CPU Multithreading Topic, User runtime

[grkM3]

My 10.20 is legit it was done at 5.3 ghz with my 2600k i was going for higher and system crashed but my 3 benches in the shot are mine.notice it says exact system below.

My ram was at like 27000 mb sec with 7-7-7 timings and ill prolly never get that run again

 

[Makaveli]

So question do runs like these count since it almost like a LN2 suicide run benchmark?

Do we want just the highest scores possible or does the machine have to be stable?

Or does it not matter?

 

[BallaTheFeared]

We should keep it limited to 200w for all setups.

 

[Makaveli]

We should keep it limited to 200w for all setups.

haha if that was the case BD would be out and so would the 6 core intel chips and that 16 core chip at the top also out.

And it would pretty much just be SB and IB

 

[BallaTheFeared]

Can't go wrong with either choice then, :thumbsup:

 

[blckgrffn]

Q6700 @ 3.0 Ghz, 4GB DDR2 RAM @ 800 Mhz

Cinebench: 3.31

***

x264 HD BENCHMARK 4.0 RESULTS

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Results for x264.exe r1913
==========================

Pass 1
------
encoded 1442 frames, 88.63 fps, 3912.26 kb/s
encoded 1442 frames, 89.04 fps, 3912.26 kb/s
encoded 1442 frames, 88.80 fps, 3912.26 kb/s
encoded 1442 frames, 87.73 fps, 3912.26 kb/s

Pass 2
------
encoded 1442 frames, 19.40 fps, 3961.59 kb/s
encoded 1442 frames, 19.40 fps, 3961.51 kb/s
encoded 1442 frames, 19.17 fps, 3961.34 kb/s
encoded 1442 frames, 19.35 fps, 3961.60 kb/s

***

POV-Ray (1920*1080, AA 0.3) Balcony: 1921 PPS, 18M 04S

7 Zip:

POV Ray:

TrueCrypt:

I figured this might be useful for the "Haswell or Bust from Core2Quad" folks out there...

I am willing to hit my i3 SB if anyone cares to have that in the mix. I am actually curious how it fairs against the Core 2 Quad...

 

[Don Karnage]

Q6700 @ 3.0 Ghz, 4GB DDR2 RAM @ 800 Mhz

Cinebench: 3.31

***

x264 HD BENCHMARK 4.0 RESULTS

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Results for x264.exe r1913
==========================

Pass 1
------
encoded 1442 frames, 88.63 fps, 3912.26 kb/s
encoded 1442 frames, 89.04 fps, 3912.26 kb/s
encoded 1442 frames, 88.80 fps, 3912.26 kb/s
encoded 1442 frames, 87.73 fps, 3912.26 kb/s

Pass 2
------
encoded 1442 frames, 19.40 fps, 3961.59 kb/s
encoded 1442 frames, 19.40 fps, 3961.51 kb/s
encoded 1442 frames, 19.17 fps, 3961.34 kb/s
encoded 1442 frames, 19.35 fps, 3961.60 kb/s

***

POV-Ray (1920*1080, AA 0.3) Balcony: 1921 PPS, 18M 04S

Will post up the pics from 7zip and TrueCrypt if there is any interest.

I figured this might be useful for the "Haswell or Bust from Core2Quad" folks out there...

I am willing to hit my i3 SB if anyone cares to have that in the mix. I am actually curious how it fairs against the Core 2 Quad...

Ouch.....

 

[exar333]

My results with 3939K @ 4.6ghz:

Cinebench

7Zip

TrueCrypt

POVRay

X264

CPUZ

 

[podspi]

Q6700 @ 3.0 Ghz, 4GB DDR2 RAM @ 800 Mhz

Cinebench: 3.31

***

x264 HD BENCHMARK 4.0 RESULTS

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Results for x264.exe r1913
==========================

Pass 1
------
encoded 1442 frames, 88.63 fps, 3912.26 kb/s
encoded 1442 frames, 89.04 fps, 3912.26 kb/s
encoded 1442 frames, 88.80 fps, 3912.26 kb/s
encoded 1442 frames, 87.73 fps, 3912.26 kb/s

Pass 2
------
encoded 1442 frames, 19.40 fps, 3961.59 kb/s
encoded 1442 frames, 19.40 fps, 3961.51 kb/s
encoded 1442 frames, 19.17 fps, 3961.34 kb/s
encoded 1442 frames, 19.35 fps, 3961.60 kb/s

***

POV-Ray (1920*1080, AA 0.3) Balcony: 1921 PPS, 18M 04S

Will post up the pics from 7zip and TrueCrypt if there is any interest.

I figured this might be useful for the "Haswell or Bust from Core2Quad" folks out there...

I am willing to hit my i3 SB if anyone cares to have that in the mix. I am actually curious how it fairs against the Core 2 Quad...

I'd be interested...

 

[blckgrffn]

I'd be interested...

Core i3-2100, 8GB DDR3 1333

http://ark.intel.com/products/53422/

ESXi 4.1 U1, Server 2008 R2 SP1 VM (no AVX pass through)

Cinebench - 2.86 (vs 3.31 Core 2 Quad Q6700 @ 3 Ghz)

***
x264 HD BENCHMARK 4.0 RESULTS


^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Results for x264.exe r1913
==========================

Pass 1
------
encoded 1442 frames, 85.53 fps, 3912.26 kb/s (vs 88.63 Core 2 Quad Q6700 @ 3 Ghz)
encoded 1442 frames, 85.61 fps, 3912.26 kb/s
encoded 1442 frames, 85.30 fps, 3912.26 kb/s
encoded 1442 frames, 84.51 fps, 3912.26 kb/s

Pass 2
------
encoded 1442 frames, 15.89 fps, 3962.12 kb/s (vs 19.4 Core 2 Quad Q6700 @ 3 Ghz)
encoded 1442 frames, 15.91 fps, 3961.65 kb/s
encoded 1442 frames, 15.86 fps, 3961.75 kb/s
encoded 1442 frames, 15.84 fps, 3961.24 kb/s

***

POV-Ray:

13m 10s (Now 3M 8S with correct POV-Ray Version! Thanks Maximilian) (vs 18M 04S Core 2 Quad Q6700 @ 3 Ghz)

7zip:

TrueCrypt:

Looks like a bit of a toss up here, pretty amazing how well a 2C/4T part can hang with a 4C/4T part. A few years brings a bit of progress after all

The Core 2 edges out the i3 in most tests, but gets murdered in POV-Ray for whatever reason.

Of course, my POV-Ray results are utter crap compared to others here (less than 1M?) so I wonder what I am doing wrong. I install it, edit the resolution.ini file, restart POV-Ray and then use the new preset I created. The image is rendered at the desired resolution and quality.

Also, dropbox has made image linking trivial, you can generate the link right from the file in the local filesystem by right clicking on it. Super handy for this forum...

Ouch.....

Heh, pretty much masochism at this point, but I made an agreement with my wife to freeze my main rig for two years so that I could acquire a 1080p projector. She made choose between a new computer or projector when I graduated with my MS. I chose projector, and she made sure to nail down that meant incremental internal upgrades as well Yes, the projector is awesome.

That new 2600k I bought is going to make a swell ESXi box or HTPC for the basement (and full time cruncher/folder)

At least I haven't played a game where I could tell I was CPU limited - and I don't play BF3. The 5870 @ 1080p is also holding up like a champ.

 

[Ferzerp]

This was just a practice graph but I think this looks good.

Neat, if I have some time this weekend, I'll try to see why cinebench would only use 32 threads (it should have access to 40), and maybe run a few of the other benchmarks. Also, technically, there are 4 procs in that thing, each 10c/20t, so really that score was 1.6 of one of those procs (as it was a 32 thread run). That makes the OC'ed 970 look impressive indeed.

What is the consensus on the processor group issue? Run two concurrently or what? I'll be honest, I couldn't get the x264 benchmark to utilize even half a processor group's worth of cores. I ran one instance, and got in the low 60's on pass two, then I ran 4 instances, and all 4 instances also got the low 60's on pass two at the same exact time. Not sure what to do there. Not sure how many instances I can run before the max throughput is met.

I should be able to do POVray as well, but I have to be careful not to step on any of the real work that system is doing.


edit: got the current release of cinebench, was able to do 40 threads, maybe I did before as well and forgot. It gets 20.48 with 40 threads. Cinebench won't run twice concurrently though, so I can't run two at the same time to add the result together (one per proc group).

 

[AtenRa]

This topic is not about competition, we are not hwbot.org.
I would like to ask to only post desktop CPU results and 24/7 Overclocks. It would be very helpful to have a couple of Server CPU results for reference but plz keep the number low.
This topic would be helpful for people trying to upgrade/purchase a new CPU, so let’s keep it like that.

Thank you all for participating :thumbsup:

 

[AtenRa]

Core i7 920 @ 4GHz
CPU Cooler : Noctua NH-D14
Motherboard: ASUS Rampage II Gene
Memory: 3x 4GB Kingston DDR3 1333MHz @ 1533MHz 9-9-9-24

TrueCrypt v7.1a

Cinebench 11.529

7-zip v9.20

Pov-Ray 3.7 RC5

x264 HD v4.0

 

[IntelUser2000]

I only ran cinebench, what do I win? It also doesn't detect properly, but I guess 16/32 is enough for this.

Sorry for being off-topic but the Task Manager is very funny to me for some reason.

 

[Maximilian]

Of course, my POV-Ray results are utter crap compared to others here (less than 1M?) so I wonder what I am doing wrong. I install it, edit the resolution.ini file, restart POV-Ray and then use the new preset I created. The image is rendered at the desired resolution and quality.

You need to use the new POVray 3.7 RC5, i dont think the old 3.6 version is multi threaded at all because it was taking forever on my 2600k as well.

18 min / 4, you should get ~4.5mins for POVray on the Q6700. Give it a shot and see :thumbsup:

 

[blckgrffn]

You need to use the new POVray 3.7 RC5, i dont think the old 3.6 version is multi threaded at all because it was taking forever on my 2600k as well.

18 min / 4, you should get ~4.5mins for POVray on the Q6700. Give it a shot and see :thumbsup:

Will do! Thanks!

I just followed the link in the OP and took the first link I saw. Different versions available? Details.

 

[Makaveli]

I've updated the chart with some of the newer post.

My 10.20 is legit it was done at 5.3 ghz with my 2600k i was going for higher and system crashed but my 3 benches in the shot are mine.notice it says exact system below.

My ram was at like 27000 mb sec with 7-7-7 timings and ill prolly never get that run again

grkm3 can you provide me with your 4.8 24/7 overclock score. I've been on the fence about adding your 5.3 ghz score since you said the system is not stable. The OP has updated the first post stating that he only wants 24/7 results so the system needs to be stable.

You need to use the new POVray 3.7 RC5, i dont think the old 3.6 version is multi threaded at all because it was taking forever on my 2600k as well.

18 min / 4, you should get ~4.5mins for POVray on the Q6700. Give it a shot and see :thumbsup:

I have a question about pov ray when looking at the results is the pps the only thing that matters or the time aswell? I ask this because from all the results i've seen only two people have completed the full render while I see results that only 30 seconds in or 1 min it. While I posted a 12 min run so are both numbers important or just the pps?

 

[Maximilian]

I have a question about pov ray when looking at the results is the pps the only thing that matters or the time aswell? I ask this because from all the results i've seen only two people have completed the full render while I see results that only 30 seconds in or 1 min it. While I posted a 12 min run so are both numbers inmportant or just the pps?

Im sure both numbers are related, PPS is pixels per second (i guess) so the more PPS it can do the shorter the time will be. Both numbers are directly linked so they're both important.

Did you use the older 3.6 version? It will only use 1 core so itll take ages on any quad/hexacore, it was only rendering 1 line at a time on my 2600k . The new beta versions like the 3.7 RC5 i mentioned use all cores, 8 lines at a time for me so use that instead.

 

[Makaveli]

Im sure both numbers are related, PPS is pixels per second (i guess) so the more PPS it can do the shorter the time will be. Both numbers are directly linked so they're both important.

Did you use the older 3.6 version? It will only use 1 core so itll take ages on any quad/hexacore, it was only rendering 1 line at a time on my 2600k . The new beta versions like the 3.7 RC5 i mentioned use all cores, 8 lines at a time for me so use that instead.

I just RDP's to the machine at home yes I did use version 3.6 doh.

You are correct about it being single threaded cause as I was watching core temp while it was running, system was at 4ghz yet I barely saw any activity across all 6 cores. Will have to rerun it later.

 

[AtenRa]

Sorry guys,

I have updated the OP with the new Pov-Ray 3.7 RC5 link

http://www.povray.org/beta/

 

[grkM3]

I will re run at a more stable setting and the cpu can do 5.2-5.4 all day and even game.the ram at almost 2300 at cl7 timings was the unstability on my run and that high bandwith at those clocks is how i got my high score.

Mind you this is all done with my case closed and running silent water cooling

 

[AtenRa]

Try the best stable OC that will finish all five apps.

 

[Makaveli]

I will re run at a more stable setting and the cpu can do 5.2-5.4 all day and even game.the ram at almost 2300 at cl7 timings was the unstability on my run and that high bandwith at those clocks is how i got my high score.

Mind you this is all done with my case closed and running silent water cooling

Great update when you can I would love to get SB chip in the 5Ghz range in the chart.

Also is cinebench really that sensitive to memory bandwidth ? I would figure it cares more about clockspeed and core count?

 

[grkM3]

Try the best stable OC that will finish all five apps.

cinebench is the hardest bench out of all those to finish,I was trying to break 10.21 because that would of put a 4c sandy higher than the old 6 core I7s

HAHA I ran my ram at 2133 with looser timings and just nailed a 10.24 at only 5.2ghz!!!! This is stable and I can use my computer at these speed for months.

 

[Phynaz]

You don't want that battle. I have 96 12C/24t 3.33Ghz Westmere systems that aren't currently production 1152c/2304t@3.33Ghz 9TB ram total.

Nice! All my big machines are Linux / Unix, wish some of them were Windows.

Hopefully you are running DC on those