Hi there,

first something about this forum: I find it quite strange to read the original question on top, then have to scroll down for the first answers and follow them back up again. Took some time till I understood. Who invented this confusing way of (de-)organizing a thread?

Now, yesterday I downloaded the BOINC 5.2.13 client and libraries for Mac OS X 10.4 Tiger and installed everything. The system is up and running, though my impression is that the software is far from being optimized for Mac.

I am running it on a Dual 1,25 GHz G4 with 1,25 GB RAM, and the two simulations (one per clock, running them at 100% for about 30 hours now) rarely get faster than about 9.x sec/Timestep. The \"remaining time\" shown in the \"work\" window never reduces? For testing I stopped my Apache and MySQL daemon, which usually are running on this machine, no difference.

I saw a single 2 or 2.3 GHz G5 on top of all processors, doing something like 0.2 sec/TS. I worked with a Dual 1,8 GHz with 4 GB RAM, a G5 is not 20 x faster. What is this?

There are several more problems: Whenever I close the BOINC manager, an information comes up that the program has unexpectedly crashed, no matter if I stop the running processes before in BOINC or not.

What about these \"trickles\"? I have a DSL line, but no matter if it is open or not, wether I actively connect or wait: Nothing is transmitted.

One project died in May 1811, the planet froze to -42°C, no clouds, no rain, no pressure, error could not be retrieved. The other run has by now crept to sumthing like 10477 timesteps and is going, but doesn\'t transmit nothing.

Whenever I stopped and restarted the program, it downloaded two new runs, though one of the two first died, while the other is ongoing, now at 0.81% in July 1811. I got about 400 MB and growing on my HD. I now clicked the \"no more jobs\" in the Project window, will that stop the request for new runs?

Do these runs really have to be that big? Doesn\'t that increase the chance of loss? Do I need a G5 or 4Gig Pentibum to take part?

I am concerned about climate change. To me, contributing to this DC project is just a minor way to act accordingly. Right now, I feel like wasting resources. This program does not make me feel like my participation is wanted.

Am I wrong?

Hi there again,

just the recent outcomes, after crunching went on through the night: Somewhen around 7:30am this morning the BOINC client failed to contact the scheduler, as I had closed the line. Obviously this caused both runs to terminate with \"error 10\", as the notice on abnormal termination directly follows. Unfortunately it was not possible to copy&paste the messages from the BOINC windows. However, I have saved the detailed Mac OS 10.4.3 crash report as *.txt, which may possibly be helpful to debug. Is anybody from the project interested to have that?