Twisted Logic new version!

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
bob
Posts: 20923
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Twisted Logic new version!

Post by bob » Thu Oct 01, 2009 8:10 pm

Eizenhammer wrote:
bob wrote:
Audmeister wrote:Sorry Bob, this was my fault, I was running a rybka because I was testing a book. I had it on my notes for a little while, then I reverted back to TwistedLogic and took it off.
There you go...

This is a real no-no on ICC. And they will actually ban you for doing so. ...
But making clone accusations based on nothing is a yes-yes?
Before starting such crap it should be yes-yes to download the "suspicious" version and have a look, instead of making up things based on some games played by whoever with whatever engine on just some server.
But of course you are not going to apologize, well, things never change ...
I made a very specific statement, about a very specific engine, on a very specific chess server. I _clearly_ said that _this_ particular engine, which is supposedly "blessed" by the author, is doing something bad. I don't see a single thing to apologize for. I said the NPS kibitzed was garbage, as was the search depth. And I was correct. I posted examples. Turns out the account _was_ running something other that what was claimed.

So exactly what is your problem with what I wrote? (besides the obvious one, that is). It was extremely specific, and provided exact details. All of which have been confirmed.

bob
Posts: 20923
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Twisted Logic new version!

Post by bob » Thu Oct 01, 2009 8:11 pm

Eizenhammer wrote:
Matthias Gemuh wrote:
Try to be a bit more precise.
Bob has not accused TwistedLogic of being a clone.
He merely said TwistedLogic was displaying dishonest search infos on ICC.

Matthias.
Mr. Hyatt: "While it is hard to get a handle on the "is this a clone" question, clearly when a program kibitzes nonsense, suspicion is more than justified."

I thought this was clear enough, but reading skills are a rare thing here.
Then you should work on yours. "TwistedLogicCCT" is the account used in the CCT events for TwistedLogic. It says it is twisted logic. It was running Rybka. Nothing I wrote was the least bit misleading.

Eizenhammer

Re: Twisted Logic new version!

Post by Eizenhammer » Thu Oct 01, 2009 9:28 pm

bob wrote:
Eizenhammer wrote:
Matthias Gemuh wrote:
Try to be a bit more precise.
Bob has not accused TwistedLogic of being a clone.
He merely said TwistedLogic was displaying dishonest search infos on ICC.

Matthias.
Mr. Hyatt: "While it is hard to get a handle on the "is this a clone" question, clearly when a program kibitzes nonsense, suspicion is more than justified."

I thought this was clear enough, but reading skills are a rare thing here.
Then you should work on yours. "TwistedLogicCCT" is the account used in the CCT events for TwistedLogic. It says it is twisted logic. It was running Rybka. Nothing I wrote was the least bit misleading.
Everything you wrote was complete nonsense, and totally misleading. I'll tell you what a naive reader like me had to think when reading your posting:

It started with this theme:
"Edsel Apostol wrote:
'There's a new Twisted Logic version here:
http://www.geocities.com/ed_apostol/index.htm
Just take a look at the readme file for more information.' "

Fine, a new version, always a pleasure for everyone.
Then the shock, all of a sudden:
Bob writes:
"I am not quite sure what to make of this program, but I can tell you one thing for certain. There is a certain level of dishonesty surrounding the thing."

Wow, I was shattered. Bob is certain, something is wrong with Twisted Logic. I had not expected this, but Bob would not write such a thing without hard evidence, of course.

He goes on:
"For example, I watched Crafty lose 3 of 4 games to it last weekend."

This seemed not too convincing to me, only 4 games, does not tell you a lot, so he must mean anything else.

" It was kibitzing (TwistedLogicCCT) and claimed to be searching 80K nodes per second. "

So it is about games on ICC, hardly a serious testing environment, one should think, so what is he after?

"I'd be willing to play _any_ program really searching 80K, when the box I was using was searching 20M nodes per second. And I would not expect to lose 3 of every 4 games. Period."

Period always convinces me: People say it rarely and only when there is no room for further debate left. And I understand now, of course, TL must be Rybka, this is what Bob is after, isn't he?

"That is a speed difference of 250X. Or with an EBF of 2, about 8 plies. The kibitzes are pure crap.
I don't believe the kibitzing info. Nor the depth info. In short, it sounds so much like the original Rybka fiasco I am not sure what to think."

The Rybka fiasco, where a giant group of highly competent elite programmers spent months and months to prove that there might be some similarities in the declaration of the uci parameters between rybka beta and fruit 2.1. Further evidence would have been so overwhelming that the group decided better not to show it.

Bob continues like this:
"One thing is for absolute certain, however. It is absolutely _not_ what it claims to be, in terms of the info it is kibitzing. I find it both amusing and disappointing that this kind of stuff continually comes up. And then people get upset when someone uses the "clone/illegal-copy" tag." As if they can't understand why such a claim would be made."

This is of course the very clever kind of saying that one is suspicous that leaves all the room in the world to say that it is not, never an accusation, look it up please.

It turns out that some guy played with rybka, indeed, on the very reliable testing environment ICC. A simple download of the public TL would have been possible, and probably a personal mail might have been more than enough to show that someone just made a mistake, but Bob has greater things in mind, listen:

"I have suspicions about _several_ current programs that are actively being tested and even participating in CCT-type events. My intent is to start naming names before the next event, to get some minimal level of honesty back into the process. Right now, it is at rock-bottom, IMHO.
While it is hard to get a handle on the "is this a clone" question, clearly when a program kibitzes nonsense, suspicion is more than justified. What's to hide? Why hide it? <sigh>

Wow, there is a flood of programs who showed strange behaviour when playing Crafty on ICC, this is so hard an evidence you just have to start naming names, everything else would be a complete no-no for an honest programmer.

Audmeister
Posts: 135
Joined: Sun Jun 01, 2008 7:38 am

Re: Twisted Logic new version!

Post by Audmeister » Thu Oct 01, 2009 10:21 pm

His accusation was reasonable at the time. I was playing rybka online to test a some books and it played crafty. Therefore his suspicion is reasonable.

But I have seen Mr. Hyatt accuse Edsel's work as a clone before, but that was back then. Now I made him re-think it. But I have to say that it was my fault for using rybka in some games to test out my new book.

Eizenhammer

Re: Twisted Logic new version!

Post by Eizenhammer » Thu Oct 01, 2009 10:45 pm

Audmeister wrote:His accusation was reasonable at the time. I was playing rybka online to test a some books and it played crafty. Therefore his suspicion is reasonable.

But I have seen Mr. Hyatt accuse Edsel's work as a clone before, but that was back then. Now I made him re-think it. But I have to say that it was my fault for using rybka in some games to test out my new book.
Thanks, but this changes nothing regarding such a public suspicion.
Anybody can play anything on an obscure server account, this happens all the time and everybody knows it. The author may set it up right, and in comes his evil grandmother and starts her beloved rybka, who cares anyway, it just does not matter and is no base for going public with some strong suspicions.

Audmeister
Posts: 135
Joined: Sun Jun 01, 2008 7:38 am

Re: Twisted Logic new version!

Post by Audmeister » Fri Oct 02, 2009 2:11 am

I agree, I do not like when people accuse people of things. I know that Edsel is a very honest person so it is a bit much when you accuse him of it :)

Oh and pray for him and everyone back in the Philippines due to the flood!

Werewolf
Posts: 1351
Joined: Thu Sep 18, 2008 8:24 pm

Re: Twisted Logic new version!

Post by Werewolf » Fri Oct 02, 2009 8:31 am

Eizenhammer wrote:
bob wrote:
Eizenhammer wrote:
Matthias Gemuh wrote:

"I'd be willing to play _any_ program really searching 80K, when the box I was using was searching 20M nodes per second. And I would not expect to lose 3 of every 4 games. Period."


"That is a speed difference of 250X. Or with an EBF of 2, about 8 plies. The kibitzes are pure crap.
So basically Bob's logic is that if a program is a truely slow-searcher it would not fair well against super-fast Crafty?

But then how does he explain HIARCS which is slow and strong?

Gian-Carlo Pascutto
Posts: 1238
Joined: Sat Dec 13, 2008 6:00 pm
Contact:

Re: Twisted Logic new version!

Post by Gian-Carlo Pascutto » Fri Oct 02, 2009 10:13 am

Werewolf wrote:
So basically Bob's logic is that if a program is a truely slow-searcher it would not fair well against super-fast Crafty?

But then how does he explain HIARCS which is slow and strong?
Emphasis mine :)

bob
Posts: 20923
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Twisted Logic new version!

Post by bob » Fri Oct 02, 2009 5:49 pm

Eizenhammer wrote:
bob wrote:
Eizenhammer wrote:
Matthias Gemuh wrote:
Try to be a bit more precise.
Bob has not accused TwistedLogic of being a clone.
He merely said TwistedLogic was displaying dishonest search infos on ICC.

Matthias.
Mr. Hyatt: "While it is hard to get a handle on the "is this a clone" question, clearly when a program kibitzes nonsense, suspicion is more than justified."

I thought this was clear enough, but reading skills are a rare thing here.
Then you should work on yours. "TwistedLogicCCT" is the account used in the CCT events for TwistedLogic. It says it is twisted logic. It was running Rybka. Nothing I wrote was the least bit misleading.
Everything you wrote was complete nonsense, and totally misleading. I'll tell you what a naive reader like me had to think when reading your posting:
Again, absolutely _none_ of what I wrote was "complete nonsense". The only nonsense here is what you are writing, without any thought or consideration first. The old "put mind in gear before putting fingers in motion" would be good advice. Point by point below:

It started with this theme:
"Edsel Apostol wrote:
'There's a new Twisted Logic version here:
http://www.geocities.com/ed_apostol/index.htm
Just take a look at the readme file for more information.' "

Fine, a new version, always a pleasure for everyone.
Then the shock, all of a sudden:
Bob writes:
"I am not quite sure what to make of this program, but I can tell you one thing for certain. There is a certain level of dishonesty surrounding the thing."

Wow, I was shattered. Bob is certain, something is wrong with Twisted Logic. I had not expected this, but Bob would not write such a thing without hard evidence, of course.

He goes on:
"For example, I watched Crafty lose 3 of 4 games to it last weekend."

This seemed not too convincing to me, only 4 games, does not tell you a lot, so he must mean anything else.

" It was kibitzing (TwistedLogicCCT) and claimed to be searching 80K nodes per second. "

So it is about games on ICC, hardly a serious testing environment, one should think, so what is he after?
"Hardly a serious testing environment?" When the _author_ uses that account to play in online tournaments? When the notes for the program say it is using twisted logic? When this account is entered in events with the authors claim that this is not a copy of any other existing program?


"I'd be willing to play _any_ program really searching 80K, when the box I was using was searching 20M nodes per second. And I would not expect to lose 3 of every 4 games. Period."

Period always convinces me: People say it rarely and only when there is no room for further debate left. And I understand now, of course, TL must be Rybka, this is what Bob is after, isn't he?
And was I wrong? You seem to miss that _tiny_ detail. It _was_ Rybka. It was claiming to be twisted logic. You do the math.

"That is a speed difference of 250X. Or with an EBF of 2, about 8 plies. The kibitzes are pure crap.
I don't believe the kibitzing info. Nor the depth info. In short, it sounds so much like the original Rybka fiasco I am not sure what to think."

The Rybka fiasco, where a giant group of highly competent elite programmers spent months and months to prove that there might be some similarities in the declaration of the uci parameters between rybka beta and fruit 2.1. Further evidence would have been so overwhelming that the group decided better not to show it.
Please either grow up or stay out of these discussions. The "original Rybka fiasco" was a clear reference to Rybka's displaying bogus node counts, bogus NPS values, bogus depth values. Anyone can see _exactly_ what I referred to since I used those explicit terms in my post, in a clear way. You are apparently just another "program xxx supporter" that wants to jump into waters where you don't belong.


Bob continues like this:
"One thing is for absolute certain, however. It is absolutely _not_ what it claims to be, in terms of the info it is kibitzing. I find it both amusing and disappointing that this kind of stuff continually comes up. And then people get upset when someone uses the "clone/illegal-copy" tag." As if they can't understand why such a claim would be made."

This is of course the very clever kind of saying that one is suspicous that leaves all the room in the world to say that it is not, never an accusation, look it up please.
What is to look up? I meant _exactly_ what I wrote. The person running the account them posted that he _was_ running Rybka while claiming to be TwistedLogic. So exactly _what_ is your complaint here? It should be with the person running one program while claiming to be another, on an account used by the author of the program to play in online events.


It turns out that some guy played with rybka, indeed, on the very reliable testing environment ICC. A simple download of the public TL would have been possible, and probably a personal mail might have been more than enough to show that someone just made a mistake, but Bob has greater things in mind, listen:

"I have suspicions about _several_ current programs that are actively being tested and even participating in CCT-type events. My intent is to start naming names before the next event, to get some minimal level of honesty back into the process. Right now, it is at rock-bottom, IMHO.
While it is hard to get a handle on the "is this a clone" question, clearly when a program kibitzes nonsense, suspicion is more than justified. What's to hide? Why hide it? <sigh>

Wow, there is a flood of programs who showed strange behaviour when playing Crafty on ICC, this is so hard an evidence you just have to start naming names, everything else would be a complete no-no for an honest programmer.
Again, read what I wrote, not what you imagine I wrote and meant. There _are_ several programs that are not what they claim. In CCT/ACCA events. In past WCCC events. And this was yet another case on ICC. Like it or not.

bob
Posts: 20923
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Twisted Logic new version!

Post by bob » Fri Oct 02, 2009 5:50 pm

Audmeister wrote:His accusation was reasonable at the time. I was playing rybka online to test a some books and it played crafty. Therefore his suspicion is reasonable.

But I have seen Mr. Hyatt accuse Edsel's work as a clone before, but that was back then. Now I made him re-think it. But I have to say that it was my fault for using rybka in some games to test out my new book.
I don't recall making very many clone claims over the years, with the exceptions that were copies of Crafty. Or the case of Rybka/Fruit.

Post Reply