I just Downloaded more WU's to 3 of my Computers & a lot of therm Erred out with this Message ...

11/22/2005 6:29:05 PM|rosetta@home|Checksum or signature error for 1di2.pdb.gz
11/22/2005 6:29:06 PM|rosetta@home|Unrecoverable error for result 1di2__abrelax_rand_len10_jit02_omega_sim_03735_0 (WU download error: couldn't get input files:<file_xfer_error> <file_name>1di2.pdb.gz</file_name> <error_code>-200</error_code> <error_message></error_message></file_xfer_error>)

A batch of workunits will have this download error. Please just ignore them. Eventually you will get a work unit that does not give this error. We had to modify an input file which changed the size of the file and therefore, work units that were already queued (and thus have the wrong checksum) will error out.

Sorry about this mishap.

Okay David, I just kept Updating until I got the Amount of WU's I wanted without any Download error's ...

Sorry about this mishap.

May I say... this is why I call this the "best run BOINC project". Not that everything goes perfectly - but that when something goes wrong, as it inevitably will, someone FROM the project bothers to come on the boards and give us information. Thank you David, and thanks as well to all your coworkers, who are equally available to us.

I'll now take this opportunity, after shameless brown-nosing, to ask about the status of the Mac application. :-) I'd love the ability to double my Rosetta RAC, but previous problems with the Mac app (31-hour WUs...) have me running it only on my one lonely little AMD.

---

Opps, sorry guys (and gals) I thought I was on the E@H site <blushes>

Maybe boinc could make each project forums site look different, to keep people like me from doing this...

Lines below belong on the E@H forum site...

Yes, I would probably participate in this project due to my interest in space, Einstein, etc...

But the feedack from the 'guys in the trenches' makes this project much more satisfying.

---

I'll now take this opportunity, after shameless brown-nosing, to ask about the status of the Mac application. :-) I'd love the ability to double my Rosetta RAC, but previous problems with the Mac app (31-hour WUs...) have me running it only on my one lonely little AMD.

We may have to resort to supporting only OSX 10.4+ to consider adding altivec optimization. I have to be honest, it's a pain for us to make a build that will run 10.3.9 also and we can't optimize it. However, other projects seem to be doing so by having the user install some libraries. I don't know what they specifically are doing.

We may have to resort to supporting only OSX 10.4+ to consider adding altivec optimization.

vDSP problems? If you're using GCC 4, you need to link with the 10.3.9 SDK version of the dylibs. If you're using XCode, compiling under 10.3.9 SDKs won't be as "optimized" as compiling under 10.4.x, but should run under either. Likewise, compiling a separate app for G5 might be "nice", but the G5's will run G4 apps perfectly well.

All above info comes from watching Team MacNN do their SETI app; see the devs site...

---

Ok, I know I am now on the rosetta site :)

If the source code could be released to some 'trusted' volunteers to do ports, tweeks, etc. that would be another way 'volunteers' can participate...

I know cheating, bad intentions are a worry that could pollute scientific results, but there should be some method that could accomodate this?

Disclaimer: I am a C/Unix programmer for the past 20+ years, but don't have the time (taking graduate level courses right now) to contribute to the cause.

---