Hey Adrian,
Yes, I checked it in. It doesn't compile by default with the new code.
If you want to compile ptraj with the new code cd into src/ptraj
and change the Makefile line from: (Adding the -D flags turns it on)
.c.o:
$(CC) -c $(CFLAGS) $(CPPFLAGS) -o $. $<
to
.c.o:
$(CC) -DEXPFRAME -DEXPHO -c $(CFLAGS) $(CPPFLAGS) -o $. $<
Then just compile ambertools as normal e.g. make -f Makefile_at
and then new version of ptraj will be in the exe directory.
Alternatively you could cd into src/ptraj and type make
and then the binary is in src/ptraj/ptraj instead of exe/ptraj
.
On Wed, Mar 4, 2009 at 10:06 PM, Adrian Roitberg <roitberg.qtp.ufl.edu>wrote:
> Hey Eric,
> This is great !
>
> Is the code in the amber11 tree ?
> a.
>
>
> Eric Absgarten wrote:
>
>> Hi,
>>
>> It was nice to meet many of you down at the developers meeting this
>> weekend.
>>
>> After talking with Tom and Ross and others down at Sanibel, I have tried
>> to
>> improve the speed of ptraj, especially on large datasets.
>> By default the experimental code isn't turned on. Two flags turn it on
>> -DEXPFRAME and -DEXPHO
>>
>> You have to add -DEXPFRAME and -DEXPHO to turn it on
>> for example you can change the Makefile in src/ptraj/Makefile like:
>> .c.o:
>> $(CC) -DEXPHO -DEXPFRAME -c $(CFLAGS) $(CPPFLAGS) -o $. $<
>>
>>
>> Unfortunately I don't have any really large datasets to test it out on
>> with.
>> Please test it out, side by side with the flags unset (original ptraj) and
>> let me know if it works and if its faster by how much.
>>
>> Appreciate any feedback.
>> Thanks,
>> Eric
>>
>>
>> **** test runs follow ****
>>
>> Test Speed runs on 50k frame trajectory with 304 atoms:
>>
>> Original amber11/ptraj
>> r1
>> real 0m43.068s
>> user 0m40.319s
>> sys 0m2.124s
>>
>> r2
>> real 0m42.861s
>> user 0m40.155s
>> sys 0m2.152s
>>
>> ********************************************
>> New Ptraj w/ -DEXPFRAME and -DEXPHO
>>
>> r1
>> real 0m26.306s
>> user 0m24.550s
>> sys 0m1.712s
>>
>> r2
>> real 0m26.455s
>> user 0m24.798s
>> sys 0m1.616s
>>
>> ***********************************************
>> New Ptraj w/ -DEXPFRAME only
>> r1
>> real 0m27.171s
>> user 0m24.614s
>> sys 0m2.092s
>>
>> r2
>> real 0m27.042s
>> user 0m24.810s
>> sys 0m1.916s
>> _______________________________________________
>> AMBER-Developers mailing list
>> AMBER-Developers.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber-developers
>>
>>
> --
> Dr. Adrian E. Roitberg
> Associate Professor
> Quantum Theory Project
> Department of Chemistry
>
> Senior Editor. Journal of Physical Chemistry
> American Chemical Society
>
> University of Florida PHONE 352 392-6972
> P.O. Box 118435 FAX 352 392-8722
> Gainesville, FL 32611-8435 Email adrian.qtp.ufl.edu
>
> _______________________________________________
> AMBER-Developers mailing list
> AMBER-Developers.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber-developers
>
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Fri Mar 06 2009 - 01:17:13 PST
