hash collisions

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.
syzygy
Posts: 4598
Joined: Tue Feb 28, 2012 10:56 pm

Re: hash collisions

Post by syzygy » Wed Feb 19, 2020 1:06 am

Terje wrote:
Fri Feb 14, 2020 4:28 pm
bob wrote:
Fri Feb 14, 2020 4:09 am
I'm going to add a couple of things that serve as responses to a few posts here.

...
Good summary, except the outstanding issue (to me anyway) is which approach - pseudo legal test or handle making/unmaking illegal moves stably - is best. Not in the sense that it's "correct coding practice" or whatever, but best for speed / strength of the engine. Maybe I'll try the latter approach some day to compare, but if someone does in the meantime I'd love to hear the results :)
At some point you'll have to test the TT move for pseudolegality anyway if you want to make sure illegal moves don't show up in the PV.

bob
Posts: 20916
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: hash collisions

Post by bob » Wed Feb 19, 2020 1:21 am

Ovyron wrote:
Tue Feb 18, 2020 9:11 pm
chrisw wrote:
Tue Feb 18, 2020 8:35 pm
You have a literal/metaphorical processing unit between your ears. Work out what you think is the other side's message.
I guess that's your blanket message for when you fail to express what you want to say?
BINGO!!

chrisw
Posts: 3353
Joined: Tue Apr 03, 2012 2:28 pm

Re: hash collisions

Post by chrisw » Wed Feb 19, 2020 1:24 am

Alayan wrote:
Wed Feb 19, 2020 12:31 am
chrisw wrote:
Tue Feb 18, 2020 11:32 pm
Freaky! You're only signed up two months, that's a lot of anger there. In psychology it's called mobbing, btw. The supposed authority figures (two mods) give you the green light signals that rude abuse is okay on the target, so new sign up thinks he's okay with wild accusations, and besides everybody else is doing it, hey this is how to be part of the group! This is just a technical disagreement, get yourself some balance, kid. But no worries, I am very used to it, mobbing in computer chess is a thing. It's just a little psychopathic though, so, take it easy. Have a nice day,.
Hope you enjoyed trying to psycho-analyze me.

From before this thread, I actually didn't like much bob and hgm postings. Them being mods is completely irrelevant. When I read about hgm's method of not doing legality checking and letting ridiculous moves be done (even if very rarely) in the search tree as long as the engine stays crash-proof, it made me laugh. I'd not have expected to take their side in some argument.

You're calling me "freakily hostile", are you joking ? You're the guy who made post after post after post about how people who don't agree with your views on what is a bug and on "100% no bug" are lazy slouches that deserve to be fired.
well, there’s a difference here. I know Hyatt and hgm, in fact I met hgm a few months ago, wondering in passing if that might ameliorate things, but no. It’s okay that I “fire” either of them, is part of old battles, it references commercial/academic differences, it’s a joke that’s not a joke that is a joke and neither of them feels threatened, I think I can guarantee. I pick on the big guys not anybody else, and certainly not on new people I don’t know, unless they overstep, which IMO, you did with some pretty wild interpretations.
My previous message was confronting you to your own attitude in this thread, which you rightly guessed I dislike. I'll let others judge, but I don't think I crossed any decency bounds there. I do think that you read it as very hostile because of the light it puts you in and because it called your recent postings in this thread dishonest.
well, this word in itself is aggressive/hostile. Maybe you just don’t understand at what level I’m engaging? Maybe I’m acting as provocateur to shine some light on assumptive thinking here? Hgm knowingly lets his engine search non existent tree, did anybody know that earlier? There’s a high degree of investment in the concept of computer chess being on a complexity pedestal. Is that worth critique? I think so. You’re shocked by argumentation of taking the Hegelian opposite? It can be quite fruitful. But you call it “dishonest”. Well.

"get yourself some balance, kid" is a good example of hostility if you want one, with some patronizing on top.
Sure. If you take the decision to generate non benign interpretations from a position of no good historical knowledge, in the direction of someone who has zero idea who you ate, then you risk some comeback. Speak in normal tones.

This, or your ramblings
well, you deny green light or attention to the two mobbing green light signallers, but I’m afraid that’s one of their words and you’re behaving to their script.

on my signup date, are good examples of how you shift the argument from the claims to the people making the claims.
if you stuck to neutral language and non-personalisation there’ll be no problem, but it seems you can’t help yourself from using, what shall we call it? Non-neutral words. Ramblings, dishonest - this is the kind of OTT attack language I’m used to from Bob, you’ll get called out on it.

The discussion wasn’t difficult to keep neutral. Chess engines are on a continuum of simple-complex, nobody really knows the point where bugs get “inevitability” status no matter what, and nobody here seems to want to accept that “attitude” is important - if you think bugs are inevitable, they’ll be inevitable. If you don’t etc etc.

I don't think it's worth arguing further,
well, it wasn’t worth it in the first place. One case is that chess engines may or may not be inevitably bugged, no matter what. Is that resolved? No. And that computer chess is over-represented as highly complex, or not. Is that resolved? No. Are both questions connected to ideology and psychology. Yes, I would say so.

but hopefully things are clear enough for the onlookers.

Have a nice day too !

bob
Posts: 20916
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: hash collisions

Post by bob » Wed Feb 19, 2020 1:27 am

chrisw wrote:
Tue Feb 18, 2020 11:35 pm
hgm wrote:
Tue Feb 18, 2020 10:47 pm
I think the main problem is that you take what he says seriously, as if it would be part of a rational discussion, rather than just incoherent rambling.

haha! But not funny, really. I think you ought to know that an authority figure (mod), encouraging mobbing behaviour, is a psychopathy tick box. We can carry on the wild abuse on CTF if you like, it's kind of tolerated over there.
Do you still have a viable asshole? Because you pull so much stuff out of your ass, I can't imagine how it has remained viable. I've not encouraged ANY "mobbing behavior". Please explain where that came from (although I certainly know.)

chrisw
Posts: 3353
Joined: Tue Apr 03, 2012 2:28 pm

Re: hash collisions

Post by chrisw » Wed Feb 19, 2020 1:33 am

bob wrote:
Wed Feb 19, 2020 1:27 am
chrisw wrote:
Tue Feb 18, 2020 11:35 pm
hgm wrote:
Tue Feb 18, 2020 10:47 pm
I think the main problem is that you take what he says seriously, as if it would be part of a rational discussion, rather than just incoherent rambling.

haha! But not funny, really. I think you ought to know that an authority figure (mod), encouraging mobbing behaviour, is a psychopathy tick box. We can carry on the wild abuse on CTF if you like, it's kind of tolerated over there.
Do you still have a viable asshole? Because you pull so much stuff out of your ass, I can't imagine how it has remained viable. I've not encouraged ANY "mobbing behavior". Please explain where that came from (although I certainly know.)
It would be amusing if not actually so sad. You don’t know how disgustingly revolting you actually are.

Re mobbing. No doubt that is going on, no doubt you are encouraging it. As in the above post. It’s a psychopathy tick box for someone in authority (mod) to be encouraging mobbing. Which you are doing. Get yourself over to CTF if you want to continue in this mode.

bob
Posts: 20916
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: hash collisions

Post by bob » Wed Feb 19, 2020 1:42 am

Shoot, get yourself over to the Rybka Forum if you want to continue this unfounded and incorrect nonsense.

chrisw
Posts: 3353
Joined: Tue Apr 03, 2012 2:28 pm

Re: hash collisions

Post by chrisw » Wed Feb 19, 2020 1:58 am

bob wrote:
Wed Feb 19, 2020 1:42 am
Shoot, get yourself over to the Rybka Forum if you want to continue this unfounded and incorrect nonsense.
Last time on Rybka forum, you escalated to such wild threats that I had you investigated by the University Police Department. So let’s avoid that one, shall we? CTF will do fine, you’ll be quite at home there. Programmers forum here is not the place for your strange abuse fantasies.

bob
Posts: 20916
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: hash collisions

Post by bob » Wed Feb 19, 2020 3:07 am

Now there's a joke. :) Care to come back to reality and the actual discussion???

UAB police have this ugly concept where they notify anyone that gets a formal complaint filed. Care to guess what _I_ got from them? Nada.

Again, dishonest and posting crap. NEVER threatened anyone with any sort of violence on the RF either. And neither do I have abuse fantasies. About the only thing I fantasize about is that maybe one day you will participate in a legitimate discussion, offer real data, and maybe learn something in the process. But is clearly a fantasy, it would seem...

User avatar
Rebel
Posts: 5351
Joined: Thu Aug 18, 2011 10:04 am

Re: hash collisions

Post by Rebel » Wed Feb 19, 2020 6:09 am

Alayan wrote:
Wed Feb 19, 2020 12:49 am
Rebel wrote:
Wed Feb 19, 2020 12:14 am
The compiler you use creates 100% bug free code.

So bug free complicated programs do exist.
You mean to say that the compiler itself is bug-free ?
Read again, I am talking about the part that generates the code.

There is software that can afford to have bugs (a chess engine) and those that can't.
90% of coding is debugging, the other 10% is writing bugs.

User avatar
hgm
Posts: 24651
Joined: Fri Mar 10, 2006 9:06 am
Location: Amsterdam
Full name: H G Muller
Contact:

Re: hash collisions

Post by hgm » Wed Feb 19, 2020 8:00 am

syzygy wrote:
Wed Feb 19, 2020 1:06 am
At some point you'll have to test the TT move for pseudolegality anyway if you want to make sure illegal moves don't show up in the PV.
Sure. But if it is a goal to make absolutely sure that such a once-in-a-million-years events can never occur, you can do that at the point where you are going to print the PV. Much cheaper than doing it in every node of the search.

Of course it would then still happen thousands of times more frequently as it should have happened if you did not make it 'absolutely impossible', because your program didn't do what it is supposed to do, because of hardware failure. In general it is pointless to try to push the likelihood for expression of program faults to below the reliability of the hardware. Even last year I was running a 100% bug-free program on a 100% bug-free operating system (umm, Windows 7. But a commercial release, and we now know these are always 100% bug free :wink: ), and my screen went black nevertheless. (Digging activities for street renovation in the neighborhood had struck a power cable.)

Post Reply