AMD Athlon 64 running 36 h and 61-62 h to completion?
6 May 2007 9:54:23 UTC
Topic 192684
(moderation:
I would like to ask about the running time for my newly downloaded work unit. I am using an AMD Athlon 64 3200+ and it has now been running a h1_0217.95_S5R2__55_S5R2c WU for 36 hours, and there is still 61-62 hours to completion. At least according to my BOINC Manager version 5.8.11.
As I am not allowed to ask questions in the most appropriate thread in Cruncher's Corner (about S5R2 WU running times) with my present amount of credits I am inclined to leave Einstein@home to its own peculiar message board rules and the new WU's inexplicable running times - even if this means that 36 hours of donated CPU time to this project should go to the waste basket.
Rgds Kirsten, founder of BOINC@Denmark
AMD Athlon 64 running 36 h and 61-62 h to completion?
)
There have been reports of tasks granting 500 cobblestones, so those estimated run times are not too far outside of the realm of possiblilty.
Most BOINC projects have a minimum amount of credit and/or RAC required to post to the normal forum types but the questions & problem type forums usually have no minimum. This was done to limit the damage certain individuals could do by making multiple accounts to make improper posts.
BOINC WIKI
BOINCing since 2002/12/8
RE: I would like to ask
)
Never mind answering. I have detached the project.
RE: There have been reports
)
One of my team members have run an almost identical WU on almost same sort of machine in 109.109 secs.
Never mind, I shall not trouble a project - Einstein@home - with such unfriendly message board rules again. I have posted in many other BOINC message boards like SETI@home, ClimatePrediction, uFluids, and QMC@home, just to mention a few, with no such rules. Therefore, it would be interesting to learn which BOINC project with similar message board rules as Einstein's you refer to.
Fortunately, I can choose between several other, friendly, projects to give my idle CPU time.
Deleted
)
Deleted
To stay on topic in case
)
To stay on topic in case others are interested: This completion times look too long to me (except maybe running Windows, but I would say even then it's too much). My 3500+ (Venice) takes 100k seconds (about 28 hours) for a task which gives me more than 500 credits (under Linux), and a friend's 3200+ (Clawhammer) would do those tasks in 107/108k CPU secs under Linux if it ever got them (I calculated that from the credit/hour ratings from smaller tasks). So, anything more than about 40 hours should be too long for this kind of CPU even when it gets one of the really big WUs, except there are real monster WUs out there which I haven't heard of yet. Possible, but unlikely. Besides, to run 90 hours on such a machine, the thing had to be more than 1000 credits worth... wow... Someone ever sees one of those, please tell me :-D
RE: To stay on topic in
)
I am running an E@H 100 hours WU on my 400 MHz PII and a monster 300 hours QMC@H WU also, with a lower share. I should get about 230 credits on Einstein. QMC gives less credits than Einstein.
Tullio
Well, I have detached the
)
Well, I have detached the project. My temperament is not for 100 hours long work units, which should not be running for more than around 30 hours on my type of machine: an AMD 64 3200+.
This machine is quite fast and in SETI@home a 9.704 second WU (=2 hour 41 min) is 'rewarded' with 64,20 credits. Even if I am not in BOINC for the credits, I would not like to waste my idle CPU cycles on a questionable work.
I know some will not like
)
I know some will not like what I'm about to say, but this is a classic example of why it would've been better to do S5R2 as a traditional beta test. The project is getting a reputation that is not really deserved...all because the project team didn't take the app to beta with a subset of users to sort out the problems and to at least get FPU/MMX optimization running (SSE could wait)...
RE: I know some will not
)
Does the project have the servers, manpower, etc. to do it as a traditional? Think about the project resources, etc. They may not be able to do things the way other people may think they should.
I personally do not have any problems with Einstein on any of my 5 machines. No errors have occurred, the data is getting done, and I am getting credit.
There are several ways to look at things. I personally love challenge, and am glad I am able to work on this. This is Science, and science is not like Windbowz, etc. There have been many programs, OSs, etc. that have gone out that were less than Beta. Look at Windows ME, Linux is always changing and there have been many bad releases of that, etc.
Just remember that the project is doing what it can, to do what it needs to keep going.
RE: I personally do not
)
This is very similar to your rose-colored view of the scheduler problems over at SAH... You based your opinion there for not extending deadlines on the sheer luck of you not having issues with reporting work. Both of my machines had problems. Others also posted problems. Eric reported that the scheduler was getting pounded. For you though, everything was coming up roses... :sigh:
I'm sure you mean well, but I would suggest that you try the same "look at the big picture" view that you suggested that I do about "servers, manpower, etc."...