AMBER Archive (2006)

Subject: Re: AMBER: Difference in results for TI calculation

From: David A. Case (case_at_scripps.edu)
Date: Tue May 30 2006 - 13:01:49 CDT


On Mon, May 29, 2006, brmeher_at_iitg.ernet.in wrote:

Please send e-mail to the amber list as plain text; not as html.

> I was performing the same calculation (Toluene---> nothing) in the tutorial
> given. But unfortunately I got some differnt results as compared to the
> tutorial.

It looks like you got results that are the same as the tutorial when you
did a similar calculation:

                                                                      
|clambda|DV/DL from my calculation |DV/DL from tutorial |
| |> |> |
|0.11270|> |> |
| |4.4439 +/- 2.1355 |4.13 +/- 2.26 |
| |> |> |
|0.5 |> |> |
| |1.5345 +/- 1.7721 |1.52 +/- 1.80 |
| |> |> |
|0.88729|> |> |
| |-0.3335 +/- 1.42 |-0.26 +/- 1.49 |

The results are certainly well within the statistical errors cited. If you
want to get results that are statistically converged to higher precision, you
need to run much longer runs: see the Shirts et al. paper cited in the
tutorial.

[I had trouble reading and understanding the rest of the original post,
perhaps because of the html encoding, perhaps because there was not a clear
comparison of actual vs. expected results.

...good luck...dac
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu