I have noticed that my Macs claim half the credit that they deserve with the beta. I am granted the correct credit, but it throws off the pending credit.
Other than that, it works pretty great. Very fast! (Although I think it doesn't play as nice with other processes as the non beta, but I am ok with that because of the tremendous speed increase!) Good work!!!
For the record I am just running Einstein and I am using the Boinc version from the E@H webpage...
OK, here's what I've found so far with the new Einstein:
Mac iBook G4 1.33 GHz, running BOINC 4.44 (Team MacNN optimized), SETI 4.30 (Team MacNN Optimized for G4 7450 Altivec, and Einstein 0.02. Tried running under BOINC Manager and found SETI would in fact pause when it said it did (checked with Activity Monitor), but Einstein did not. Einstein said in Messages it had paused, but Activity Monitor showed both clients running at the same time, each with around 45% of the the CPU.
Just tried the same setup except running under MenuBar with the same results: Einstein says it paused, but it didn't. I had the exact same problem on my work Powerbook 1.0 GHz. My Powerbook is running the BOINC 4.44 client and SETI 4.30 client, and sharing nicely with CPDN; both clients pause when they should. When I ran Einstein (older version) with these two, it would refuse to pause then also.
I have screen shots showing the BOINC Manager messages and Activity Monitor, if anyone on the Einstein team wants them. I can also try special runs, if you want me to test something.
OK, here's what I've found so far with the new Einstein:
Mac iBook G4 1.33 GHz, running BOINC 4.44 (Team MacNN optimized), SETI 4.30 (Team MacNN Optimized for G4 7450 Altivec, and Einstein 0.02. Tried running under BOINC Manager and found SETI would in fact pause when it said it did (checked with Activity Monitor), but Einstein did not. Einstein said in Messages it had paused, but Activity Monitor showed both clients running at the same time, each with around 45% of the the CPU.
Just tried the same setup except running under MenuBar with the same results: Einstein says it paused, but it didn't. I had the exact same problem on my work Powerbook 1.0 GHz. My Powerbook is running the BOINC 4.44 client and SETI 4.30 client, and sharing nicely with CPDN; both clients pause when they should. When I ran Einstein (older version) with these two, it would refuse to pause then also.
I have screen shots showing the BOINC Manager messages and Activity Monitor, if anyone on the Einstein team wants them. I can also try special runs, if you want me to test something.
Sounds like several of us are seeing the same thing. I've turned off "keep in memory" in preferences, and zeroed my cache for predictor, to see if I can remotely force it to switch back over to Einstein. I'll know within the hour if that works or not.
Anyone running the CLI client who can see if this problem exists there or not? I won't be able to test it myself for quite a while. Clearly the problem does exist for both GUI clients. It happened sometimes with the regular app too.
I've now crunched my way through about 90 units with 0.02, and while many of them are still pending I have only had one unit with credit denied. Terrific beta. :)
Another quick question. What files are suppose to be in the einstein project directory? Here's what is in mine, but 1 is a WU, I think, not sure if that is suppose to be there.
Config_Fin1
earth
H1_1138.5 (.....what is this?)
sun
plus the beta version
einstein_0.02_powerpc-apple-darwin
app_info.xml
Is this proper? Do I need to delete something before I start einstein up again?
Do I need to delete something before I start einstein up again?
No. You may want to delete the old 4.78 App if you are short on diskspace, but it's not necessary.
The H1... file btw. is the file that actually contains the real data from the detector, all other files ther are just support files.
I was having a problem with getting BOINC to recognize the presence of the new Einstein science application. I realized after a while the directory I had copied the new files to was left over from a previous BOINC client...
I found that under "/Library/Application Support/BOINC Data" there was both an einstein.phys.uwm.edu directory and a projects/einstein.phys.uwm.edu directory.
Likewise there were directories 0 and 1 there, and a slots directory that subdirectories had 0, 1, and 2. In both cases, the ones off "BOINC Data" were not current.
So if you've ever ran more than one version of the client (I've run at least 4.25 and 4.43 in this directory, and there may be remnents of 4.19 also) be very careful where you copy the executable and the app_info.xml files!
I've now deleted all the old directories and things are humming along.
I have noticed that my Macs
)
I have noticed that my Macs claim half the credit that they deserve with the beta. I am granted the correct credit, but it throws off the pending credit.
Other than that, it works pretty great. Very fast! (Although I think it doesn't play as nice with other processes as the non beta, but I am ok with that because of the tremendous speed increase!) Good work!!!
For the record I am just running Einstein and I am using the Boinc version from the E@H webpage...
OK, here's what I've found so
)
OK, here's what I've found so far with the new Einstein:
Mac iBook G4 1.33 GHz, running BOINC 4.44 (Team MacNN optimized), SETI 4.30 (Team MacNN Optimized for G4 7450 Altivec, and Einstein 0.02. Tried running under BOINC Manager and found SETI would in fact pause when it said it did (checked with Activity Monitor), but Einstein did not. Einstein said in Messages it had paused, but Activity Monitor showed both clients running at the same time, each with around 45% of the the CPU.
Just tried the same setup except running under MenuBar with the same results: Einstein says it paused, but it didn't. I had the exact same problem on my work Powerbook 1.0 GHz. My Powerbook is running the BOINC 4.44 client and SETI 4.30 client, and sharing nicely with CPDN; both clients pause when they should. When I ran Einstein (older version) with these two, it would refuse to pause then also.
I have screen shots showing the BOINC Manager messages and Activity Monitor, if anyone on the Einstein team wants them. I can also try special runs, if you want me to test something.
C
Join Team MacNN
OK, here's what I've found so
)
OK, here's what I've found so far with the new Einstein:
Mac iBook G4 1.33 GHz, running BOINC 4.44 (Team MacNN optimized), SETI 4.30 (Team MacNN Optimized for G4 7450 Altivec, and Einstein 0.02. Tried running under BOINC Manager and found SETI would in fact pause when it said it did (checked with Activity Monitor), but Einstein did not. Einstein said in Messages it had paused, but Activity Monitor showed both clients running at the same time, each with around 45% of the the CPU.
Just tried the same setup except running under MenuBar with the same results: Einstein says it paused, but it didn't. I had the exact same problem on my work Powerbook 1.0 GHz. My Powerbook is running the BOINC 4.44 client and SETI 4.30 client, and sharing nicely with CPDN; both clients pause when they should. When I ran Einstein (older version) with these two, it would refuse to pause then also.
I have screen shots showing the BOINC Manager messages and Activity Monitor, if anyone on the Einstein team wants them. I can also try special runs, if you want me to test something.
C
Join Team MacNN
Sounds like several of us are
)
Sounds like several of us are seeing the same thing. I've turned off "keep in memory" in preferences, and zeroed my cache for predictor, to see if I can remotely force it to switch back over to Einstein. I'll know within the hour if that works or not.
Anyone running the CLI client who can see if this problem exists there or not? I won't be able to test it myself for quite a while. Clearly the problem does exist for both GUI clients. It happened sometimes with the regular app too.
Team MacNN - The best Macintosh team ever.
I've now crunched my way
)
I've now crunched my way through about 90 units with 0.02, and while many of them are still pending I have only had one unit with credit denied. Terrific beta. :)
Quick Question. Are we
)
Quick Question. Are we supposed to remove the eintein_4.78_powerpc-apple-darwin when we drop in the einstein_0.02_powerpc-apple-darwin version?
Another quick question. What
)
Another quick question. What files are suppose to be in the einstein project directory? Here's what is in mine, but 1 is a WU, I think, not sure if that is suppose to be there.
Config_Fin1
earth
H1_1138.5 (.....what is this?)
sun
plus the beta version
einstein_0.02_powerpc-apple-darwin
app_info.xml
Is this proper? Do I need to delete something before I start einstein up again?
Thanks!
MacStef
RE: Is this
)
Yes.
No. You may want to delete the old 4.78 App if you are short on diskspace, but it's not necessary.
The H1... file btw. is the file that actually contains the real data from the detector, all other files ther are just support files.
BM
BM
Great job this Beta. 100% WU
)
Great job this Beta.
100% WU successfully completed (including H1_0096..., H1_0957... and l1_0869... files)
I was having a problem with
)
I was having a problem with getting BOINC to recognize the presence of the new Einstein science application. I realized after a while the directory I had copied the new files to was left over from a previous BOINC client...
I found that under "/Library/Application Support/BOINC Data" there was both an einstein.phys.uwm.edu directory and a projects/einstein.phys.uwm.edu directory.
Likewise there were directories 0 and 1 there, and a slots directory that subdirectories had 0, 1, and 2. In both cases, the ones off "BOINC Data" were not current.
So if you've ever ran more than one version of the client (I've run at least 4.25 and 4.43 in this directory, and there may be remnents of 4.19 also) be very careful where you copy the executable and the app_info.xml files!
I've now deleted all the old directories and things are humming along.