Toga II 3.0 released

[Martin Thoresen]

Hi Jerry,

It seems to work fine with up to 16 cores, but the node speed is not correct when using more than 4 cores.

Is this something you could look at?

I am planning of including this for Season 2 of TCEC in that case.

Thanks,
Martin

[melajara]

[deleted msg]

[melajara]

I simply defended Mel Cooper.

I did not claim that the improvement is not significant.

I only claimed that the estimate of 100 elo or 80 elo was probably too optimistic that was Mel Cooper's point based on my understanding.

Well, if you read my older post, you will see that I said that I thought 80 elo was too optimistic: "I doubt it will be 80 elo". But at least the improvement is significant this time.

Jerry

Agreed, my apologies and thank you for the good work

[jd1]

178.5 - 358.5 33.24% Perf=2726 537 out of 2850 games played

Yes, it is better than Zappa Mexico II for instance if it keeps this up for Ingo's list. Not bad for a program that does not have bitboards. I'm not sure Zappa does have bitboards actually but its SMP implementation was certainly pretty advanced compared with Toga SMP implementation does not count though for Ingo's list.

bigger rating than zappa Mexico but interesting that it does not prevent zappa to win convincingly

Toga II 3.0 32b - Zappa Mexico II (2703) 54.0 - 82.0 39.71%

Yes, shows the effect of randomness in short matches!

[jd1]

I simply defended Mel Cooper.

I did not claim that the improvement is not significant.

I only claimed that the estimate of 100 elo or 80 elo was probably too optimistic that was Mel Cooper's point based on my understanding.

Well, if you read my older post, you will see that I said that I thought 80 elo was too optimistic: "I doubt it will be 80 elo". But at least the improvement is significant this time.

Jerry

Agreed, my apologies and thank you for the good work

No problems

[jd1]

Hi Jerry,

It seems to work fine with up to 16 cores, but the node speed is not correct when using more than 4 cores.

Is this something you could look at?

I am planning of including this for Season 2 of TCEC in that case.

Thanks,
Martin

Hi Martin,

Thank you! Yes, I will certainly have a look at this. What node counts do you see with 16 cores?

Regards,
Jerry

[IWB]

178.5 - 358.5 33.24% Perf=2726 537 out of 2850 games played

Yes, it is better than Zappa Mexico II for instance if it keeps this up for Ingo's list. Not bad for a program that does not have bitboards. I'm not sure Zappa does have bitboards actually but its SMP implementation was certainly pretty advanced compared with Toga SMP implementation does not count though for Ingo's list.

bigger rating than zappa Mexico but interesting that it does not prevent zappa to win convincingly

Toga II 3.0 32b - Zappa Mexico II (2703) 54.0 - 82.0 39.71%

Yes, shows the effect of randomness in short matches!

Yes, 150 games is nothing! I never draw conclusions out of that - as this is simply wrong.

I am waiting for the 150 games vs DF13 but puting the 2850 games into Bayeselo the current rating of Toga II 3.0 is 2714! That is 7 Elo below the etimation and just 3 Elo better than Zappa ...

Tomorrow I will have some final results for the full list and as well for the IPON-RRRL of the TOP 20.

What is interesting is that Toga is putting Zappa on the 20th position. We need a bold, honest and talented programmer to throw out the oldest engine in the Top 20!

Bye
Ingo

[Martin Thoresen]

Hi Martin,

Thank you! Yes, I will certainly have a look at this. What node counts do you see with 16 cores?

Regards,
Jerry

Jerry,

From the start position with 1 GB hsah (can't set it higher, I assume 2 GB should work even with a 32-bit binary, ref Hiarcs) after about 1 minute of infinite analysis, with reload after each test:

1 core: ~1250 KN/s
2 cores: ~2300 KN/s
4 cores: ~3750 KN/s
6 cores: ~4500 KN/s
8 cores: ~4600 KN/s
12 cores: ~4550 KN/s
16 cores: ~4600 KN/s

It seems to scale reasonably well up to 4 cores, then the rest doesn't give any meaningful numbers...

Hope this helps.

Best,
Martin

[geots]

Hi everyone!

This is this new release of Toga II 3.0, thanks to Jim Ablett for the compilation.

http://www.computerchess.info/tdbb/phpB ... hp?id=1402

Features:
- scores 64% against previous Toga versions at various time controls, so should be a decent strength improvement.
- Multi-PV fixed

For more details see readme.txt where a full list of changes may be found.

Hope you enjoy using Toga

Jerry

To clarify: it scored +100 elo (64%) against Toga II 2.02 in 6000 games, 2000 at 10' + 0.1" and 4000 at 1 min + 0.1 s / move. It also defeated Toga Returns 1.1 and Toga 1.4.1 SE by similar margins.

If you will be patient, I will give you results against 2.02 and Zappa Mexico that you can take to the bank. I mention Zappa after reading Ingo's thread. After looking at CCRLs list- if 3.0 doesn't score better than Zappa Mexico- that would mean you almost have no increase with 3.0. Which I don't buy for a minute.


Best,

george

[jd1]

Thanks Martin!

I'll have a look at it. Could you please do one more thing for me (I don't have a 16 core machine), could you verify it is actually using 16 cores?

Thanks,
Jerry