What are the minimum system requirements (or tell me where to look)?

How have you gathered 300 threads?!?!?!

FAHB is not ported to Android devices, yet:
https://secure.worldcommunitygrid.org/help/viewTopic.do?shortName=minimumreq

not sure about if it will be?!




check sys reqs:
https://secure.worldcommunitygrid.org/help/viewTopic.do?shortName=minimumreq

Same here... got only 4 tasks at one machine... maybe only the big boys get all the tasks :-p

Congratulations to all of the people who've worked to get this new phase of FAAH launched.
I hope that the "high level physical theory and computations" that are new to FAAH will improve the accuracy of the modelling of the ligand-target binding energies.

The fact that modelling with AutoDock and VINA has produced false-positives (FPs) proves that there is significant divergence between reality and predictions computed so far. Thus it is very likely that there are many false-negatives (FNs) as well as false-positives, and for some time I have thought that eliminating some of the former would be even more important than eliminating the latter. With the FPs there is the possibility of applying refined computational modelling as is being done with FAHB, or of testing them the hard way, ie by experiment with the actual ligand molecules. However, I assume that the positive results from FAAH Phase 1, both false and real, are only a small proportion of the ligands screened. Screening the negatives using the real chemicals is not feasible and is of course what FAAH is designed to replace, while it would seem that using BEDAM on our CPUs may be too slow. The FNs could be lost from further testing "forever".

Running BEDAM on GPUs may provide an answer, and according to the target of the link cited above ( http://www.compmolbiophysbc.org/project-updat...enmmagbnpbedammini-summit ), there is some hope of this in the foreseeable future: "We have recently visited Vijay Pande's lab at the SimTk Institute at Stanford University to brain storm on the porting of our BEDAM and AGBNP models to GPU's in OpenMM. We have made a lot of progress and we hope to have an initial prototype by the end of the summer."

Meanwhile, let's hope that running BEDAM with our CPUs on the false-positives will show that it is a much more accurate modelling system, and that there are some real, active and therapeutically excellent ligands out there.
---
I noted a post from knreed, who's still shown as a WCG Tech. Welcome back to the forum! Are you back at WCG, or just a cruncher who also keeps a fatherly eye on happenings in FAAH?