Ah ok, fair enough, I'll add the 660 Ti times then :)

---

Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7

I let the 660TI run some Milkyway, almost all times are at 383.2x seconds. You can check with the link from your last post.

97 valid WUs just now.

Nope, that's odd, I only see 2 in progress mod fit WUs.....

---

Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7

There are currently 291 valid WUs from the 660TI GPU. But i let 7 CPU cores run Asteroids at home. So there is some upwards variation from the 383 seconds.

AMD FX 8150 CPU "8 Cores / 4 Modules" I have no means to let the GPU run on one Module exclusively. This is why the times vary.

7 out of 8 cores on Asteroids?
So 1 left free for MW GPU then?
Should be ok, what's the modules thing about??

384s seems to cover most WUs btw :)

I'd already added the 660's time ;)

---

Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7

An AMD FX 8150 Desktop CPU has 4 Modules. A Module are 2 Integer processing units and one Floating point processing unit. But windows sees a module as 2 cores. As long as integer stuff is running it works fine this way. But when you run Floating Point things, you have to share the one unit for 2 cores.

When i run Asteroids@home on 7 cores then one Floating Point unit is shared with the one Milkyway@home needs from time to time. So GPU-Times go up.

The windows sceduler is also a mess. It works ok for real cores. But when you have Hyper Threading on Intel CPUs enabled, it does not make a difference between real and virtual cores. Windows also trows tasks around on the cores at will, which causes delays (empty cpu pipeline on a core and reload the stuff on another one). This is quite frustrating in my opinion, since a lot of time for work on the cpu is wasted.

Ah ok, so when you dedicate 1 module to MW GPU you then get the consistent 343s time on the 660 Ti?

---

Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7

Theoretical yes. But windows throws around the Threads at random. I don't know any option or program that can reserve one Module.

The 323 second times apply when i only run Milkyway@home, no CPU tasks from Boinc. But i should keep the times, when i could reserve one Module exclusively.

Theoretical yes. But windows throws around the Threads at random. I don't know any option or program that can reserve one Module.

The 323 second times apply when i only run Milkyway@home, no CPU tasks from Boinc. But i should keep the times, when i could reserve one Module exclusively.

I use process lasso on my Intel i7 - W7 system, don't know if it can help you here ?

dunx

This Processs Lasso tool seems interesting, but i think it is way overkill for me. Since most my computers only have windows and boinc running.

The tool seems to avoid hyper treading, which i also do in only simulating 4 CPU cores in boinc, instead of the 8 cores Windows shows me.

I haven't read something about the AMD CPU architecture on the tool website. The AMD CPUs are a little complicated to optimize on because of their design. I don't think the tool would help me there, since only a program which is programmed for the Module design would fully utilize it.

By the way, does anyone know, what AMD driver to use for the 6950 / 6970 GPUs, so that all Milkyway WUs run fine. With the current driver the long tasks just create Computation Errors after a few seconds. And i didn't have time yet to search the forum.

See this thread for 6900 v1.02 crunching http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=3400

That thread tells you how to clean out previous drivers properly, & then you can either install cat 13.1 (old I know!) or use a method to alter later drivers (posted later in the thread).

---

Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7

Hi Folks,
This is slightly off topic:-/ BUT its advance notice to anyone who is a wingman of my 3rd rig [TPMO3] About 20 mins ago there was a loud crack and TPMO3 is now defunct, I don't as yet know what exactly blew out and gave up its magic smoke, but its a dodo.

Anyone who is running as wingman to the WU that rig had in its cache, will have quite a wait:-((

Just thought I should let folks know the situation.

Regards

---

Regards,
Cliff.
--
Been there Done That, still no Damn T-Shirt

Hi,

Ref the defunct rig, I'm hoping to fit the hdd to another rig and access that drive.

whats the best way to clear the cache? disconnect from project or about WU?

Regards,

---

Regards,
Cliff.
--
Been there Done That, still no Damn T-Shirt

Hi Folks,

I've managed to boot to the HDD from the defunct rig and aborted all WU for the projects it was running.
Hopefully everyone who was winging it with that rig will soon have the WU re-issued.

Regards

---

Regards,
Cliff.
--
Been there Done That, still no Damn T-Shirt

Thanks for the heads up Cliff. Aborting is the best way, glad you could do that.

Good luck with the rest of the components, i hope most survived.


[TA]Assimilator1 i am using the 13.1 driver on the 6950s. I haven't tested other projects yet with it. But what is intresting, if you run a 5970 (2 5870s in one card) then you have to go back to 12.4. Because the other drivers have trouble with the integrated Crossfire when running milkyway.

But for games works the 15.7 fine.

I know lots of people running MW@H like to know how their rigs compare to other peoples & would like to see how an upgrade would improve output.
Whilst you can look through the MW BOINC stats the info is rather vague, it doesn't give clock speeds or specific GPUs for example, nor does it account for anything that might add to WU times.

So please share your new scores for old & new GPUs & CPUs alike!

Please note new requirements for the benchmark :-

Validated 106.88 credit WUs only & must be from the MilkyWay@Home v1.02 app*.

Average of 5 WU times (not cherry picked please! ;) ).

A dedicated CPU core for each GPU, please state clock speeds if overclocked (including factory overclocks) or state 'stock'.

Please only crunch 1 WU at a time per GPU otherwise it will massively increase WU time! (even if it does increase output, the WU times seem to fluctuate much more than singly crunched WUs so you can't 1/2 the time either).

For CPUs you'll want to crunch 1 WU per real core.

It would also be handy if you could state your BOINC & driver version & OS, incase it does make any odds.

* the separation mod fit ones are slower. ~20s on my HD 5850 @850 MHz.

Please check in the latest posted benchmark table to see if their are any updates to the benchmark requirements.

*****************************************************************************

If your wondering why I'm starting over again it's because the original (long) version of the 213.76 credit WU used for the benchmark has become only sporadically available now.
At 1st a short version joined it & so for a while we had both (which confused some people) & then the long 1 disappeared altogether for a time! It is sporadically back now but it just makes things too difficult.

Also I later found out that some people don't run the mod fit WUs (like the 213.76s) because of lower credit & some people can't run them due to hardware limitations (e.g HD 3800s). I did ask originally but no one said anything :P.
So this time I've chosen MW v1.02 106.88 credit WU for the benchmark, although it's shorter run times give the illusion of smaller performance gaps between different hardware it'll be easier overall.

Hi guys, I'd like to contribute by testing my system, but I am a little bit confused. Where shall I download the MilkyWay@Home v1.02 app, and which options should I check / uncheck in the small BOINC messages window? How do I perform the benchmark? Thanks for your help in advance. :-)

Cookiemonsta

---

The v1.02 app is automatically downloaded when you attach to MW@H, unless you have disabled it in the online MW preferences.

I don't know what you are referring to re 'small BOINC message window', their is nothing to check/uncheck, unless you did the above.

To get the benchmark simply look at you online MW account > computers > view > tasks > milkyway@home > valid, & then just look at at least 5 WUs (can be much more if you like) runtimes (MW v1.02 for GPU, v1.00 CPU) & simply average the result :).

Apart from the last dozen or so WUs your GPU seems to be averaging 141-142s, but the last dozen seems to be about 145-146s, have you changed the clock rate or drivers?
Your CPU seems to be mostly doing them in about 4100-4200s, but I'll let you calculate an accurate average for that one, as your times seem to vary quite a bit for it.

---

Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7

To get the benchmark simply look at you online MW account > computers > view > tasks > milkyway@home > valid, & then just look at at least 5 WUs (can be much more if you like) runtimes (MW v1.02 for GPU, v1.00 CPU) & simply average the result :).

Ha, THERE it is! I was desperately looking for this in the BOINC manager itself (Advanced View, checked message windows etc...). Thought that "app" was a plugin n stuff... Thanks mate!

Apart from the last dozen or so WUs your GPU seems to be averaging 141-142s, but the last dozen seems to be about 145-146s, have you changed the clock rate or drivers?

The calculation times of my computer may tend to vary because I also let it calculate when I am working at the computer (surfing, playing browsergames etc.). During that time, the BOINC CPU usage drops. When I am not at the computer, I give it 99 %. My system is NOT tweaked (yet), CPU and GPU (MSI GTX 980) are normal sale versions. Here is the average of 5 WUs when the PC is not used by me, just Windows and the normal background processes running:

Calculation of GPU (v1.36 thingy):
runtime 40.08 s
CPU time 5.45 s
Cobblestones 26.74

Calculation of CPU (v1.02 thingy, 11+1 logical cores doing that simultaneously):
runtime 145.19 s
CPU time 10.3 s
Cobblestones 106.88

Computer Stats (nothing overclocked):
CPU: GenuineIntel Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz [Family 6 Model 63 Stepping 2] (12 processors) [physically a Hexacore I guess... strange]
GPU: MSI NVIDIA GeForce GTX 980 (4095MB) driver: 353.06 OpenCL: 1.2
RAM: 16 GB DDR5
HD: calculating on 512 GB SSD
OS: Microsoft Windows 8.1 Core x64 Edition

Greetings from Germ... ehm Bhutan (guess it's permitted to calculate for other countries hehe)

---

Calculation of GPU (v1.36 thingy):
runtime 40.08 s
CPU time 5.45 s
Cobblestones 26.74

Calculation of CPU (v1.02 thingy, 11+1 logical cores doing that simultaneously):
runtime 145.19 s
CPU time 10.3 s
Cobblestones 106.88

Just found out that these stats are with a CPU usage limitation of 70 %. It seems that the limitation setting in the simple view is totally different from the SAME setting in the advanced view. So when you just switch the BOINC manager view, different CPU usage (and other) settings apply! Very confusing...

Remind me of posting stats with 99 or 100 % usage. Which settings are actually intented for the benchmarks of this thread?

---

You've got your times mixed up.

v1.36 is the GPU WUs, but the mod fit ones, as I mentioned earlier I'm not using those but the v1.02.

The CPU WUs are v1.00 not v1.02 (also mentioned), the MW v1.0x WUs have a credit of 106.88, also mentioned ;).
And as I also mentioned (did you read my post? :P ), your CPU is crunching WUs in 4000-4300s.

Are you sure the simple view settings is different? I've not come across that before......

If you want your CPU to crunch on all cores then set CPU usage (multiprocessor systems setting) to 100 or 0%, in advanced view.
But if you want to crunch with your GPU you'll want to leave a virtual CPU core free for it.

Anyway, I can use your GPU's time (145.2s) :).

What's strange about a hexacore? It's cool! :)

[update]
Their is something wrong with your CPU times anyway, my old C2Q @3.6 GHz does them in ~3159s, your CPU should easily beat mine.
If you look in the table you'll see a 5820k there, although it's o/ced to 4.1 GHz, your CPUs times should be fairly close to it.

Oh wait, I bet you have hyperthreading on don't you?
As per my benchmark requirements you need to turn HT off, although it does indeed improve output in most DC projects it also messes up the times.
(I guess seeing as the CPU table is small I could have times with HT on & off).

---

Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7