| Morelia-Linares (2007) |
|
The 24th edition of the Linares tournament was held in two different countries (for the second time), with the first half in Morelia, Mexico (February) and the second half in Linares, Spain (March). The participants were: Veselin Topalov (Elo ranked #1 in the world), Viswanathan Anand (#2), Vassily Ivanchuk (#5), Peter Leko (#6), Levon Aronian (#7), Alexander Morozevich (#8), Peter Svidler (#12) and the prodigy Magnus Carlsen (#24). The most notable absentees were World Champion Vladimir Kramnik (#3) and Shakhriyar Mamedyarov (#4). Teimour Radjabov (#11) was scheduled to participate, but had to withdraw because of an upsetting burglery in his hotel room. Ivanchuk, who had been invited to give a simul and take part in some other events, was in town and agreed to jump in at short notice. He picked IM Manuel Leon Hoyos as his second. The opening ceremony was held 16 February at the playing venue Theatro de Campo. Games started the next day at 3:30 pm local time. Time control: 2 hours 40 moves + 1 hour 20 moves + 30 minutes. Leaving for Linares, Anand and Carlsen shared the lead with 4.5/7. Anand stated it was hard to imagine Carlsen would not become World Champion one day. In Linares, Carlsen's father returned home, and was replaced by Peter Heine Nielsen. In the end, Anand won by a whole point, while Carlsen took second place based on his win against Morozevich in Round 1. The prodigy had arrived at the top of the tables. Morelia, Mexico (17-25 Feb) and Linares, Spain (2-10 March) Age Elo 01 02 03 04 05 06 07 08
1 Anand 37 2779 ** 11 1½ 0½ ½½ ½½ ½½ 1½ 8½
2 Carlsen 16 2690 00 ** 1½ ½½ ½½ 11 1½ ½0 7½
3 Morozevich 29 2741 0½ 0½ ** ½½ ½1 ½1 01 ½1 7½
4 Aronian 24 2744 1½ ½½ ½½ ** ½½ 0½ ½½ ½½ 7
5 Svidler 30 2728 ½½ ½½ ½0 ½½ ** ½½ ½½ ½1 7
6 Ivanchuk 37 2750 ½½ 00 ½0 1½ ½½ ** 1½ ½½ 6½
7 Topalov 31 2783 ½½ 0½ 10 ½½ ½½ 0½ ** ½½ 6
8 Leko 27 2749 0½ ½1 ½0 ½½ ½0 ½½ ½½ ** 6
Category: XX (2746). Arbiters: Juan Vargas Sánchez and Faik Gasanov.The 2nd Morelia Open was played 15-18 February. It was won by Vadim Markovich Milov (6/7) ahead of Ivan Cheparinov (also 6/7). Sources
Spanish ChessBase, 17/02/2007 (https://es.chessbase.com/post/morel...)
ChessBase, 2/16/2007 (https://en.chessbase.com/post/teimo...)
ChessBase, 2/26/2007 (https://en.chessbase.com/post/morel...)
ChessBase, 3/10/2007 (https://en.chessbase.com/post/linar...)
FIDE rating list January 2007 (http://fidelists.blogspot.com/2008/...)
Ian Rogers in Tidskrift för Schack, 3/2007, pp. 4-16 (http://www.schack.se/tfsarkiv/histo...)
Dylan McClain in New York Times, 4 March 2007 (https://www.nytimes.com/2007/03/04/...)
Henrik Carlsen in Norsk Sjakkblad, 2/2007, pp. 10-17 (http://www.sjakk.no/filarkiv/nsf/no...)
Raul Ocampo Vargas at Chess Coach, 12 March 2007 (http://chesscom-chesscoach.blogspot...)
Various authors in Peón de Rey, No. 65 (April 2007), pp. 5-31 (https://e-nautia.com/santiago/disk/...) Round dates: February 17, 18, 19, 21, 22, 24, 25 in Morelia; March 2, 3, 4, 6, 7, 9, 10 in Linares (from ChessBase). Previous: Morelia-Linares (2006). Next: Morelia-Linares (2008)
|
|
|
|
< Earlier Kibitzing · PAGE 286 OF 287 ·
Later Kibitzing> |
| Mar-13-07 | | barbababa: <s4life> <You are referring to ELO scores there.>
No, I mean chessmetrics ratings.
<mean and standard deviation are usually the parameters that describe distributions... not the extreme values as you seem to think.> They both tell something of a distribution. If you have some clever idea, why don't you tell it? <whatthefat> Sonas fixes the average rating of #3-#20 players. If the players today are stronger (weaker) than 100 years ago, then there certainly is deflation (inflation) in chessmetrics ratings. |
|
| Mar-13-07 | | s4life: <barbababa: <s4life> <You are referring to ELO scores there.> No, I mean chessmetrics ratings.> Lol.. ok. So chessmetrics ratings are not comparable across time according to you. So be it then.. <They both tell something of a distribution. If you have some clever idea, why don't you tell it?> Extreme values of a distribution are not really informative, suffice to say smart people thought long ago (it's not my 'clever idea') about how to parametrize a distribution and came up with momentums. I'd love to explain you this further but I'll give your stats teacher a chance. |
|
| Mar-13-07 | | barbababa: <s4life> <Lol.. ok. So chessmetrics ratings are not comparable across time according to you. So be it then..> Finally you understood something. <I'd love to explain you this further but I'll give your stats teacher a chance.>
Could you then finally explain what did you mean by this: <take the first 2 momentums of both distributions. That should be plenty enough to make comparisons regarding strength.If you are willing, compute those momentums for the top10 and then for the top5. I am sure that's gonna be even more revealing...> |
|
| Mar-13-07 | | maxfrank: Moments (momentums???) are textbook MathStat 100. Can also be funky in practice. Understanding the limitations of parametric distributions, difficulties of analyzing tails of distributions, stochastics of dynamical systems, etc. is way beyond that - in particular, beyond the financial means of the chess economy. Sonas is doing OK with a difficult problem, and handles some of the technical challenges by using mathematical intuition properly. If he's tried to improve his system over time, that is much to his credit. Nobody has a better answer to comparison of chess performance over time. In effect, this is reduced to an indexing problem, but the value of the index (rating) is revealed by a probability of winning, as a function of the ratings of both players. The probabilistic nature of this relationship is part of the difficulty. As with other indexing problems (CPI, etc.) one can attempt to adjust the index for inflation. |
|
| Mar-13-07 | | yalie: < slomarko: I've to say I generaly agree with those who say that Anand's victory while deserved wasnt really impressive. Especially in Morelia, he has achieved nothing of out the openings, in fact against Svidler he a pawn down with white, against Moro he was almost lost with white, against Ivanchuk he was only saved by Chucky's time trouble and again it was with white. Then he was totaly outplayed by Aronian, and while in the last round he managed to turn the tables on Leko after coming out of the opening much worse. In Linares he changed tactics and started to play for short draws, apart from the game with inexperienced Carlsen where he played really well. Add to all that he had a really bad position against Svilder. I mean i dont want to take anything away from Anand he made the fewest errors of them all and deserved to win. And yet personaly i think Anand is on a descending curve. > Has it ever occurred to people that Anand didnt get anythiing outa openings bcos he is saving em for Mexico? |
|
| Mar-13-07 | | slomarko: <Has it ever occurred to people that Anand didnt get anythiing outa openings bcos he is saving em for Mexico?> Interesting idea. Only time will tell. |
|
| Mar-13-07 | | TheGladiator: <zarg: Rybka running on a laptop, has a great chess playing strength too, the main reason for this is not 2-4 GHz CPUs, but the improved algorithms, e.g. better evaluating function, used today.> First, about being a layman, it only pertains to cluster-computing. And I'm no hardware expert either, but I have a decent background in parallellization, distributed computing, algorithm theory, data communication _and_ general AI algorithms, with a focus on search (A* and friends.) I attended (and completed) master-level classes on most of these subjects while at the University, and of course I've got basic courses in hardware architecture and design too. I agree that my initial comment to <demondays> probably wasn't really called for (I think it was somewhat vague in several respects), but I meant no harm by it and I attached a smiley too :) Regarding the quote above, I'm not sure that the most important difference in playing strength between a current Rybka/Fritz 10 on current hardware with 1 GB ram, 2-4Ghz CPU and higher bus bandwidth and e.g. fritz 8 on say a Pentium 133Mhz box w/64Mb ram actually is the heuristic algorithms. Intuitively, based on general knowledge of computer chess and the importance of seeing some plies deeper on average, I think a major part of the increase in playing strength is due to the increase in "brute force elements" of the hardware. Of course I know that even 40x increase (100Mhz vs 4Ghz) in clock speed doesn't amount to very much deeper searches on average (due to the exponentially increasing number of nodes to search), but I still think this is a major factor in the improvement of the chess programs. The good thing about this, however, is that we need not merely "think" (have opinions) about this stuff, we could check it out. Which do you think wins in a 5 game match?
Match 1:
Fritz 10 on Pentium 133Mhz/64Mb ram
vs
Fritz 8 on Pentium 4 3.00Ghz/2 Gb ram
or Match 2:
Rybka 1.2 32-bit on Pentium 133Mhz/64Mb ram
vs
Fritz 8 on Pentium 4 3.00Ghz/2 Gb ram
On equal hardware (Athlon 64 X2 4600+ (2.4 GHz)) - see http://computerchess.org.uk/ccrl/40... - these engines were rated as such: Fritz 8: 2801
Fritz 10: 2873
Rybka 1.2 32bit: 2978
*Rybka 2.2 32bit: 2996
I'm not able to test any of the 64bit versions or 2CPU or 4CPU versions or commercial versions of Rybka, but the difference between Rybka 1.2 and 2.2 (32bit) in terms of rating (see above) isn't big. Game? |
|
| Mar-13-07 | | slomarko: <TheGladiator> what time control do you have in mind? |
|
| Mar-13-07 | | s4life: <barbababa: <s4life> <Lol.. ok. So chessmetrics ratings are not comparable across time according to you. So be it then..> Finally you understood something.
>
No, I am just treating you like a kid.. so you dont start crying. |
|
| Mar-13-07 | | TheGladiator: <s4life>
Nice comment. Keep it up. You said you considered "domination" as defined by Sonas to be a satisfactory substitute for "chess strength". Well, IMNSHO it's not - it's something very different - since you can't compare the players' strength over time by comparing their relative domination over contemporaries. However, by studying some rating details (games rated for various players and tournaments), I just observed some really serious flaws with the chessmetrics ratings, especially for the age listings of preteens and players in their early teens. With such flaws, Sonas' formulas become of secondary importance. But perhaps you know this already, with your excellence in statistics, s4life - or is it Sonas4Life ;) |
|
| Mar-13-07 | | barbababa: Elo rating tries to describe players absolute strength, but has problems with inflation/deflation. The rating floor causes inflation because low rated players can only be correctly rated or over rated. If they lose their rating they have added some points to the rating pool thus causing inflation in ratings. The players who stop playing chess cause on average deflation, because their rating is usually higher when they quit playing than when they started playing. I wonder why FIDE does not try to fix these inflation/deflation sources. Simply remove rating floors and let the bad players ratings drop as low as they go. To get rid of the deflation they could save the players initial ratings and when a player has been inactive long enough (5-10 years) he loses his rating. Then calculate the difference between the initial and final rating of the player who has stopped playing and add it to retired players pool. In the end of the year there should be as much points in retired players pool as the retired players have taken out of the system. Then simply add these extra points to all active players, either equally or according to some distribution. Next year repeat the same. |
|
| Mar-13-07 | | TheGladiator: <If they lose their rating they have added some points to the rating pool thus causing inflation in ratings.> This is true, and would create real inflation (more rating points in the pool without increasing the number of players, which increases the average). However: <Elo rating tries to describe players absolute strength, but has problems with inflation/deflation.><I wonder why FIDE does not try to fix these inflation/deflation sources.> Removing the inflation/deflation sources that you mention, will not make for a system where the ELO rating describes "absolute strength". The reason is, I think, quite simple - experts in statistics like <s4life> might even describe it mathematically for us - but in a system like this which theoretically is based on Gaussian distribution (due to the historical rating floors, this isn't the exact distribution in FIDE's system of today), the more ppl that enter the system, the higher and lower the maximum and minimum values will become. This will happen, even though the rating average of the pool stays the same. Trying to adjust these relative measures doesn't make much sense, since they will still be relative measures of results between players who actually played each other (directly or "indirectly"). Unless rating was a measure of "quality, knowledge and smartness/sporting ability", I see no point in trying to artifically adjust the relative numbers. I'm not sure if I understood Sonas correctly, but it seems to me that what he does, in practical terms could be seen as introducing and enforcing not only a rating floor ("limit") but also a rating ceiling. This is a simplification, but the rating then just becomes a number on a scale from A to B, where those closest to A are the least dominating and the ones closest to B are the most dominating. This could easily be projected in to a number in the range between 0 and 1000, for instance - which I think would've been a more honest way of representing these "inflation/deflation-adjusted" ratings. Then ppl would at least not compare them to FIDE ratings in the same way as many tend to today. But again - what could I read out of such a system if say one player was rated 450 in 1980 and another (or the same) was rated 540 in 2005? I think such a comparison would tell me zilch and be really worthless. |
|
| Mar-13-07 | | barbababa: <TheGladiator> <the more ppl that enter the system, the higher and lower the maximum and minimum values will become. This will happen, even though the rating average of the pool stays the same.>
This is true, but more players also mean more stronger players (in sense of their real playing stregth). Imagine that we have a perfect rating system that describes players absolute strength and has no inflation/deflation. If we today, with the current number of active players, should have on average one player rated above 2800, then in the future when the number of players has increased tenfold, we should have on average 10 players above 2800 (assuming that the new payers follow the same distribution as old players). The average of todays top ten players would be clearly under 2800 and the average of future top ten players would be over 2800. However, this increase would not be due to rating inflation, but because the future top ten players absolute strength would really be better. |
|
| Mar-13-07 | | sheaf: <csmath> dortmund 1996 category 18 anand score a +5 and was joint winner with kramnik, key point is that there were only 9 rounds ! |
|
| Mar-13-07 | | botevist: Since there is a lot of nonsense regarding ELO inflation, etc., let me illuminate the matter. Of course ratings are comparable between eras, if you apply the correct approach. It would be impossible if a group of players played 1960-1970 only and a completely different group of players played 1970-1980 with no overlap. In reality generations overlap and there is enough for statistics. Disregarding age, style, etc., which can be averaged out with more players, if in a reasonably big number of games A beats B beats C, then it means A beats C. Besides, what does inflation mean? Is there any doubt that a top modern player will destroy any master from 100 years ago? |
|
| Mar-13-07 | | Pawnsgambit: <botevist> The things are not that simple. In most of the cases A beats B. beats C, however C beats A. or consider A beats B.B beats C. C beats D. D beats both A & B. its gets very complex. |
|
| Mar-13-07 | | TheGladiator: <botevist: Of course ratings are comparable between eras, if you apply the correct approach.> No matter what "approach you apply", of course ratings aren't comparable between eras. Disregarding age, style, etc. - which can't be averaged out with any number of players - still, even given a reasonably high number of games where A beats B beats C, then you wouldn't know much about A and C, if A beat B 5 years or more before B played the first game against C. In the general case, the outcome of chess games aren't transitive unless there is a real huge difference in strength between the players involved. And in reality, you don't play a relatively high number of games against a player out of your league. If A only narrowly beats B (like in a 15-10 result), then differences in style and the very realistic possibility of evolution/change of/in strength between the time A plays B and B plays C, will guarantee lack of transitivity. Besides, changes in the administration of rating (for instance rating frequency), notable increase in number of players in the system, and reasons mentioned earlier, will make the nominal numbers change anyway, even if chess completely stopped evolving in the scientific sense. <botevist> - I'm sorry, but I really can't agree that this was very illuminating. Referring to previous debate as "nonsense", kindly offering to "illuminate matters", and stating that the view you offer is "obvious" ("Of course <claim A, claim B, etc.>") don't make your claims more true, unfortunately. Not less either, but you get my drift... |
|
| Mar-13-07 | | TheGladiator: <barbababa: (assuming that the new players follow the same distribution as old players).> Is this something we safely can assume? Just consider what is happening now with the FIDE system - 10 000s of new players enter the system, but the majority of them are weaker than what the lowest rated players used to be when the rating floor was e.g. 2000. Also consider what happened on ICC when its audience was widened some years ago - the difference between max and min ratings has increased a lot, but it seems that there are quite many who used to be 3000-3100 who now are 3300-3400. Of course, some of these have improved, but this doesn't tell everything. The thing about ICC compared to FIDE ELO, though, is that the number of games played is very much higher, and so one would believe that the effects of the increase in total # of players are seen faster. <However, this increase would not be due to rating inflation, but because the future top ten players absolute strength would really be better.> I wouldn't call it inflation anyway, if the average of the pool didn't change, but I'm not so sure that the absolute strength of all of the future top ten players would really be better, for reasons stated above. While I partly agree in principle, I think there at least is some limit to this, and that changes of the distribution of real playing strength must be anticipated. |
|
| Mar-13-07 | | whatthefat: <barbababa: Sonas fixes the average rating of #3-#20 players. If the players today are stronger (weaker) than 100 years ago, then there certainly is deflation (inflation) in chessmetrics ratings.> Indeed, and I've tried asking him about this directly, but I'm yet to receive a reply. I suspect that in the end, it's very difficult (impossible?) to build a retroactive ratings system that copes with inflation 100%. |
|
| Mar-13-07 | | barbababa: <TheGladiator> <Is this something we safely can assume?> Nope. I think it is the simplest possible assumption one can make, but somehow I doubt it is that simple. I have no idea how to prove or disprove it. Today people have much more free time and money and it is much more easier to go to an international chess tournament and to get an Elo rating. Earlier money and time problems might have caused some kind of threshold. There are many other possible reasons that can affect the strength distribution of new players. Still, unless there is some clear, strong reason that changes the distribution, it could be close to strength distribution of old players. I don't know much about ICC. If the number of players increases suddenly, then I would expect that the the average rating of top 10 increases and the average rating of bottom ten decreases. ICC probably don't have a rating floor, the deflation mechanism of retired players they probably have, but I am not sure ICC is old enough that it has allready started to effect. Has ICC published any data about this that could clarify the situation? <... but I'm not so sure that the absolute strength of all of the future top ten players would really be better..> In my imaginary case, the average rating of top ten players would be higher and since we have perfect rating system their absolute strength would also be higher than top ten players of today. This is simply because if you add enough players, sooner or later you will find another kasparov and even later you will find ten kasparovs (or better) and then the top ten's absolute strength is certainly better than today. |
|
| Mar-13-07 | | TheGladiator: <This is simply because if you add enough players, sooner or later you will find another kasparov and even later you will find ten kasparovs> Yeah, but like you say, this is an imaginary case. In reality there are only so many "gifted ppl" that potentially will take up chess (for lots of reasons - lack of money/sponsors for more than 50-100 professional players (to be really generous) is one, for instance), while we can expect that more and more tournaments between "normal club players" will get rated by FIDE. Maybe some or most national federations will decide to do away with "legacy" (custom) rating systems, and rather have FIDE rate their "national" games? This is of course a practical consideration more than a theoretical one, but that's the nature of it, IMHO. And by the way - Sonas' system gets into real trouble with lots of players whos games only occasionally appear in international game collections (as ChessBase's Big Database or their Online base). The effect is that active chess players (nationally or even internationally) still are treated as unrated when they occasionally show up as opponents of more published players. This despite of having active ratings both nationally and in FIDE, so they are inactive and unrated only to the eyes of ChessMetrics. Due to geographically differences and such similiar (games by german "club players" are for instance overrepresented in ChessBase's Databases), even how two contemporaries are rated and compare to eachother, might hing on a more or less systematic difference in which or how many of their games (in international tournaments) will be rated. To this comes the number of non-classical games that probably still are included in the source data. As an example, one can consider a superstrong international event like Aeroflot Open (A-group, ELO 2550-) where lots of players end up with only a small number of games against "rated opponents", and who gets for instance the highest TPR there is decided just as much by who was "luckiest about which games got rated" than each players actual result. With this kind of "bad data", the relevancy of the formulas' "correctness" kind of fades, to me at least... |
|
| Mar-14-07 | | zarg: <TheGladiator:
Of course I know that even 40x increase (100Mhz vs 4Ghz) in clock speed doesn't amount to very much deeper searches on average (due to the exponentially increasing number of nodes to search), but I still think this is a major factor in the improvement of the chess programs.>The IBM Power2 Super Chip of Deep Blue, was running at low clock speed (135 MHz), but you cannot say a 4 GHz CPU of a different architecture is 40 times faster. The P2SC chip could execute 8 instructions simultaneously. However, the IBM CPU was targeted for super-computing (highly optimized for floating-point math), but had 16 coprocessors, where each could search 2+ million np/s. So, the performance of the 32x P2SC chips, wasn’t that important, since there was 512x coprocessors (ASICs), executing "the inner chess loop". Today, AFAIK only Hydra can match Deep Blue brute-force search of 200 million np/s. < The good thing about this, however, is that we need not merely "think" (have opinions) about this stuff, we could check it out.> Well, Deep Blue was far from running Fritz 8 software and Pentium 133 MHz was not the same thing as a Power2 CPU with 16 coprocessors. :-) However, still it would be interesting to check the advance in HW vs chess SW in the last 10 years. IIRC, my home PC in 1997 was rather a Pentium 2, so Pentium 133 MHz would be pre-97 hardware. Was Fritz 8 really available back in 1997? My guess would be Fritz 5 or 6. If doing a match between Rybka 1.2 32-bit on Pentium 2 (200-300 MHz), vs Fritz "97" on Pentium 4 (3 GHz), my bet would be on Rybka. |
|
| Mar-14-07 | | slomarko: I'd bet on Rybka too! |
|
| Mar-14-07 | | TheGladiator: <zarg: <Of course I know that even 40x increase (100Mhz vs 4Ghz) in clock speed doesn't amount to very much deeper searches on average (due to the exponentially increasing number of nodes to search), but I still think this is a major factor in the improvement of the chess programs.> The IBM Power2 Super Chip of Deep Blue, was running at low clock speed (135 MHz), but you cannot say a 4 GHz CPU of a different architecture is 40 times faster.> I wasn't referring to Deep Blue anymore in the quote above; I was talking about faster CPUs on (basically) the same architecture. And for the rest of that post about how much improvement of the hardware part of the equation possibly adds to the chess strength compared to better heuristics (and better tuning of the basic depth-first with pruning algorithms) for PC programs. <Well, Deep Blue was far from running Fritz 8 software and Pentium 133 MHz was not the same thing as a Power2 CPU with 16 coprocessors. :-)> Well, of course! I wasn't suggesting anything of the kind :) <However, still it would be interesting to check the advance in HW vs chess SW in the last 10 years. IIRC, my home PC in 1997 was rather a Pentium 2, so Pentium 133 MHz would be pre-97 hardware. Was Fritz 8 really available back in 1997? My guess would be Fritz 5 or 6.If doing a match between Rybka 1.2 32-bit on Pentium 2 (200-300 MHz), vs Fritz "97" on Pentium 4 (3 GHz), my bet would be on Rybka.> Wasn't Pentium 2 (first time around) 166/200/233 Mhz? Even though I've got both fritz 5.32 and fritz 6 available, I don't have any (rating) estimates of difference in strength beetween rybka 1.2 and those old fritzes, so I think I still would prefer using fritz 8. The software alone contributes to a 200 point difference in strength on current hardware, but does hardware do the trick also? Whether we test against pentium 2 (got one 166Mhz in the attic), pentium 3 (I've also got a 500 Mhz box back home) or my ancient 133Mhz box (which I got chiep and was a faithful router (running linux) until recently, now retired ;) I don't think is very important - it just was a scenario in which we knew the "before" relationship between the programs where we also can get at least _some_ intuition about the hardware part of the equation. If you don't mind, I'll use two of my old boxes for rybka and run fritz 8 on "modern" hardware. Is your bet still on Rybka? |
|
| Mar-14-07 | | zarg: <TheGladiator: <zarg: <Of course I know that even 40x increase (100Mhz vs 4Ghz) in clock speed doesn't amount to very much deeper searches on average (due to the exponentially increasing number of nodes to search), but I still think this is a major factor in the improvement of the chess programs.>
The IBM Power2 Super Chip of Deep Blue, was running at low clock speed (135 MHz), but you cannot say a 4 GHz CPU of a different architecture is 40 times faster.> I wasn't referring to Deep Blue anymore in the quote above; I was talking about faster CPUs on (basically) the same architecture.> OK, but context of this discussion, was comparing current desktop CPU with Deep Blue hardware. Anyway, even when comparing different models of the same CPU family, care must be taken, benchmarking is better than looking at just the clock speed. <Wasn't Pentium 2 (first time around) 166/200/233 Mhz?> IIRC, the slowest Pentium 2 ran at 233 Mhz, you might be thinking of Pentium Pro. Pentium 133 MHz sound like 1995 hardware. :) <I think I still would prefer using fritz 8.> Which year did Fritz 8 arrive? Then I suggest you run Rybka on hardware available at that point. You might get in trouble running Rybka on a very old Pentium box, because MMX instructions was not available before Pentium Pro (AFAIK). I don't know if Rybka uses such speedup trick though. <If you don't mind, I'll use two of my old boxes for rybka and run fritz 8 on "modern" hardware. Is your bet still on Rybka?> I will still bet on Rybka....
IF you run it on the "best" available desktop PC hardware, when Fritz 8 hit the market AND
Fritz 8 is run on the "best" available desktop PC hardware, when Rybka 1.2 hit the market. :) |
|
 |
|
|
< Earlier Kibitzing · PAGE 286 OF 287 ·
Later Kibitzing> |
|
|
|
NOTE: Create an account today
to post replies and access other powerful features which are available only to registered users.
Becoming a member is free, anonymous, and takes less than 1 minute! If you already have a username,
then simply login login under your username now to join the discussion.
|
Please observe our posting guidelines:
- No obscene, racist, sexist, or profane language.
- No spamming, advertising, duplicate, or gibberish posts.
- No vitriolic or systematic personal attacks against other members.
- Nothing in violation of United States law.
- No cyberstalking or malicious posting of negative or private information (doxing/doxxing) of members.
- No trolling.
- The use of "sock puppet" accounts to circumvent disciplinary action taken by moderators, create a false impression of consensus or support, or stage conversations, is prohibited.
- Do not degrade Chessgames or any of it's staff/volunteers.
Please try to maintain a semblance of civility at all times.
NOTE: Please keep all discussion on-topic.
This forum is for this specific tournament only. To discuss chess or this site in general,
visit the Kibitzer's Café.
|
Messages posted by Chessgames members
do not necessarily represent the views of Chessgames.com, its employees, or sponsors.
All moderator actions taken are ultimately at the sole discretion of the administration. |
Spot an error? Please suggest your correction and help us eliminate database mistakes!
Copyright 2001-2025, Chessgames Services LLC
|
Kibitzing
