Some Notes about Hyper-Threading

[Sedat Canbaz]

Hello Dear Chess Friends !

This time my goal is to help to Computer Chess Fiends,who are not sure regarding to use Hyper Threading ON or OFF

For More Information/Details:
http://www.sedatcanbaz.com/chess/hyper-threading/

Best Regards,
Sedat

[ernest]

regarding to use Hyper Threading ON or OFF

Since this is MP, shouldn't your results be based on many identical tests (and averaged), rather than only one?

[Sedat Canbaz]

regarding to use Hyper Threading ON or OFF

Since this is MP, shouldn't your results be based on many identical tests (and averaged), rather than only one?

Dear my friend Ernest,

As far as i know you are owner of Intel Core 2 Duo E6600,which does not support HT and i doubt that you have a lot of experience with HT technology

And if you really still dont have i7 machine...then think at least 7 times before make any postings regarding about current issue


Really i dont understand your goal...what is your problem with me ?

Really i wonder too,what would be your reaction,if my work was comercial ?

BTW,i noticed several negative statements (not critics) about my work on CSS forum;Rybka forum and here too

Look what i tell you:
1)If dont like my work:simply i suggest you to ignore my postings
or
2)If you really have better idea than my work,then come here and prove but with a serious data

Note:without comments next time please:its will be great if you share your games,benchmarks...
Where other chess friends can benefit from your work

One thing more:It does not seem your comments to be 'critics' over me
Even your replays look like: you are trying to find holes in my work,but especially you cant !


Another interesting note:The main difference between you and me:
-You are trying to be a 'Hole Finder'(unfortunately so far i could not find any useful data posted by you)

-About my work:its true that my work is not perfect,still i should improve some things...but anyway i am trying to do my best and since many years my work is shared free to public

Btw, looking at most postings...i am quite satisfied by the reaction of many good chess friends and that really makes me feel proud and give a lot of positive energy to do my best for Comupterchess

And last note:if all Computerchess Friends will be like you,believe me one day i will be not in Computerchess

Have a nice day/evening,
Sedat

[rodolfoleoni]

Hello Dear Chess Friends !

This time my goal is to help to Computer Chess Fiends,who are not sure regarding to use Hyper Threading ON or OFF

For More Information/Details:
http://www.sedatcanbaz.com/chess/hyper-threading/

Best Regards,
Sedat

Very interesting.

I own a core i7 2630 QM, and I tried to disable HT, but I didn't find the option in BIOS. So I'm running my tests with 8 cpus, but I think it'd be better with 4 cpus and HT Off.

Only one thing. I notice HT on + 8 cpus gave a better result on speed test. I thought HT off + 4 cpus was better on this test, just because to scale from 4 to 8 cpus would have a cost in terms of node rate. Can you confirm the data are correct?

[Sedat Canbaz]

Hello Dear Chess Friends !

This time my goal is to help to Computer Chess Fiends,who are not sure regarding to use Hyper Threading ON or OFF

For More Information/Details:
http://www.sedatcanbaz.com/chess/hyper-threading/

Best Regards,
Sedat

Very interesting.

I own a core i7 2630 QM, and I tried to disable HT, but I didn't find the option in BIOS. So I'm running my tests with 8 cpus, but I think it'd be better with 4 cpus and HT Off.

Only one thing. I notice HT on + 8 cpus gave a better result on speed test. I thought HT off + 4 cpus was better on this test, just because to scale from 4 to 8 cpus would have a cost in terms of node rate. Can you confirm the data are correct?

Yes...and i can confirm again that the data is correct (teted several times)

Houdini 2.0C 8 threads HT ON generates approx.7% higher kns values than Houdini 2.0C 4 physical cores HT OFF


About Fritz Chess Benchmark:
Fritz Benchmark 8 threads HT ON generates much higher,approx.20% more kns values than Fritz Benchmark 4 physical cores HT OFF


But as i mentioned before,HT ON chess speed performance is worse than HT OFF
The higher kns values dont mean that HT ON is faster for Chess
Please check more carefully and compare the results between both systems-HT OFF and HT ON
And you will notice that the chess speed performance is better when Hyper-Threading is disabled

Note:of course these values are correct for my i7 920 @3.30GHz (not sure exactly what will be the results about rest processors)
But however,i strongly believe that HT OFF is better for Chess


Best Regards,
Sedat

[Albert Silver]

Hello Dear Chess Friends !

This time my goal is to help to Computer Chess Fiends,who are not sure regarding to use Hyper Threading ON or OFF

For More Information/Details:
http://www.sedatcanbaz.com/chess/hyper-threading/

Best Regards,
Sedat

Very interesting.

I own a core i7 2630 QM, and I tried to disable HT, but I didn't find the option in BIOS. So I'm running my tests with 8 cpus, but I think it'd be better with 4 cpus and HT Off.

Only one thing. I notice HT on + 8 cpus gave a better result on speed test. I thought HT off + 4 cpus was better on this test, just because to scale from 4 to 8 cpus would have a cost in terms of node rate. Can you confirm the data are correct?

Yes...and i can confirm again that the data is correct (teted several times)

Houdini 2.0C 8 threads HT ON generates approx.7% higher kns values than Houdini 2.0C 4 physical cores HT OFF


About Fritz Chess Benchmark:
Fritz Benchmark 8 threads HT ON generates much higher,approx.20% more kns values than Fritz Benchmark 4 physical cores HT OFF


But as i mentioned before,HT ON chess speed performance is worse than HT OFF
The higher kns values dont mean that HT ON is faster for Chess
Please check more carefully and compare the results between both systems-HT OFF and HT ON
And you will notice that the chess speed performance is better when Hyper-Threading is disabled

Note:of course these values are correct for i7 920 @3.33GHz (not sure exactly what will be the results about rest processors)
But however,i strongly believe that HT OFF is better for Chess


Best Regards,
Sedat

The biggest advantage the i7-2600K has over the i5-2500K, when both are overclocked to the same speeds, is HT, so the results of the 2600K with HT=On should be identical or worse with the i5-2500K if HT=ON is indeed worse as you state.

[bob]

Hello Dear Chess Friends !

This time my goal is to help to Computer Chess Fiends,who are not sure regarding to use Hyper Threading ON or OFF

For More Information/Details:
http://www.sedatcanbaz.com/chess/hyper-threading/

Best Regards,
Sedat

Very interesting.

I own a core i7 2630 QM, and I tried to disable HT, but I didn't find the option in BIOS. So I'm running my tests with 8 cpus, but I think it'd be better with 4 cpus and HT Off.

Only one thing. I notice HT on + 8 cpus gave a better result on speed test. I thought HT off + 4 cpus was better on this test, just because to scale from 4 to 8 cpus would have a cost in terms of node rate. Can you confirm the data are correct?

Yes...and i can confirm again that the data is correct (teted several times)

Houdini 2.0C 8 threads HT ON generates approx.7% higher kns values than Houdini 2.0C 4 physical cores HT OFF


About Fritz Chess Benchmark:
Fritz Benchmark 8 threads HT ON generates much higher,approx.20% more kns values than Fritz Benchmark 4 physical cores HT OFF


But as i mentioned before,HT ON chess speed performance is worse than HT OFF
The higher kns values dont mean that HT ON is faster for Chess
Please check more carefully and compare the results between both systems-HT OFF and HT ON
And you will notice that the chess speed performance is better when Hyper-Threading is disabled

Note:of course these values are correct for my i7 920 @3.30GHz (not sure exactly what will be the results about rest processors)
But however,i strongly believe that HT OFF is better for Chess


Best Regards,
Sedat

Problem with HT on is that if you have 4 physical cores, and search X NPS, when you go to 8 cores (HT on) the tree will grow by 30%. If your NPS doesn't grow by MORE than 30%, you see a net loss.

NPS is NOT the way to measure parallel search performance. It provides completely bogus comparisons...

[bob]

regarding to use Hyper Threading ON or OFF

Since this is MP, shouldn't your results be based on many identical tests (and averaged), rather than only one?

Dear my friend Ernest,

As far as i know you are owner of Intel Core 2 Duo E6600,which does not support HT and i doubt that you have a lot of experience with HT technology

And if you really still dont have i7 machine...then think at least 7 times before make any postings regarding about current issue


Really i dont understand your goal...what is your problem with me ?

Really i wonder too,what would be your reaction,if my work was comercial ?

BTW,i noticed several negative statements (not critics) about my work on CSS forum;Rybka forum and here too

Look what i tell you:
1)If dont like my work:simply i suggest you to ignore my postings
or
2)If you really have better idea than my work,then come here and prove but with a serious data

Note:without comments next time please:its will be great if you share your games,benchmarks...
Where other chess friends can benefit from your work

One thing more:It does not seem your comments to be 'critics' over me
Even your replays look like: you are trying to find holes in my work,but especially you cant !


Another interesting note:The main difference between you and me:
-You are trying to be a 'Hole Finder'(unfortunately so far i could not find any useful data posted by you)

-About my work:its true that my work is not perfect,still i should improve some things...but anyway i am trying to do my best and since many years my work is shared free to public

Btw, looking at most postings...i am quite satisfied by the reaction of many good chess friends and that really makes me feel proud and give a lot of positive energy to do my best for Comupterchess

And last note:if all Computerchess Friends will be like you,believe me one day i will be not in Computerchess

Have a nice day/evening,
Sedat

I have access to HUNDREDS of HT-enabled machines. And I have yet to find one single example where hyperthreading on provides faster time-to-depth results than hyperthreading off. NPS is irrelevant. Time to a fixed depth is much more important, as that is how you measure parallel search performance.

[diep]

Hello Dear Chess Friends !

This time my goal is to help to Computer Chess Fiends,who are not sure regarding to use Hyper Threading ON or OFF

For More Information/Details:
http://www.sedatcanbaz.com/chess/hyper-threading/

Best Regards,
Sedat

Very interesting.

I own a core i7 2630 QM, and I tried to disable HT, but I didn't find the option in BIOS. So I'm running my tests with 8 cpus, but I think it'd be better with 4 cpus and HT Off.

Only one thing. I notice HT on + 8 cpus gave a better result on speed test. I thought HT off + 4 cpus was better on this test, just because to scale from 4 to 8 cpus would have a cost in terms of node rate. Can you confirm the data are correct?

Yes...and i can confirm again that the data is correct (teted several times)

Houdini 2.0C 8 threads HT ON generates approx.7% higher kns values than Houdini 2.0C 4 physical cores HT OFF


About Fritz Chess Benchmark:
Fritz Benchmark 8 threads HT ON generates much higher,approx.20% more kns values than Fritz Benchmark 4 physical cores HT OFF


But as i mentioned before,HT ON chess speed performance is worse than HT OFF
The higher kns values dont mean that HT ON is faster for Chess
Please check more carefully and compare the results between both systems-HT OFF and HT ON
And you will notice that the chess speed performance is better when Hyper-Threading is disabled

Note:of course these values are correct for i7 920 @3.33GHz (not sure exactly what will be the results about rest processors)
But however,i strongly believe that HT OFF is better for Chess


Best Regards,
Sedat

The biggest advantage the i7-2600K has over the i5-2500K, when both are overclocked to the same speeds, is HT, so the results of the 2600K with HT=On should be identical or worse with the i5-2500K if HT=ON is indeed worse as you state.

Speaking is silver, testing is gold.

i5 has less memory channels than i7. 2 versus 3, so i5 is a lot worse than i7.

for diep HT works magnificent, and when overclocking a cpu to 4.5Ghz or so it works even better, just like hiarcs team also reported they turned on HT for hiarcs on the overclocked 12 core @ 24 logical core box. 4.65ghz or so overclocked during tournament?

At those speeds HT gives for diep 30% or so. At 3.xGhz it's more like 20%.

And yes - it does search deeper.

By the way you can also see it in testresults from Lostcircuits how bad i5 is.

At 3.7Ghz (turboboost) it's 1.0M nps versus C2Q doing better there if you extrapolate its speed.

http://www.lostcircuits.com/mambo//inde ... itstart=16

Of course the gulftown and sandy-bridge 6-core cpu's total dominate, as they have 50% more cores and in case of sandy bridge 4 memory channels.

But that's another story...

[rodolfoleoni]

Problem with HT on is that if you have 4 physical cores, and search X NPS, when you go to 8 cores (HT on) the tree will grow by 30%. If your NPS doesn't grow by MORE than 30%, you see a net loss.

NPS is NOT the way to measure parallel search performance. It provides completely bogus comparisons...

And here's my problem: I tried to disable HT but I didn't find any option in BIOS-Advanced. There's only an utility, "Easy Flash", and I should only use it to browse and find the BIOS file.... but I've no idea about where to search for it.

It's an Asus laptop, X53S series. Any guess?

Thanks in advance.