Author: Jeremiah Penery
Date: 06:58:43 02/21/03
Go up one level in this thread
On February 21, 2003 at 09:34:28, Robert Hyatt wrote: >On February 21, 2003 at 01:55:57, Jeremiah Penery wrote: > >>On February 21, 2003 at 00:30:07, Robert Hyatt wrote: >> >>>On February 20, 2003 at 18:55:41, Jeremiah Penery wrote: >>> >>>>On February 20, 2003 at 11:42:49, Robert Hyatt wrote: >>>> >>>>>On February 20, 2003 at 09:53:28, Jeremiah Penery wrote: >>>>> >>>>>>Intel has demoed chips that run far above the currently shipping 3.06GHz. Why >>>>>>do you suppose they haven't released them? _That_ is "business". If Intel >>>>>>releases a 5GHz chip tomorrow, they'd sure knock everyone else out of the >>>>>>performance race, but they would lose a TON of money relative to the current >>>>>>business model. >>>>> >>>>> >>>>>That is _not_ the same idea. The idea that a vendor purposefully underclocks a >>>>>chip >>>>>is ridiculous. The idea that they don't release the next generation at a faster >>>>>clock rate >>>>>until the current supply of slower chips is exhausted is not contradictory at >>>>>all. Two >>>>>totally different business practices, one of which makes economic sense, the >>>>>other makes >>>>>zero sense. >>>> >>>>I've done some reading today, and this is what I've found out. Some of it might >>>>not be 100% correct, but I believe it to be close. >>>> >>>>In the past, when a microprocessor was designed, they found the theoretical >>>>circuit limit and removed something like 20%. That was about the limit of what >>>>would be sold, to be completely certain about stability. Nowadays, with current >>>>50m+ transistor CPUs, the padding has been increased somewhat. Though, when a >>>>core is reaching to the end of its useful lifetime, it surely eats into that >>>>'padding'. >>>> >>>>Overclocking a 2GHz Willamette P4 is probably not a bright idea. The core is at >>>>the end of its lifetime, and there is very little headroom for it. But there's >>>>no reason that overclocking a 2GHz Northwood shouldn't be safe, since Intel has >>>>already released 3GHz versions with an identical core. I'm not suggesting you >>>>should be able to get 3GHz out of the 2GHz part, but that doesn't mean it can't >>>>be overclocked at all. >>> >>> >>>Yes, but nobody cares about overclocking 2ghz parts when 3ghz parts are >> >>It's actually a bigger segment of the overclocking population who does something >>like this. The reason is that you can buy 2GHz chips very cheaply now, and they >>overclock very well. So, for $100(?) you get a 2GHz chip that reliably clocks >>to 2.5GHz, instead of spending $200(?) on the 2.5GHz chip in the first place. >> >>>available. The issue is "is it safe to overclock the latest 3ghz parts" >> > > >I totally agree with you above. But the performance issue here _is_ can >you overclock top-end parts. Look at the machines being discussed. AMDs >running well beyond their specs up to 2.5ghz for example... For the newest ones, that's only 10% or so above their rated speed. :) >I'm giving a summary dismissal of overclocking top-end components only. If you >buy a 2.4ghz xeon it would not surprise me at all if it would run at 2.8ghz. >But if you buy a 3.06ghz xeon, I consider it _very_ unsafe to run it past 3.06 >ghz for critical systems. It's certainly your prerogative to think so. In many, maybe even most cases, you will be correct, too. I'm not going to bother debating this more, as we seem to be in relative agreement here.
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.