ElectricFox
I folded Myself
10/12/2004
New core: AMBER core 82
We are releasing a new core, based on the AMBER MD code (see http://amber.scripps.edu/ .) The AMBER core will give us functionality not present in either the Tinker or Gromacs cores.
AMBER uses double precision only. AMBER does not currently have SSE2 specific assembly code and so it will likely run closer to Tinker than Gromacs in general performance. We are planning on adding SSE2 support in future revs, but to keep the core as stable as possible in the initial rollout, we've left it out. Keep in mind that since AMBER uses double precision, SSE2 won't help *that* much.
Since this is a new core, there may be unexpected problems. However, this core has been extensively tested in house and has done extremely well in beta testing. Also, since it doesn't use SSE2 as heavily as Gromacs or other issues which push the limits on CPUs, it will likely be less troublesome to release than the Gromacs core was.
We are very excited about the new scientific possibilities that are posssible with the new core. More news on these results as they come in!
Source: Folding@Home News
10/12/2004
-Major Update To EM
-AMBER Core Now Recognized
-New Protein Data File Structure
-New EM Calculator
Even though there have been some major changes to EM internally, I always try and make upgrades as easy and simple as possible. This update is no exception. If you are already using EM, you only need to get the upgrade version. There is no need to reinstall EM using the full install zip file.
The new AMBER core is now recognized. I revamped the protein data file. I am no longer basing the protein value on the protein name. I am now using the project number. This should get rid of a lot of those unknown protein point values because the protein name on the Stanford web site doesn't match the actual protein name in the work unit data file. This is a major update to the program so I have gone from version 2 to version 3. I still have more changes to make, but this converts EM to work with this new protein data file structure and recognize the new AMBER core. The calculator has also been updated to work with this new structure also. I expect that because EM has gotten very complicated internally, there may be something I missed. Please post anything you notice that doesn't act correctly or as expected. Provide the work unit name/project number when you post.
The old protein data file will no longer be supported or updated. The old file has been removed from the site and the new one has been put up in its place. If this new method works out, I may be able to make EM update automatically and the days of posting new protein data files may be over... I hope. Only time will tell.
Thanks very much to those of you who have donated a couple of $$$ to the EM-DC.COM web site. I really appreciate the help.
Source: Electron Microscope Homepage
10 caracteres como é habitual...