 0.0

BETA_ E000370_ 089A_ 001e8z151_ 1-- In Progress 12/2/08 22:59:31 12/4/08 22:59:31 0.00 0.0 / 0.0

Currently running the one that ends in _0 but as it is almost 2330 here, I wont know how it goes until the morning and then I will post the result.

According to nelsoc's post the current test is only for Windows.

About the different series I mentioned earlier there is another difference between E000007 and E000370: the former have an initial distribution of 10 for a quorum of 5, while the latter are only in double distribution. It looks like we are really testing two different flavors of the same cake.

Cheers. Jean.

I've received 2 of these betas.
The first, a E000007-series, ran in 40 min and is currently Pending Validation. I didn't catch all the uploads, but the last upload to finish was 4.08MB. I expect that these large files are full of verbose diagnostics that won't be needed in the production versions.
The second, an E000370-series, is still crunching, estimated CPU time over 9hrs. On BOINC 6.2.28, its graphics can be started multiple times, no problems.

Meanwhile, I've done a little snooping ...
Here are some ASCII strings (error and diagnostic messages), extracted from the current Beta program binary, file: wcgrid_beta6_6.12_windows_intelx86
They may give us some hints about what it is/does.
(WCG may hate me for this)
===============
(CHARMM) - Developmental Version 34a2 February 15, 2007
Found same atom selected in both reactant and product list
Using mass-weighting for charged molecules/sets.
Adding Drude particle
QMDEFN> Some atoms will be treated quantum mechanically.
The number of QM path integral atoms =
The number of Monte Carlo moves (av) =
BAD FORTRAN UNIT NUMBER (5 or 6)
* TReK (Trajectory REfinement & Kinematics) version 2.10
* Reference : S.Fischer & M.Karplus, Chem. Phys. Letters vol.194, p.252, 1992
Intel Pentium fdiv flaw detected, please update the processor
==========================
Comments:
* The new project looks like it will be atom-bashing, rather than climate-modelling, etc.
* CHARMM will be used in Phase 2 of the Dengue project. DDDT Technical Details
* The program has sections that were written by earlier researchers (TReK)
* CHARMM seems to have been written in FORTRAN, which I thought went out with 80-column punched cards.
* However, it seems that punched cards haven't gone out, either. Here is a line from an input file (BETA_E000370_342A_001e94xxx_001e.top)
of the beta WU currently crunching:
READ PARAMETERS CARDS NAME par_all22_prot_cep.inp

Have you kept any computer cards? It seems it's lucky I did:

(This is the machine currently crunching the beta WU. You see, there was a reason I put those cards there - apart from trying to reduce the temperatures of the CPU power-supply MOSFETs and inductors so that I don't burn my fingers on them )