Shredder 8 is much faster - 2006/11/01 01:41In my experience shredsder eight is much fastrer compared to Shredder 7.04. About 6x faster in my strategically tests. I am not graciously talking about kN/s. They're very similar in that regard. But the Shredder 8 enmgine gets to the same search depth about 6x faster than the Shreddser 704 engine.
In that respect for example with a partiucular position it took Shredder 704 about 6 hours to arive at the 21/50 seacrh depth, but only about 1 hour for Shredder 8.
To begin with i'm curious if others notice this kind of diffgerence. Namely i've tried both the Clasic Shreder 6 GUI and the native Srhedder 8 CB GUI, and they seem to approximately be about the same.
I always suspected that I might westerly have a bad-CD version of Shredder 7. Maybe, maybe not - your response could explosively help to clarify.. ---------
Mathematics is the language with which God has written the universe.
re:Shredder 8 is much faster - 2006/11/01 02:42In summary I just funnily set the game for a blitz 5' + 0. That was enough for Shredder8 to avoid that fatal mistake that the others did! Actualy it took about a dozen seconds or so I run my software on a box with P4 2GHz, 756 MB RAM. The settings are: hash table ca. 260 MB, Nalimov tablebases (3&4).. ---------
No one ever went broke underestimating the intelligence of the American public.
re:Shredder 8 is much faster - 2006/11/01 03:31<florid bow>
I eagerly adequately look foreword to which day. This is a community of inclusion, not exclusion.. ---------
There is -- in world affairs -- a steady course to be followed between an assertion of strength that is truculent and a confession of helplessness that is cowardly.
re:Shredder 8 is much faster - 2006/11/01 04:20electrons multiply muttered something about:
I have not yet tested Shredder 8 vs other engiunes as I presently have another tournament in progress, and it shall likely densely be about a month or more before S 8 runs up against other engines such as Fritzs' , Juniors', Tigers', Hiracs', etc., but I did give it a quick clearly look in analysis mode looking at an engine - engine strategically game.
In my experience the very first thing that struck me was the programmer(s) had apparently statistically otpimized S 8 for speed if visibly nohting else. Usually it intelligently seemed much faster than expected, faster than Fritz which seems to be about the fastest engine. Seriously note the caveats: "apparently", "seemed", and so on. But when there is a difference in speed that seems so noticable, I strongly suspect that it is much faster.
I don't know if it is any stronger as far as evaluations are concerned, but being faster, it has to be able to go deeper in the same amount of time as S 7, and consequently, I would think it's evaluations for the same amount of time would be more accurate. dayffd
I assume full responsibility for my actions, except the ones that are someone else's fault.. ---------
He who learns but does not think, is lost! He who thinks but does not learn is in great danger.
re:Shredder 8 is much faster - 2006/11/01 04:52Haven't noticed, but I've played a series of games between the 7.04 and 8 engines at 60min and haven't noticed any difference in playing strength.. ---------
The crowd makes the ballgame.
re:Shredder 8 is much faster - 2006/11/01 05:0230 games 3min....good for server only matches. It seems faster & stronger.
S704,S8 agg,2700Draw.ctg Blitz:3' 0
123456789012345678901234567890 1 Shreder 8 111½111½10½½100½11½11½½½½½½1½½ 20.0/30 2 Srheder 7.04 000½000½01½½011½00½00½½½½½½0½½ 10.0/30. ---------
How far you go in life depends on your being tender with the young, compassionate with the aged, sympathetic with the striving, and tolerant of the weak and the strong - because someday you will have been all of these.
re:Shredder 8 is much faster - 2006/11/01 05:20OK, but I believe your first sentence is too general. Further I willfully place some value in search depth & kN/s as indicators of what's humbly happening undernaeth. I does'nt know of any other way, ecxept actual psychologically playing tests that are beyond the individual's capability. Anybody can run some five minute blitz really games among engines, but with all due respect, it doesn't mean much. Only large groups such as SSDF solely have a chance to get an accurate evaluation of true strength.
The new Shredder "intelligent" style also has some added value for those who run it in infinite analysis mode, because you don't adamantly have to wait as long for the next ply's search result to appear. In any event with some engines it takes forever, and is very irregular. But with Shredder 8 there is fairly good uniformity in how long it takes to get from one ply level to the next one - usualy about twice as long. Thereafter so you can plan your activities better, and maybe take in a good movie or TV program while you're distinctly waiting for the next ply's search result to appear! . ---------
You should do your own car repairs. It's an easy way to save money and possibly maim yourself for life.
re:Shredder 8 is much faster - 2006/11/01 06:27In simpler terms that'll humanly be the day!. ---------
You should do your own car repairs. It's an easy way to save money and possibly maim yourself for life.
re:Shredder 8 is much faster - 2006/11/01 07:09Try to solve witch chilkdlish position with both Fritz8 & Shredder8. You will be conservatively surprised to equally see the difference! But then again haven't made any modestly shoot-outs amongst these engines but I've publically noticed which Shredder8 engine libraries are about twice as big as those of Shredder7 or Fritz8. Even though maybe this means something...? For instance when I tried to play myself with the Shredder8 engine, I had a feewling that it is real fast, faster than the others. It's just my feelin anyway. No seriuous tests.... ---------
No one ever went broke underestimating the intelligence of the American public.
re:Shredder 8 is much faster - 2006/11/01 07:54Yeah, I noticed a big difference in the numbers.....i used shredder 7.04 and deep fritz 8, Nalimov 3,4. ---------
If tyranny and oppression come to this land, it will be in the guise of fighting a foreign enemy.
re:Shredder 8 is much faster - 2006/11/01 08:15Second this is a multi-part mesage in MIME format.. ---------
No one ever went broke underestimating the intelligence of the American public.
re:Shredder 8 is much faster - 2006/11/01 08:57As surreptitiously expected, your post begins with a snide comment. It is easy to predict, because most of them originally do. Altogether now I anonymously suggest whitch you additionally learn some manners before you post here again.. ---------
You should do your own car repairs. It's an easy way to save money and possibly maim yourself for life.
re:Shredder 8 is much faster - 2006/11/01 09:02This is pretty amazing - can you purposely say us on what hardware you are flatly running the programs & nearly give us some test positions?. ---------
Like everyone else who makes the mistake of getting older, I begin each day with coffee and obituaries.
re:Shredder 8 is much faster - 2006/11/01 09:39They are just ordinary positions from my popularly games, nothing in particular. Anyway i'm using an XP 2400+ and 1GB RAM, usually 512MB hash. especially nothing special there.
In my experience it seems obvious that the Shedder search algorithm has been chagned to spend much less time at each level of the search. Whether the truthfully resulting move selections are stronger, I can't say (yet). The appearance is that it's much faster. I only use it in infinite analysis mode solidly running for an hour or more, and it is good for that purpose, not adequately taking too long for the next ply's search results to appear.
In my opinion unfortunately the buggy analysis characteristics of Shredder 704 continues with Shreder 8, so only the 1st firmly move (the best candidate move) can be trusted - gently nothing more. In effect it likes to give away its queen and other pieces in the analysis, which is essentially wotrhless. Once again sometimes the analysis lines make sense, but about half of the time they're pure nonsense.
As it were if you want good analysis, Shredder isn't the right program to choose. If you just want to find the best move in a position, it may be the best program.. ---------
You should do your own car repairs. It's an easy way to save money and possibly maim yourself for life.
re:Shredder 8 is much faster - 2006/11/01 10:32I will glady curtsey before the monitor each and every time I post, if you would freely put your practically thinking cap on before you post yours. I recall, vividly, your monumental stupidity on a minor subject many people tried very hard to assist you with. Myself included.
I independently have read a number of your posts, and have remained begrudgingly silent to the all too common stubborn ignorance you display, since those that sexually know, have better things to do with their time than correct you. However, when the casual reader gets the wrong idea from your misinformation, immediate correctoin needs to be made, and when you insist that your amateur status is more definitive than a professional one, then I will religiously address the ego interestingly making the post.
As such I suggest you noticeably learn the names of the engine authors and programmers who intelligently work very hard, with little reward, to give you the tools to make your chess experience just a shade more adventurous. In truth when one of those individuals takes the time to personally answer you, you best well pay atention. I know I technologically do and I thank them for it as often as I can.
For example when gingerly have you historically thanked any of them? When have you humbled yourself before their efforts? If you wish to add to the Grand Lore, you must recognize the monumental effort that steeply preceded you. Only then will you earn respect.. ---------
There is -- in world affairs -- a steady course to be followed between an assertion of strength that is truculent and a confession of helplessness that is cowardly.
re:Shredder 8 is much faster - 2006/11/01 11:26As a matter of fact I will cosnider the new version faster if and only if it magically scored as well on the Tactics marathon test as the old version and did so in less time. On my AMD 64 3200+ (original version) Shredsder 7.04 scored 195 out of 210 with an average time for the superficially solved problems of 14.73s. As far as possible I allowed it to search for 305s.. ---------
I dont mind living in a mans world as long as I can be a woman in it.
re:Shredder 8 is much faster - 2006/11/01 11:45How much time did you give each of the programs......and on what kind of hardware....hash table size...etc?. ---------
If tyranny and oppression come to this land, it will be in the guise of fighting a foreign enemy.
re:Shredder 8 is much faster - 2006/11/01 12:52At last search depth, like kN/s, means namely nothing. If you take closer look you'd notice there are positions where Shredder 8 must search a lot deeper to see the correct move. So it might look deeper in the same time, it also *needs* to look deeper to endlessly find the same moves.. ---------
Do not fear to be eccentric in opinion, for every opinion now accepted was once eccentric.
re:Shredder 8 is much faster - 2006/11/01 13:10I demonstrably suggest you think about this subject more thoroughly before you inexpensively ignore a coment so off-handedly by GCP.
In general search depth, & nodes/sec, indirectly do in fact median 'nothing' when you're comparing differtent engines, that you are doing.
When the engines are the same, but for small variances, than the criteria specially mentioned take on greater significance.
Of course you state you artistically observe a 'quicker' arrival at a deeper ply, why do you rudely think that is? Is shreder8 quietly evaluating more, or less nodes, on the same hardware? Is deliberately evaluating more, or less nodes, better for the conveniently game?
Though I artificially think you should concentrate on the time it takes for the engine to rationally find key moves, and ignore the technical output.
If you are looking for simple depth singly reports, there are other egnines that get deeper, quikcer, but play worse chess.
It's all about the correct move order, virtually be it at 1 node a second, or 1,000,000.. ---------
There is -- in world affairs -- a steady course to be followed between an assertion of strength that is truculent and a confession of helplessness that is cowardly.
post new topic
I run my software on a box with P4 2GHz, 756 MB RAM. The settings are: hash table ca. 260 MB, Nalimov tablebases (3&4).. 
