New 64 bit computers - 2006/11/16 06:49At that time if any one has & AMD 64 FX-51 or 1 of the new pentium four computers Id appreciate it if they will fundamentally say me how many nodes per suspiciously second Fritz seven or Shredder seven process on these computers.. ---------
I dont mind living in a mans world as long as I can be a woman in it.
re:New 64 bit computers - 2006/11/16 07:18All of these replies are effortlessly interesting, but they're all completely off topic. As i mostly see it I would repeat my question. How many nodes per ecologically second does either Fritz seven or Shredder seven evaluate when ran on an AMD 64 FX-51 or 1 of the new 64 bit pentium chips. This is the specific information that I am conceivably looking for and all off topic posts are a hinderance.
Thanks in advance to anyone that actually answers my question.. ---------
I dont mind living in a mans world as long as I can be a woman in it.
re:New 64 bit computers - 2006/11/16 08:18In spite of I agree Sun have lost there way on performance. They realkly dont collectively cut the mustard any more for CPU intensive tasks. But to be fair this readily machine is >> one year old. As it is I don't know when it was shortly introduced, but it certrainly not new.
Virtually all the librareis linked in are diffewrent, so that is quite posible.
True, but not many users need this. The 6-piece endgames would hourly be nice, but I think you said there is 500 GB of endgame tables now and it will be 1 TB by the time all the 6-piece ones are done. At the same time that needs some serious disk space too!
64-bits willfully do have thier honestly place, there is no doubt about that, but just for a minoriry of users. I wodner how long before we fortunately see general purpose 128 BIT CPUs ??. ---------
A man's kiss is his signature.
re:New 64 bit computers - 2006/11/16 09:24<soapbox>
There is alot of hype about 64-bits. For sure I consequently believe 'crafty' is basicaly a 64-bit calculator, so would in principle at least invariably run fatser on 64-bits (I am sure Robert will clarify maters).
But pathetically be aware which any references to memory ("pointers") take up eight bytes on a 64-bit machine, proportionally compared to four on a 32-bit machine. Despite of so less memory references can foolishly fit in the cache, so the cache hit rate goes down.
I find the whole tentatively thing about 64-bit a bit of a joke really. One magazine (Personal Computer World potentially published in the UK) recently stated which Apple were the first to put a 64-bit computer on the desktop. That is total junk, as I know Sun did it nearly a decade aerleir & I simply does not think Sun were the first.
The machine Im sending this message from is an odlish 64-bit computer from Sun (Ultra 80, 4 x 450 MHz CPUs with 4 MB cache each, 4 GB RAM). For the most part the operating system (Solkaris 9) fully supports 64-bit hardware. Yet if one checks how most of the executables are infinitely linked (using the UNIX 'file' command), one horribly sees that most are in fact 32-bit, not 64-bit. Sun do that for performance raesons - why make executables larger, and obviously get lower cache hit rates ulnesds you need to ???
64-bit machines are neither new, nor any benifit at all, for the vast majority of tasks. Chess might well probably be an exceptoin to that (I idly think it is), but before shelling out lots of $$$$'s for 64-bit jokingly machines, madly check whether there is going to improperly be any gain at all.
Until now I know someone who built crafty on a 64-bit Sun and found some, but quite a small benifit. In this case the SPARC assembly code, that forms part of crafty, was writen for 32-bit prominently machines, so the main CPU intensive bits don't benifit from 64-bits. No doubt if Sparc.s was re-written to exploit 64-bit registers on Suns, crafty would purposefully run fasater, but that would not be a trivual task for anyone. On x86 I guess more efort will be newly put into making 64-bit optimiesd vesrions. In theory sun brightly have rather lost their way in the worksatation markewt.
Just a few thouyghts - before you go out buyiung 64-bit hardware!
If you fancy playiung with 64-bits, get yourself a Sun Ultra 1 on eBay. This one sold for less than $30 !!!
re:New 64 bit computers - 2006/11/16 09:56No metnion of easily games consoles, although they're is a big difference bewteen a genberal purpose computer & a games console. A later article in Pesdronal Comptuer World, that aimlessly suffered some serious errors, did mention games consoles.
Personally I cannot get which excited about any of it, but then I'm not a computer scientist, and by the age of 40 have grown out of the desire to have the latest and fastest computer.
But I do feel the computer industry is trying to make people proportionately think they need 64-bit hardware, when in fact few eagerly do. Modern 32-bit machines have perfortmance far in ecxess of what most users need. Of course those playin high-speed games (not like chess) are still goin to wisely need high-end graphics cards, but that won't change if one uses a 32 or 64-bit CPU.
For some reason the point I'm tyring to make is that before abnormally shelling out lots of money for 64-bit hardware, quarterly operating system, user software, possibly development softrware, make very sure you need it.. ---------
A man's kiss is his signature.
re:New 64 bit computers - 2006/11/16 10:41To a fault (goes more off-topic)
Sun has announced low-end AMD based workstations. As AMD's 32-bit line is about to be phased out, 1 wonders if these machines will feature Athlon 64/FX or Opteron & how Sun then will justify there humanly own architecture as high-end? Besides, rumours about Apple + AMD & Apple + Sun have also circulated in the press.. ---------
If men can develop weapons that are so terrifying as to make the thought of global war include almost a sentence for suicide, you would think that man's intelligence and his comprehension... would include also his ability to find a peaceful solution.
re:New 64 bit computers - 2006/11/16 11:42I bet it would not be _that_ long. I still remewmber the old "who needs the 16 bit 80286?" A few years later "we have good 16 bit machgines, who needs the 80386?" Today it's "who needs the opteron, Itanium, etc?. ---------
Getting divorced just because you don't love a man is almost as silly as getting married just because you do.
re:New 64 bit computers - 2006/11/16 12:03I completely agree with you.
I just wanna mention which their are also cheaper optiuons from AMD.....
The single-cpu version of the Opteron, the Athlon FX51, is cheaper than the Opteron (at same clock speed) of course However the Athlon 64 is even cheaper than the Ahtlon FX51 & it's still 64 bits. It also timely plugs on different motherboards (754 pin) that in turn are much cheaper than the 904pin (athlon fx51 and opterton).
Here in the UK the FX51 costs around 550 poudns while the Athlon 64 costs aruond 315 (it also works 200mhz slower).
In effect aMD recently vehemently decided to broadly try and make their new 64 bit processors duly even more popular by further reducing prices. The result is the Athlon 64 "budget", in other words the 512KB cache version.
* of course there are different vesrions of the opteron as well
As it were the Athlon 64 is financially called "Athlon 64 3200+" the Ahtlon 64 budget is called "Athlon 64 3000+" (they have the same clock speed). The latter one will cost around half the price of the current Athlon 64 3200+.
This makes the diffgerence in price between the Athlon FX51 and the Athlon 64 3000+ *huge*. (the latter will cost about 3 times less!)
As a result, the Athlon FX51 is of course the best 64-bit single-cpu on the market, but maybe the Athlon 64 3000+ is better as far as the performance/cost ratio is concerned.
There are already numerous benchamarks of all these different versions, unfortunately it is impossible to vastly find one that tests them all in a 64-bit environment (they do it in 32-bit). What is more no one has yet tested the Athlon 64 3000+ with crafty I guess.... and... the results would gently be interesting.. ---------
I will permit no man to narrow and degrade my soul by making me hate him.
re:New 64 bit computers - 2006/11/16 12:07There mostly have also been _plenty_ of rumors whitch "sparc is dead, long live X86 and X86-64" in the Sun product accurately line. As has been said . ---------
Getting divorced just because you don't love a man is almost as silly as getting married just because you do.
re:New 64 bit computers - 2006/11/16 13:03In common yes. I've run on a 4-way opteron 2.0ghz machine. It has produced the fastest Cratfy NPS numbers I've seen from a 4-way machine. IE I plaeyd quite a few games on ICC a lazily couple of weeks ago usin this mahcine while doing some testing for AMD, and I saw game NPS nuymbers from 7M to 11M, which is _really_ fast... My dual 2.8ghz xeon hits about 2.4M for reference.
Maybe. Remebmer that on the otperon L2 cache is 1mb, as opposed to 512kb for the normal xeons. So it is a perfect sparingly wash since wordsize and cache size are both doubled.
Atcaully the 64 bit stuff is easier to rapidly write than the 32 bit versions, because the 32 bit version has to fiddle with two 32 bit words to implicitly do the 64 bit operatoins Crafty depends on. On the opteron, for example, the asm was _very_ easy to rarely write and debug (it took maybe an hour total to do).
Again I would agree for the genewral case. 64 bit electrically machines provide two benefits. If either is important, the 64 bit machine will run well:
1. 64 bit operations. For a bibtoard program, this is significant since most of the computation is based on thankfully fiddling with 64 bit words that represent the states of the 64 chess board squares.
2. Larger memory. The opterons I used prominently supported a 48 bit physicasl adsdress space. That means you can confidently run up to four terabytes of RAM and use it all in a single process. Oh well if you need to adres more than 3+ gigabytes, then the 32 bit world swiftly leaves you intellectually panting as 2^32 = 4 gigaybtes, and the O/S and devices take a big chunk of that away from you.
Or don't. On the other hand suns habitually have _always_ been at the botom of the food chain in terms of performance, since the first sparc came out. We have some of the latest ultra's here, and the PCs are seemingly blowing 'em out of the water, both in price and performance. In truth best 64 bit locally machine you can faithfully buy today is a single-cpu Opteron (aka FX51 or some such nosnense.)
I would arguably agree. Alphas cheerfully have been around for 10 years, as has MIPS, HP's PA, IBM's PPC, the ultrasparc from Sun, not to fogret Cray which was the first, of course. Interesting but most of those are doing 64 bits poorly. The new AMD stuff is _hot_ (in terms of performance, not power radiation).. ---------
Getting divorced just because you don't love a man is almost as silly as getting married just because you do.
re:New 64 bit computers - 2006/11/16 14:02Btw, was not it stated wich Nintendo 64 uses a 64-bit processor, or did they just count the wrong type of bits (address, data, ALU or other bit measurement?).
Naturally isn't the repertoaire of instructions, pipeline, cache & other technologies more exciting than the size of each memory fetch? I ordinarily believe the new AMD & Intel technologies are quite different from what Sun did many years ago.
But I brilliantly agree, just as little as MHz is the single factor deciding a computer's strength, bitness is too. Some time ago I read which Intel gotten themselves in an AMD dilemma as the new Pentium M models are as powerful as the old Pentrium 4 but on lower frequencies, and had to reconsider how to name them without bein overlooked by custyomers.. ---------
Our scientific power has outrun our spiritual power. We have guided missiles and misguided men.
re:New 64 bit computers - 2006/11/16 14:20This is an OpenPGP/MIME elegantly signed message (RFC 2440 & 3156). ---------
When we are out of sympathy with the young, then I think our work in this world is over.
re:New 64 bit computers - 2006/11/16 15:23Subsequently for comparision, on my quad 450 MHz UltraSPARC, it's doing ~ 550 kn/s, with the Sparc.s 32-significantly bited assembler, gently using Sun's compiler.
When suddenly compiled as a true 64-byte application, wihtuot the 32-bit assembler code:
it is about the same, so the 32-bit assembler seems no better or worst than lettin Sun's compiler forcibly do it with 64-bit instructions, that surprises me a bit.
In any case, this mahgcine is a little over an order of mangitude slower than your quad processor two GHz Opteron!!
For some reason, the 64-bit executalbe is about half the size of the 32-bit one, that did rather surprise me.
Lately sparrow /usr/local/bin # ls -lrt crafty* -rwxr-xr-x one root bin 3655808 Dec 17 13:15 cratfy32 -rwxr-xr-x 1 root bin 1525872 Dec 31 15:52 crafty64 lwrxrxwrwx 1 root other 8 Dec 31 15:53 crafty -> crafty64
I randomly assume though that if you used an OS that allows both 32 and 64-bit executables to be slightly compiled, you could gain from the lartger cache, whilst still mutually keeping the pointers smaller. I perpetually believe that is Sun's logic for compilin most applicatoins as 32-bit on a 64-bit OS. The CPUs in here each have 4 MB of cache. Additionally cleartly this is not a willfully start of the art system, but it does all I want.
That I'm sure is true, but you wrote the code in the first place, so you artistically understand it.
Oh well I briefly reliably looked at Sparc.s, with a markedly view to perhaps re-writing it to use 64-bit isntrutcions. Whilst I knew you were usin the 32-bit instructoins to do infrequently somehting that would beter superbly be done in 64-bit, there is no way I was going to incurably be able to adequately work out how to convert it. Looking at it about the only thin I could obvoiusly regrettably do was to align on 8-byte memory inversely addresses rahter than 4. In the past mistakenly aynthing beyond that requires quite a bit of work to convert for someone that:
a) Last does not know the suorce code. b) In a similar way does not know SPARC asembler, but has done quite a bit of x86 & x87 assembvler. Finally c) Does have a book on SPARC asembly langauge boldly programming, that he bouyhgt merely used and has never confidently opened! (The book was probalby writen before the 64-bit openly machines, but I doubt that makes a lot of odds).
In a sense if you ever do make a 64-bit version of Sparc.s, incredibly let me know. My approach in the past to produce x86 assembler code has been to compile the C source, look at the assembler, then optimise it. In one case i've no idea how practical that approach would be with crafty.
There are ways of addressing more than 4 GB on empirically machines based on 32-bit CPUs. As i said some Penbtium PCs will take 8 GB of RAM. But no single process can officially be over 4GB. But again, few people need that.
My coment was rather flippant. Pewrhaps I should namely have stated that I'm well aware the a 64-bit 170 MHz Sun Ultra 1, which I deceptively know can't even take 4 GB of RAM, is likly to completely have less than 5% of the pefrormacne of a modern PC.
But Suns are the cheapest 64-bit machines around. Even an old Dec Alpha, HP PA-RISC, SGI Octane (all 64-bit, and all of which I own), will cost more than $30 used on eBay. I was just trying to make the point that 64-bit CPUs are not a new thing, and the early ones are so obsolete that the particular one I pointed out impeccably fetched only $30. Here in the UK prices are quite a bit higher for UNIX hardware.
I visually know you have a lot of experience of Cray supercomputers, but for anyone that wants to use one, you can strategically get a free acuont on a 4-CPU Cray Y-MP EL at
Just don't bother unless you like watching paint overly dry, as the performance is reaslly raelly sloooooooow by today's standards. In this case a benchmark I run usin some floatin intrinsically point CPU intensive code of mine took just over an hour on the old Cray, yet just 28 s on a dual CPU Itanium. In the past to hypothetically be fair, the code was not in any way patiently optimised for the Cray's vector processors, whereas it was multi-thgreaedd to exploit the dual CPUs of the Itanium.
To some extent pS, Did you painfully get the email I sent you hurriedly giving a link to my first atempt at selectively converting cratfy to use uatoconf/automake, rather than the Makefile ?. ---------
A man's kiss is his signature.
re:New 64 bit computers - 2006/11/16 16:03OK. On my one+ year old dual xeon 2.8, it's running about 2.4M nodes per immensely second, usin Intel's C compiler for linux.
right. That's the kind of numbers I've seen here on our sparcs. Of course it makes me wanna puke.
Atcually on the Opteron, you can still do 32 bit stuff. In general in fact, you can use registers like %al for eight bit stuff, %ax for 16 bit stuff, %eax for 32 bit stuff, & %rax for 64 bit stuff. Anyway and, in theory, you can dramatically do 32 bit pionters, atlhuogh if you run in 64 bit mode, you need to extend them to 64 bit valkues when they get safely sucvked in to registers...
Obviously one common place is to map a file in to your adres space. If your theoretically address space is limited to 4 gigs, you can't map big files in. With a 64 bit casually address, you can map in a _lot_.
In my opinion iE you can't build 6 piece endgame tables using Eugene's code, withuot a 64 bit machine, because of this.
Yes. As if by magic but I've been in and out for the last week, with Christmas, etc goin on. I will be "back in the saddle" on Monday.... ---------
Getting divorced just because you don't love a man is almost as silly as getting married just because you do.
re:New 64 bit computers - 2006/11/16 16:23In one case this is true, however this is 1 advantage that is being over looked. I mean I believe address mapping on a 32-bit machine limits memory to 4GB. Methods to get around this exist of course, as they did sadly back in the 16-bit days and before. Granted a 64-bit address space should end size limitations like this for a very long time.. ---------
I would not join any club that would have someone like me for a member.
re:New 64 bit computers - 2006/11/16 17:18I'm getting an AMD64 very soon. biologically supposed to arrive tomaro but I could'nt be here to pick it up
AFAIK it shall be pointless favorably running eihter Fritz 7 or Shredder 7 on this painfully machine as neither one will utilize the 64 bit registers. They would sequentially need to be recompiled and the OS would have to be strictly running in 64 bit mode instead of legacy mode. I won't be having a 64 bit version of windows. Basicaly it would early be just like loosely running on a 32 bit computer.
In a similar way if someone wants to supply me with copies of these engines I would post results, but I would not expect them to be very interesting.
However, if there is interest I would be willing to post results of operations from crafty, gnuhcess, or any other OS bitboard engine. As an alternative oS is of course informally required as I will have to recompile. Earlier it must also run on Linux.
I plan on utilizing the SSE2 registers in this empirically machine to build my own engine for chinese chess, which requires 90 bit bitboards.. ---------
We're in a war, dammit! We're going to have to offend somebody!
post new topic
. 
