Whewwww!!! That made my day! now on to the next problem... email to
come...
On Fri, Jun 1, 2012 at 12:35 AM, Ross Walker <ross.rosswalker.co.uk> wrote:
> Hi Jonathan,
>
> Good to know you got it working in the end. Good news is a 560 should
> outperform the C1060.
>
> > 61 file comparisons passed
> > 13 file comparisons failed
> > 7 tests experienced errors
>
> The 7 'HARD' errors you got are all cuda malloc errors - this basically
> means your card doesn't have enough memory to run that specific test. For
> example the nucleosome runs need 1.5GB+ GPU memory. So this is nothing to
> worry about, just a limitation of your graphics card.
>
> The differences are:
>
> Minor round off differences - nothing to worry about.
> irest1_ntt0_igb8_ntc2
> irest1_ntt0_igb5_ntc2
> mdout.dhfr_charmm_min
> mdout.dhfr_charmm_min
> mdout.tip5p_box_npt
>
> Large differences:
> irest1_ntt2_igb1_ntc2
> irest1_ntt3_igb1_ntc2
> 4096wat:mdout.vrand
> 4096wat:mdout.pure_wat_oct_nvt_ntt2 + mdcrd file
> mdout.dhfr.ntb2_ntt3
> mdout.ips_ntt3
> npt_ntr_ntt3_oct:mdout.ntp1
>
> These ones all occur when the random number generator is triggered. The
> random number stream is different on different GPU models. ARRGHHH!!! I
> could have sworn I removed these from the default tests for the release
> version of the code. Somehow they are still in there.... :-( I guess I'll
> disable them with the next patch in order to avoid confusion.
>
> Short answer though is that YES EVERYTHING IS OK.
>
> Have fun...
>
> All the best
> Ross
>
> /\
> \/
> |\oss Walker
>
> ---------------------------------------------------------
> | Assistant Research Professor |
> | San Diego Supercomputer Center |
> | Adjunct Assistant Professor |
> | Dept. of Chemistry and Biochemistry |
> | University of California San Diego |
> | NVIDIA Fellow |
> | http://www.rosswalker.co.uk | http://www.wmd-lab.org/ |
> | Tel: +1 858 822 0854 | EMail:- ross.rosswalker.co.uk |
> ---------------------------------------------------------
>
> Note: Electronic Mail is not secure, has no guarantee of delivery, may not
> be read every day, and should not be used for urgent or sensitive issues.
>
>
>
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jun 01 2012 - 08:30:03 PDT
