I'll stick around and bide my time.

Well, I don't _want_ to chase you off, but you know as well as I do that a new version of the app isn't going to happen in a few hours, and maybe not even in a few days. If you're getting nothing but errors, I really would recommend going on to another project, then coming back later. (Or maybe putting just the G5 on Einstein, where it really shines, even though that's not your normal approach.) There's just no point in spending the G5 CPU time for nothing - it's not helping the project, as they have to reissue the WU, and it's not helping your credits. Obviously your P4's are fine, I just hate to see a G5 either "sitting idle", or throwing trash, when it _should_ be the star of your farm.

---

I'll stick around and bide my time.

Well, I don't _want_ to chase you off, but you know as well as I do that a new version of the app isn't going to happen in a few hours, and maybe not even in a few days. If you're getting nothing but errors, I really would recommend going on to another project, then coming back later. (Or maybe putting just the G5 on Einstein, where it really shines, even though that's not your normal approach.) There's just no point in spending the G5 CPU time for nothing - it's not helping the project, as they have to reissue the WU, and it's not helping your credits. Obviously your P4's are fine, I just hate to see a G5 either "sitting idle", or throwing trash, when it _should_ be the star of your farm.

I took your suggestion and my PowerMac is now 90% Einstein@Home ...

Still puzzled why my run times for SETI@Home stink so badly (2.5 hours plus) when my Xeon can do them in and hour 20 min ... and I am using the altivec Team MacNN's client (not the best one??) ...

---

I'll stick around and bide my time.

Well, I don't _want_ to chase you off, but you know as well as I do that a new version of the app isn't going to happen in a few hours, and maybe not even in a few days. If you're getting nothing but errors, I really would recommend going on to another project, then coming back later. (Or maybe putting just the G5 on Einstein, where it really shines, even though that's not your normal approach.) There's just no point in spending the G5 CPU time for nothing - it's not helping the project, as they have to reissue the WU, and it's not helping your credits. Obviously your P4's are fine, I just hate to see a G5 either "sitting idle", or throwing trash, when it _should_ be the star of your farm.

I took your suggestion and my PowerMac is now 90% Einstein@Home ...

Still puzzled why my run times for SETI@Home stink so badly (2.5 hours plus) when my Xeon can do them in and hour 20 min ... and I am using the altivec Team MacNN's client (not the best one??) ...

That's weird, my PowerMac with the MacNN client took 72 minutes on average to crunch Seti with an optimized app. I've come to the conclusion that the client is not the problem with Rosetta Mac crunching, but lies with the app.

Dogbytes,

were you using the G4 or G5 version?

I have been running the one done by Javalizard.

I just noticed they have a different version too. This not a great time to be trying an expiriment. But, I have my PowerMac on drain and will drop in the new client when the work I have is all done. From there it will be an interesting test.

I had found another version elsewhere but now have lost the source so I doubt I will run it.

---

Paul, mine is a G5 2.5. I got the client and app over on team MacNN. Rattle around you'll fine it.

Still no luck for me.....looks like Im back to Predictor for awhile....

2005-12-03 02:43:42 [rosetta@home] Unrecoverable error for result 1di2__abrelax_rand_len10_jit02_omega_sim_filters_14018_0 ( - exit code -1073741819 (0xc0000005))
2005-12-03 02:54:28 [rosetta@home] Unrecoverable error for result 1di2__abrelax_rand_len10_jit02_omega_sim_filters_31863_0 ( - exit code -1073741819 (0xc0000005))
2005-12-03 08:25:35 [rosetta@home] Unrecoverable error for result 1dtj__abrelax_rand_len10_jit02_omega_sim_filters_34935_0 ( - exit code -1073741819 (0xc0000005))
2005-12-04 00:18:06 [rosetta@home] Unrecoverable error for result 1dcj__abrelax_rand_len10_jit02_omega_sim_filters_56056_0 ( - exit code -1073741819 (0xc0000005))
2005-12-04 10:23:55 [rosetta@home] Unrecoverable error for result 1ogw__abrelax_rand_len10_jit02_omega_sim_filters_59247_0 ( - exit code -1073741819 (0xc0000005))
2005-12-04 15:14:37 [rosetta@home] Unrecoverable error for result 1dcj__abrelax_rand_len10_jit02_omega_sim_filters_60619_0 (There are no child processes to wait for. (0x80) - exit code 128 (0x80))
2005-12-04 16:34:37 [rosetta@home] Unrecoverable error for result 1ogw__abrelax_rand_len10_jit02_omega_sim_filters_62554_0 (There are no child processes to wait for. (0x80) - exit code 128 (0x80))
2005-12-05 03:53:51 [rosetta@home] No work from project
2005-12-05 06:19:39 [rosetta@home] No work from project
2005-12-05 10:33:37 [rosetta@home] Unrecoverable error for result 1di2__abrelax_rand_len10_jit02_omega_sim_filters_65876_0 ( - exit code -1073741819 (0xc0000005))
2005-12-06 00:04:32 [rosetta@home] Unrecoverable error for result 1ogw__abrelax_rand_len10_jit02_omega_sim_filters_74993_0 (There are no child processes to wait for. (0x80) - exit code 128 (0x80))
2005-12-06 00:43:27 [rosetta@home] Unrecoverable error for result 1dtj__abrelax_rand_len10_jit02_omega_sim_filters_76401_0 (There are no child processes to wait for. (0x80) - exit code 128 (0x80))
2005-12-06 06:02:27 [rosetta@home] Unrecoverable error for result 1dtj__abrelax_rand_len10_jit02_omega_sim_filters_77260_0 ( - exit code -1073741819 (0xc0000005))
2005-12-06 06:03:01 [rosetta@home] Unrecoverable error for result 1ogw__abrelax_rand_len10_jit02_omega_sim_filters_79819_0 ( - exit code -164 (0xffffff5c))
2005-12-06 07:43:20 [---] Couldn't resolve hostname [boinc.bakerlab.org]
2005-12-06 07:43:23 [rosetta@home] Scheduler request to https://boinc.bakerlab.org/rosetta_cgi/cgi failed with a return value of -113
2005-12-06 07:43:23 [rosetta@home] No schedulers responded
2005-12-06 08:55:24 [rosetta@home] Unrecoverable error for result 1di2__abrelax_rand_len10_jit02_omega_sim_filters_81033_0 ( - exit code -1073741819 (0xc0000005))
2005-12-06 08:56:30 [rosetta@home] No work from project
2005-12-06 10:30:07 [rosetta@home] Unrecoverable error for result 1ogw__abrelax_rand_len10_jit02_omega_sim_filters_81465_0 ( - exit code -1073741819 (0xc0000005))
2005-12-06 11:15:02 [rosetta@home] Unrecoverable error for result 1dtj__abrelax_rand_len10_jit02_omega_sim_filters_83468_0 ( - exit code -1073741819 (0xc0000005))
2005-12-06 20:42:47 [rosetta@home] Unrecoverable error for result 1dcj__abrelax_rand_len10_jit02_omega_sim_filters_88920_0 ( - exit code -1073741819 (0xc0000005))
2005-12-07 03:06:52 [rosetta@home] Unrecoverable error for result 1ogw__abrelax_rand_len10_jit02_omega_sim_filters_92366_0 ( - exit code -1073741819 (0xc0000005))
2005-12-07 06:22:00 [rosetta@home] Unrecoverable error for result 1dtj__abrelax_rand_len10_jit02_omega_sim_filters_91478_1 ( - exit code -1073741819 (0xc0000005))
2005-12-07 07:20:08 [rosetta@home] Unrecoverable error for result 1dtj__abrelax_rand_len10_jit02_omega_sim_filters_95528_0 ( - exit code -1073741819 (0xc0000005))
2005-12-07 09:12:10 [rosetta@home] Unrecoverable error for result 1dtj__abrelax_rand_len10_jit02_omega_sim_filters_97171_0 ( - exit code -1073741819 (0xc0000005))
2005-12-07 09:15:44 [rosetta@home] Unrecoverable error for result 1ogw__abrelax_rand_len10_jit02_omega_sim_filters_97229_0 ( - exit code -1073741819 (0xc0000005))
2005-12-07 12:37:17 [rosetta@home] No work from project
2005-12-07 12:41:27 [rosetta@home] No work from project
2005-12-07 14:39:06 [rosetta@home] Unrecoverable error for result 1dcj__abrelax_rand_len10_jit02_omega_sim_filters_98583_0 (There are no child processes to wait for. (0x80) - exit code 128 (0x80))

---

Paul, mine is a G5 2.5. I got the client and app over on team MacNN. Rattle around you'll fine it.

Paul, here is the kink you wanted, I found it. http://members.dslextreme.com/~readerforum/forum_team/boinc.html
This is the optimized client. I think it is javalizard.

---

This is the optimized client. I think it is javalizard.

That link has the optimized clients AND the optimized applications - but the javalizard one is quite old, and not the fastest. The alpha apps right above it are much, much faster. Note that those clients are 4.44, NOT 5.x - I would NOT recommend using the clients, only the applications.

---

This is the optimized client. I think it is javalizard.

That link has the optimized clients AND the optimized applications - but the javalizard one is quite old, and not the fastest. The alpha apps right above it are much, much faster. Note that those clients are 4.44, NOT 5.x - I would NOT recommend using the clients, only the applications.

http://members.dslextreme.com/~readerforum/forum_team/boincbeta.html v5's (AYOR)

Thanks guys. I was using the Javalizard one. I have the other one now and am waiting to process the last couple work units with the old application. I will drop in the new version then and TRY to get work and see what happens.

I don't use the optimized BOINC Clients as I don't see the point, and Bill is right, I like to be able to run the newer versions over the older ones. As a matter of fact, I have 2 now running 5.2.13 now, probably will do one or two computers a day till I am all on that version or later (from the older 4.72).

I really do like the speed of the Einstein@Home Science Application though. It illustrates my point about Steve's decision is not a very good one ... of course, it may be that the Power versions will still be offered at the top end ... we will see ... I really want one of the dual-dual machines ... pant, pant ... :)

---

I have no problems with the optimized clients. The Boinc V5.x.x releases are designed to work with the widest possible amount of different CPU's out on the market. An optimized client is CPU specific, whether it be mmx, sse, sse2, sse3, etc. It would be nice if Boinc were dispensed in a variety of flavors, rather than plain vanilla, one size fits all.

---

I have no problems with the optimized clients. The Boinc V5.x.x releases are designed to work with the widest possible amount of different CPU's out on the market. An optimized client is CPU specific, whether it be mmx, sse, sse2, sse3, etc. It would be nice if Boinc were dispensed in a variety of flavors, rather than plain vanilla, one size fits all.

THe only thing that an optimized BOINC client does is speed up the benchmarks. It will make no appreciable difference in the running of applications. Optimized Science Applications is where you make money with optimization.

So, though you don't have to take my word for it, there is no gain to be made in making multiple BOINC Client versions. The only place optimization gives you gains is when you have a lot of repitition ... like in the science applications.

Most of the activities in the BOINC client are "linear" with little looping.

---

I have no problems with the optimized clients. The Boinc V5.x.x releases are designed to work with the widest possible amount of different CPU's out on the market. An optimized client is CPU specific, whether it be mmx, sse, sse2, sse3, etc. It would be nice if Boinc were dispensed in a variety of flavors, rather than plain vanilla, one size fits all.

THe only thing that an optimized BOINC client does is speed up the benchmarks. It will make no appreciable difference in the running of applications. Optimized Science Applications is where you make money with optimization.

So, though you don't have to take my word for it, there is no gain to be made in making multiple BOINC Client versions. The only place optimization gives you gains is when you have a lot of repitition ... like in the science applications.

Most of the activities in the BOINC client are "linear" with little looping.

One thing to take into account is take HT CPU benchmarks are improperly low with the generic client.