Cyrix chips overheat!
Matt McDevitt
The chips manufactured by Cyrix....not necessarily the Cyrix processors
manufactured by IBM and others....are having some serious problems with
overheating. The average Cyrix computer, as a matter of fact, has at least
2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
heat sinks had better be ceramic or they'd melt! I heard this from
WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
processor (to save money) and end up getting an Intel IN ADDITION TO the
Cryix, after it caught on fire or just burned up :)
--
Matt McDevitt
mak...@ix.netcom.com
j-rex
Matt McDevitt (mak...@ix.netcom.com) wrote:
: The chips manufactured by Cyrix....not necessarily the Cyrix processors
matt, why do you keep posting stuff you read on the ZDnet? you're not
helping to inform anyone. please stop posting your off topic spam.
later
--
jason keeney /~\_/~\ Paper Rat Co.
\`O o`/
mailto:jr...@netcom.com | | "Makers of fine OS/2 games"
http://www.webcom.com/jrex ==\o/== -Since 1996-
Daniel Phillips
---------------
Daniel Phillips
phil...@dowcow.com
Daniel Phillips
>Matt McDevitt (mak...@ix.netcom.com) wrote:
>
>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>manufactured by IBM and others....are having some serious problems with
>overheating.
This is new to me. What I see on the chat sessions is that people
who have Cyrix chips are generally "ecstatic" about them. It seems
the performance claims are largely true.
>The average Cyrix computer, as a matter of fact, has at least
>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>heat sinks had better be ceramic or they'd melt! I heard this from
>WWW.ZDNET.COM and other places.
I have had a lot of problems trying to relate things that I find on
ZDNET and in Ziff-Davis magazines such as PC Mag and Computer
Gaming World to reality. Just my experience.
---------------
Daniel Phillips
phil...@dowcow.com
Jaguar
phil...@dowco.com (Daniel Phillips) wrote:
Well, I own a Cyrix...It's pretty nice so far, havnt really put it
through the ropes yet. Just got it like 2 or 3 days ago. Only bad
thing I notice about it so far is the fact that most benchmarks and
other autodetectors like Windows 95, will detect it as a 486. I assume
that is because the chip is newer than the benchmark or autodetecter,
but it's got me pretty worried
The All Mighty
On 26 Aug 1996 00:12:53 GMT, "Matt McDevitt" <mak...@ix.netcom.com>
wrote:
>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>manufactured by IBM and others....are having some serious problems with
>overheating. The average Cyrix computer, as a matter of fact, has at least
>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>heat sinks had better be ceramic or they'd melt! I heard this from
>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>processor (to save money) and end up getting an Intel IN ADDITION TO the
>Cryix, after it caught on fire or just burned up :)
>--
>Matt McDevitt
>mak...@ix.netcom.com
You idiot, Cyrix 6x take a difrent <lower> voltage than a pentium, if
you use the wrong voltage, it will work, but overheat. my sugestion is
to get a voltage regulator and set it to the recomended cyrix
setting... <if your motherboard doesn't support it>
eni...@gunhed.demon.co.uk
Matt McDevitt (mak...@ix.netcom.com) wrote:
> The chips manufactured by Cyrix....not necessarily the Cyrix processors
> manufactured by IBM and others....are having some serious problems with
> overheating. The average Cyrix computer, as a matter of fact, has at least
> 2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
> heat sinks had better be ceramic or they'd melt! I heard this from
> WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
> do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
> processor (to save money) and end up getting an Intel IN ADDITION TO the
> Cryix, after it caught on fire or just burned up :)
Sounds like Matt has 'Intel Inside'......
Jason von Nieda
Matt McDevitt wrote:
>
> The chips manufactured by Cyrix....not necessarily the Cyrix processors
> manufactured by IBM and others....are having some serious problems with
> overheating. The average Cyrix computer, as a matter of fact, has at least
> 2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
> heat sinks had better be ceramic or they'd melt! I heard this from
> WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
> do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
> processor (to save money) and end up getting an Intel IN ADDITION TO the
> Cryix, after it caught on fire or just burned up :)
> Matt McDevitt
> mak...@ix.netcom.com
Cyrix processors generally come with a Cyrix Heat Sink and fan which is
very
high power. My Cyrix 6x86-133 runs very cool with the supplied heat
sink.
--
Jason von Nieda PC / Network Specialist
E-Mail: age...@idir.net WWW:
http://www.idir.net/~ageless
I represent only myself, because no one else trusts me :)
Charles D. Laderoute
eni...@gunhed.demon.co.uk wrote:
>
> Matt McDevitt (mak...@ix.netcom.com) wrote:
>
> > The chips manufactured by Cyrix....not necessarily the Cyrix processors
> > manufactured by IBM and others....are having some serious problems with
> > overheating. The average Cyrix computer, as a matter of fact, has at least
> > 2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
> > heat sinks had better be ceramic or they'd melt! I heard this from
> > WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
> > do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
> > processor (to save money) and end up getting an Intel IN ADDITION TO the
> > Cryix, after it caught on fire or just burned up :)
>
I have used a Cyril kick-up chip for 3 yrs or so that changed a 386/33 to a 486/66.
Never had a single problem. Runs great. Don't know about their MBs or machines.
cdl
Keith Lucas
In article <01bb92e3$02f40a80$5104b8cd@maktos>,
Matt McDevitt <mak...@ix.netcom.com> wrote:
>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>manufactured by IBM and others....are having some serious problems with
>overheating. The average Cyrix computer, as a matter of fact, has at least
>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>heat sinks had better be ceramic or they'd melt! I heard this from
>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>processor (to save money) and end up getting an Intel IN ADDITION TO the
>Cryix, after it caught on fire or just burned up :)
So you've not actually USED a Cyrix -- just read about them ?
They do not overheat. Well they do if the fan breaks, but not
otherwise. And I'm speaking from the experience of running a 120MHz
one for the last couple of months -- I run Unix, so the thing is just
left on all the time and I have had no problems with overheat. Nor
have I had any reports of overheat from people I know in the hardware
business who are routinely supplying 120+ Cyrixs all over the place.
Maybe Ziff-Davis & co have been infiltrated by Wintel agents...
-----------------------------------------------------------------------------
"It's not a personality.. it's a bulldozer" sill...@wardrobe.demon.co.uk
Current project: Computer wargaming's next generation...
------------------------------------------------------------------------------
Paul Garceau
The Illustrious "Matt McDevitt" <mak...@ix.netcom.com> wrote:
>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>manufactured by IBM and others....are having some serious problems with
>overheating. The average Cyrix computer, as a matter of fact, has at least
>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>heat sinks had better be ceramic or they'd melt! I heard this from
>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>processor (to save money) and end up getting an Intel IN ADDITION TO the
>Cryix, after it caught on fire or just burned up :)
>--
>Matt McDevitt
The Cyrix Kit I am considering comes with a 283W fan. Haven't seen
the box yet, so am unsure as to the ventilation they've put with it.
Though it would seem rather illogical to supply a kit without the
proper cooling facilities. Oh yes, The Cyrix supplied motherboard is
as pure as Cyrix could make it I would think.
Who would know more about their own processor and its environmental
requirments other than the original manufacturer? It may be cheaper
to only buy the processor, but it seems rather self-defeating to do so
unless you are making motherboards for people to use. And, if you
are, it doesn't make any sense to _not_ consider the environmental
requirements for the main processor while creating its box.
Paul G.
RatBoy
On Mon, 26 Aug 1996 08:33:06 GMT, T...@all.mighty (The All Mighty)
wrote:
>On 26 Aug 1996 00:12:53 GMT, "Matt McDevitt" <mak...@ix.netcom.com>
>wrote:
>
>>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>>manufactured by IBM and others....are having some serious problems with
>>overheating. The average Cyrix computer, as a matter of fact, has at least
>>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>>heat sinks had better be ceramic or they'd melt! I heard this from
>>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>>processor (to save money) and end up getting an Intel IN ADDITION TO the
>>Cryix, after it caught on fire or just burned up :)
>>--
>>Matt McDevitt
>>mak...@ix.netcom.com
>You idiot, Cyrix 6x take a difrent <lower> voltage than a pentium, if
>you use the wrong voltage, it will work, but overheat. my sugestion is
>to get a voltage regulator and set it to the recomended cyrix
>setting... <if your motherboard doesn't support it>
>
>
I am one of the device engineers working for a foundary for Cyrix.
Their chips do not overheat when operated at the proper voltage. One
of the tests performed on the chips before shipping is a 168 hour 150
degree celcius burn in (that's 1.5 times the boiling point of water,
Matt). It's statistically unlikely that room temp will will kill it
after that. Now quit spamming!
jmi...@linknet.net
I have a DFI mobo with a Cyrix P166+ that has not been shut down for
23 days so far. It has the Cyrix heatsink and fan. I can put my
fingers on the heatsink and leave them there. The fan is not any
noisier than Pentium fans. I think that the people who are having
overheat problems are using cheap after market heatsink/fan
combinations. No problems here whatsoever!! Save yourself some money
and get Cyrix, but only use the Cyrix heatsink and fan!!
Your Name Here
>>On 26 Aug 1996 00:12:53 GMT, "Matt McDevitt" <mak...@ix.netcom.com>
>>wrote:
>>
>>>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>>>manufactured by IBM and others....are having some serious problems with
>>>overheating. The average Cyrix computer, as a matter of fact, has at least
>>>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>>>heat sinks had better be ceramic or they'd melt! I heard this from
>>>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>>>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>>>processor (to save money) and end up getting an Intel IN ADDITION TO the
>>>Cryix, after it caught on fire or just burned up :)
>>>--
>>>Matt McDevitt
What's it called, disinformation maybe?
Gary Barber
: Well, I own a Cyrix...It's pretty nice so far, havnt really put it
: through the ropes yet. Just got it like 2 or 3 days ago. Only bad
: thing I notice about it so far is the fact that most benchmarks and
: other autodetectors like Windows 95, will detect it as a 486. I assume
: that is because the chip is newer than the benchmark or autodetecter,
: but it's got me pretty worried
That is because the Cyrix chips don't support the CPUID and CPUID1
functions correctly.
-gb
the
On Mon, 26 Aug 1996 05:58:08 GMT, phil...@dowco.com (Daniel Phillips) wrote:
>>Matt McDevitt (mak...@ix.netcom.com) wrote:
>>
>>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>>manufactured by IBM and others....are having some serious problems with
>>overheating.
>
>This is new to me. What I see on the chat sessions is that people
>who have Cyrix chips are generally "ecstatic" about them. It seems
>the performance claims are largely true.
>
>>The average Cyrix computer, as a matter of fact, has at least
>>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>>heat sinks had better be ceramic or they'd melt! I heard this from
>>WWW.ZDNET.COM and other places.
>
>I have had a lot of problems trying to relate things that I find on
>ZDNET and in Ziff-Davis magazines such as PC Mag and Computer
>Gaming World to reality. Just my experience.
>
>---------------
>Daniel Phillips
>phil...@dowcow.com
> if you mean the 486 series maybe but weve dealt with them for a couple years and the 686 series is an excellent performer at
> < intel$$$$
J. Boer
On 28 Aug 1996, Matt McDevitt wrote:
total crap deleted...
> But that article in a reputable publication (Ziff-Davis publishes about 10
> magazies...the 10 most popular, I think) I doubt they'd publish something
> without checking it out first.
you are a blind turd matt.
> But it was logical though....why WOULD it make sense that a smaller company
> could produce cheaper chips, and have them be faster as well? I just
> can't accept that. Intel must be making theirs cheaper, and if they felt
> threatened they'd drop the prices.
by that logic GM would have made the fastest coolest cars
and NBC the best TV show...
> Now if Cyrix chips weren't as reliable...now that would be logical.....
> --
> Matt McDevitt
> mak...@ix.netcom.com
Matt,
Get a life matt. If you don't want a Cyrix don't buy one.
Don't just repeat what you heard with no first hand knowledge
or experience to share or backup your empty spew.
jason
Matt McDevitt
Due to overwhelming response to my original posting, I will go into more
detail on what I said....
They cited in particular the P166+ and they mentioned a couple of quotes,
one of them was
"They should downgrade the P166+ to a P133"
I don't know about the slower chips, they might be just fine.
I used to think it was nice that someone was giving Intel a run for their
money, to promote lower prices (heaven knows I'm waiting till the Pentium
200 comes down in price).
But that article in a reputable publication (Ziff-Davis publishes about 10
magazies...the 10 most popular, I think) I doubt they'd publish something
without checking it out first.
could produce cheaper chips, and have them be faster as well? I just
can't accept that. Intel must be making theirs cheaper, and if they felt
threatened they'd drop the prices.
HA5MRC
Matt McDevitt wrote:
>
> The chips manufactured by Cyrix....not necessarily the Cyrix processors
> manufactured by IBM and others....are having some serious problems with
Don't worry guys, real men know how to cool this F1 engine.
I run it OVERCLOCKED without problems, but I use home-made cooling
system.
(I also used it for my previous Intel processors)
(Sorry, no photo)
Just FYI:
An absolutely not overcrowded PC box: http://teto.sch.bme.hu/club/
(the first picture)
Info about voltages, dissipation and other technical stuff:
ftp://ftp.cyrix.com/6x86/
73 de Sanyi HG5CRS
Francisco Cortina
I own a Cyrix 120 and leave it on days on end. It has never glitched
once.
Neil Graham
Maxwell Smart wrote:
>
> >On 28 Aug 1996, Matt McDevitt wrote:
> >> could produce cheaper chips, and have them be faster as well? I just
> >> can't accept that. Intel must be making theirs cheaper, and if they felt
> >> threatened they'd drop the prices.
> >
> >and NBC the best TV show...
>
> way, attempting to point out that economies of scale often allow a
> larger manufacturer to produce products at a lower per-unit cost.
>
> This has not, however, always held true. Once a company gets TOO
> large and lazy (GM in before the Japanese invasion), their
> cost-per-unit may rise. This also ignores the fact that Big Blue and
> SGS/Thompson is handling the manufacturing.
Economies of scale also does not apply well to design. It's greatest
effects can be seen in production. Better chips are made by better design.
and their design also has an impact on per unit cost (component count etc.).
Another key factor to dedsigning these processors is that you can only
have a few key designers. Because so much is at stake Intel are more
likely to hire the designers who are most likely to get the job done and
done well. Other companies might have taken the risk to hire more inovative
designers with the possibility of a late shipping date or no product at the
end at all. There are several clone processors here or on the horizon, I
am sure more companies were devloping their own but scrapped them when they
fell behind the pace.
--
"Glook!" le...@aurora.co.nz
Like defender? try -> http://www.laserpoint.com/mswin/glook/glook.shtml
That which does not kill us, has made its last mistake
John Hardaway
>>Matt McDevitt (mak...@ix.netcom.com) wrote:
>>
>>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>>manufactured by IBM and others....are having some serious problems with
>>overheating.
>
>This is new to me. What I see on the chat sessions is that people
>who have Cyrix chips are generally "ecstatic" about them. It seems
>the performance claims are largely true.
Being one of the Cyrix P166+ users who, while not quite ecstatic, is
nevertheless quite pleased, I've never had a problem with overheating.
This is on a system that is on 24/7. It has been the most stable
system I've had in a long time.
>
>>The average Cyrix computer, as a matter of fact, has at least
>>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>>heat sinks had better be ceramic or they'd melt! I heard this from
>>WWW.ZDNET.COM and other places.
>
>I have had a lot of problems trying to relate things that I find on
>ZDNET and in Ziff-Davis magazines such as PC Mag and Computer
>Gaming World to reality. Just my experience.
>
Guess I don't have one of the "average" Cyrix systems as it only has 1
fan (running at the same speed as the one on my Intel P90 system) and
1 heat sink. Still haven't seen any sign of a heat problem.
Bill Laspas
On 28 Aug 1996 04:40:28 GMT, "Matt McDevitt" <mak...@ix.netcom.com>
wrote:
>Due to overwhelming response to my original posting, I will go into more
>detail on what I said....
>They cited in particular the P166+ and they mentioned a couple of quotes,
>one of them was
>"They should downgrade the P166+ to a P133"
>I don't know about the slower chips, they might be just fine.
>I used to think it was nice that someone was giving Intel a run for their
>money, to promote lower prices (heaven knows I'm waiting till the Pentium
>200 comes down in price).
>But that article in a reputable publication (Ziff-Davis publishes about 10
>magazies...the 10 most popular, I think) I doubt they'd publish something
>without checking it out first.
>But it was logical though....why WOULD it make sense that a smaller company
>could produce cheaper chips, and have them be faster as well? I just
>can't accept that. Intel must be making theirs cheaper, and if they felt
>threatened they'd drop the prices.
>Now if Cyrix chips weren't as reliable...now that would be logical.....
>--
>Matt McDevitt
>mak...@ix.netcom.com
>
>
You can get the info this dude is writing about at:
http://www.zdnet.com/zdi/special/chips/
Two months ago I built myself a computer using the Cyrix P166+ and had
no problems what so ever. Put my finger on the heat sink and its warm
but that's normal. I used the same parts Cyrix use in their bench mark
system. All I can think of is PC mag is probably run by Republicans.
I been building clones for about ten years now the first one was a
8088 with a clock speed of 4.77 mhz. At that time IBM PCs where going
for two grand.
Later,
Bill
Matt McDevitt
Don't believe me? PC Week explains all at
http://www.pcweek.com/news/0819/19cyr.html
--
Matt McDevitt
mak...@ix.netcom.com
ji...@notintel.com
I have a cyrix p166+ and its never overheated, in fact it runs faster
than shit (which would be intel). I can vouch, for anyone thinking
about a cyrix, there great!
On Wed, 28 Aug 1996 15:54:24 -0500, "J. Boer"
<jb...@blue.weeg.uiowa.edu> wrote:
>
>On 28 Aug 1996, Matt McDevitt wrote:
>
>total crap deleted...
>
>> But that article in a reputable publication (Ziff-Davis publishes about 10
>> magazies...the 10 most popular, I think) I doubt they'd publish something
>> without checking it out first.
>
>you are a blind turd matt.
>
>> But it was logical though....why WOULD it make sense that a smaller company
>> could produce cheaper chips, and have them be faster as well? I just
>> can't accept that. Intel must be making theirs cheaper, and if they felt
>> threatened they'd drop the prices.
>
>by that logic GM would have made the fastest coolest cars
>and NBC the best TV show...
>
>> Now if Cyrix chips weren't as reliable...now that would be logical.....
>> --
>> Matt McDevitt
>> mak...@ix.netcom.com
>
CC
Francisco Cortina wrote:
>
> I own a Cyrix 120 and leave it on days on end. It has never glitched
> once.
Same for me. My friend has the Cyrix 133. No problems their either.
--
Ellsworth Hall,
CyberComposer
http://www.bcpl.lib.md.us/~ehall/aml.html
Patrick Jenkins
>I own a Cyrix 120 and leave it on days on end. It has never glitched
>once.
Same here! I have a Cyrix 166. It has not been turned off for close
to 6 weeks. No problems yet! Not even a crash.
NEENER NEENER
OX-11
Well, perhaps the 'overheating' problem also has something to do with the
fact that Cyrix chips fail to run NT4.0 very well. The chips reportedly
are 25% slower than the equivalent intel chips. Cyrex says not to worry
though-- there's a software fix for the problem, and its foundry is now
correcting the problem in silicon. I wonder if Cyrex will replace any of
the older defective chips with new ones for free the way intel did? If I
were a Cyrex chip owner who just upgraded to NT4 and found out Cyrex wont
even bother replacing the old chip, I would wish I had remained with
intel. A 'software fix' wouldn't be good enough for me.
MDL
snipped...
>>Just thought I'd save some of you the 150 you'd spend on a Cyrix
>>processor (to save money) and end up getting an Intel IN ADDITION TO the
>>Cryix, after it caught on fire or just burned up :)
OK, so Cyrix chips do get hot, however I'm running a P120+ clocked to
a P150+ with no loss in performance or crashes. I have installed a
secondary fan to improve airflow in my case - but even when the
ambient temperature is 70-75F, all is well. The chip cost £75 in the
UK - for P150+ performance who cares about a little heat - if the chip
gets too hot it will cut out (usually before any damage is done).
Mike
Jaguar
CC <eh...@mail.bcpl.lib.md.us> wrote:
>Francisco Cortina wrote:
>>
>> I own a Cyrix 120 and leave it on days on end. It has never glitched
>> once.
>Same for me. My friend has the Cyrix 133. No problems their either.
>--
Another here, I got a Cyrix 120 and no probs
martian
Keith Lucas <sill...@wardrobe.demon.co.uk> wrote:
>In article <01bb92e3$02f40a80$5104b8cd@maktos>,
>Matt McDevitt <mak...@ix.netcom.com> wrote:
>>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>>manufactured by IBM and others....are having some serious problems with
>>overheating. The average Cyrix computer, as a matter of fact, has at least
>>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>>heat sinks had better be ceramic or they'd melt! I heard this from
>>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>>processor (to save money) and end up getting an Intel IN ADDITION TO the
>>Cryix, after it caught on fire or just burned up :)
>So you've not actually USED a Cyrix -- just read about them ?
>They do not overheat. Well they do if the fan breaks, but not
>otherwise. And I'm speaking from the experience of running a 120MHz
>one for the last couple of months -- I run Unix, so the thing is just
>left on all the time and I have had no problems with overheat. Nor
>have I had any reports of overheat from people I know in the hardware
>business who are routinely supplying 120+ Cyrixs all over the place.
>Maybe Ziff-Davis & co have been infiltrated by Wintel agents...
> -----------------------------------------------------------------------------
>"It's not a personality.. it's a bulldozer" sill...@wardrobe.demon.co.uk
> Current project: Computer wargaming's next generation...
> ------------------------------------------------------------------------------
Buy a genuine Intel chip and you'll have no problems plus intel chips
are more faster and reliable
Christian
gsc...@edm.net (Jaguar) wrote:
So do with my Cyrix P166+ never had even the smallest problem with it
and its damn fast.
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
If its not grounded there will be lots of hum but
if there is no hum the circuit may be dead !!!
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
Chris VE2DR
cv...@total.net
VE...@qbgate.ampr.org
The Music Lovers Pages
http://www.total.net/~cverr
Frank Scrooby
Hi all
In article <81d7cc$1339...@news.martian.bc.ca>, now...@man.com (martian)
wrote:
> Buy a genuine Intel chip and you'll have no problems plus intel chips
> are more faster and reliable
Unless you're talking about floating point operations here you have absolutely
no clue what you are talking about. The Cyrix 686 processors drill any Pentium
processor of equivilent megahertz outrageously.
If however you are using floating point operations (as in Quake, Formula One
Grand Prix II and probably CAD packages) then the Pentiums DO win because the
Cyrix chips have an inferior floating point unit.
Another factor is of course cost. The Cyrix processors (looking at some local
prices) are between 40% and 60% the price of a similiar performing Pentium. So
for the price of 1.5 Pentiums you can afford two Cyrix chips. For people trying
to built a decent performance machine on a budget that is important.
More reliable ? I don't know. I've seen a couple of dead processors. I've seen
quite a few fail from overheating. Never seen or heard of a Cyrix dying from
overheating yet (but then again its only been on the market awhile). I've seen a
couple of dead Pentiums and heard of plenty more.
Basically there is no reason to kiss ass to Intel just because they claim there
Pentium is better (the floating point performance IS but so what) If you're
looking for a low end processor the AMD K5 pretty much out performs the P75 with
out much difference in the floating point performance.
Regards
FrankS
Terry Martin
On Thu, 29 Aug 1996 01:05:38 -0700, OX-11 <ja...@omicron.csustan.edu>
wrote:
Well you are not very versed are you or else you would know the
FASTEST chip on the market right now is the CYRIX P200+ so knock off
the ingnorant shit (oh yea NT sucks too only a complete moron would
purchase it)
Words of Wisdom or my .02 worth
SeRVaNT91
martian wrote:
>
> Keith Lucas <sill...@wardrobe.demon.co.uk> wrote:
>
> >In article <01bb92e3$02f40a80$5104b8cd@maktos>,
> >Matt McDevitt <mak...@ix.netcom.com> wrote:
> >>The chips manufactured by Cyrix....not necessarily the Cyrix processors
> >>manufactured by IBM and others....are having some serious problems with
> >>overheating. The average Cyrix computer, as a matter of fact, has at least
> >>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
> >>heat sinks had better be ceramic or they'd melt! I heard this from
> >>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
> >>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
> >>processor (to save money) and end up getting an Intel IN ADDITION TO the
> >>Cryix, after it caught on fire or just burned up :)
>
> >So you've not actually USED a Cyrix -- just read about them ?
>
> >They do not overheat. Well they do if the fan breaks, but not
> >otherwise. And I'm speaking from the experience of running a 120MHz
> >one for the last couple of months -- I run Unix, so the thing is just
> >left on all the time and I have had no problems with overheat. Nor
> >have I had any reports of overheat from people I know in the hardware
> >business who are routinely supplying 120+ Cyrixs all over the place.
>
> Buy a genuine Intel chip and you'll have no problems plus intel chips
> are more faster and reliable
Oh people I'm tired of people makin assumptions on shit they don't even
have. Oh but I've "Read about it"... yeah from another person who also
doesn't have the product. Well I have a 1 month old Cyrix P-166 and
the only thing that it lacks is the FPU speed which is one of the FEW
catagories the intel shit beats it on. I work with Intel Pentium's all
day and yes ones a 166 and the Cyrix is faster. So until you have had
hands on experience with both shut the hell up. And quite posting
dumb ass posts that aren't backed up by SHIT!!!
mu student
> I run it OVERCLOCKED without problems, but I use home-made cooling
> system.
> (I also used it for my previous Intel processors)
What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
you describe your cooling system?
Alejandro Dubrovsky
"Matt McDevitt" <mak...@ix.netcom.com> writes:
>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>manufactured by IBM and others....are having some serious problems with
>overheating. The average Cyrix computer, as a matter of fact, has at least
>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>heat sinks had better be ceramic or they'd melt! I heard this from
>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>processor (to save money) and end up getting an Intel IN ADDITION TO the
>Cryix, after it caught on fire or just burned up :)
>--
>Matt McDevitt
>mak...@ix.netcom.com
As far as i know, cyrix doesn't make any of their chips, that is, they are
a fabless company and ALL of their chips are manufactured by IBM. They
split the chips 50-50 and put their names on it.
And might as well add i got a C6x86 100, no problems
chau
Robert Neely
In article <81d7cc$1339...@news.martian.bc.ca>, now...@man.com says...
>
>Keith Lucas <sill...@wardrobe.demon.co.uk> wrote:
>
>>In article <01bb92e3$02f40a80$5104b8cd@maktos>,
>>Matt McDevitt <mak...@ix.netcom.com> wrote:
>>>manufactured by IBM and others....are having some serious problems with
>>>overheating. The average Cyrix computer, as a matter of fact, has at least
>>>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>>>heat sinks had better be ceramic or they'd melt! I heard this from
>>>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>>>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>>>processor (to save money) and end up getting an Intel IN ADDITION TO the
>>>Cryix, after it caught on fire or just burned up :)
>
>
>>They do not overheat. Well they do if the fan breaks, but not
>>otherwise. And I'm speaking from the experience of running a 120MHz
>>one for the last couple of months -- I run Unix, so the thing is just
>>left on all the time and I have had no problems with overheat. Nor
>>have I had any reports of overheat from people I know in the hardware
>>business who are routinely supplying 120+ Cyrixs all over the place.
>
>
>>
-----------------------------------------------------------------------------
>>"It's not a personality.. it's a bulldozer" sill...@wardrobe.demon.co.uk
>> Current project: Computer wargaming's next generation...
>>
------------------------------------------------------------------------------
>
>are more faster and reliable
>
>
>
I think it is about time for this thread to die. If you read the original
post, there are a couple of glaring errors that make me think this was just
flame fodder.
First, a properly designed system should only need one fan (well two if you
have one on the heat sink), unless it's something like a specialized server
with with a lot of extra hardware (multiple disk arrays or the sort). I have
three fans in my computer, but parts of it (case and power supply in
particular) started as a '286 system back in 1987 and it has been evolving ever
since. In other words, it was not designed for a 5th or 6th generation x86.
Second, I don't think it is possible to put 2 heat sinks on a CPU. It would be
sort of like putting 8 tires on an auto. And "caught fire and burned up"? In
20 years of tinkering I've smoked a few circuits, but I've never seen a 5V (or
3.3V as the case is here) circuit actually burst into flame.
Personally, I've enjoyed reading everyone's reports about the Cyrix chips, but
these are primarilly binary newsgroups and discussions should be taken
elsewhere. That's just proper netiquette.
Paul Garceau
The Illustrious "Matt McDevitt" <mak...@ix.netcom.com> wrote:
>Due to overwhelming response to my original posting, I will go into more
>detail on what I said....
>They cited in particular the P166+ and they mentioned a couple of quotes,
>one of them was
>"They should downgrade the P166+ to a P133"
>I don't know about the slower chips, they might be just fine.
>I used to think it was nice that someone was giving Intel a run for their
>money, to promote lower prices (heaven knows I'm waiting till the Pentium
>200 comes down in price).
>But that article in a reputable publication (Ziff-Davis publishes about 10
>magazies...the 10 most popular, I think) I doubt they'd publish something
>without checking it out first.
>But it was logical though....why WOULD it make sense that a smaller company
>could produce cheaper chips, and have them be faster as well? I just
>can't accept that.
Big is not necessarily better.
A smaller company, with a much smaller overhead doesn't have to charge
such high prices for their chips, simply because they don't have the
need to support a multi-billion dollar operation.
Cyrix makes chips and, I believe, the mother-boards for their chips.
They have no other costs outside of normal operating costs (paying the
utilities, minimal advertising and customer support). Small is
better, and more able to operate with efficiency than a large company
is able to.
A smaller company has no need to charge high prices for their product
(not sure of the figures, but at least 10-15 million dollars a week of
Intels yearly budget goes into their prime-time television commercials
-- Cyrix can probably operate on that amount of money for a number of
years). Especially if they only need to cover their basic operating
costs (including manufacturing).
> Intel must be making theirs cheaper, and if they felt
>threatened they'd drop the prices.
I would have to disagree with this. Intel charges what they do
because they have no choice. Cyrix knows this, and so, Cyrix can
undercut Intel while at the same time providing a better, more
reliable and more flexible chip. Simply, Cyrix can, and does, provide
a better cpu.
>Now if Cyrix chips weren't as reliable...now that would be logical.....
And as you know, the Cyrix _is_ more reliable. Small _is_ better.
Paul G.
Laine Jones
OX-11 <ja...@omicron.csustan.edu> wrote in article
<Pine.SUN.3.91.96082...@omicron.csustan.edu>...
> Well, perhaps the 'overheating' problem also has something to do with the
> fact that Cyrix chips fail to run NT4.0 very well. The chips reportedly
> are 25% slower than the equivalent intel chips. Cyrex says not to worry
> though-- there's a software fix for the problem, and its foundry is now
> correcting the problem in silicon. I wonder if Cyrex will replace any of
> the older defective chips with new ones for free the way intel did? If I
> were a Cyrex chip owner who just upgraded to NT4 and found out Cyrex wont
> even bother replacing the old chip, I would wish I had remained with
> intel. A 'software fix' wouldn't be good enough for me.
If I remember right (remember reading numerous articles) Intel's Pentium
PRO 200 also runs WIN95/NT 4.0 SLOW due to the internals of the Chip.. Now
if you look at the speck sheets for the Cyrix/IBM 6x86 chips you will find
it's has the same architecture of a Pentium pro.. so it's NOT an isolated
problems with Cyrix/IBM chips
Architectural Comparison
6x86 Pentium Pro Pentium
Processor Processor Processor
Full x86 Instruction Set
Optimization X
Superscalar X X X
Superpipelined X X
Register Renaming X X
Data Dependency Removal X X
Multi-Branch Prediction X X
Speculative Execution X X
Out-of-Order Completion X X
80-Bit Floating Point Unit X X X
16K Primary Cache X X X
So in other words quit posting stupid message about "well I heard from my
buddy's sister's boy-friend's father's brother's Son who is a wanna-be tech
thats it's does some things SLOW". Maybe Intel with have to replace their
PP-200s as well due to the SLOW speed running WIN95. Or maybe Microsoft
will have to look into what Superscalar architecture does to performance on
some operating systems? (once again Thats my 2 bits...)
Brian Salagaj
ji...@notintel.com wrote:
>
> I have a cyrix p166+ and its never overheated, in fact it runs faster
> than shit (which would be intel). I can vouch, for anyone thinking
> about a cyrix, there great!
>
> On Wed, 28 Aug 1996 15:54:24 -0500, "J. Boer"
> <jb...@blue.weeg.uiowa.edu> wrote:
>
> >
> >On 28 Aug 1996, Matt McDevitt wrote:
> >
> >total crap deleted...
> >
> >> magazies...the 10 most popular, I think) I doubt they'd publish something
> >> without checking it out first.
> >
> >
> >> could produce cheaper chips, and have them be faster as well? I just
> >> threatened they'd drop the prices.
> >
> >and NBC the best TV show...
> >
> >> Matt McDevitt
> >> mak...@ix.netcom.com
> >
> >Matt,
> > Get a life matt. If you don't want a Cyrix don't buy one.
> >Don't just repeat what you heard with no first hand knowledge
> >or experience to share or backup your empty spew.
> >
> >jason
had a problem with my 6x86 133
--
Brian
======================================================================
* USS VOYAGER *
* ___ *
* __.----'---`----.__ * Alien245
* _.--'=========O=========`--._ * br...@mars.superlink.net
* _.-' o `-._
*http://mars.superlink.net/brian
* ===================================== *
* .----.__`---._________.---'__.----. *
* |/__\|------/___ _ ___\------|/__\| *
* `----' | .' `. | `----' *
* \ `._.' / *
* `-------' *
"Time is the fire in which we burn" Soren-Generations
======================================================================
Mike O'Connell
>Keith Lucas <sill...@wardrobe.demon.co.uk> wrote:
>>In article <01bb92e3$02f40a80$5104b8cd@maktos>,
>>Matt McDevitt <mak...@ix.netcom.com> wrote:
>>>I heard this from
>>>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>>>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>>>processor (to save money) and end up getting an Intel IN ADDITION TO the
>>>Cryix, after it caught on fire or just burned up :)
>>So you've not actually USED a Cyrix -- just read about them ?
>>They do not overheat. Well they do if the fan breaks, but not
>>otherwise. And I'm speaking from the experience of running a 120MHz
>>one for the last couple of months -- I run Unix, so the thing is just
>>left on all the time and I have had no problems with overheat. Nor
>>have I had any reports of overheat from people I know in the hardware
>>business who are routinely supplying 120+ Cyrixs all over the place.
>>Maybe Ziff-Davis & co have been infiltrated by Wintel agents...
>are more faster and reliable
Are you speaking from experience with the full range of Cyrix chips? Or
are you just another mindless drone who reads press releases and thinks he
knows something?
Me, I've worked with nigh-everything (except the P200, but it doesn't
count - it's shipping, but only barely) including most of Cyrix's chips.
Their chips don't run nearly as hot as the P60 and P66's, and Ziff
definately didn't run articles like this...
BTW, Matt, Ziff has a reputation in the industry for espousing half-truths
as fact - something that's endeared them to some companies. PC/Mac Week
are the worst (stating nothing more than what a company's spin doctor
wants them to print), but their monthly publications sometimes get slimy,
too.
BTW, stop crossposting to binary groups. There's enough crap floating
around in them.
--
Mike O'Connell
ocon...@icomsim.com
"Intel Inside" - the world's most widely used warning label
None of your damn business
catf...@iofc.com (Terry Martin) wrote:
>On Thu, 29 Aug 1996 01:05:38 -0700, OX-11 <ja...@omicron.csustan.edu>
>wrote:
>>Well, perhaps the 'overheating' problem also has something to do with the
>>fact that Cyrix chips fail to run NT4.0 very well. The chips reportedly
>>are 25% slower than the equivalent intel chips. Cyrex says not to worry
>>though-- there's a software fix for the problem, and its foundry is now
>>correcting the problem in silicon. I wonder if Cyrex will replace any of
>>the older defective chips with new ones for free the way intel did? If I
>>were a Cyrex chip owner who just upgraded to NT4 and found out Cyrex wont
>>even bother replacing the old chip, I would wish I had remained with
>>intel. A 'software fix' wouldn't be good enough for me.
>>
> Well you are not very versed are you or else you would know the
>FASTEST chip on the market right now is the CYRIX P200+ so knock off
>the ingnorant shit (oh yea NT sucks too only a complete moron would
>purchase it)
>Words of Wisdom or my .02 worth
Your .02 isn't worth a pot of piss...Obviously you have no clue what
NT is or have even seen it. As someone who has hated virtually
Microsoft product so far, it pains me to say that NT 4.0 rocks.
Obviously you are some sort of 12 year old moron whose mommy and daddy
won't buy him NT and now you're pissed about it
hack_willy
On Fri, 30 Aug 1996 10:59:38 -0700, SeRVaNT91
<inden...@earthlink.net> wrote:
>martian wrote:
>>
>> Keith Lucas <sill...@wardrobe.demon.co.uk> wrote:
>>
>> >In article <01bb92e3$02f40a80$5104b8cd@maktos>,
>> >Matt McDevitt <mak...@ix.netcom.com> wrote:
>> >>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>> >>manufactured by IBM and others....are having some serious problems with
>> >>overheating. The average Cyrix computer, as a matter of fact, has at least
>> >>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>> >>heat sinks had better be ceramic or they'd melt! I heard this from
>> >>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>> >>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>> >>processor (to save money) and end up getting an Intel IN ADDITION TO the
>> >>Cryix, after it caught on fire or just burned up :)
>>
>> >So you've not actually USED a Cyrix -- just read about them ?
>>
>> >They do not overheat. Well they do if the fan breaks, but not
>> >otherwise. And I'm speaking from the experience of running a 120MHz
>> >one for the last couple of months -- I run Unix, so the thing is just
>> >left on all the time and I have had no problems with overheat. Nor
>> >have I had any reports of overheat from people I know in the hardware
>> >business who are routinely supplying 120+ Cyrixs all over the place.
>>
>
>> Buy a genuine Intel chip and you'll have no problems plus intel chips
>> are more faster and reliable
>
>Oh people I'm tired of people makin assumptions on shit they don't even
>have. Oh but I've "Read about it"... yeah from another person who also
>doesn't have the product. Well I have a 1 month old Cyrix P-166 and
>the only thing that it lacks is the FPU speed which is one of the FEW
>catagories the intel shit beats it on. I work with Intel Pentium's all
>day and yes ones a 166 and the Cyrix is faster. So until you have had
>hands on experience with both shut the hell up. And quite posting
>dumb ass posts that aren't backed up by SHIT!!!
I have a Cyrix P120+ ,its been on for about 3 months straight ,no
problems.I'm going to get a P200+ for work.
joels
>ji...@notintel.com wrote:
>>
>> I have a cyrix p166+ and its never overheated, in fact it runs faster
>> than shit (which would be intel). I can vouch, for anyone thinking
>> about a cyrix, there great!
>>
>> On Wed, 28 Aug 1996 15:54:24 -0500, "J. Boer"
>> <jb...@blue.weeg.uiowa.edu> wrote:
>>
>> >
>I have to agree with Jimmy, Cyrix are faster than intel and I haven't
>had a problem with my 6x86 133
>--
>Brian
I would like to set the record straight based on my experience. I own
a computer store and I build many Pentium and Cyrix 6x class machines.
Yes, the Cyrix chips run significantly hotter than Pentium chips. No,
this is NOT a problem if you put the chip on a well-made motherboard.
We sell 2 motherboards. One for the Pentium and one for the Cyrix.
The Cyrix will flake in Win95 if the board is not up to spec. This is
NOT a problem with the chip but with boards that are not manufactured
with the Cyrix voltage requirements. If you're in the market I would
recommend staying away from cheap chinese "generic" boards with
the Cyrix. We use the FIC 2200 (www.fic.com.kr) 430HX chipset
boards and have 0% failure with the Cyrix. YES, the Cyrix is faster
(except in Quake and programs that use the FPU intensively, which
account for maybe 1% of all programs), especially in Win95. NO
it will not burn up or 'turn to glass'. That idea is comical at best.
I sincerely wonder who is behind all the yellow journalism lately
concerning the 6x86. These ppl are being paid or they really
don't know much about CPUs or motherboards. Unfortunately,
ppl never question the magazines.
Joel
Wabbit
On 31 Aug 1996 04:26:29 GMT, "Laine Jones" <ljo...@direct.ca> wrote:
>
>
>OX-11 <ja...@omicron.csustan.edu> wrote in article
><Pine.SUN.3.91.96082...@omicron.csustan.edu>...
>> Well, perhaps the 'overheating' problem also has something to do with the
>
>> fact that Cyrix chips fail to run NT4.0 very well. The chips reportedly
>> are 25% slower than the equivalent intel chips. Cyrex says not to worry
>> though-- there's a software fix for the problem, and its foundry is now
>> correcting the problem in silicon. I wonder if Cyrex will replace any of
>> the older defective chips with new ones for free the way intel did? If I
>> were a Cyrex chip owner who just upgraded to NT4 and found out Cyrex wont
>
>> even bother replacing the old chip, I would wish I had remained with
>> intel. A 'software fix' wouldn't be good enough for me.
>
**********************************************
Keep the Peace
Watch the Cross-Posting
"I'll get you, you cwazy, wascawly Wabbit"
The Energizer Wabbit keeps going and going....
**********************************************
Paul Garceau
I would not encourage anyone to overclock anything. Stay within specs
and you can count on support when it's needed. Go outside of specs,
and you're on your own.
My $.02 worth.
Paul G.
David Pugmire
The old Windows vs. everyone else is back in this post.
My reply is this. First I use a Cyrix P166+ and have never had a problem
As for Windows, NT , etc. I have used ALL versions of windows that are
on the market, I do not like them, but that is my personal choise. Not
everyone has to like the same product, and it does not mean you do not
know what you are talking about if your choise is say OS/2 and not
WINNT.
Peter F. Curran
In article <322776...@zansiii.millersv.edu>,
mu student <bo...@zansiii.millersv.edu> writes:
>> I run it OVERCLOCKED without problems, but I use home-made cooling
>> system.
>> (I also used it for my previous Intel processors)
>
>What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
>you describe your cooling system?
If you overclocked that much you'd never get the initial bios prompt. In
my experience 133% is the maximum speed you can usually get. Remember,
when you overclock you are trying to run a chip at a speed that other chips
have been able to run off which were made using the same manufacturing
process. There has never been a 486DX 50, (A DX2 50 is a different chip),
so it probably wont work.
-Pete Curran
Peter F. Curran
In article <50ajc2$b...@nadine.teleport.com>,
pa...@teleport.com (Paul Garceau) writes:
>The Illustrious mu student <bo...@zansiii.millersv.edu> wrote:
>
>>> system.
>>> (I also used it for my previous Intel processors)
>
>>What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
>>you describe your cooling system?
>
>and you can count on support when it's needed. Go outside of specs,
>and you're on your own.
>
> My $.02 worth.
>
> Paul G.
>
>
The only REAL danger with overclocking is the possibility of unknowingly
corrupting data. If your chip failed to pass at a higher speed rating
than the one printed on it's chip due to a floating point operation
error, you may risk bad calculations. Thoroughly testing a CPU at the
overclocked speed is essential.
Remember, these chips came off the same "assembly line", (manufacturing
process), and so a P75 is virtually identical to a P90, except for the
markings. The difference is the P75 either failed to run at 90 MHz
during a torture test at elevated temperatures, OR perhaps Intel just
needed more P75 chips than P90 ones!
Also, although I overclock myself, I support Intel's use of over-clock
blocking circuitry. It is because overclocking works so well on most
chips that you have the illegal practice of people relabeling chips for
a higher speed.
Peter Curran
Rensselaer Polytechnic Institute
Peter F. Curran
In article <01bb949a$b1e6fbc0$4796d9ce@maktos>,
"Matt McDevitt" <mak...@ix.netcom.com> writes:
>Due to overwhelming response to my original posting, I will go into more
>detail on what I said....
>They cited in particular the P166+ and they mentioned a couple of quotes,
>one of them was
>"They should downgrade the P166+ to a P133"
>I don't know about the slower chips, they might be just fine.
>I used to think it was nice that someone was giving Intel a run for their
>money, to promote lower prices (heaven knows I'm waiting till the Pentium
>200 comes down in price).
>magazies...the 10 most popular, I think) I doubt they'd publish something
>without checking it out first.
>could produce cheaper chips, and have them be faster as well? I just
>can't accept that. Intel must be making theirs cheaper, and if they felt
>threatened they'd drop the prices.
>--
>Matt McDevitt
>mak...@ix.netcom.com
>
>
Plain and simple truth? The Cyrix has a better architectural design! Now
don't get me wrong, in many ways the Pentium is a superior product. Their
floating point unit is more efficient and they use a more sophisticated
manufacturing process which allows them to make smaller, less power
consuming, and more efficient transistors. If you could use Intel's fab
sites to make the Cyrix design, you'd see an enormous increase in speed.
The reason the Intel chip was more expensive, (and note they recently
slashed prices), was because they could charge whatever the market would
bear.
Pete Curran
Rensselaer Polytechnic Inst.
K.Y. Jerry
How do you make a "home-made" extra cooling system for a CPU?
Peter F. Curran
In article <81d7cc$1339...@news.martian.bc.ca>,
now...@man.com (martian) writes:
>Keith Lucas <sill...@wardrobe.demon.co.uk> wrote:
>
>>In article <01bb92e3$02f40a80$5104b8cd@maktos>,
>>Matt McDevitt <mak...@ix.netcom.com> wrote:
>>>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>>>manufactured by IBM and others....are having some serious problems with
>>>overheating. The average Cyrix computer, as a matter of fact, has at least
>>>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>>>heat sinks had better be ceramic or they'd melt! I heard this from
>>>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>>>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>>>processor (to save money) and end up getting an Intel IN ADDITION TO the
>>>Cryix, after it caught on fire or just burned up :)
>
>>So you've not actually USED a Cyrix -- just read about them ?
>
>>They do not overheat. Well they do if the fan breaks, but not
>>otherwise. And I'm speaking from the experience of running a 120MHz
>>one for the last couple of months -- I run Unix, so the thing is just
>>left on all the time and I have had no problems with overheat. Nor
>>have I had any reports of overheat from people I know in the hardware
>>business who are routinely supplying 120+ Cyrixs all over the place.
>
>
>>"It's not a personality.. it's a bulldozer" sill...@wardrobe.demon.co.uk
>> Current project: Computer wargaming's next generation...
>> ------------------------------------------------------------------------------
>
>are more faster and reliable
>
>
>
The above comment is just plain WRONG. Although Intel chips are reliable
for the most part, they are not infalable!!!! Anyone remember the
infamous Intel FPU bug!?!?!?! Yes it was fast, but it gave the wrong
answer!
The Cyrix processors are a viable alternative to Intel's and that is
why Intel was forced to slash prices. This is a GOOD THING because
competition is just what end users need in order to keep the prices
reasonable.
Gene Irwin
ji...@notintel.com wrote:
>
> I have a cyrix p166+ and its never overheated, in fact it runs faster
> than shit (which would be intel). I can vouch, for anyone thinking
> about a cyrix, there great!
>
> On Wed, 28 Aug 1996 15:54:24 -0500, "J. Boer"
> <jb...@blue.weeg.uiowa.edu> wrote:
>
> >
> >
> >total crap deleted...
> >
> >> magazies...the 10 most popular, I think) I doubt they'd publish something
> >> without checking it out first.
> >
> >
> >> could produce cheaper chips, and have them be faster as well? I just
> >> can't accept that. Intel must be making theirs cheaper, and if they felt
> >> threatened they'd drop the prices.
> >
> >and NBC the best TV show...
> >
> >> --
> >> Matt McDevitt
> >> mak...@ix.netcom.com
> >
> > Get a life matt. If you don't want a Cyrix don't buy one.
> >Don't just repeat what you heard with no first hand knowledge
> >or experience to share or backup your empty spew.
> >
> >jason
Has anyone ever looked at the history of PC-MAG? A few years there was a
review of printers. they bashed a particular epson model and raved about
another brand (Okidata, I think). 2 Issues later, they trashed the other
brand and sang the praises of the epson (both printers the exact same
model). I have no proof, just a suspicion. I believe they rave about
the one who advertises on more pages/with bigger ads. (that did seem to
be the case with the printer review confusion).
I no longer believe a word they print.
BTW, does anyone remember that INTEL had a problem with chips
overheating?
Food for thought.
Thomas Djafari
>
> Has anyone ever looked at the history of PC-MAG? A few years there was a
> review of printers. they bashed a particular epson model and raved about
> another brand (Okidata, I think). 2 Issues later, they trashed the other
> brand and sang the praises of the epson (both printers the exact same
> model). I have no proof, just a suspicion. I believe they rave about
> the one who advertises on more pages/with bigger ads. (that did seem to
> be the case with the printer review confusion).
>
> I no longer believe a word they print.
>
> BTW, does anyone remember that INTEL had a problem with chips
> overheating?
>
> Food for thought.
I worked in a French company that publishes several magazines :
If u want a good article :
1) spend a lot of money in advertising, they NEVER bash good
advertisers.
2) take the journalists to eat or offer them a week end (ex promote a
jet sky game ? take them to a week end of jet sky with the product's
presentation)
Mr.P
On 1 Sep 1996 06:17:52 GMT, Je...@unknown.address.com (K.Y. Jerry)
wrote:
>How do you make a "home-made" extra cooling system for a CPU?
Create an extra small fan and clip it or glue it onto the CPU.
Better yet, create an extra small fan, glue a heatsink to the CPU, and
glue the fan to the heatsink so that the air gets blown directly onto
the heatsink.
Or maybe an ice-pack will do. :)
Mr.P
On 1 Sep 1996 06:05:07 GMT, cur...@remove.rpi.edu (Peter F. Curran)
wrote:
>>> system.
>>> (I also used it for my previous Intel processors)
>>What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
>>you describe your cooling system?
>my experience 133% is the maximum speed you can usually get. Remember,
>when you overclock you are trying to run a chip at a speed that other chips
>have been able to run off which were made using the same manufacturing
>process. There has never been a 486DX 50, (A DX2 50 is a different chip),
>so it probably wont work.
There has been a 486 DX 50. No DX 66, but a DX 50 has existed.
Alex Noble
In article <50b92j$8...@usenet.rpi.edu>, "Peter F. Curran"
<cur...@remove.rpi.edu> writes
>If you overclocked that much you'd never get the initial bios prompt. In
>my experience 133% is the maximum speed you can usually get. Remember,
I've got amotherboard which can run at 25, 33, 40 or 50 MHz, and a TI (I
think) DX2-80 processor. Currently I'm running the board at 33MHz which
gives a total of 66MHz processor speed because when I step the MB speed
upto 40MHz (which should let the processor run at maximum speed) the
system doesn't work at all. Would it be viable to try and overclock the
MB and run it at 50MHz, thereby running the chip at 100MHz? What would
be the danger, _what might get toasted?_ Is there danger to the rest of
the system components? And if anybodies there, how come it doesn't work
in 40MHz?
>process. There has never been a 486DX 50, (A DX2 50 is a different chip),
>so it probably wont work.
There has been an Intel 486DX-50, where the motherboard _and_ the chip
ran at 50MHz un-clockdoubled, but there were some problems with them
because they ran very hot and many MBs didn't like being ran at 50MHz.
Plenty people used them though, they _were_ the fastest of their time.
Alex Noble.........................................alex@elbon.demon.co.uk
Coming soon : www.fatcitynews.com
Stephen Worthington
In article <50b92j$8...@usenet.rpi.edu>,
Peter F. Curran <cur...@remove.rpi.edu> wrote:
>In article <322776...@zansiii.millersv.edu>,
> mu student <bo...@zansiii.millersv.edu> writes:
>>> I run it OVERCLOCKED without problems, but I use home-made cooling
>>> system.
>>> (I also used it for my previous Intel processors)
>>
>>What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
>>you describe your cooling system?
>
>my experience 133% is the maximum speed you can usually get. Remember,
>have been able to run off which were made using the same manufacturing
>so it probably wont work.
>
Tell that to my CPU. It is a 486DX 50 MHz (not clock doubled), and
it was made by Intel before they started to produce the clock doubled
chips. Few motherboard manufacturers were able at that time to make
the 50 MHz clocks work correctly, so hence the clock doubling so they
could run the RAM bus at 33 MHz and be able to make the motherboards
easily.
My PC has been running very happily at 50 MHz for years now, probably
because I read the chip specs and quickly added a CPU fan to keep it
within temperature specs (even though I was assured everything was OK
by the retailer). But the spec sheet clearly showed that in summer
when the ambient temperature got over 25 degrees Celcius, I would be
in trouble. And my past experience with Intel processors (I am a
software engineer speciallising in embedded systems) said that
overheating was always the problem with them. The secret to higher
clock speeds (and overclocking) seems always to be "keep it cool".
--
Stephen Worthington Telephone: +64-4-569-6764 (home)
Digi-Tech Communications Ltd +64-4-389-8909 (work)
ste...@digitech.co.nz (work) Fax: +64-4-389-9901 (work)
ste...@inisant.actrix.gen.nz (home)
Yettie
In message <322776...@zansiii.millersv.edu> - mu student
<bo...@zansiii.millersv.edu>Fri, 30 Aug 1996 19:16:23 -0400
Laine Jones
Peter F. Curran <cur...@remove.rpi.edu> wrote in article
<50b92j$8...@usenet.rpi.edu>...
> In article <322776...@zansiii.millersv.edu>,
> mu student <bo...@zansiii.millersv.edu> writes:
> >> I run it OVERCLOCKED without problems, but I use home-made cooling
> >> system.
> >> (I also used it for my previous Intel processors)
> >
> >What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
> >you describe your cooling system?
>
> If you overclocked that much you'd never get the initial bios prompt. In
> my experience 133% is the maximum speed you can usually get. Remember,
> when you overclock you are trying to run a chip at a speed that other
chips
> have been able to run off which were made using the same manufacturing
> process. There has never been a 486DX 50, (A DX2 50 is a different
chip),
> so it probably wont work.
>
> -Pete Curran
Sorry Pete There was an Intel 486 DX-50 for a short while (plagued with
timing problems on bus speed)
So Intel scrapped it and went with DX/2 chips. I have seen and upgraded
these boards for people many times!
-Laine
Faker Mage
David Matiskella
Liquid cooling is definitely the way to go. I think Freon used to be the
fluid of choice since if it leaked it wouldn't fry your computer. Liquid
nitrogen would be cool also except that its a bit to expensive for normal
use. Besides with all that heat you will not have to by a space heater:)
David Matiskella
mati...@aa.washington.edu
beavis
On Mon, 02 Sep 1996 02:15:08 +0000, ste...@inisant.actrix.gen.nz
(Stephen Worthington) wrote:
>In article <50b92j$8...@usenet.rpi.edu>,
>Peter F. Curran <cur...@remove.rpi.edu> wrote:
>>In article <322776...@zansiii.millersv.edu>,
>> mu student <bo...@zansiii.millersv.edu> writes:
>>>> I run it OVERCLOCKED without problems, but I use home-made cooling
>>>> system.
>>>> (I also used it for my previous Intel processors)
>>>
>>>What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
>>>you describe your cooling system?
>>
>>If you overclocked that much you'd never get the initial bios prompt. In
>>my experience 133% is the maximum speed you can usually get. Remember,
>>when you overclock you are trying to run a chip at a speed that other chips
>>have been able to run off which were made using the same manufacturing
>>process. There has never been a 486DX 50, (A DX2 50 is a different chip),
> ^^^^^^^^^^^^^^^^^^^^^
>>so it probably wont work.
>>
>> -Pete Curran
>
>Tell that to my CPU. It is a 486DX 50 MHz (not clock doubled), and
>it was made by Intel before they started to produce the clock doubled
>chips. Few motherboard manufacturers were able at that time to make
>the 50 MHz clocks work correctly, so hence the clock doubling so they
>could run the RAM bus at 33 MHz and be able to make the motherboards
>easily.
>
>My PC has been running very happily at 50 MHz for years now, probably
>because I read the chip specs and quickly added a CPU fan to keep it
>within temperature specs (even though I was assured everything was OK
>by the retailer). But the spec sheet clearly showed that in summer
>when the ambient temperature got over 25 degrees Celcius, I would be
>in trouble. And my past experience with Intel processors (I am a
>software engineer speciallising in embedded systems) said that
>overheating was always the problem with them. The secret to higher
>clock speeds (and overclocking) seems always to be "keep it cool".
>
>--
>Stephen Worthington Telephone: +64-4-569-6764 (home)
>Digi-Tech Communications Ltd +64-4-389-8909 (work)
>ste...@digitech.co.nz (work) Fax: +64-4-389-9901 (work)
>ste...@inisant.actrix.gen.nz (home)
I've always loved it when someone starts talking about something they
know nothing about, then someone who know something coes along and
blows them out of the water.I have been running a 200+ system lan with
over half of the systems using 486DX50 CPU's for the better part of 3
years now.. As far as Overclocking is concerned. I agree If you can
keep it cool enough it will work up to about 140% of its rating after
that, all I've seen is smoke city.....I would like to get any feedack
by anyone who has been able to excede that mark. and how they did it
short of active refrigerants..
Jaymes Schooler
ITOS Computer Services
Honolulu Hi
David Scott
On 30 Aug 1996 13:49:36 GMT, rob...@poultry.poulsci.ncsu.edu (Robert
Neely) wrote:
I have been using a Cyrix Chip and Board for 4 months now. I am an
A+ certified technician that has been working in computers for about
10 years.
My system is a Cyrix 686 @ 133 Mhz with 3 1.6 gig drives, an 8X
Creative cd-rom, a Mitsumi 6X cd-rom,(one for games.. One for a
dedicated data cd.) a wimpy 380 meg drive I am playing with, 50 Meg of
Ram 32 meg edo and 32 meg non-edo. 512 Cache, a Creative labs awe32
sound card, that runs both cd-roms, a Tseng w32/64 2mg pci video card,
a Future Domain SCSI controller(just to see if I could find an IRQ for
it.)and a Maxitech PCi NIC.
My system is on 24x7 and I have (get this..) ONE real fan...That is
the fan in the power supply. The cpu has a little Heast sink and fan
on it.
What every one seems to forget is that the INTEL units HAVE THE
EXACT SAME PROBLEM! When the P-60 First came out I test 3 units.
They all Died because Intel had a heat problem.. If you compare the
Heat sink that comes on an intel and a cyrix... you wonder why the
intel even has a chip.. It is at least twice the surface area.
Do Not believe everything that Ziff-Davis(microsoft bed buddies)
tells you.
Dave
>In article <81d7cc$1339...@news.martian.bc.ca>, now...@man.com says...
>>
>>Keith Lucas <sill...@wardrobe.demon.co.uk> wrote:
>>
>>>In article <01bb92e3$02f40a80$5104b8cd@maktos>,
>>>Matt McDevitt <mak...@ix.netcom.com> wrote:
>I think it is about time for this thread to die. If you read the original
>post, there are a couple of glaring errors that make me think this was just
>flame fodder.
>
>Second, I don't think it is possible to put 2 heat sinks on a CPU. It would be
>sort of like putting 8 tires on an auto. And "caught fire and burned up"? In
>20 years of tinkering I've smoked a few circuits, but I've never seen a 5V (or
>3.3V as the case is here) circuit actually burst into flame.
>
>Personally, I've enjoyed reading everyone's reports about the Cyrix chips, but
>these are primarilly binary newsgroups and discussions should be taken
>elsewhere. That's just proper netiquette.
>
David Scott
I am sorry, Robert did not write this.. I did.
The Postman
On Sun, 01 Sep 1996 10:28:02 -0700, Gene Irwin <gir...@frugal.com> wrote:
:ji...@notintel.com wrote:
:>
:> I have a cyrix p166+ and its never overheated, in fact it runs faster
:> than shit (which would be intel). I can vouch, for anyone thinking
:> about a cyrix, there great!
:>
Yeah, lemme see how well the 6X86-166 stacks against a REAL P5-166 in all tests-
sure the Pcbench tests come out a little better, but I dont see them running
Spec95 or Spec92 on these chips- Cyrix only compares them in Application power
and not in raw computing power. Any performance gain is not over all areas,
just "Windows 95" machines. Anyone who is going to run a UNIX OS or Windows NT
is not going to want a cyrix chip- on NT the cyrix units suffered a huge
performance hit. I've tried a Cyrix-5x86-120 and compared it against a
DX4-120(AMD) and found that although its application speed was a little faster,
the RAW CPU SPEED tests all indicated that it was slower, even with the branch
prediction turned on- The only tester that said that it was faster was Landmark
6.0- and that's because the special branch thingie in the cyrix chip fools the
crap out of landmark tester, because landmark uses the same calc scheme over and
over again. Cyrix chips are fine for applications, but they still don't offer
more chip for the money.
:> >
:> >> But that article in a reputable publication (Ziff-Davis publishes about 10
:> >> magazies...the 10 most popular, I think) I doubt they'd publish something
:> >> without checking it out first.
:> >
:> >you are a blind turd matt.
ZD is quite reputable in most respects- they gave a honest comparison of both
the intel and Cyrix chips.
:> >
:> >> But it was logical though....why WOULD it make sense that a smaller company
:> >> could produce cheaper chips, and have them be faster as well? I just
:> >> can't accept that. Intel must be making theirs cheaper, and if they felt
:> >> threatened they'd drop the prices.
:> >
Well, you're looking at it the wrong way. Intel knows they own the X86 CPU
market. So why are they going to drop prices and risk getting into the same
"tit-for-tat" cutthroat pricewar that Cyrix and AMD are in right now. Cyrix is
not producing a "Faster" chip. They are producing an equivalent chip at a lower
clock speed.
:> >by that logic GM would have made the fastest coolest cars
:> >and NBC the best TV show...
:> >
:> >> Now if Cyrix chips weren't as reliable...now that would be logical.....
:> >> --
If you dont have a mobo thats blessed by cyrix, your cyrix chip will "skip a
beat" I've had it happen to me with my 5X86, every now and then there would be a
glitch of some sort that never happend with my AMD chip. Both chips had nice
cool heatsinks and fans as well.....
:> >
:> >Matt,
:> > Get a life matt. If you don't want a Cyrix don't buy one.
:> >Don't just repeat what you heard with no first hand knowledge
:> >or experience to share or backup your empty spew.
:> >
:> >jason
:
:
:Has anyone ever looked at the history of PC-MAG? A few years there was a
:review of printers. they bashed a particular epson model and raved about
:another brand (Okidata, I think). 2 Issues later, they trashed the other
:brand and sang the praises of the epson (both printers the exact same
:model). I have no proof, just a suspicion. I believe they rave about
:the one who advertises on more pages/with bigger ads. (that did seem to
:be the case with the printer review confusion).
:
:I no longer believe a word they print.
:
See, thats the problem with the market today, every brand has its own series of
lemons. (Except for Western Digital, for whom any of their stuff could be a
lemon at any time...)
:BTW, does anyone remember that INTEL had a problem with chips
:overheating?
Hell yeah. It happened with the P60/P66 chips alot. However these were shitty
5V chips for which at the time, the manufacturer of the pcnever even told you
the clock speed- It was just referred to as a "Pentium" system. Well, to be
perfectly honest though, any chip from a 486SX25 on up can overheat without at
least a good heatsink!
Zman
Peter F. Curran wrote:
>
> In article <322776...@zansiii.millersv.edu>,
> mu student <bo...@zansiii.millersv.edu> writes:
> >> I run it OVERCLOCKED without problems, but I use home-made cooling
> >> system.
> >> (I also used it for my previous Intel processors)
> >
> >What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
> >you describe your cooling system?
>
> If you overclocked that much you'd never get the initial bios prompt. In
> my experience 133% is the maximum speed you can usually get. Remember,
> when you overclock you are trying to run a chip at a speed that other chips
> have been able to run off which were made using the same manufacturing
> process. There has never been a 486DX 50, (A DX2 50 is a different chip),
>
> -Pete Curran
Intel did make a 486DX-50
I believe some had over heat problems, but I
know of a couple i built that are running fine.
Eric
Al Lanc
I'm also running a Cyrix 120MhZ running Unix on 24 hours/day and it
overheats every time I put the cover on -- so I just run it with the cover
off. Cyrix chips _DO_ overheat! But other than that, compatibility has been
no problem.
-Allan
Keith Lucas <sill...@wardrobe.demon.co.uk> writes:
>So you've not actually USED a Cyrix -- just read about them ?
>They do not overheat. Well they do if the fan breaks, but not
>otherwise. And I'm speaking from the experience of running a 120MHz
>one for the last couple of months -- I run Unix, so the thing is just
>left on all the time and I have had no problems with overheat. Nor
>have I had any reports of overheat from people I know in the hardware
>business who are routinely supplying 120+ Cyrixs all over the place.
>Maybe Ziff-Davis & co have been infiltrated by Wintel agents...
> -----------------------------------------------------------------------------
>"It's not a personality.. it's a bulldozer" sill...@wardrobe.demon.co.uk
> Current project: Computer wargaming's next generation...
> ------------------------------------------------------------------------------
--
----------------------------------------------------------------------
What George Washington did for us was to throw out the British, so
that we wouldn't have a fat, insensitive government running our
country. Nice try anyway, George.
n...@gwe486.cb.att.com
In article <50b92j$8...@usenet.rpi.edu>,
Peter F. Curran <cur...@remove.rpi.edu> wrote:
>In article <322776...@zansiii.millersv.edu>,
> mu student <bo...@zansiii.millersv.edu> writes:
>>> I run it OVERCLOCKED without problems, but I use home-made cooling
>>> system.
>>> (I also used it for my previous Intel processors)
>>
>>What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
>>you describe your cooling system?
>
>If you overclocked that much you'd never get the initial bios prompt. In
>my experience 133% is the maximum speed you can usually get. Remember,
>when you overclock you are trying to run a chip at a speed that other chips
>have been able to run off which were made using the same manufacturing
>process. There has never been a 486DX 50, (A DX2 50 is a different chip),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Damn! The one sitting behind me for the last few years seems very
substantial! The 50 MHz bus speed was not real popular with board makers,
because the difficulties in getting a 486 era bus to run reliably at that
speed. The 33 MHz bus speed was preferred, and so you could get DX2/66
boards cheaper than a DX/50. If your process is off-chip bandwidth bound
instead of in-cache CPU bound, the 50 outperforms the DX2/66. The type of
work I do involves tens of gigabytes of data being crunched with countless
mega-flops of processing, and so we use the DX50.
>so it probably wont work.
>
> -Pete Curran
I don't know if over clocking a DX33 to 50 MHz would work, but a DX50 works
very well.
---
Neil Kirby DoD# 0783 n...@gwe486.cb.lucent.com
Lucent Technologies - Home of Bell Labs Innovations
(Formerly AT&T) Bell Labs Columbus OH USA +1 (614) 860-5304
President Internet BMW Riders
The BMW R1100RSL - Red Bike, Blue Skies; Smiling Rider.
David Scott
On Mon, 02 Sep 1996 13:23:18 -0700, Zman <ea...@seanet.com> wrote:
I have Clocked an Intel 486 dx 33 Mhz at 71 mhz for 6 years...Still
runs without a problem and I DO NOT have a Heatsink.
Victor Healey
In article <322b5b1a...@news.atl.mindspring.com>,
dave...@atl.mindspring.com says...
What every one seems to forget is that the INTEL units HAVE THE
EXACT SAME PROBLEM! When the P-60 First came out I test 3 units.
pushed to the limit. Intel didn't do well with the pentium until
they abbandoned that process for the way they manufacture chips now.
Intel is reliable. Cyrix is...
Take a survey of reliable computer shops and most will cuss their
involment with Cyrix chips.
No I don't work for Intel.
--
Victor Healey 4-Phun Info Servers
Java Alert:
Malicious hackers have unleashed a brood of deadly "black widow"
Java applets on the Net.
HA5MRC
> The Illustrious mu student <bo...@zansiii.millersv.edu> wrote:
>
> >> I run it OVERCLOCKED without problems, but I use home-made cooling
> >> system.
> >> (I also used it for my previous Intel processors)
>
> >What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
> >you describe your cooling system?
OK. The drawing is on:
http://teto.sch.bme.hu/~hg5crs
(Sorry, there is no photo yet...)
The size of the fan is about 4 by 4 inches, 18 Watts (220V).
And made of metal, not plastic...
No heatsink, because it would be just thermal resistance at this
air speed. The only problem with this fan: the CPU will be covered by
dust very soon (and a little bit more noisy).
The only problem is: the voltage dropper still dissipates about 12
Watts,
(without forced cooling) so its temperature is still above 100 deg. C.
Have anyone seen the heatsink of a DEC Alpha APX?
> I would not encourage anyone to overclock anything. Stay within specs
> and you can count on support when it's needed. Go outside of specs,
> and you're on your own.
> My $.02 worth.
> Paul G.
Support? This means nothing for me, where I live, there is no support on
components.
Anyway, who can prove that a CPU went wrong just because of
overclocking?
Overclocking worth much more than $0.02 for me (exactly the price
difference).
There were much more crashes because of software made by xxx than
hardware failures.
My overclocking experience:
AMD 486 DX2-50 -> DX2-66 2 years (normal heatsink and fan)
AMD 486 DX2-66 -> DX2-80 1.5 years (extra fan, see http page)
Intel P75 -> I could not overclock it
Cyrix 6x86 120MHz -> 133 MHz 2 months.
73s again,
Sandor
ps: thanks for visiting the HA5KFU pages!
Chim Chim
On Mon, 02 Sep 1996 13:23:18 -0700, Zman <ea...@seanet.com> wrote:
>Peter F. Curran wrote:
>>
>> In article <322776...@zansiii.millersv.edu>,
>> mu student <bo...@zansiii.millersv.edu> writes:
>> >> I run it OVERCLOCKED without problems, but I use home-made cooling
>> >> system.
>> >> (I also used it for my previous Intel processors)
>> >
>> >What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
>> >you describe your cooling system?
>>
>> If you overclocked that much you'd never get the initial bios prompt. In
>> my experience 133% is the maximum speed you can usually get. Remember,
>> when you overclock you are trying to run a chip at a speed that other chips
>> have been able to run off which were made using the same manufacturing
>> process. There has never been a 486DX 50, (A DX2 50 is a different chip),
>> so it probably wont work.
>>
>> -Pete Curran
>
>Intel did make a 486DX-50
>I believe some had over heat problems, but I
>know of a couple i built that are running fine.
> Eric
That's what I heard too. I have a 486DX-50 [True DX 50, c/o Intel],
and I've never run into any problems with it overheating before...
Ever. Now, on the other hand, I have a 486 SLC/2 66, and that's
overheated a couple times... [If Pete had any truth to what he was
saying, I wouldn't have a computer right now]
Chim Chim
no_hope
I Worked Beta testing the Cyrix 120+ - 166+ and the betas worked, but
after production was released we got many call and returnes for
overheating problems, hundreds were returned, and they did require two
fans one high torque on the cpu and the other on the heatsink of the
regulator set at 3.6 volts,, i dont care what these mags says i saw it
a had to try to support it and back it up,,, i almost bought one,,,
On Mon, 26 Aug 1996 09:24:47 GMT, Keith Lucas
<sill...@wardrobe.demon.co.uk> wrote:
>In article <01bb92e3$02f40a80$5104b8cd@maktos>,
>Matt McDevitt <mak...@ix.netcom.com> wrote:
>>The chips manufactured by Cyrix....not necessarily the Cyrix processors
>>manufactured by IBM and others....are having some serious problems with
>>overheating. The average Cyrix computer, as a matter of fact, has at least
>>2 fans (running about 3X as fast as an Intel fan) and 2 heat sinks...the
>>heat sinks had better be ceramic or they'd melt! I heard this from
>>WWW.ZDNET.COM and other places. I didn't know they ran that hot..but they
>>do. Just thought I'd save some of you the 150 you'd spend on a Cyrix
>>processor (to save money) and end up getting an Intel IN ADDITION TO the
>>Cryix, after it caught on fire or just burned up :)
>
John Nash
On 1 Sep 1996 06:17:52 GMT, Je...@unknown.address.com (K.Y. Jerry)
wrote:
>How do you make a "home-made" extra cooling system for a CPU?
>
Well first you start with a bucket of ice cold water....
Winter
Mr.P (m...@trance.pei.edu) wrote:
: On 1 Sep 1996 06:05:07 GMT, cur...@remove.rpi.edu (Peter F. Curran): >>What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
: >If you overclocked that much you'd never get the initial bios prompt. In
: >my experience 133% is the maximum speed you can usually get. Remember,
: >when you overclock you are trying to run a chip at a speed that other chips
: >have been able to run off which were made using the same manufacturing
: >process. There has never been a 486DX 50, (A DX2 50 is a different chip),
: >so it probably wont work.
: There has been a 486 DX 50. No DX 66, but a DX 50 has existed.
True. It had problems using the VLB bus, which was (and is) rated for only
33 MHz (40 if its buffered, which i had with my DX2-80 and DX4-120).
Mike
Francis Tang
The Postman wrote:
>
> On Sun, 01 Sep 1996 10:28:02 -0700, Gene Irwin <gir...@frugal.com> wrote:
>
> :ji...@notintel.com wrote:
> :>
> :> I have a cyrix p166+ and its never overheated, in fact it runs faster
> :> than shit (which would be intel). I can vouch, for anyone thinking
> :> about a cyrix, there great!
> :>
> Yeah, lemme see how well the 6X86-166 stacks against a REAL P5-166 in all tests-
> sure the Pcbench tests come out a little better, but I dont see them running
> Spec95 or Spec92 on these chips- Cyrix only compares them in Application power
> and not in raw computing power. Any performance gain is not over all areas,
> just "Windows 95" machines. Anyone who is going to run a UNIX OS or Windows NT
> is not going to want a cyrix chip- on NT the cyrix units suffered a huge
> performance hit. I've tried a Cyrix-5x86-120 and compared it against a
> DX4-120(AMD) and found that although its application speed was a little faster,
> the RAW CPU SPEED tests all indicated that it was slower, even with the branch
> prediction turned on- The only tester that said that it was faster was Landmark
> 6.0- and that's because the special branch thingie in the cyrix chip fools the
> crap out of landmark tester, because landmark uses the same calc scheme over and
> over again. Cyrix chips are fine for applications, but they still don't offer
> more chip for the money.
>
Excuse me, what ARE benchmarks? (No, I'm not a newbie - it's a
rhetorical question.) CPU Benchmarks are designed to compare the speed
of CPU's which are ultimately designed for APPLICATIONS. It's all very
well having a chip that excels at benchmarks but what use is it if it's
crap at the application that you use? Yes, I do admit that benchmarks
are supposed to give an accurate indication of application-crunchability
but the truth is that this is not always the case: different CPUs may be
good at different applications, and so there can be no ONE
benchmark-number by which you can compare all chips for all
applications.
By the way, what exactly do you mean by "more chip for the money"?
Personally, I'd understand that as "the ability to run my applications
faster for less money", and it would be different for different people
who use different applications.
Frank.
Francis Tang
Mr.P wrote:
>
> On 1 Sep 1996 06:05:07 GMT, cur...@remove.rpi.edu (Peter F. Curran)
> wrote:
> >In article <322776...@zansiii.millersv.edu>,
> >>> I run it OVERCLOCKED without problems, but I use home-made cooling
> >>> system.
> >>> (I also used it for my previous Intel processors)
> >>you describe your cooling system?
> >If you overclocked that much you'd never get the initial bios prompt. In
> >my experience 133% is the maximum speed you can usually get. Remember,
> >when you overclock you are trying to run a chip at a speed that other chips
> >have been able to run off which were made using the same manufacturing
> >process. There has never been a 486DX 50, (A DX2 50 is a different chip),
> >so it probably wont work.
>
> There has been a 486 DX 50. No DX 66, but a DX 50 has existed.
Yes they do exist. There's one in a Netware 3.11 server at my old
place-of-work.
Elliott Oti
Francis Tang wrote:
>
> Excuse me, what ARE benchmarks? (No, I'm not a newbie - it's a
> rhetorical question.) CPU Benchmarks are designed to compare the speed
> of CPU's which are ultimately designed for APPLICATIONS. It's all very
> well having a chip that excels at benchmarks but what use is it if it's
> crap at the application that you use? Yes, I do admit that benchmarks
> are supposed to give an accurate indication of application-crunchability
> but the truth is that this is not always the case: different CPUs may be
> good at different applications, and so there can be no ONE
> benchmark-number by which you can compare all chips for all
> applications.
While this is a good point: benchmark do not always accurately reflect
how well a chip will perform in the "real" world; may I point out
that the performance of most apps running across different CPU's
depends more on the quality of the programmer and compiler targeted
for that particular CPU than the CPU as such. The 80X86 has an appalling
dearth of general registers, and a wierd mix of opcodes, but is the
most common CPU in use and programmers have learned over the years to
squeeze an amazing amount of performance out of it. There are probably
more man-hours put into optimizing 80x86 code worldwide than all other
CPU's put together. This tends to mask or even reverse differences in
performance.
Elliott
Scott
beavis...
care to e-mail me some info on overclocking ...I have a 486 dx-4 100
Thanks...
Spirit
Follow-up to quin...@primenet.com (no_hope), who wrote:
> I Worked Beta testing the Cyrix 120+ - 166+ and the betas worked, but
> after production was released we got many call and returnes for
> overheating problems, hundreds were returned, and they did require two
> fans one high torque on the cpu and the other on the heatsink of the
> regulator set at 3.6 volts,, i dont care what these mags says i saw it
> a had to try to support it and back it up,,, i almost bought one,,,
>
I am using one now, a 5x86/100 clocked to 120 and it runs perfectly
WITHOUT OVERHEATING!
I suggest you try a latest one before you continue to condemn them.
Spirit
Man is but an entity of his own choosing.
But oh, how we would all be gods if we could so choose.
Compuserve: 100636,3581 Email:wie...@kentnet.co.uk
Statto
On Fri, 30 Aug 1996 15:25:55 GMT, catf...@iofc.com (Terry Martin)
wrote:
> Well you are not very versed are you or else you would know the
>FASTEST chip on the market right now is the CYRIX P200+ so knock off
>the ingnorant shit (oh yea NT sucks too only a complete moron would
>purchase it)
>
>Words of Wisdom or my .02 worth
I take it you've never heard of the Pentium Pro then????
It leaves the Cyrix P200+ standing.
My daughter's Cyrix P200+ with 16mb ram and Win95 is blisteringly fast
but next to my Pentium Pro 200 with 96mb ram and Win NT4, it is dead
slow.
You obviously think you know everything there is to know but really
you know FUCK ALL...
Statto
Archmage
>Has anyone ever looked at the history of PC-MAG? A few years there
was a
>review of printers. they bashed a particular epson model and raved
about
>another brand (Okidata, I think). 2 Issues later, they trashed the
other
>brand and sang the praises of the epson (both printers the exact same
>model). I have no proof, just a suspicion. I believe they rave about
>the one who advertises on more pages/with bigger ads. (that did seem
to
>be the case with the printer review confusion).
>
>I no longer believe a word they print.
>
>BTW, does anyone remember that INTEL had a problem with chips
>overheating?
>
>Food for thought.
Hear hear! About time someone brought that up...Personally, I'm
suspect of any publication that accepts advertising...did anyone notice
that nice big, shiny Intel pullout in that issue? =) Naw, couldn't be
biased at all.....I've used Cyrix for years, no problems, and my order
is in for a 6x86 cpu....
Paul Shirley
In article <3232d6dd...@news.rapid.co.uk>, Statto
<da...@rapid.co.uk> writes
>My daughter's Cyrix P200+ with 16mb ram and Win95 is blisteringly fast
>but next to my Pentium Pro 200 with 96mb ram and Win NT4, it is dead
>slow.
My home P75 with 40mb ram, 0 cache runs compilers, Netscape and most
other stuff I use faster than the 16 meg P100 at work. Any speed
comparisons using Win95 with <24Meg of ram are *invalid*.
...and you forgot to tell us how fast the P6 ran Win95 compared to the
Cyrix... but of course that would give you the 'wrong' result wouldn't
it?
>You obviously think you know everything there is to know but really
>you know FUCK ALL...
...and he's still smarter than you.
--
Paul Shirley
Matt McDevitt
People are going to buy what they like....I don't know yet conclusively if
Cyrix chips overheat or not and I was the original creator of this thread!
--
Matt McDevitt
mak...@ix.netcom.com
"Can't we all just get along!"
Matt Gischer
Statto (da...@rapid.co.uk) wrote:
: On Fri, 30 Aug 1996 15:25:55 GMT, catf...@iofc.com (Terry Martin)
: wrote:
:
: > Well you are not very versed are you or else you would know the
: >FASTEST chip on the market right now is the CYRIX P200+ so knock off
: >the ingnorant shit (oh yea NT sucks too only a complete moron would
: >purchase it)
: >
: >Words of Wisdom or my .02 worth
:
:
: I take it you've never heard of the Pentium Pro then????
: It leaves the Cyrix P200+ standing.
: My daughter's Cyrix P200+ with 16mb ram and Win95 is blisteringly fast
: but next to my Pentium Pro 200 with 96mb ram and Win NT4, it is dead
: slow.
I'm sure having 6 times as much RAM has nothing to do with it.
Tony Doimeadios
your Cyrix CPU with ZERO cache cannot run faster than an Intel CPU..
period. That was an attempt to sell cheap computers to people who want
the CPU instructions but didnt care about performance... although they
never said that. If anyone goes in and disables their internal cache,
the system creeps to a halt! The power behind 32-bit cpu's in the
Pentium class and beyond relies heavily on having instructions force-fed
down it's pipe - thus the need for cache for high-performance systems.
I do a LOT of 3D animation work and the FPU in the Cyrix brand CPU's just
cannot perform to the standards of the Intel chips. Cyrix re-organized
the instruction order so that its chips favored integer math and
sacraficed floating point math. normally not bad, but all of your
high-end apps MUST have a powerful FPU.
The Cyrix is optimized for 16-bit instructions - which is what dominates
the market right now. Most of my existing apps are 16-bit, but I'm
buying all new, 32-bit code... The lesson: Cyrix is a chip for low-end,
$$$ saving buyers who use mainly 16-bit code. I make my living using
32-bit code, so I use an Intel Pentium with a 32-bit OS (Win NT 3.51).
-tony
Paul Shirley (Pa...@chocolat.foobar.co.uk) wrote:
: My home P75 with 40mb ram, 0 cache runs compilers, Netscape and most
: other stuff I use faster than the 16 meg P100 at work. Any speed
: comparisons using Win95 with <24Meg of ram are *invalid*.
:
: ...and you forgot to tell us how fast the P6 ran Win95 compared to the
: Cyrix... but of course that would give you the 'wrong' result wouldn't
: it?
: ...and he's still smarter than you.
:
:
: --
: Paul Shirley
Paul Garceau
The Illustrious "Matt McDevitt" <mak...@ix.netcom.com> wrote:
>People are going to buy what they like....I don't know yet conclusively if
>Cyrix chips overheat or not and I was the original creator of this thread!
There's been a pretty comprehensive discussion about it to me. Or so
it would seem.
If you started this thread with "Cyrix chips overheat!" then you had
already decided that they do.
Many people went on to proove that Cyrix chips do not overheat if they
have the proper cooling (whether over-clocked or not it seems). Very
few people said that Cyrix 6x86 does. Most people agreed that the
Cyrix was no more prone to overheating than any other chip.
Bottom line is, Cyrix, like Intel or any other chip will not overheat
if properly cooled.
Paul G.
Nothing real can be threatened.
Nothing unreal exists.
Marco Ramius@Krazny.Oktiabr
David Pugmire <ner...@inconnect.com> wrote:
>
>
>> catf...@iofc.com (Terry Martin) wrote:
>>
>>snip. snip.
I use cyrix 5x86 120 with no problem. You wizards have loose screw
from too much magazine ads.
I don't use fan. Can touch heat sink while running burn-in algorithym
with no pain. Can you do same with original pentium from Intel?
Could fry chicken at 55 mhz. You use wrong VOLTAGE is all your
problem. Chip runs at 3.3 - 3.45 volts not 5 like 486 and old
pentium. You need reconfigure mainboard is all.
Americans are not so smart wizards eh?
P.S. PC magazine is for children and housewives running home greeting
card buisness. Not for wizards. Read byte, much better.
Dosvedanya.
"Marco Ramius, Capitain, Krazny Octiabr, Red Banner, Northern Fleet".
Joe Husosky
I resell PC's for a living and have never had any problems with Cyrix
686 chips either during setup or after the sale. But now 3 different
suppliers have now discontinued selling Cyrix chips due to overly high
RMAs on them.
Vishnu Verheer
>: >>What would happen if I over clocked a 486 DX 33 at 50 MHZ? also could
>: >If you overclocked that much you'd never get the initial bios prompt. In
>: >my experience 133% is the maximum speed you can usually get. Remember,
>: >when you overclock you are trying to run a chip at a speed that other
chips
>: >have been able to run off which were made using the same manufacturing
>: >process. There has never been a 486DX 50, (A DX2 50 is a different chip),
>: >so it probably wont work.
>:
>: There has been a 486 DX 50. No DX 66, but a DX 50 has existed.
>
>33 MHz (40 if its buffered, which i had with my DX2-80 and DX4-120).
Yes, a 50 Mhz chip exists. I have one (or HAD) in my Compaq Deskpro.
It ran at truly 50Mhz on the EISA bus. They stopped making these though.
This was one of the original 486/50's. Not a DX2.
Christopher LaSota
da...@rapid.co.uk (Statto) wrote:
>On Fri, 30 Aug 1996 15:25:55 GMT, catf...@iofc.com (Terry Martin)
>wrote:
>> Well you are not very versed are you or else you would know the
>>FASTEST chip on the market right now is the CYRIX P200+ so knock off
>>the ingnorant shit (oh yea NT sucks too only a complete moron would
>>purchase it)
>>
>>Words of Wisdom or my .02 worth
>I take it you've never heard of the Pentium Pro then????
>It leaves the Cyrix P200+ standing.
>My daughter's Cyrix P200+ with 16mb ram and Win95 is blisteringly fast
>but next to my Pentium Pro 200 with 96mb ram and Win NT4, it is dead
>slow.
>You obviously think you know everything there is to know but really
>you know FUCK ALL...
>Statto
Thats right, my 386SX 20 is a speed monster compared to my P-Pro 250
with 64 megs of ram.
Vishnu Verheer
t...@visi.net
san...@winternet.com (Tony Doimeadios) wrote:
>-tony
tony, did anyone tell you that you are stupid? I run a 686-150+
with os/2, I guess that is a 16 bit monster ehh? I can run circles
around the intel chip that was in here, a p150, same apps, same setup
down to the last chip, just swirched out the chips on a tyan mother
board, wow, BIG diffrence in preformance. NO overheating, it is left
on all the time. You people arguing over the fpu are idiots also. The
FPU in cyrix chips are 80 bits, that is why they have trouble, they
are a little slower, note I said, a little slower, than the Intel. If
the software writers were to write around this MUCH superior FPU, we
would all dance around the headquarters of cyrix. I have used cyrix
products from the 486-66 to the 586/120 , which by the way was an
awesome little monster, and now on to the 686/150 and I am waiting on
the 200 mhz chip to come down in price a little. not a damn lick of
trouble.
David Piette
It sounds like you guys know about software but are fairly ignorant about
hardware. The base prosessing units in the Pentium and the Cyrix 6x86 chips
are 128bit, NOT 32 BIT, 32 bit prossors havn't been around since the stone
age. As for the cache you were talking about, the Cyrix chips have a huge
amount of cache intigrated right into the chip, this is why the Cyrix CPU
benchmarks faster than a Pentium running at the same clock speed. The Cyrix
CPU's also have better internal bus structure which alows the Cyrix to do
more work for every clock stroke.
Paul Garceau
Out of curiousity. What defines whether RMAs are high or not? Intel?
I don't know what they are exactly (RMAs). Even so, logic can still
be useful.
Unknown
pa...@teleport.com (Paul Garceau) wrote:
> Paul G.
RMA's are returned merchandise. Logic is useful. 3 suppliers out of 4
or 40,000? A friend of mine sells systems using both Intel 586 and
Cyrix 686 chips, he sets the motherboards up correctly and has had
little or no probs with either cpu's. His Cyrix supplier has
mentioned that unscrupulous suppliers strip the correct cooling fans
from the cpu's to sell seperately elsewhere for $15 and replace them
with generics which cost them $2.00. This practice has caused a lot of
problems. Also motherboard voltage must be correct, lots of people
including techs screw up there.
The new VX Triton 2 boards work great with both Intel and Cyrx. I
have the 686-150 and have had no problems with any of my applications
(Word 7, Win 95, VistaPro 3.0, Photoshop, Excel, etc.) or games
(Duke3d, Quake, Doom2, Descent 2,Warcraft2, Sierra miscellaneous) For
scanning and digital imaging and printing at work I plan to build a
system using the Cyrix 686-200mz as its cpu.
If there is anything else to watch out for, watch out for bad RAM,
there is a ton of it out there. A lot are mislabeled/rescreened/etched
as normal RAM and is really EDO and VISA VERSA. This will give you
GPF's, fpu errors, memory failures, and a ton of other probs.
Vishnu Verheer
Paul Garceau
The Illustrious (Dangerous Dee) wrote:
>>pa...@teleport.com (Paul Garceau) wrote:
>> Out of curiousity. What defines whether RMAs are high or not? Intel?
>> I don't know what they are exactly (RMAs). Even so, logic can still
>>be useful.
>RMA's are returned merchandise. Logic is useful. 3 suppliers out of 4
>or 40,000? A friend of mine sells systems using both Intel 586 and
>Cyrix 686 chips, he sets the motherboards up correctly and has had
>little or no probs with either cpu's. His Cyrix supplier has
>mentioned that unscrupulous suppliers strip the correct cooling fans
>from the cpu's to sell seperately elsewhere for $15 and replace them
>with generics which cost them $2.00. This practice has caused a lot of
>problems. Also motherboard voltage must be correct, lots of people
>including techs screw up there.
A paranoid person might think Intel had a hand in this, if Intel felt
threatened by Cyrix...but only a paranoid person would think so. It's
sad to see that there are so many people trying to make an
unscrupulous buck.
> The new VX Triton 2 boards work great with both Intel and Cyrx. I
>have the 686-150 and have had no problems with any of my applications
>(Word 7, Win 95, VistaPro 3.0, Photoshop, Excel, etc.) or games
>(Duke3d, Quake, Doom2, Descent 2,Warcraft2, Sierra miscellaneous) For
>scanning and digital imaging and printing at work I plan to build a
>system using the Cyrix 686-200mz as its cpu.
I remember reading about a Cyrix 6x86/166 kit that was available for
around $850 (Tiger, out of Miami, FL), including box back in July
(PC-Magazine). Now, the 6x86/200 from the same company is less than
$760. They think that they'll have the 6x86/200 towards the end of
the month.
Is anyone aware of any other kits?
> If there is anything else to watch out for, watch out for bad RAM,
>there is a ton of it out there. A lot are mislabeled/rescreened/etched
>as normal RAM and is really EDO and VISA VERSA. This will give you
>GPF's, fpu errors, memory failures, and a ton of other probs.
Good advice. Just because memory is cheap does not mean it is 'good'
or 'reliable'.