Opteron versus Xeon

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

Moderators: hgm, Rebel, chrisw

Brian

Opteron versus Xeon

Post by Brian »

Hi,

I have a bot account at FICS and I am going to buy a new computer. What CPUs are you recommending as the best for computer chess?

1) Opteron Quad Core

2) Xeon Quad Core

What level of calculations could I expect from one of these CPUs and if I got dual CPUs?

For the dual core CPUs is AMD or Intel better for computer chess calculations ?

Thank you in advance.
User avatar
hgm
Posts: 27790
Joined: Fri Mar 10, 2006 10:06 am
Location: Amsterdam
Full name: H G Muller

Re: Opteron versus Xeon

Post by hgm »

Xeon can mean anything, from Pentium 4 (absolute crap) to Core 2 Duo (top of the bill).
jdart
Posts: 4366
Joined: Fri Mar 10, 2006 5:23 am
Location: http://www.arasanchess.org

Re: Opteron versus Xeon

Post by jdart »

Well, quad core means a modern Xeon (3200-series).

There is no quad core Opteron available yet. I believe they are slated to be released next month.

From what I have heard the initial opteron parts will not have a clock frequency as high as had been hoped, and so may have trouble competing with Intel for a while.

Also Intel is supposed to be moving to 45nm Xeons (Penryn), with a larger cache later this year. I would wait a few months and get one of those, personally.

--Jon
Terry McCracken

Re: Opteron versus Xeon

Post by Terry McCracken »

jdart wrote:Well, quad core means a modern Xeon (3200-series).

There is no quad core Opteron available yet. I believe they are slated to be released next month.

From what I have heard the initial opteron parts will not have a clock frequency as high as had been hoped, and so may have trouble competing with Intel for a while.

Also Intel is supposed to be moving to 45nm Xeons (Penryn), with a larger cache later this year. I would wait a few months and get one of those, personally.

--Jon
I read that Intel will be moving to 32 nano meter process at their site! :shock:

Revolutionary Regards,
Terry
bedouin

Re: Opteron versus Xeon

Post by bedouin »

Such a powerful machine to play humans at FICS? Why not playchess where the guys with big iron assemble and where games contribute to the rybkaII.ctg? There is no point being a cheapskate and then shelling out for big iron whose price will drop faster than the stock market. It doesn't make cents.
Nid Hogge

Re: Opteron versus Xeon

Post by Nid Hogge »

Right now anything from the core 2 family is better by a large margin from any AMD offering.

However you may wanna wait a month or so for AMD's Barcelona to see how it performs.. although it's gonna take a while before youll see them running in high clocks.
Nid Hogge

Re: Opteron versus Xeon

Post by Nid Hogge »

I read that Intel will be moving to 32 nano meter process at their site!

Revolutionary Regards,
Terry
Yes, although not that fast. 45nm is next and should arrive soon(Q407-Q108), Then there's another 45nm based on a new Arch(Nehalem), And then probably by Very late 2009 we start seeing 32nm westmere.

The shrinks plan is 45nm-32nm-22nm-16nm and that's as far as they currently go. http://en.wikipedia.org/wiki/16_nanometer

What comes after that?
Now then it starts to get really interesting :P
Terry McCracken

Re: Opteron versus Xeon

Post by Terry McCracken »

Nid Hogge wrote:
I read that Intel will be moving to 32 nano meter process at their site!

Revolutionary Regards,
Terry
Yes, although not that fast. 45nm is next and should arrive soon(Q407-Q108), Then there's another 45nm based on a new Arch(Nehalem), And then probably by Very late 2009 we start seeing 32nm westmere.

The shrinks plan is 45nm-32nm-22nm-16nm and that's as far as they currently go. http://en.wikipedia.org/wiki/16_nanometer

What comes after that?
Now then it starts to get really interesting :P
I'll have to check with Intel, but I believe they may already have the 32nm process going or very shortly.

What comes after 16nm? Eventually 1nm, the first true Nanoprocessor!; If it's possible to scale down to that size?

Futuristic Regards,
Terry
bob
Posts: 20943
Joined: Mon Feb 27, 2006 7:30 pm
Location: Birmingham, AL

Re: Opteron versus Xeon

Post by bob »

Terry McCracken wrote:
Nid Hogge wrote:
I read that Intel will be moving to 32 nano meter process at their site!

Revolutionary Regards,
Terry
Yes, although not that fast. 45nm is next and should arrive soon(Q407-Q108), Then there's another 45nm based on a new Arch(Nehalem), And then probably by Very late 2009 we start seeing 32nm westmere.

The shrinks plan is 45nm-32nm-22nm-16nm and that's as far as they currently go. http://en.wikipedia.org/wiki/16_nanometer

What comes after that?
Now then it starts to get really interesting :P
I'll have to check with Intel, but I believe they may already have the 32nm process going or very shortly.

What comes after 16nm? Eventually 1nm, the first true Nanoprocessor!; If it's possible to scale down to that size?

Futuristic Regards,
Terry
looks to be impossible. An atom is a half-nanometer in diameter, so you are talking about features made up of two or less atoms. Very small. Very difficult to manufacture. And then there's the reliability issue.
Terry McCracken

Re: Opteron versus Xeon

Post by Terry McCracken »

bob wrote:
Terry McCracken wrote:
Nid Hogge wrote:
I read that Intel will be moving to 32 nano meter process at their site!

Revolutionary Regards,
Terry
Yes, although not that fast. 45nm is next and should arrive soon(Q407-Q108), Then there's another 45nm based on a new Arch(Nehalem), And then probably by Very late 2009 we start seeing 32nm westmere.

The shrinks plan is 45nm-32nm-22nm-16nm and that's as far as they currently go. http://en.wikipedia.org/wiki/16_nanometer

What comes after that?
Now then it starts to get really interesting :P
I'll have to check with Intel, but I believe they may already have the 32nm process going or very shortly.

What comes after 16nm? Eventually 1nm, the first true Nanoprocessor!; If it's possible to scale down to that size?

Futuristic Regards,
Terry
looks to be impossible. An atom is a half-nanometer in diameter, so you are talking about features made up of two or less atoms. Very small. Very difficult to manufacture. And then there's the reliability issue.
Yes this is all true, but not necessarily impossible. New Nano Technologies may be quite impressive!

Back to the Future Regards,
Terry