On Feb 21, 2008, at 8:05 AM, Peter Maas wrote: > Just for your information on the BCUT descriptor > > On my windowsXP Intel 2.4 GHz, 1GB RAM, JDK 1.6) > > 50 structures > was 1 min 15 sec > Now 12 sec > > 5059 structures > was 1h 32m 44s > Now 16m 38s > > So big improvement.
The current SVN has an even bigger improvement - path length calc's are much faster with the updated polarizability code. I see at least a 2x speedup, sometimes more. It's still going to be slow for very large molecules. CID 3807 (http://pubchem.ncbi.nlm.nih.gov/summary/summary.cgi? db=pccompound&term=3807) takes about 1.5s and CID 24763 (http:// pubchem.ncbi.nlm.nih.gov/summary/summary.cgi?cid=24763) takes along time - though I'd regard the latter as pathological. In any case,I could process 24K structures from PubChem in under 4 min Let me know how the latest code works on your molecules ------------------------------------------------------------------- Rajarshi Guha <[EMAIL PROTECTED]> GPG Fingerprint: 0CCA 8EE2 2EEB 25E2 AB04 06F7 1BB9 E634 9B87 56EE ------------------------------------------------------------------- CChheecckk yyoouurr dduupplleexx sswwiittcchh.. ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Cdk-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/cdk-user
