home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!dtix!darwin.sura.net!wupost!usc!rpi!usenet.coe.montana.edu!news.u.washington.edu!ogicse!das-news.harvard.edu!cantaloupe.srv.cs.cmu.edu!crabapple.srv.cs.cmu.edu!andrew.cmu.edu!sm86+
- From: sm86+@andrew.cmu.edu (Stefan Monnier)
- Newsgroups: comp.sys.atari.st
- Subject: Re: New operating systems (OS/2-TOS)
- Message-ID: <QebvgW_00VpFAFiUYX@andrew.cmu.edu>
- Date: 29 Aug 92 10:00:02 GMT
- References: <H.JHXj&6Ouz4Y@marsu.tynet.sub.org> <1992Aug26.182907.28894@ra.msstate.edu> <1992Aug26.202952.26085@nosc.mil>
- <4278@wet.UUCP>
- Organization: Junior, Math/Computer Science, Carnegie Mellon, Pittsburgh, PA
- Lines: 39
- In-Reply-To: <4278@wet.UUCP>
-
- Excerpts from netnews.comp.sys.atari.st: 28-Aug-92 Re: New operating
- systems (.. Ed Krimen@wet.UUCP (849)
- > Can someone explain what a heat sink does and how it works? I get the
- > impression that it provides a larger surface area to the chip so that it
- > can dissipate heat faster. I think I've seen what they look like too;
- > it looks kinda like a vent, with the tall ridges. But how does it work?
- > It only looks like it's clamped onto the chip. And why is it that only,
- > at least from what I've heard, 80486 and 68040 chips need these?
-
- > --
- > ||| Ed Krimen
- > ||| ekrimen@wet.com
- > / | \ E.KRIMEN (GEnie)
-
-
- The power dissipation of a chip is proportional to his frequency and to
- the number of tramsistors packed on it (approximately)
- It also depends of the use of all those transistors: on a highly parallel
- CPU (superscalar or just pipelined), transistors are likely to be all (or
- almost) used at all times. So that on such new CPU, the consumption
- is really high (can go above 1A).
- An extreme example is a recent VAX CPU running at 100Mhz presented
- by DEC and which is likely to be comparable (for power consumption)
- to the alpha CPU (from the same company): It drinks about 6A on average!
- (and as consumption is not constant but is concentrated at each tic of the
- clock, at each tic the intensity goes up two about 30-40A).
-
- All this power has to be dissipated for the simple reason that silicium
- transistors' reliability limit is at about 150-200 C ! (and those
- chips are not really big)
-
- That's why !
-
- Stefan Monnier
-
-
- -----------------------------------------------------
- -- On the average, people seem to be acting normal --
- -----------------------------------------------------
-