Why are the Ippo derivative stronger than Stockfish?

Discussion of chess software programming and technical issues.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
lkaufman
Posts: 3724
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Why are the Ippo derivative stronger than Stockfish?

Post by lkaufman » Sat Feb 26, 2011 3:29 pm

Houdini wrote:
rvida wrote:I would feel very uncomfortable revealing in public what I have learned from H1.5...
Just give appropriate credit in the next Critter release, for example "without ideas obtained by disassembling Houdini 1.5 this Critter version would not nearly have been as strong".

Robert
But he said that what he learned from Houdini 1.5 was Ippo-specific and not applicable to Critter. So either you are accusing him of lying, or you are asking him to lie!

User avatar
rvida
Posts: 481
Joined: Thu Apr 16, 2009 10:00 am
Location: Slovakia, EU

Re: Why are the Ippo derivative stronger than Stockfish?

Post by rvida » Sun Feb 27, 2011 11:32 pm

Houdini wrote: Just give appropriate credit in the next Critter release, for example "without ideas obtained by disassembling Houdini 1.5 this Critter version would not nearly have been as strong".

Robert
Sure, I would not have problem with that.

However, I like to implement things my own way. Chess programming became my hobby, it gives me a lot of joy (and a lot of frustration), and I don't want throw away 2,5 years of work because of using some "tainted" code. Nowadays the CC community is more sensitive to "reusing ideas" than anytime before.

If I wanted to be a copycat I would use my time more wisely - I could start from Ippolit and not wasting energy on developing my own engine from scratch...

P.S.: I am very competent at introducing my own bugs, I don't need to copy them from others :)

Richard

Roger Brown
Posts: 782
Joined: Wed Mar 08, 2006 8:22 pm

Re: Why are the Ippo derivative stronger than Stockfish?

Post by Roger Brown » Mon Feb 28, 2011 12:41 am

rvida wrote:
Sure, I would not have problem with that.

However, I like to implement things my own way. Chess programming became my hobby, it gives me a lot of joy (and a lot of frustration), and I don't want throw away 2,5 years of work because of using some "tainted" code. Nowadays the CC community is more sensitive to "reusing ideas" than anytime before.

If I wanted to be a copycat I would use my time more wisely - I could start from Ippolit and not wasting energy on developing my own engine from scratch...

P.S.: I am very competent at introducing my own bugs, I don't need to copy them from others :)

Richard


Hello Richard,

That last is certainly quotable!

Thanks for the cool...

Later.

User avatar
rvida
Posts: 481
Joined: Thu Apr 16, 2009 10:00 am
Location: Slovakia, EU

Re: Why are the Ippo derivative stronger than Stockfish?

Post by rvida » Mon Feb 28, 2011 1:18 am

lkaufman wrote: ...
Anyway you confirm my opinion that Houdini is an optimized/improved Ippo derivative
I think it is not appropriate to belittle Robert's contribution.

Please keep in mind that there was a lot of people trying to improve upon Ippo, but after Ivanhoe.xx <I do not remember the version number> only Robert has achieved a significant progress so far

1. optimized - yes, there are speed optimizations all over the place. But they give at most 10-12% speedup over latest Ivanhoes (I have not measured anything, it is just my upper-bound guesstimate).

2. improved - yes, in almost all areas. What is different you ask? Well, material balance, rook/queen mobility, contempt, king safety (safe queen contact checks & mate-in-1 threats), lazy eval and LMR in PV nodes, D.C. style smooth null-R scaling, null move threats canceling reductions, new (and very clever) way of updating "posgain" table, and -what I like most- is a never-seen-before new kind of 'horizon' extension.

This is just a brief overview - there is a lot more.
Conclusion? All I can say that nobody should deny Robert's talent.

Oh, I forgot to mention some endgame evaluators quite far beyond the scope of what SF or Critter has... (hint: esp. bishop related)

Richard

User avatar
Don
Posts: 5106
Joined: Tue Apr 29, 2008 2:27 pm

Re: Why are the Ippo derivative stronger than Stockfish?

Post by Don » Mon Feb 28, 2011 2:10 am

rvida wrote:
lkaufman wrote: ...
Anyway you confirm my opinion that Houdini is an optimized/improved Ippo derivative
I think it is not appropriate to belittle Robert's contribution.
Which contribution is that?

As far as his talents are concerned, I don't think anyone is saying Robert is not talented. I don't think Larry said that either did he?

Anyway, you just verified what we all know anyway, that Robert worked from Ippolito and made good progress working on someone else's program.

Please keep in mind that there was a lot of people trying to improve upon Ippo, but after Ivanhoe.xx <I do not remember the version number> only Robert has achieved a significant progress so far

1. optimized - yes, there are speed optimizations all over the place. But they give at most 10-12% speedup over latest Ivanhoes (I have not measured anything, it is just my upper-bound guesstimate).

2. improved - yes, in almost all areas. What is different you ask? Well, material balance, rook/queen mobility, contempt, king safety (safe queen contact checks & mate-in-1 threats), lazy eval and LMR in PV nodes, D.C. style smooth null-R scaling, null move threats canceling reductions, new (and very clever) way of updating "posgain" table, and -what I like most- is a never-seen-before new kind of 'horizon' extension.

This is just a brief overview - there is a lot more.
Conclusion? All I can say that nobody should deny Robert's talent.

Oh, I forgot to mention some endgame evaluators quite far beyond the scope of what SF or Critter has... (hint: esp. bishop related)

Richard

User avatar
Don
Posts: 5106
Joined: Tue Apr 29, 2008 2:27 pm

Re: Why are the Ippo derivative stronger than Stockfish?

Post by Don » Mon Feb 28, 2011 3:37 am

rvida wrote:
lkaufman wrote: ...
Anyway you confirm my opinion that Houdini is an optimized/improved Ippo derivative
I think it is not appropriate to belittle Robert's contribution.

Please keep in mind that there was a lot of people trying to improve upon Ippo, but after Ivanhoe.xx <I do not remember the version number> only Robert has achieved a significant progress so far

1. optimized - yes, there are speed optimizations all over the place. But they give at most 10-12% speedup over latest Ivanhoes (I have not measured anything, it is just my upper-bound guesstimate).

2. improved - yes, in almost all areas. What is different you ask? Well, material balance, rook/queen mobility, contempt, king safety (safe queen contact checks & mate-in-1 threats), lazy eval and LMR in PV nodes, D.C. style smooth null-R scaling, null move threats canceling reductions, new (and very clever) way of updating "posgain" table, and -what I like most- is a never-seen-before new kind of 'horizon' extension.
None of these are new ideas including horizon extensions. Whether it's never seen before is not clear since I don't know what it is. But if you are impressed with those other things, then chances are this is also not a new idea. Houdart strikes me as a really good engineer, not an innovator and this does nothing to refute that notion. The cloning reinforces that notion.

This is just a brief overview - there is a lot more.
Conclusion? All I can say that nobody should deny Robert's talent.

Oh, I forgot to mention some endgame evaluators quite far beyond the scope of what SF or Critter has... (hint: esp. bishop related)

Richard

benstoker
Posts: 342
Joined: Tue Jan 19, 2010 1:05 am

Re: Why are the Ippo derivative stronger than Stockfish?

Post by benstoker » Mon Feb 28, 2011 4:39 am

Don wrote:
rvida wrote:
lkaufman wrote: ...
Anyway you confirm my opinion that Houdini is an optimized/improved Ippo derivative
I think it is not appropriate to belittle Robert's contribution.

Please keep in mind that there was a lot of people trying to improve upon Ippo, but after Ivanhoe.xx <I do not remember the version number> only Robert has achieved a significant progress so far

1. optimized - yes, there are speed optimizations all over the place. But they give at most 10-12% speedup over latest Ivanhoes (I have not measured anything, it is just my upper-bound guesstimate).

2. improved - yes, in almost all areas. What is different you ask? Well, material balance, rook/queen mobility, contempt, king safety (safe queen contact checks & mate-in-1 threats), lazy eval and LMR in PV nodes, D.C. style smooth null-R scaling, null move threats canceling reductions, new (and very clever) way of updating "posgain" table, and -what I like most- is a never-seen-before new kind of 'horizon' extension.
None of these are new ideas including horizon extensions. Whether it's never seen before is not clear since I don't know what it is. But if you are impressed with those other things, then chances are this is also not a new idea. Houdart strikes me as a really good engineer, not an innovator and this does nothing to refute that notion. The cloning reinforces that notion.
Houdart a really good engineer ... guess that makes you a really good bug squasher.

Do pray tell, what innovation hath thy bestowed upon mankind, other than that find hand for sweeping?

This is just a brief overview - there is a lot more.
Conclusion? All I can say that nobody should deny Robert's talent.

Oh, I forgot to mention some endgame evaluators quite far beyond the scope of what SF or Critter has... (hint: esp. bishop related)

Richard
[/quote]

lkaufman
Posts: 3724
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Why are the Ippo derivative stronger than Stockfish?

Post by lkaufman » Mon Feb 28, 2011 4:40 am

rvida wrote: 1. optimized - yes, there are speed optimizations all over the place. But they give at most 10-12% speedup over latest Ivanhoes (I have not measured anything, it is just my upper-bound guesstimate).

2. improved - yes, in almost all areas. What is different you ask? Well, material balance, rook/queen mobility, contempt, king safety (safe queen contact checks & mate-in-1 threats), lazy eval and LMR in PV nodes, D.C. style smooth null-R scaling, null move threats canceling reductions, new (and very clever) way of updating "posgain" table, and -what I like most- is a never-seen-before new kind of 'horizon' extension.

This is just a brief overview - there is a lot more.
Conclusion? All I can say that nobody should deny Robert's talent.

Oh, I forgot to mention some endgame evaluators quite far beyond the scope of what SF or Critter has... (hint: esp. bishop related)

Richard
After reading this, I wonder why you said the improvements were Ippo-specific and not useful for other programs?Surely for example a "never-before-seen new kind of horizon extension" should be useful in Critter, SF etc., no?

I don't deny Robert's talent, but he should admit the obvious truth that Houdini started as Ivanhoe (or some Ippo derivative) and he has improved it, rather than pretending that he wrote Houdini himself.

benstoker
Posts: 342
Joined: Tue Jan 19, 2010 1:05 am

Re: Why are the Ippo derivative stronger than Stockfish?

Post by benstoker » Mon Feb 28, 2011 4:42 am

lkaufman wrote:
Houdini wrote:
rvida wrote:I would feel very uncomfortable revealing in public what I have learned from H1.5...
Just give appropriate credit in the next Critter release, for example "without ideas obtained by disassembling Houdini 1.5 this Critter version would not nearly have been as strong".

Robert
But he said that what he learned from Houdini 1.5 was Ippo-specific and not applicable to Critter. So either you are accusing him of lying, or you are asking him to lie!
Ahh Larry! Anger and jealousy can no more bear to lose sight of their objects than love.

benstoker
Posts: 342
Joined: Tue Jan 19, 2010 1:05 am

Re: Why are the Ippo derivative stronger than Stockfish?

Post by benstoker » Mon Feb 28, 2011 4:55 am

mcostalba wrote:
Houdini wrote: Again, I didn't blame, I suggested giving credit.
Robert, I think you've done a good job with Houdini, actually you are the only one that was able to obtain such a big boost out of Ippo. So I have already written and I have no problem to write again that I think you deserve sure credit for the improving you got with Houdini.

But I don't think would be correct to align all the developers on a single line: "let everybody give credit to Ippo becuase everybody took ideas from that !".

I think is not correct doing so because we blur the line between taking ideas and taking more than that.

If Richard says that Houdini was easy to disassemble because he already got 90% of source code, this for me is an important fact, also because I trust a lot Richard and I know he is not the guy that gives random numbers....

Now I stop because I think you already got the message. Anyhow let's make a funny deal: you write on your site that you started out of Ippo sources and we write on our that we have took ideas from Ippo :-) What do you think ?
Until such time as YOU or the ICGA panel fashions a sufficient definition of an impermissible derivative based on a set of specific engineering criteria, this casting of petty apsersions is just so much bullshit. You know nothing. You don't know what it means to say "got 90% of source code ...". It means absolutely nothing. If you cannot provide a definition and set of criteria for determining an impermissible derivative, then shut up. If you can provide such, then offer it to the ICGA to be vetted.

Post Reply