Author: Dann Corbit
Date: 16:16:42 02/24/99
Go up one level in this thread
On February 24, 1999 at 18:40:30, Bruce Moreland wrote: > >On February 24, 1999 at 17:08:41, Dann Corbit wrote: > >>Why is there a 1 CPU limit for microcomputers? This is an artificial >>distinction, since they are clearly being sold in bulk right now. I suggest, >>instead, to have a limit on the retail cost of the machine. If the machine can >>be bought for $10,000 from the local computer outlet, then it is a >>microcomputer. Have an official date that the price (whatever you might choose) >>must be under a certain level. > >The same arguments happen every year. Here was my argument against a dollar >limit on hardware for a WMCCC last time, and I think it is still good. > >I sit in my house and occasionally buy a computer, which I take to these events. > When I come back, I have to eat the thing for the next year or two. It becomes >my development machine or my ICC machine or whatever. > >If I want to buy a computer, I should be able to buy a quality computer. I >shouldn't be essentially forced to buy a crap machine so I can get an extra few >mhz out of it. I don't want to have to buy a crappy case and a crappy >motherboard and a crappy hard-disk, just so I can go 10% faster until the >machine breaks. I agree with the sentiment. But if your machine is a 'microcomputer' with a 1GHz Alpha chip which would cost $10,000 for the chip alone (were it available to the public but isn't) is your machine a microcomputer while the 2 CPU PII 300 MHz machine costing $4000 is not? I don't buy it. But I'm listening. >There are lots of other reasons why this is unworkable, in my opinion. I don't >think it is possible to put a limit on the machines in the WMCCC, the only >effective limit is the single-processor limit, possibly supplemented by a >requirement that the machine be approximately PC-sized (a restriction based upon >the size of the box, inherent in the term "microcomputer"). If you can put in a super hot chip that no one else can possibly compete against, then is that more fair than multiple cpu's that come in a cheaply available machine? >Limitations on the chip you can use will lead to an Intel-centric competition. >I shouldn't have to use the Pentium architecture if I don't think it suits me. > >Attempts to allow other processor types by implementing a performance ceiling >once again force me to buy an inferior development machine. Again: if I have a >few thousand for a new machine, and my old machine is a 200, and the new >machines are 450's, and the limit is a 300, then I either have to go on the 200 >or buy a useless 300. If the ceiling is a 450 and there is an Alpha chip that >is faster, and maybe is even cheaper than the 450, I can't upgrade to the new >machine, I have to go on last year's machine, so the Intel boxes get an unfair >advantage. I see no reason to limit the MHz figure. But I think that if no one else on earth can come up with the same machine or if the machine costs $100,000 then it is not a microcomputer, no matter how small it is or how many chips are in it. >This goes on and on. It's not terribly important to me, especially since I have not even written a chess program and am therefore rather unlikely to get an invite. ;-) However, I do think that in the name of fairness we should think about what the "Microcomputer" championship means. If Joe can compete with a $50,000 machine that no one can get even a full year later and Pete is kicked out because his $4,000 machine has 2 CPU's, then something seems intrinsically unfair about it. I do realize that it is an opportunity for chipmakers to show off. I imagine that they sponsor the thing in certain ways and if that is driving it then fine. But we should call it the "Single CPU champion that fits in a cubic foot" champion of the world. Calling it the microcomputer champion seems a bit of a sham to me.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.