Sorry about that last post, I replied without entering anything.

My system is running 24/7.

This system does little else, except for running an AVGFree process every night.

I normally have a Seti process running on each CPU.

Because of heat issues, I've got "Use at most" set to 50%.

When two Seti processes are running, Task Manager shows a little over 50% CPU usage.

The Progress column for the AP WU shows 19%, so, I assume this WU will use about 350 hours of CPU before completing. Is that about right?

I suspended the Seti Enhanced 6.03 WU a couple of hours ago.

We'll see if that helps.

Thanks.

Hi, one of my systems has been crunching an AP WU for several weeks now.

It shows 69 1/2 hours of CPU time to this point, and 1007 hours to completion.

Its Status is "Running, high priority."

Its report deadline is 9/10/2008.

It'll never make it at this rate.

I'd hate to have all this CPU time wasted.

My system is a Intel Dual Core D805 running at 2.66GHz with 2Gig of RAM installed.

Normal Seti@Home WU's take about 6 hours to complete on these CPU's.

Any suggestions?

Thanks.

The 1007 hours to completion isn't really an important figure - that's only an estimate, and all it tells us is that BOINC has got the basis for its estimation wrong - no AP unit should take anything like that long on a machine like yours.

The useful figure that could help us advise you is the "Progress" - the %age complete shown in BOINC Manager.

One thing that worries me is that you've only done 69.5 hours (under three days) of work in the 'several weeks' it's been on your machine. Either that means your computer isn't turned on for very many hours each day, or that the CPU spends a high proportion of its time on other BOINC tasks - a low SETI resource share. Neither is good news for completing AP tasks.

But let's wait for that %age figure before rushing to conclusions.

---

That sounds more reasonable, but still seems on the long side. For comparison, I've recently reported (details are on this message board somewhere) that one of my P4 single cores, running at 2.0 GHz, completed an AP task in 146 hours. Yours should be faster.

You haven't mentioned your operating system, but 2 GB should be enough for anything - even Windows Vista ;-) !

So unless you need to run applications which need all that RAM, I would suggest you find the preference for "Leave applications in memory when suspended?", and ensure that it's ticked. (You should find it in much the same area as the CPU usage that you're already using). The only other thing would be to minimise your use of the SETI graphics or screensaver - they can eat into your CPU usage as well.

Intel being faster than AMD in these kinds of scenarios are due to Intel's large L2, and that the L2 is shared amongst the cores. About 90% of the crunch time can be decided with the built-in benchmark of BOINC. Integer math doesn't do anything for crunching, it's the floating point benchmark that matters.

AMD CPUs are more efficient in their design, but Intel's large L2 and extremely high FSB are what make the difference here.

I posted about this back in...earlier this year, a pretty detailed breakdown of the difference between the two.

Found it.. http://setiathome.berkeley.edu/forum_thread.php?id=45616&nowrap=true#721263

Hope this answers your question.

This is so innacurate, i can't let it be.

Core 2 has much more resources to process float that K8 or K10, the pipeline is 4 large on Core 2 while it is only 3 large on K10 and previous.
Macrofusion allow Core 2 to decode up to 5 instructions, in the case of Compare and jump for example ...
This allow Core 2 to process more in parallel, and increase float Instruction per clock ratio.

Memory deambiguation in core 2 accelerate SETI quite a bit too, it allow the processor to keep the memory streams linear, while other processor need to use the CAS of their memory controler to seek several times.

The cache of Core 2 is only a good resources for the large 4 wide pipeline. If you take time to vTune seti, you ll see that the cache miss ratio is low, and the prefetch success is almost 90%.

Grrrrrrr......

who?

Alright then, I stand corrected. I was going off of the information I had and at the time, didn't feel like doing research to get new information. Didn't think there was anything new. I figured Intel would just keep continuing on with their technique from the P4s.

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

You should complain to Rom and David on the BOINC forums (http://boinc.berkeley.edu). The BOINC windows client doesn't actually try to figure out the total cache. It just puts in the value 1,000,000 bytes, and then divides that by the number of cores.

If you look at our host stats you'll see that this "feature" of boinc makes it look like cache sizes are decreasing as the number of cores increases.

Eric

Have a QX9650 @ 3500MHz, FSB 1400, running UBUNTU 8.04 (x86), for a change.
An AP WU takes about 20-28 hours, to complete.
Does anyone know, why LINUX is showing the real amount off L2 cache ½ per 2 core's off total L2, while WINDOWS only shows 244KB instead off (¼ or ½ off total cache amount)?
I don't run an optimized app. yet, on this one.

---

@SETIEric@qoto.org (Mastodon)

You should complain to Rom and David on the BOINC forums (http://boinc.berkeley.edu). The BOINC windows client doesn't actually try to figure out the total cache. It just puts in the value 1,000,000 bytes, and then divides that by the number of cores.

If you look at our host stats you'll see that this "feature" of boinc makes it look like cache sizes are decreasing as the number of cores increases.

Eric

Have a QX9650 @ 3500MHz, FSB 1400, running UBUNTU 8.04 (x86), for a change.
An AP WU takes about 20-28 hours, to complete.
Does anyone know, why LINUX is showing the real amount off L2 cache ½ per 2 core's off total L2, while WINDOWS only shows 244KB instead off (¼ or ½ off total cache amount)?
I don't run an optimized app. yet, on this one.

Same thing with OSX, the Manager says I have 972k of cache when I really have 4MB. I only have a C2D.

---

That sounds more reasonable, but still seems on the long side. For comparison, I've recently reported (details are on this message board somewhere) that one of my P4 single cores, running at 2.0 GHz, completed an AP task in 146 hours. Yours should be faster.

You haven't mentioned your operating system, but 2 GB should be enough for anything - even Windows Vista ;-) !

So unless you need to run applications which need all that RAM, I would suggest you find the preference for "Leave applications in memory when suspended?", and ensure that it's ticked. (You should find it in much the same area as the CPU usage that you're already using). The only other thing would be to minimise your use of the SETI graphics or screensaver - they can eat into your CPU usage as well.

OK, I made the changes suggested, and increased the "Use at most" percentage back up to 100% despite CPU heating concerns.

These changes were made on 8/22 and BOINC started showing better progress almost immediately.

My AP WU finally completed sometime this morning.

It took almost 227 hours of CPU time to complete the WU, or about 9.5 days of CPU time.

The WU took just over 20 days wall clock time to complete with the Task Manager CPU showing just over 50% usage during the entire period.

All is well.

---

Intel being faster than AMD in these kinds of scenarios are due to Intel's large L2, and that the L2 is shared amongst the cores. ...

AMD CPUs are more efficient in their design, but Intel's large L2 and extremely high FSB are what make the difference here. ...

This is so innacurate, i can't let it be.

Core 2 has much more resources to process float that K8 or K10, the pipeline is 4 large on Core 2 while it is only 3 large on K10 and previous.
Macrofusion allow Core 2 to decode up to 5 instructions, in the case of Compare and jump for example ...
This allow Core 2 to process more in parallel, and increase float Instruction per clock ratio.

Memory deambiguation in core 2 accelerate SETI quite a bit too, it allow the processor to keep the memory streams linear, while other processor need to use the CAS of their memory controler to seek several times.

The cache of Core 2 is only a good resources for the large 4 wide pipeline. If you take time to vTune seti, you ll see that the cache miss ratio is low, and the prefetch success is almost 90%.

And that's only part of the 'story'.

The main aspect for a particular class of CPU is how well the application is 'sympathetic' to the CPU and the system hardware. This is where the Chicken Coop optimisers have made such fantastic improvements in the s@h code for various CPUs.

[edit] This is also where some "Who?"-esq specialised CPU instructions can help with streamlining any time consuming or 'expensive' code sequences used by certain applications. This could improve on the existing assembler speed-up tricks used. [/edit]

Another (and to my mind a very devious and wasteful) aspect is whether a certain series of Intel compilers or code has been used that unnecessarily and unfairly 'sabotages' AMD CPUs with the well known "Naughty Intel" trick...

Has the Astropulse client been compiled with the 'Naughty Intel' silliness included, or is it just a question of how the client uses the different sizes of CPU cache?

Happy crunchin',
Martin

(All just my opinions and observations as ever.)

---

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

Has the Astropulse client been compiled with the 'Naughty Intel' silliness included, or is it just a question of how the client uses the different sizes of CPU cache?

MS VC++ build has comparable to stock times. No IPP or Intel-specific code used on stock AP.
Not only cache sizes different. Quality of reordering, branch prediction and so on and so forth. Unfortunately (because I like AMD ;) ) Intel clearly wins this round for SETI-like apps...

Same thing with OSX, the Manager says I have 972k of cache when I really have 4MB. I only have a C2D.

Arkayn, you have the same cache as the Mac Pro's according to the manager so look on the bright side.

---

You can choose not to run astropulse in the future, by unselecting it in your account, in your SETI@home preferences section

---

the truth is out there

I have 5 computers running SETI 24/7, 4 are AMD and 1 is an Intel. I have come to the conclusion that I will only run Astrokitty on my Laptop which is an Intel C2D.

And I have set my preferences to do that.

---

I have 5 computers running SETI 24/7, 4 are AMD and 1 is an Intel. I have come to the conclusion that I will only run Astrokitty on my Laptop which is an Intel C2D.

And I have set my preferences to do that.

I'm going to be doing the same with my AMD Athlon 64 Processor 3800+ 2.4 GHz. An average AP unit on my Intel Pentium Dual CPU E2180 @ 2.00GHz takes under 60 hours to complete. On the AMD I'm at 85 hours at almost 89% complete. I just feel I accomplish more on this sys doing the MB's. Also with some of the problems I've seen with the AP's I just don't want to also risk having the sys tied up for 100 hours or so and have problems with it in the end. If/When they come out with an app that allows AMD to compete with my Intel I'll be putting the sys back on track with AP.

---

On the AMD I'm at 85 hours at almost 89% complete.

Sorry should have read 80% complete....Now at almost 87 hours at about 82.6%

---