Milkyway@Home - GPU & CPU performance stats wanted

[Assimilator1]

Hi guys, longtime since I've posted in this section! Anyone remember me?

[Final update 6-4-15, not using the 213.76 credit WU anymore as it has gone scarce, see post #18 for a link to a new table using a different WU.]

Anyway, I've created a thread over at AnandTechs, Milkyway, & KWSN forums gathering stats for various GPUs & CPUs. Still missing quite a few different types of GPUs & CPUs. Was hoping you guys could add to it so I'll add a thread here too .

Btw even if you have a GPU already listed, if it's at a different clock speed or driver revision etc & it gives a different time then LMK I may well post your time. Please read the benchmark requirements carefully if you want to post a time for the table.

I think this forum s/w is the same as ATs so hopefully little editing is needed & the links will work!

So please share your new scores for old & new GPUs & CPUs alike!
I will update the table as necessary.

Requirements for benchmark, validated 213.76 credit WUs only, average of 5 WU times, dedicated CPU core for the GPU, please state clock speeds if overclocked (including factory o/cs) or state 'stock'. It would also be handy if you could state your BOINC & driver version & OS, incase it does make any odds.
Since 20th Feb. (ish) MW released new faster 213.76 credit WUs along with the previously 'standard' speed WUs, so going by the benchmarks gathered upto the 17/2/14, you need to average 5 of the longer 213.76 credit WUs only. Each 'speed' WU should only vary by a few seconds or so where their is nothing else working the GPU.
[update] As of 21/4/14 the long 213.76 credit WUs are extremely rare now so dig deep to find them! The 2 times I got today I found 5 in 307 & 5 out 1020 results!! Once these WUs are gone it'll be the end of this benchmarking run.

Oh & for the benchmark please only crunch 1 WU at a time per GPU otherwise it will massively increase WU time, naturally! (even if it does increase output), and the WU times seem to fluctuate much more than singly crunched WU so you can't 1/2 the time either.
For CPUs you'll want to crunch 1 WU per real core.

Current GPU statistics ~ Average Time to Complete 1 'long' 213.76 credit WU :-

HD 7970 (GPU 1150 MHz) ....................................... 75s Sunny129
R 280X (GPU 1100 MHz) ......................................... 75s Mumak
R 280X (GPU 1030 MHz) ......................................... 77s Dunx
HD 7970 (GPU 1100 MHz) ....................................... 77s Sunny129
HD 7970 (GPU 1000 MHz, RAM 1400 MHz, Cat 14.1) .. 78s Mesyn191
HD 7950 (GPU 1100 MHz) ....................................... 80s Leonheart
HD 7950 (GPU 1100 MHz) ....................................... 81s Mumak
HD 7970 (GPU 1050 MHz) ....................................... 81s Sunny129
HD 7950 (GPU 900 MHz) ......................................... 82s WES
HD 7970 (stock) .................................................... 83s mikey
HD 7970 (GPU 1000 MHz) ....................................... 85s Sunny129
HD 7950 (GPU 1000 MHz, RAM 1500 MHz,Cat 14.1 b) . 87s Leonheart
HD 7970 (GPU 950 MHz) ......................................... 90s Sunny129 (same time when RAM u/c to 800 MHz!)
HD 7950 (GPU 1000 MHz, RAM 1500 MHz) ................ 95s Leonheart
HD 7870 XT (stock) .............................................. 120s Matt
HD 7950 (GPU 850 MHz) ....................................... 121s salvordorhardin
HD 6990 (stock) .................................................. 165s Alez [TSBT's Pirate] (dual GPU card)
HD 6970 (stock) .................................................. 165s Mikey
HD 6950 (stock) .................................................. 176s Icecold
HD 5870 (GPU 868 MHz, RAM u/c 900 MHz) ............ 187s Brickhead
HD 5870 (stock) .................................................. 192s mikey
HD 5850 (GPU 850 MHz) ....................................... 210s Assim1
HD 5850 (GPU 850 MHz, RAM u/c 500 MHz) ............ 211s Assim1
HD 5850 (GPU 775 MHz) ...................................... 231s Assim1
HD 5850 (GPU 750 MHz) ...................................... 245s salvordorhardin
HD 5850 (stock) ................................................. 246s Assim1 (this post)
HD 5850 (GPU 800 MHz) ...................................... 246s petrusbroder
HD 5830 (stock) ................................................. 282s Pheonix
GTX 780 Ti (stock) .............................................. 403s ... Linux - 395s biodoc
R9 270X ............................................................ 407s Alez [TSBT's Pirate]
GTX Titan .......................................................... 414s Yankton (GPU load only 18% despite a free CPU core)
HD 4870 512 MB (stock) ...................................... 444s JumpinJohnny
GTX 560 Ti 448c (GPU 880 MHz) ........................... 471s GleeM
HD 4870 1GB (stock) ........................................... 503s Assim1
GTX 570 (stock) .................................................. 520s - Linux biodoc
GTX 480 (GPU 750 MHz) ...................................... 520s Dunx
HD 7970M (TBC) ................................................. 525s Spencer
HD 4850 1 GB (stock) .......................................... 553s wayliff
GTX 770 (GPU 1333 MHz) .................................... 609s Stojag
HD 4830 512 MB (GPU 670 MHz) .......................... 615s Assim1 (this post)
HD 7770 GHz ed. (GPU 1100 MHz) ........................ 724s Mumak
HD 7770 GHz ed. (stock) ..................................... 807s Deerslayer
GTX 660 Ti (GPU 1046 MHz) ................................. 816s - Linux biodoc
GTX 560 Ti (stock) .............................................. 836s Deerslayer
HD 7750 (GPU 900 MHz, RAM 1300 MHz) ............. 1096s branjo
GTX 460 (GPU 750 MHz) 768 MB ......................... 1127s - Linux Ken g6

Current CPU statistics ~ Average run Time* to Complete 1 'long' 213.76 credit WU :-

Intel Xeon E3-1230 v3 (Turboing to 3.7 GHz) ...... 9881s Stojag
Intel Core 2 Q9550 (3.6 GHz, 424 MHz FSB) ..... 12,532s Assim1
Intel Core i7-3612QM ..................................... 13,068s mikey
AMD Phenom II X4 965 (3.41 GHz) .................. 13,774s JumpinJohnny
Intel Core 2 Q8300 ........................................ 14,949s mikey
AMD Phenom II X6 1090T ............................... 14,950s mikey
AMD FX 6300 ................................................ 15,521s mikey
AMD FX 6100 ................................................ 17,449s mikey
Intel Core 2 Q8200 ........................................ 19,542s mikey
AMD Phenom 9850 ........................................ 25,844s mikey

OS Windows, unless otherwise stated. CPU at stock clock where no speed shown.

* I latter realised for the CPU tests only that I should have used & asked for the 'CPU time' rather than 'run time' which would include any slow downs due to running other processes. I think that largely due to the use of averaging at least 5 WU times that it will keep the inaccuracy low & that the times are still useful, just bear in mind their will be a little inaccuracy there. Just re-iterate that this refers to the CPU benchmarks only.

MW forums thread, although the daft 1hr edit limit might means I can only post updated tables at the end of that thread!

KWSN forum thread

AT thread

 

[Assimilator1]

No MW crunchers posting?

 

[Assimilator1]

Oh well, I guess no one's interested here , I shan't be updating the table here then (unless theirs interest latter on).

 

[MGP]

According to Boincstats OcUK has 37 users crunching Milkyway at the moment. I recognise very few of those names, probably many represent fire and forget crunchers. Certainly although we still have an active user base for DC projects, there are unfortunately very few who are involved in forum activity. I'd suspect very few of the active players are even aware of this thread.

 

[Assimilator1]

Yea roger that, had meant to mention that in my previous post, as in their not interested or not active here.

What project(s) do you crunch?

Btw I see you're in Surrey too , where abouts?

 

[Biffa]

Sometimes you need to wait a little while to get a reply

Everyones busy coin farming these days.

And us Surrey guys need to stick together.

 

[Assimilator1]

I thought 2 days was long enough to wait for at least a minimal reply .

Where abouts are you Biffa?
Yey to Surrey OCers

 

[Assimilator1]

Hi Biffa & MGP, don't want to answer?

Oh btw I've updated the AT table with a HD 6990 & a R9 270X times.

 

[MGP]

Hi as my sig shows, I generally fold. There is an office borg doing a bit of boinc stuff, mostly GPUGrid, Einstein and Asteroids, which is more than enough for a Dell desktop.

 

[Assimilator1]

Your sig doesn't say which project that's for , & I was also wondering roughly where you are (we're in the same county ).

Done quite a bit of folding myself, ran it for a few years 24/7 on my old C2Q until last autumn when I switched to A@H & a little LHC.
Still my 7.1 million points is nowhere near your 101 million! lol
What do you crunch on?

 

[Biffa]

I'm in Farnham

 

[MGP]

I'm using one of Biffa's cast off supercrunchers that has 4 AMD CPUs on board. That let's me do the BigAdv WUs, through a customised linux setup, until such time as Stanford cuts them off at which time I'll probably throw in the towel on folding.

And I'm in dreary Addlestone.

 

[Biffa]

I already threw in the towel on folding. Might get into a bit of BOINC in the winter. But I have a bit of a head on for Crypto at the mo.

 

[Assimilator1]

Crypto? Is that like RC5?
How come you gave up on F@H?
You live very close to me!

MGP
I lived in Chertsey nr the 'borders' of Addlestone until 1980 (I was little then ), off Roakes avenue.
You going to quit F@H then because of the lower score on normal WUs? (which us little people crunch ).

 

[MGP]

Once the Big Adv WUs finish I doubt running this workhorse for folding will be an economic proposition. It made sense to get it at the time in terms of cost v electric v points. It also made sense as it was preset to work efficiently (as time passes any IT gets surpassed by something else). So if Big Adv goes, it will probably not be efficient, even if I can work out how to optimise it for the more standard SMP stuff, and in any case would be loosing benefit compared to the GPU multi rigs, if I stuck with folding.

Oh and a lot of Roakes Avenue, got flattened a few years ago for more modern styled flats replacing a lot of the concrete stuff and allotments.

 

[Assimilator1]

Well you get no money for running F@H but I see that the big drop on ppd would be dis-heartening, I started with SMP when I got my previous C2Q & still crunch with SMP so no drop for me (when I run it).

Re Roaked av., interesting! I bet I wouldn't recognise it now, btw I actually used to live on Canford drv. I drove through their a few years ago & that at least looked the same, although it sounds like that was before Roakes was updated.

 

[Assimilator1]

Oh well I updated the table afterall (& description), even though I said I wouldn't.

On an unrelated note, I found some OcUK SETI stats on my HDD yesterday which I compiled for you guys back in 2000! lol.

I've just checked through 1100 of my valid results & their were only 3 results with long times but I think those were where I down clocked the GPU (for YouTube!) or where I was watching a video.

So I think it's safe to say now the 'long' 213.76 credit WUs are extinct, this benchmarking run is now over.
Oh & my RAC has now gone upto 123k thanks to the short 213s!

At some point I'll do another benchmarking thread, this time using the MW v1.02 WUs. No plans as to when yet.

 

[Assimilator1]

Incase anyone's interested I did start a new benchmarking thread using a different WU (the long 213.76s became scarce at best).

So the new benchmarking thread is in the AnandTech DC forum & the Milkyway number crunching forum.

 

[andy_taximan]

ooo were you been all year? still not much going on in here, never done milkyway more of a Malteser fan myself

 

[Assimilator1]

Lol,
You should try Milkyway

Started playing Elite Dangerous late last summer, so not been on any forums much (bar the Elite one), including my usual haunt at AnandTech.

So the DC forum died here?