Intel Celeron - The New Gaming CPU of Choice, Overclocker's Dream
Victor Healey
TIP
Run - don't walk to
http://www.tomshardware.com/cpubench.html
Look for section under updated chip family review.
The question I have is can you overclock an Intel Celeron using an ABIT LX
board?
--
Vic Healey ki4je
winndixie....@bellsouth.net
ICQ 5148591
Andrew
Why would you want a P-II without its level 2 cache.??????
Its like buying a Porsche and putting a Ford engine in it.
Andrew Wood
Senior Technical Specialist
Fujitsu (ICL) Ltd
Guldan
You won't want one... but I suggest that the performance of a PII without
'onboard' L2 but with L2 on the mainboard like with the Pentiums, will
outdo a Pentium, and it will probably be cheap as hell! I like those odds!
Andrew <andre...@cableol.co.uk> wrote in article
<6h7bm6$5...@news5-gui.server.cableol.net>...
Samuel Stoddard
Andrew (andre...@cableol.co.uk) wrote:
: Why would you want a P-II without its level 2 cache.??????
: Its like buying a Porsche and putting a Ford engine in it.
: Andrew Wood
: Senior Technical Specialist
: Fujitsu (ICL) Ltd
The big reason is the overclockablility. 400mhz w/ no problems! Hell yea
(ofcorse, I doubt this chip will be easy to find for upgraders for a
while, as it is OEM bait for a while)
David
On 17 Apr 1998 08:44:03 -0400, ssto...@freenet.columbus.oh.us (Samuel
Stoddard) wrote:
>: Why would you want a P-II without its level 2 cache.??????
>
>The big reason is the overclockablility. 400mhz w/ no problems! Hell yea
And the P2 isn't overclockable???
Antek S. Baranski
Andrew wrote:
> Why would you want a P-II without its level 2 cache.??????
How about the price and speed? Have a look at
http://sysdoc.pair.com/cpubench.html and look at the GAMES results with
a Celeron running at 400 MHz (OVER CLOCKED!!!!) A snippet from Tom's
Hardware page:
AOpen AX6B BX motherboard
Corsair PC100 64 MB SDRAM
Intel Graphic Express AGP Graphics Card with i740 chip
3Dfx Voodoo2 12 MB reference board, driver from April 8, 1998
Quantum Fireball SE 4.3 HDD using UDMA and Intel bus master driver 3.01
Quake II 3.14 at 640x480, 3Dfx GL driver
Quake II demo1.dm2
Pentium II 400/100 (Deschutes) 85.8 fps
Pentium II 350/100 (Deschutes) 80.1 fps
Celeron 400/100 (OVERCLOCKED) 79.1 fps <<==== !!!!! And if the price is
as low as INTeL says it will be then why not buy a Celeron. :-)
> Its like buying a Porsche and putting a Ford engine in it.
A GT40 engine any day. ;-)
>
> Andrew Wood
> Senior Technical Specialist
> Fujitsu (ICL) Ltd
Cheers,
--
ing. Antek S. Baranski
European Laboratory for Particle Physics (CERN) - IT division - ASD
group
Tel: +41 - (0)22 - 76 742 47 Fax: +41 - (0)22 - 76 786 30 ICQ:
1127269
Wayne Ernst
The Celeron is available for around $185 and up.
Check Pricewatch. 4*113? Anyone? I'm sure
before too long, it will be priced near it's (in quan.)
price of $156, or so - making it a disposable chip
should it sizzle.
Samuel Stoddard <ssto...@freenet.columbus.oh.us> wrote in article
<6h7iqj$7...@login.freenet.columbus.oh.us>...
> Andrew (andre...@cableol.co.uk) wrote:
> : Why would you want a P-II without its level 2 cache.??????
>
> : Its like buying a Porsche and putting a Ford engine in it.
>
> : Andrew Wood
> : Senior Technical Specialist
> : Fujitsu (ICL) Ltd
>
> The big reason is the overclockablility. 400mhz w/ no problems! Hell yea
Wayne Ernst
How come just a year and a half ago, we were
getting screwed over by motherboard manufacturers
who were putting 'fake' L2 cache on motherboards.
At the time, it was totally unacceptable to us. Now,
Intel releases a chip without L2 cache, people realize
that it's good at one game, it's overclockable - and it's
now very acceptable.
Antek S. Baranski <Antek.Baranski@"euro.lab.for.part.phys"cern.ch> wrote in
article <353748F6...@cern.ch>...
>
>
> Andrew wrote:
>
> > Why would you want a P-II without its level 2 cache.??????
>
> How about the price and speed? Have a look at
> http://sysdoc.pair.com/cpubench.html and look at the GAMES results with
> a Celeron running at 400 MHz (OVER CLOCKED!!!!) A snippet from Tom's
> Hardware page:
>
> AOpen AX6B BX motherboard
> Corsair PC100 64 MB SDRAM
> Intel Graphic Express AGP Graphics Card with i740 chip
> 3Dfx Voodoo2 12 MB reference board, driver from April 8, 1998
> Quantum Fireball SE 4.3 HDD using UDMA and Intel bus master driver 3.01
>
> Quake II 3.14 at 640x480, 3Dfx GL driver
>
> Quake II demo1.dm2
>
> Pentium II 400/100 (Deschutes) 85.8 fps
> Pentium II 350/100 (Deschutes) 80.1 fps
> Celeron 400/100 (OVERCLOCKED) 79.1 fps <<==== !!!!! And if the price is
> as low as INTeL says it will be then why not buy a Celeron. :-)
>
> > Its like buying a Porsche and putting a Ford engine in it.
>
> A GT40 engine any day. ;-)
>
> >
> > Andrew Wood
> > Senior Technical Specialist
> > Fujitsu (ICL) Ltd
>
Antek S. Baranski
David wrote:
> On 17 Apr 1998 08:44:03 -0400, ssto...@freenet.columbus.oh.us (Samuel
> Stoddard) wrote:
>
> >: Why would you want a P-II without its level 2 cache.??????
> >
> >The big reason is the overclockablility. 400mhz w/ no problems! Hell yea
>
> And the P2 isn't overclockable???
Well it is I have run my 266 at 300 for several weeks running the RC5-64 key
generator day and night (100% CPU load) and its just fine. The problem is not
the CPU CORE which is the same but the L2 cache-chips used. They aren't too
good at keeping up with the higher clock, example 400 MHz would mean 200 MHz
clock on the cache.... don't think that will work for a 266 or even a 333 P2.
Cheers,
--
ing. Antek S. Baranski
European Laboratory for Particle Physics (CERN) - IT division - ASD group
Tel: +41 - (0)22 - 76 742 47 Fax: +41 - (0)22 - 76 786 30 ICQ:
1127269
REMOVE THE "bla bla" to-reply
Luis Constantino
Hmm,
That article is interesting. But look at it again... If you will then you
will see that in almost every benchmark, the old P2 333 and 300 are around 3 or
4th. and that's with them NOT overclocked. So don't throw away your memory,
CPU's and LX6's yet... What I'd Love to see would be a descutes 333
overclocked to(83*5) 416 in there.... I've got a funny feeling it'd top
everything there no?
As far as the BX/Celeron combo, it has the potential to be a cheap
alternative, IF the PC100 ram prices drop....
Just my .02
Lue
Mike Tiller
"Andrew" <andre...@cableol.co.uk> writes:
> Why would you want a P-II without its level 2 cache.??????
>
> Its like buying a Porsche and putting a Ford engine in it.
Hey! Ford makes Jaguars and Mustang Cobras :-)
> Andrew Wood
> Senior Technical Specialist
> Fujitsu (ICL) Ltd
--
Michael Tiller
*Ford* Motor Company
Antek S. Baranski
Hi Wayne,
Wayne Ernst wrote:
> How come just a year and a half ago, we were
> getting screwed over by motherboard manufacturers
> who were putting 'fake' L2 cache on motherboards.
Because the _old_ Socket 7 CPUs needed the cache to get a decent performance.
> At the time, it was totally unacceptable to us. Now,
> Intel releases a chip without L2 cache, people realize
> that it's good at one game, it's overclockable - and it's
> now very acceptable.
Why is it acceptable? Well because you get what you need, the Celeron will give
you the best performance game-wise for the money. $185 for the Celeron is the
same price you would pay for a AMD K6-266 and the Celeron just blast the AMD
out of the water, and you don't need to overclock it for that. (If you do
overclock it you get into game speed heaven so to speak :-)
But before people all jump up and say HEY I need that thing!!!! They should
remember this:
1) The AMD K6 will drop into the existing motherboard and providing your board
has the right jumpers or Soft Menu options it will run at 266 or more.
2) The Celeron needs a SLOT 1 motherboard.
2a) An EX-chipset board **SUCKS MAJOR** but very cheap! (Castrated BX chipset)
2b) A BX-chipset board (the only choice I think) and not that much more
expensive than an EX-board.
3) The Celeron needs SDRAM, but not just any SDRAM it needs QUALITY SDRAM to
run at 100 MHz.
For more information have a look at:
http://sysdoc.pair.com/IntelsAprilBlessings.html
T. Bull
Then again....who is to say that a motherboard company might not go ahead
and put an L2 on the board just for the Celeron. Asus wasn't supposed to
put an 83 Mhz bus speed on an Intel HX board....but they did.
At 400 Mhz a celeron is getting better than K6300 benchmark scores and as
far as games are concerned ....simply put....it is a gamers (Quakers) CPU.
For the price (because I'm a budget-minded person----read Cheap
Bastard)....I'll get one now until the Katmai.
Chuck
Wayne Ernst wrote in message <01bd6a04$c8a96500$7a8345c6@default>...
>How come just a year and a half ago, we were
>getting screwed over by motherboard manufacturers
>who were putting 'fake' L2 cache on motherboards.
>At the time, it was totally unacceptable to us. Now,
>Intel releases a chip without L2 cache, people realize
>that it's good at one game, it's overclockable - and it's
>now very acceptable.
>
>Antek S. Baranski <Antek.Baranski@"euro.lab.for.part.phys"cern.ch> wrote in
>article <353748F6...@cern.ch>...
>>
>>
>> Andrew wrote:
>>
>> > Why would you want a P-II without its level 2 cache.??????
>>
>> How about the price and speed? Have a look at
>> http://sysdoc.pair.com/cpubench.html and look at the GAMES results with
>> a Celeron running at 400 MHz (OVER CLOCKED!!!!) A snippet from Tom's
>> Hardware page:
>>
>> AOpen AX6B BX motherboard
>> Corsair PC100 64 MB SDRAM
>> Intel Graphic Express AGP Graphics Card with i740 chip
>> 3Dfx Voodoo2 12 MB reference board, driver from April 8, 1998
>> Quantum Fireball SE 4.3 HDD using UDMA and Intel bus master driver 3.01
>>
>> Quake II 3.14 at 640x480, 3Dfx GL driver
>>
>> Quake II demo1.dm2
>>
>> Pentium II 400/100 (Deschutes) 85.8 fps
>> Pentium II 350/100 (Deschutes) 80.1 fps
>> Celeron 400/100 (OVERCLOCKED) 79.1 fps <<==== !!!!! And if the price is
>> as low as INTeL says it will be then why not buy a Celeron. :-)
>>
>> > Its like buying a Porsche and putting a Ford engine in it.
>>
>> A GT40 engine any day. ;-)
>>
>> >
>> > Andrew Wood
>> > Senior Technical Specialist
>> > Fujitsu (ICL) Ltd
>>
Antek S. Baranski
Luis Constantino wrote:
> Hmm,
> That article is interesting. But look at it again... If you will then you
> will see that in almost every benchmark, the old P2 333 and 300 are around 3 or
P2 333 & 300 Old? If that's old than I must be missing soemthing....... :-))
> 4th. and that's with them NOT overclocked. So don't throw away your memory,
> CPU's and LX6's yet... What I'd Love to see would be a descutes 333
> overclocked to(83*5) 416 in there.... I've got a funny feeling it'd top
> everything there no?
I don't think so because the L2 cache-ships won't work on 208 MHz! :-(( As you said
have a look at the article again and you will say that Tom's says that the problem
with overclocking the PII is not the CPU core but the chips used for the L2 cache.
:-((
> As far as the BX/Celeron combo, it has the potential to be a cheap
> alternative, IF the PC100 ram prices drop....
Exactly the only problem with the BX/Celeron combo is the price you have to pay for
memory that *will* actually work in the system, which is quite important, I think
:-))
Next thing you will see is: "SDRAM certified to work in your BX/Celeron 400 MHz
machine". hehehehehe ;-)
> Just my .02
>
> Lue
As to the original question can you overclock a Celeron in a LX6 board yes you can,
just set the clock to 75 or 83 Mhz (if you memory will work with it) and x4 will
give you some nice speed. If you can go beyond x4 say 4.5x or even 5x then you will
get even more interesting speeds.... ;-))
Cheers,
--
ing. Antek S. Baranski
European Laboratory for Particle Physics (CERN) - IT division - ASD group
Tel: +41 - (0)22 - 76 742 47 Fax: +41 - (0)22 - 76 786 30 ICQ: 1127269
mark_j...@hotmail.com
In article <01bd6a61$ea47fc00$0100...@raytodd.vianet.net.au>,
"Guldan" <ray...@vianet.net.au> wrote:
>
> You won't want one... but I suggest that the performance of a PII without
> 'onboard' L2 but with L2 on the mainboard like with the Pentiums, will
> outdo a Pentium, and it will probably be cheap as hell! I like those odds!
>
Well, If you are looking for cheap. You can Get a Pentium Pro 166 with 512K
Cache, overclock it to 233, all for about $300. (Thats including the new
Motherboard and the Peltier Cooling Fan)
My little system like that just blew away a comparably equiped Pentium II 300
in some performance test we ran.
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
Zorak
But when your objective is to pick up chicks, the Porsche with the Ford
engine will still work as good!
Andrew wrote in message <6h7bm6$5...@news5-gui.server.cableol.net>...
>Why would you want a P-II without its level 2 cache.??????
>
>Its like buying a Porsche and putting a Ford engine in it.
>
Samuel Stoddard
Wayne Ernst (wayne...@pressroom.com) wrote:
: How come just a year and a half ago, we were
: getting screwed over by motherboard manufacturers
: who were putting 'fake' L2 cache on motherboards.
: At the time, it was totally unacceptable to us. Now,
: Intel releases a chip without L2 cache, people realize
: that it's good at one game, it's overclockable - and it's
: now very acceptable.
This is true. I have heard of the horrid D3D and busniess app benchmarks,
because of the lack of a L2 cache (well, maybe horrid isnt the correct
word, but atleast not that great). The only REAL advantage is the fact
taht the FPU, what Quake-based games rely on for their speed, is the same
speed. Notice that Tom said it was ideal for a "Quake Station", but I
doubt it will run many other games as well as Quake (but there are a lot
of Quake-ish games comming out soon, which will probibly make full use of
the FPU)
: Antek S. Baranski <Antek.Baranski@"euro.lab.for.part.phys"cern.ch> wrote in
: article <353748F6...@cern.ch>...
: >
: >
: > Andrew wrote:
: >
: > > Why would you want a P-II without its level 2 cache.??????
: >
: > How about the price and speed? Have a look at
: > http://sysdoc.pair.com/cpubench.html and look at the GAMES results with
: > a Celeron running at 400 MHz (OVER CLOCKED!!!!) A snippet from Tom's
: > Hardware page:
: >
: > AOpen AX6B BX motherboard
: > Corsair PC100 64 MB SDRAM
: > Intel Graphic Express AGP Graphics Card with i740 chip
: > 3Dfx Voodoo2 12 MB reference board, driver from April 8, 1998
: > Quantum Fireball SE 4.3 HDD using UDMA and Intel bus master driver 3.01
: >
: > Quake II 3.14 at 640x480, 3Dfx GL driver
: >
: > Quake II demo1.dm2
: >
: > Pentium II 400/100 (Deschutes) 85.8 fps
: > Pentium II 350/100 (Deschutes) 80.1 fps
: > Celeron 400/100 (OVERCLOCKED) 79.1 fps <<==== !!!!! And if the price is
: > as low as INTeL says it will be then why not buy a Celeron. :-)
: >
: > > Its like buying a Porsche and putting a Ford engine in it.
: >
: > A GT40 engine any day. ;-)
: >
: > >
: > > Andrew Wood
: > > Senior Technical Specialist
: > > Fujitsu (ICL) Ltd
: >
: > Cheers,
: >
: >
: >
The Bombigator's Grandkid
In article <01bd6a03$c3e73e80$7a8345c6@default>, "Wayne Ernst" <wayne...@pressroom.com> wrote:
>The Celeron is available for around $185 and up.
>Check Pricewatch. 4*113? Anyone? I'm sure
>before too long, it will be priced near it's (in quan.)
>price of $156, or so - making it a disposable chip
>should it sizzle.
>
I'm not an Intel basher. I use Intel products at work and am very happy with
them. I own all three different colors of the Intel Bunny Suit dolls.
I've also run Cyrix, Intel and AMD chips at home. All three companies have
produced products that impressed me at different times.
But the Celeron is a BAD technical idea. Its just a marketing scheme to push
Slot I machines out into the sub-$1000 PC world. Without an L2 cache, I
don't care how fast the chip run's internally, it won't be able to efficently
run modern programs, whether games or business apps (or Operating systems).
Its probably the worst idea Intel has had since they disabled the internal
math co-processor on the 486DX and sold the resulting lobotomized monstrosity
as the 486SX.
If anyone wants verification of what I'm saying, try this experiment:
If you motherboard has a L2 cache and award bios, go into your bios settings
and set the L2 cache to disabled.
When I've tried this, I've seen actual speed of application performance drop
by 60-75%, depending on the app. Games tend to take a bigger hit than
business apps. So your overclocked Celeron running at 450 Mhz would
probably run at about half the speed of a normally clocked full-fledged P-II
at 233 Mhz.
You'd be better of getting a cheap socket 7 motherboard and a Cyrix 6x86M2
200Mhz chip. It would run as fast, wouldn't burn out, and would cost about
60% as much as your Celeron/Slot I motherboard combo.
Thats probably what the OEM's will decide, too. They'll put out Celeron
models for the rubes unwilling to shell out the dough for a real PII machine,
but willing to pay an extra $100-200 for a machine endorsed by breakdancing
idiots in bunny suits. But the Celeron still can't beat a socket 7 chip/MB
combo from AMD or Cyrix. And the AMD K6 really is perfectly adequate for most
home users needs at or above 166Mhz and probably will be for a couple more
years.
Frankly, Intel deserves the financial problems it has now. They abandoned
Socket 7 in favor of their own proprietary Slot I architechture. Every
computer company for the past 30 years that has tried to cram a proprietary
architechture down the market's throat has failed. I just don't understand
why everyone keeps making the same mistake.
This has been (yet) a(nother) communication from:
CHRIS GRAY, Man of Unlimited Potential! (Yet, oddly, of limited accomplishments)
Milk - "So-What have we learned today?" Cheese - "Nothing-And that's plenty for us!"
(Reply to me by typing out cgray athingy "beld" dot "totaltel" period "com" -
I sometimes munge my reply address to indulge my paranoia and persecution delusions)
Mike Smith
Guldan <ray...@vianet.net.au> wrote in article
<01bd6a61$ea47fc00$0100...@raytodd.vianet.net.au>...
> You won't want one... but I suggest that the performance of a PII
without
> 'onboard' L2 but with L2 on the mainboard like with the Pentiums,
will
> outdo a Pentium, and it will probably be cheap as hell! I like those
odds!
This is not an option, however. Slot 1 does not allow for front-side
L2 cache.
--Mike Smith
Richard N. Barg
<<The problem is not the CPU CORE which is the same but the L2 cache-chips
used. >>
Are you saying that Intel's cache chips are rated based on the CPU they are
tied to and that the overclockability of the CPU is not direclty
proportionate to the overclockability of the onboard cache chips..........or
put another way that the cache chips may be manufactured to a specification
which is much tighter than the CPU?
Antek S. Baranski wrote in message <35375B3D...@cern.ch>...
>
>
>David wrote:
>
>> On 17 Apr 1998 08:44:03 -0400, ssto...@freenet.columbus.oh.us (Samuel
>> Stoddard) wrote:
>>
>> >: Why would you want a P-II without its level 2 cache.??????
>> >
>> >The big reason is the overclockablility. 400mhz w/ no problems! Hell yea
>>
>> And the P2 isn't overclockable???
>
>Well it is I have run my 266 at 300 for several weeks running the RC5-64
key
>generator day and night (100% CPU load) and its just fine. The problem is
not
>the CPU CORE which is the same but the L2 cache-chips used. They aren't too
>good at keeping up with the higher clock, example 400 MHz would mean 200
MHz
>clock on the cache.... don't think that will work for a 266 or even a 333
P2.
>
>Cheers,
>
>--
>ing. Antek S. Baranski
>European Laboratory for Particle Physics (CERN) - IT division - ASD group
>Tel: +41 - (0)22 - 76 742 47 Fax: +41 - (0)22 - 76 786 30 ICQ:
>1127269
the count
Luis Constantino wrote:
> Hmm,
> That article is interesting. But look at it again... If you will
> then you will see that in almost every benchmark, the old P2 333 and
> I've got a funny feeling it'd top everything there no?
I doubt it would top a 600 MHz Alpha-PC ;)
Of course, you can't run Win95 on an AlphaPC, but you can run NT
(a 64 bit native port of NT, that is :)
John Stanton
In article <6h7bm6$5...@news5-gui.server.cableol.net>,
andre...@cableol.co.uk says...
> Why would you want a P-II without its level 2 cache.??????
>
>
> Senior Technical Specialist
> Fujitsu (ICL) Ltd
>
>
>
>
Victor Healey
In article <6h7bm6$5...@news5-gui.server.cableol.net>, andre...@cableol.co.uk says...
> Why would you want a P-II without its level 2 cache.??????
>
> Its like buying a Porsche and putting a Ford engine in it.
>
>
Because you can get performance that excells by overclocking the Celeron.
The l2 cache raises issues that prevent this in a normal P2 with L2 cache.
For a few bucks you can blow away all the socket 7 CPUs made to date.
Any questions?
--
Ki4je = spam defeat try
winndixie
@
bellsouth.net
B. Cass
On Fri, 17 Apr 1998 15:08:16 GMT,
-Y...@Like.I.Want.Spam.Read.My.Sig.Please (The Bombigator's Grandkid)
wrote:
>I'm not an Intel basher. I use Intel products at work and am very happy with
>them. I own all three different colors of the Intel Bunny Suit dolls.
>
FYI, I am sorry to inform you but your Doll collection is incomplete.
I assume you are refering to the little buggers that stores like
CompUSA had for $6 or for free w/ a pII purchase back in the fall.
The colors, as i remember them: green, yellow, purple, blue.
Sorry to burst yer bubble! <G>
PS. I worked at CompUSA for 8 months, and picked up 3 of the little
guys for free. They are hangin around my apartment now . . .
watching me .. . hmm. Marketing cutesie ploy or hi-tech spy gadgetry
to see what I use my Cyrix 200 for? <G>
Brandon
Remove SPAM ME to send me email @:
andon...@MEgte.net
Kasey Chang (fix address before replying to me!)
>Run - don't walk to
>http://www.tomshardware.com/cpubench.html
>Look for section under updated chip family review.
Absolutely BS, according to the PC World preview. The folowing points
where offered...
1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
into a M/B designed for Celeron.
2) Celeron is SLOWER than all its "cheap" clone competitors such as
the AMD K6-233 or the Cyrix 6X86MX. It has NO onboard secondary cache.
3) In business apps testing, Celeron is SLOWER THAN MOST Pentium-200
MMX's, despite its 266 MHz clock speed.
4) Intel is already designing Celeron's follow-up, codename Mendocino,
which should debut before end of the year.
Conclusion (mine, not PC World): any one who buys a Celeron-based PC
is a dumba**.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Kasey Chang DCL, Fremont, CA k a s e y c @ d i s c o p y . c o m
PDOXWIN, Star Trek, Computer Games, Science Fiction, Writing, && more
---------------------------------------------------------------------
Commercial use of this e-mail address implies your consent to pay me
amounts of up to US$100.00 per e-mail message from you received by me.
ZDBop
In article <6h7r3f$6v...@news2.ntr.net>,
-Y...@Like.I.Want.Spam.Read.My.Sig.Please (The Bombigator's Grandkid) writes:
>Frankly, Intel deserves the financial problems it has now. They abandoned
>Socket 7 in favor of their own proprietary Slot I architechture.
A-Freakin'-MEN! I hope they choke. I won't.
>Every computer company for the past 30 years that has tried to cram a
proprietary
>architechture down the market's throat has failed. I just don't understand
>why everyone keeps making the same mistake.
"Greed. Greed is good."
Gordon Gecko
Antek S. Baranski
Hi,
The Bombigator's Grandkid wrote:
> <SNIP>
>
> When I've tried this, I've seen actual speed of application performance drop
> by 60-75%, depending on the app. Games tend to take a bigger hit than
> business apps. So your overclocked Celeron running at 450 Mhz would
> probably run at about half the speed of a normally clocked full-fledged P-II
> at 233 Mhz.
Please have a look at: http://sysdoc.pair.com/IntelsAprilBlessings.html before making such a bold
statement. :-)) Especially at: http://sysdoc.pair.com/cpu-3dsm.html ,
http://sysdoc.pair.com/cpu-ws95.html and http://sysdoc.pair.com/cpu-icoming.html . The only
performance problem are the ws95 results... The rest just rocks!
> You'd be better of getting a cheap socket 7 motherboard and a Cyrix 6x86M2
> 200Mhz chip. It would run as fast, wouldn't burn out, and would cost about
> 60% as much as your Celeron/Slot I motherboard combo.
And a Cyrix CPU is definitely a NO GO for performance of any kind... :-(( And Cyrix used to be the
best FPU you could get for a 386 too bad they lost something somewhere..... :-((
> <SNIP>
>
>
> This has been (yet) a(nother) communication from:
>
> CHRIS GRAY, Man of Unlimited Potential! (Yet, oddly, of limited accomplishments)
Cheers,
Antek S. Baranski
Richard N. Barg wrote:
> <<The problem is not the CPU CORE which is the same but the L2 cache-chips
> used. >>
>
> Are you saying that Intel's cache chips are rated based on the CPU they are
> tied to and that the overclockability of the CPU is not direclty
> proportionate to the overclockability of the onboard cache chips..........or
> put another way that the cache chips may be manufactured to a specification
> which is much tighter than the CPU?
Exactly! I have several (7) PII-266 machines and the ones that were bought in
July 1997 (5) can run at 300 but the new machines bought last month can't do it.
When I get around to I will change the CPU's in two machines and have a look
what happens. I know that one of the new machines wouldn't boot at all and the
other one got screen corruption...... :-((
Luis Constantino
> P2 333 & 300 Old? If that's old than I must be missing soemthing....... :-))
>
Ok, I might have annoyed some ppl with this. So sorry about what I said with respect
to the 300 and 333.
> > 4th. and that's with them NOT overclocked. So don't throw away your memory,
> > CPU's and LX6's yet... What I'd Love to see would be a descutes 333
> > overclocked to(83*5) 416 in there.... I've got a funny feeling it'd top
> > everything there no?
>
> I don't think so because the L2 cache-ships won't work on 208 MHz! :-(( As you said
> have a look at the article again and you will say that Tom's says that the problem
> with overclocking the PII is not the CPU core but the chips used for the L2 cache.
> :-((
>
I know that the L2 cache is the limiting factor here, but you can still overclock the
original(non clock-locked) 233s and 266's to 300. there have also been reprts of many
people getting there deschutes 333 to do 416, and most can do 375. So the L2 cache,
while it might be limiting, does not preent overclocking. ps.my clocklocked 233 does
290(but the PCI stb nvidia card don't like it so much) and I always run it at 263
> > As far as the BX/Celeron combo, it has the potential to be a cheap
> > alternative, IF the PC100 ram prices drop....
>
> Exactly the only problem with the BX/Celeron combo is the price you have to pay for
> memory that *will* actually work in the system, which is quite important, I think
> :-))
>
> Next thing you will see is: "SDRAM certified to work in your BX/Celeron 400 MHz
> machine". hehehehehe ;-)
>
so true about the memory! my thoughts: after the intial rush of ppl wanting the latest
and greatest. BX board prices and PC100 memory will decrease in price. roght now
PC100 is atleast twice the price of normal(eeprom) SDRAM.
> As to the original question can you overclock a Celeron in a LX6 board yes you can,
> just set the clock to 75 or 83 Mhz (if you memory will work with it) and x4 will
> give you some nice speed. If you can go beyond x4 say 4.5x or even 5x then you will
> get even more interesting speeds.... ;-))
>
> Cheers,
>
> --
> ing. Antek S. Baranski
> European Laboratory for Particle Physics (CERN) - IT division - ASD group
> Tel: +41 - (0)22 - 76 742 47 Fax: +41 - (0)22 - 76 786 30 ICQ: 1127269
> REMOVE THE "bla bla" to-reply
Just my .02 Lue
Antek S. Baranski
Kasey Chang (fix address before replying to me!) wrote:
> On Fri, 17 Apr 1998 09:08:44 GMT, "Victor Healey"
> <fourphun....@bellsouth.net> wrote:
>
> >Run - don't walk to
> >http://www.tomshardware.com/cpubench.html
> >Look for section under updated chip family review.
>
> Absolutely BS, according to the PC World preview. The folowing points
> where offered...
To be honest I think Tom does a better job than any magazine in the world,
this is my opinion which isn't necessarily correct.
> 1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
> into a M/B designed for Celeron.
What M/B might that be? The EX-chipset is a cheap BX-chipset and they are
the only ones with support for 100 MHz front-bus speed. So what would the
problem be for a PII to go into a BX board or even an EX board? If you try
running your PII at 100 MHz front-end it might complain yes.
> 2) Celeron is SLOWER than all its "cheap" clone competitors such as
> the AMD K6-233 or the Cyrix 6X86MX. It has NO onboard secondary cache.
When is it slower? Benchmarks such as WinBench ARE NOT reliable!!! Have a
look at: http://sysdoc.pair.com/3Dhype98.html neither is WinStone reliable
because is isn't the real world in the real world the CPU waits for your
input....
> 3) In business apps testing, Celeron is SLOWER THAN MOST Pentium-200
> MMX's, despite its 266 MHz clock speed.
Correct although...... http://sysdoc.pair.com/cpu-3dsm.html The standard
Celeron 266/66 is in the middle of the pack above the rest.... :-))
> 4) Intel is already designing Celeron's follow-up, codename Mendocino,
> which should debut before end of the year.
I have no idea yet.....
> Conclusion (mine, not PC World): any one who buys a Celeron-based PC
> is a dumba**.
>
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> Kasey Chang DCL, Fremont, CA k a s e y c @ d i s c o p y . c o m
> PDOXWIN, Star Trek, Computer Games, Science Fiction, Writing, && more
Cheers,
John Michael Wilson
Victor Healey (fourphun....@bellsouth.net) wrote:
A little relevant information. Intel makes P-II 266's in both 0.35u
and 0.25u. Based on the 400Mhz results from the 266's with no L2
cache, I would assume that this is the 0.25u version. The 0.35u
verison with L2 cache will never reach 400Mhz under normal
conditions (i.e. no liquid nitrogen). As a matter of fact all
P-II 333's are 0.25u. The power dissipation is outrageous for
the 300Mhz 0.35u chips. I pulled this information from the
datasheets on Intel's Developers Site :)
john
: TIP
: Run - don't walk to
: http://www.tomshardware.com/cpubench.html
:
: Look for section under updated chip family review.
: The question I have is can you overclock an Intel Celeron using an ABIT LX
:
:
:
Mark
On Fri, 17 Apr 1998 09:43:08 -0500, "T. Bull" <CCon...@snowhill.com>
wrote:
>Then again....who is to say that a motherboard company might not go ahead
>and put an L2 on the board just for the Celeron. Asus wasn't supposed to
>put an 83 Mhz bus speed on an Intel HX board....but they did.
>
Adding L2 cache is not trivial like 83 Mhz. To get a higher speed all
you do is put a faster clock on the motherboard.
To add L2 will require a totally new chipset. Intel won't produce one
that provides L2. Nor will any of the Taiwanese chip manufacturers
since Intel banned them from producing slot1 chips.
John Smith
Sorry, I have to take issue with this. I have built some reliable, screaming
Ford engines putting out 1 hp/cu.in or more (though I am at heart a Chevy
man, not a Fordnatic. Had to clarify this ;) ) How does a 500 hp. 460 with
over 550 lbs/ft of torque sound, all within a useable RPM range? Try getting
that level of everyday reliable performance out of a Porsche for even 2x the
price! Or a 320 hp. 5.0 (not even close to its potential)? It could even be
made to fit in a Porsche, but why go thru the trouble and devaluation of
that overpriced sports car, except if you want a more reliable and
potentially as if not more powerful engine? There is a 351C (yep! Ford!) in
the Pantera. BTW, with turbocharging, even the lowly Pinto engine can put
out 300+ *reliable* horsepower. (still a Pinto, though...) That L2 cache
seems not to affect rendering performance quite so much as having a
single-issue FPU. The P2---> Celeron is sort of like dropping the
compression ratio and retarding the timing for 87 octane gasoline. By
contrast, the P2 FPU---> AMD/Cyrix FPU is like removing the spark plug wires
on four of those 8 cylinders. Maybe if AMD and Cyrix would get on the ball
before Intel flattens them with its ponderous bulk...
Andrew wrote in message <6h7bm6$5...@news5-gui.server.cableol.net>...
-J
On Fri, 17 Apr 1998 17:13:34 GMT, Kas...@spamsucker.discopy.com (Kasey
Chang (fix address before replying to me!)) wrote:
>On Fri, 17 Apr 1998 09:08:44 GMT, "Victor Healey"
><fourphun....@bellsouth.net> wrote:
>
>>Run - don't walk to
>>http://www.tomshardware.com/cpubench.html
>>Look for section under updated chip family review.
>
>Absolutely BS, according to the PC World preview. The folowing points
>where offered...
>
>1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
>into a M/B designed for Celeron.
I didnt know this... could be an interesting point. Is this true?
>
>2) Celeron is SLOWER than all its "cheap" clone competitors such as
>the AMD K6-233 or the Cyrix 6X86MX. It has NO onboard secondary cache.
So overlcock it to 400MHz and blow them all away.
>
>
>3) In business apps testing, Celeron is SLOWER THAN MOST Pentium-200
>MMX's, despite its 266 MHz clock speed.
see above
>
>4) Intel is already designing Celeron's follow-up, codename Mendocino,
>which should debut before end of the year.
>
>Conclusion (mine, not PC World): any one who buys a Celeron-based PC
>is a dumba**.
>
I will trust Toms Hardware Guide over PC World *any* day of the week.
Not to mention that PC Week had an early version of the chip, *PLUS*
PC Week did not overclock it to 400MHz with a 100MHz bus.
Enjoy,
-J
>=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>Kasey Chang DCL, Fremont, CA k a s e y c @ d i s c o p y . c o m
>PDOXWIN, Star Trek, Computer Games, Science Fiction, Writing, && more
Wayne Ernst
Overclocking this chip brings up a good point. I wonder
how the bios will identify it on the screen? Someone
could easily sell this as a 400 MHz system and get
an additional $600 profit. I would hope that someone
in the market for a 400 MHz system, would know that
they were buying a true PII with a BX chipset. But then
again...
-J <jho...@removemewhro.org> wrote in article
<3537b123....@news.whro.org>...
John King
The Bombigator's Grandkid (-Y...@Like.I.Want.Spam.Read.My.Sig.Please) wrote:
: But the Celeron is a BAD technical idea. Its just a marketing scheme to push
: Slot I machines out into the sub-$1000 PC world. Without an L2 cache, I
k
I think it just has an external L2 cache NOT zero L2 cache. The onchip
stuff is gone (for now).
--
+-- Telephone HEADSETS - corded, cellular, wireless. ----------------------+
| Plantronics, Unex, ACS, GN Netcom, VXI, and more. Repairs too. |
| Commercial, small office home office. All nations, almost all phones. |
+-- Increase productivity, ease neck pain. (253) 845-4088 jo...@eskimo.com -+
john a. mcclenny
-J (jho...@removemewhro.org) wrote:
: On Fri, 17 Apr 1998 17:13:34 GMT, Kas...@spamsucker.discopy.com (Kasey
: Chang (fix address before replying to me!)) wrote:
: >On Fri, 17 Apr 1998 09:08:44 GMT, "Victor Healey"
: ><fourphun....@bellsouth.net> wrote:
: >
: >>Run - don't walk to
: >>http://www.tomshardware.com/cpubench.html
: >>Look for section under updated chip family review.
: >
: >Absolutely BS, according to the PC World preview. The folowing points
: >where offered...
: >
: >1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
: >into a M/B designed for Celeron.
: I didnt know this... could be an interesting point. Is this true?
No :). But the Celeron Intel MB only handles 66 MHz Ram,
which defeats the whole purpose.
Mike Smith
> And a Cyrix CPU is definitely a NO GO for performance of any kind...
:-(( And Cyrix used to be the
> best FPU you could get for a 386 too bad they lost something
somewhere..... :-((
Yeah, I remember that, had a Cx387 chip myself. Definitely faster than
the Intel part. I'm sure they *could* make a faster FPU if they
wanted, but they probably decided (maybe correctly) that most people
don't care too much about FPU performance. ('Cept for N million
Quake/Q2 players... ;-)
--Mike Smith
Mike Smith
Wayne Ernst <wayne...@pressroom.com> wrote in article
<01bd6a3a$470fb0e0$ab8345c6@default>...
> Overclocking this chip brings up a good point. I wonder
> how the bios will identify it on the screen? Someone
> could easily sell this as a 400 MHz system and get
> an additional $600 profit. I would hope that someone
> in the market for a 400 MHz system, would know that
> they were buying a true PII with a BX chipset. But then
> again...
If it's an Award BIOS, the first screen after the beep will show the L2
cache size (in the table where it shows serial ports, PIO modes, etc.)
--Mike Smith
Mike Smith
> Exactly! I have several (7) PII-266 machines and the ones that were
bought in
> July 1997 (5) can run at 300 but the new machines bought last month
can't do it.
> When I get around to I will change the CPU's in two machines and have
a look
> what happens. I know that one of the new machines wouldn't boot at
all and the
> other one got screen corruption...... :-((
The one that wouldn't boot at all was probably clock-locked. Try 75x4
instead of 66x4.5. The other one, well, sometimes shit happens. ;-)
--Mike Smith
Mike Smith
mark_j...@hotmail.com wrote in article
<6h84fm$g5h$1...@nnrp1.dejanews.com>...
> Well, If you are looking for cheap. You can Get a Pentium Pro 166
with 512K
> Cache, overclock it to 233, all for about $300. (Thats including the
new
> Motherboard and the Peltier Cooling Fan)
I'm a big overclocking fan, but I wouldn't go telling people they'll be
able to overclock a PPro166 to 233. Maybe. But *not* definitely.
Where are you going to find a 166/512 these days, anyway?
--Mike Smith
Mike Smith
john a. mcclenny <j...@netcom.com> wrote in article
<jamErK...@netcom.com>...
> : >1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium
II
> : >into a M/B designed for Celeron.
>
> : I didnt know this... could be an interesting point. Is this true?
>
> No :). But the Celeron Intel MB only handles 66 MHz Ram,
> which defeats the whole purpose.
Is this really germane? The point (IMO) is that you can put a Celeron
in a BX board (Tom Pabst tested the Celeron in an AX6B, I assume the
Abit BX6 and the Asus board [P2B98?] will also take the Celeron), not
the other way around.
--Mike Smith
Foolio
So are you going to pick up chicks with your cool new Celeron P.O.S
machine???
Zorak wrote in message <6h7sl5$nli$1...@winter.news.erols.com>...
>But when your objective is to pick up chicks, the Porsche with the Ford
>engine will still work as good!
>
>Andrew wrote in message <6h7bm6$5...@news5-gui.server.cableol.net>...
>>Why would you want a P-II without its level 2 cache.??????
>>
>>Its like buying a Porsche and putting a Ford engine in it.
>>
Foolio
>I'm not an Intel basher. I use Intel products at work and am very >happy
with
>them. I own all three different colors of the Intel Bunny Suit dolls.
That's queer!
> I've also run Cyrix, Intel and AMD chips at home. All three >companies
have
>produced products that impressed me at different times.
I thought you loved Intel bunny boy?!?!
>But the Celeron is a BAD technical idea. Its just a marketing >scheme ...
agree totaly. no cache ==> go to memory _every_ _single_
_instruction_. Lame
>You'd be better of getting a cheap socket 7 motherboard and a >Cyrix 6x86M2
>200Mhz chip
forget Cyrix. They aren't worth the trouble!
>Frankly, Intel deserves the financial problems it has now. They >abandoned
>Socket 7 in favor of their own proprietary Slot I architechture
I hope it is their down fall bunny boy!!
Foolio
OK, I think he meant a ford taurus engine, not a souped up homebuilt engine
:)
John Smith wrote in message <6h8ahh$j4g$1...@news1-alterdial.uu.net>...
>Sorry, I have to take issue with this. I have built some reliable,
screaming
>Ford engines putting out 1 hp/cu.in or more (though I am at heart a Chevy
>man, not a Fordnatic. Had to clarify this ;) ) How does a 500 hp. 460 with
>over 550 lbs/ft of torque sound, all within a useable RPM range? Try
getting
>that level of everyday reliable performance out of a Porsche for even 2x
the
>price! Or a 320 hp. 5.0 (not even close to its potential)? It could even be
>made to fit in a Porsche, but why go thru the trouble and devaluation of
>that overpriced sports car, except if you want a more reliable and
>potentially as if not more powerful engine? There is a 351C (yep! Ford!)
in
>the Pantera. BTW, with turbocharging, even the lowly Pinto engine can put
>out 300+ *reliable* horsepower. (still a Pinto, though...) That L2 cache
>seems not to affect rendering performance quite so much as having a
>single-issue FPU. The P2---> Celeron is sort of like dropping the
>compression ratio and retarding the timing for 87 octane gasoline. By
>contrast, the P2 FPU---> AMD/Cyrix FPU is like removing the spark plug
wires
>on four of those 8 cylinders. Maybe if AMD and Cyrix would get on the ball
>before Intel flattens them with its ponderous bulk...
>
Lupine
> -J (jho...@removemewhro.org) wrote:
> : On Fri, 17 Apr 1998 17:13:34 GMT, Kas...@spamsucker.discopy.com (Kasey
> : Chang (fix address before replying to me!)) wrote:
>
> : >On Fri, 17 Apr 1998 09:08:44 GMT, "Victor Healey"
> : ><fourphun....@bellsouth.net> wrote:
> : >
> : >>Run - don't walk to
> : >>http://www.tomshardware.com/cpubench.html
> : >>Look for section under updated chip family review.
> : >
> : >Absolutely BS, according to the PC World preview. The folowing points
> : >where offered...
> : >
> : >1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
> : >into a M/B designed for Celeron.
>
> : I didnt know this... could be an interesting point. Is this true?
>
> No :). But the Celeron Intel MB only handles 66 MHz Ram,
> which defeats the whole purpose.
Um.. I do not know of anyone who proposes to put a Pentium II onto a M/B
designed for the Celeron. The discussion is putting the Celeron onto a
PII BX motherboard. The intel bx board has physical problems.. but as we can
see from Tom's warehouse.. on other motherboards, this is just fine.
The Bombigator's Grandkid
In article <35379122...@cern.ch>, "Antek S. Baranski" <Antek.Baranski@"euro.lab.for.part.phys"cern.ch> wrote:
>Hi,
>
>The Bombigator's Grandkid wrote:
>
>> <SNIP>
>>
>> When I've tried this, I've seen actual speed of application performance drop
>> by 60-75%, depending on the app. Games tend to take a bigger hit than
>> business apps. So your overclocked Celeron running at 450 Mhz would
>> probably run at about half the speed of a normally clocked full-fledged P-II
>> at 233 Mhz.
>
>Please have a look at: http://sysdoc.pair.com/IntelsAprilBlessings.html before
> making such a bold
>statement. :-)) Especially at: http://sysdoc.pair.com/cpu-3dsm.html ,
>http://sysdoc.pair.com/cpu-ws95.html and
> http://sysdoc.pair.com/cpu-icoming.html . The only
>performance problem are the ws95 results... The rest just rocks!
>
>> 60% as much as your Celeron/Slot I motherboard combo.
>
HeHe - OKOK - It looks like I was wrong. But If you look at the winstone
results, I was almost on the money. The Celeron is about 10% slower just
below the Cyrix 6x86M2 in performance when clocked properly, and about 15%
faster when overclocked to 400Mhz.
And in fairness to my earlier pronouncements, the last time I disabled my L2
cache, I didn't have the benefit of a high-end 3D card to offload all the 3D
polygon fill away from the CPU(and the DRAM to L2 cache bottleneck), so its
not surprising the Winstone test results are more in line with what I've seen
in the past.
But I admit, I hadn't considered the fact that most hard-core gamers ARE using
high end graphics cards nowadays. And the Slot I boards all have AGP, which
doubles video output speed by itself. Now that you've got me thinking about
it, I find the thought of overclocking a Celeron a very attractive idea (at
leat for games anyway). Maybe I'll give it a try this summer when the things
become available on the gray market in quantity.
I've got another question though: what's to stop dishonest vendors from
grinding the lettering off "Celeron" chips and relabelling them as "Pentium
II" ? Is there going to be any easy way to tell the two apart, short of
cracking their case?
This has been (yet) a(nother) communication from:
CHRIS GRAY, Man of Unlimited Potential! (Yet, oddly, of limited accomplishments)
Milk - "So-What have we learned today?" Cheese - "Nothing-And that's plenty for us!"
(Reply to me by typing out cgray athingy "beld" dot "totaltel" period "com" -
I sometimes munge my reply address to indulge my paranoia and persecution delusions)
Mark
On 17 Apr 1998 20:19:39 GMT, jo...@eskimo.com (John King) wrote:
>The Bombigator's Grandkid (-Y...@Like.I.Want.Spam.Read.My.Sig.Please) wrote:
>
>: But the Celeron is a BAD technical idea. Its just a marketing scheme to push
>: Slot I machines out into the sub-$1000 PC world. Without an L2 cache, I
>k
>I think it just has an external L2 cache NOT zero L2 cache. The onchip
>stuff is gone (for now).
You're wrong. It has ZERO L2 cache.
John King
: > Its like buying a Porsche and putting a Ford engine in it.
Looks like this Ford engine handles 10,000rpm though.
Ga'ash Soffer
the celeron looks totally different, i think.
--
Rage's Hardware
http://www.computerheaven.net/rageshardware/
Hardware reviews, previews, and news
full i740 review, AX6LC, 3D accelerator
benchmarks and image quality comparision
and more...
Robato Yao
In <MPG.fa157bff...@news.atl.bellsouth.net>, fourphun....@bellsouth.net (Victor Healey) writes:
>In article <6h7bm6$5...@news5-gui.server.cableol.net>, andre...@cableol.co.uk says...
>> Why would you want a P-II without its level 2 cache.??????
>>
>>
>>
>Because you can get performance that excells by overclocking the Celeron.
Which may be probably limited with a block that prevents overclocking by
multiplying the clock. You may have to overclock by pushing the MB bus
speed, much like the newest batches of Pentium MMX.
>The l2 cache raises issues that prevent this in a normal P2 with L2 cache.
>For a few bucks you can blow away all the socket 7 CPUs made to date.
>Any questions?
Except that .25 micon AMD K6s are also overclockable, and socket 7
motherboards still have L2 caches (as big as 1MB) that won't be affected
(since they are not backside caches) when the K6s are highly
overclocked.
Does the 440EX chipset support 100MHz? That's another question, when
you consider that the socket 7 MBs now have them.
Rgds,
Chris
>
>--
>Ki4je = spam defeat try
>winndixie
> @
>bellsouth.net
(counting down from top 50 oxymorons...)
10. Tight slacks
9. Definite maybe
8. Pretty ugly
7. Twelve-ounce pound cake
6. Diet ice cream
5. Rap music
4. Working vacation
3. Exact estimate
2. Religious tolerance
And the NUMBER ONE top oxy-MORON
1. Microsoft Works
---From the Top 50 Oxymorons (thanks to Richard Kennedy)
Robato Yao
In <MPG.fa14a1e7...@news.mindspring.com>, jlsb...@mindspring.com (John Stanton) writes:
>In article <6h7bm6$5...@news5-gui.server.cableol.net>,
>andre...@cableol.co.uk says...
>> Why would you want a P-II without its level 2 cache.??????
>>
>> Its like buying a Porsche and putting a Ford engine in it.
>>
>> Senior Technical Specialist
>> Fujitsu (ICL) Ltd
>>
>>
>>
>>
Actually, it's more like a Ford with a Porsche engine with too small
gas tank.
Rgds,
Chris
Robato Yao
In <35378b1b...@205.158.0.30>, Kas...@spamsucker.discopy.com (Kasey Chang (fix address before replying to me!)) writes:
>On Fri, 17 Apr 1998 09:08:44 GMT, "Victor Healey"
><fourphun....@bellsouth.net> wrote:
>
>>Run - don't walk to
>>http://www.tomshardware.com/cpubench.html
>>Look for section under updated chip family review.
>
>Absolutely BS, according to the PC World preview. The folowing points
>where offered...
>
>1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
>into a M/B designed for Celeron.
>
>the AMD K6-233 or the Cyrix 6X86MX. It has NO onboard secondary cache.
>
>
>MMX's, despite its 266 MHz clock speed.
>
>which should debut before end of the year.
>
>Conclusion (mine, not PC World): any one who buys a Celeron-based PC
>is a dumba**.
In my honest opinion, the Celeron is a loopsided CPU just like the Cyrix
686MX. Both sits on opposite poles---the Cyrix excels in application
and integer performance (same with the K6), but sucks in FPU
performance. The Celeron sucks in application/integer performance but
should excel in FPU performance, probably running rings around a
P233MMX.
The K6 and the 686MX achieves their tremendous application performance
without a backside L2 cache. Why Intel, with its much
bigger tremendous technological resources, failed to do what smaller
competitors have done, seems right on the brink of inexcusable in the
public eye. Sure, the Celeron will undoubtedly beat the crap of a
686MX PR266 on Quake II, but far more people are going run MS Office
(where the 686MX would undoubtedly be much faster) than Quake II. Good
FPU performance is no excuse for lousy application performance, yet the
reverse is true---good application performance makes up for lousy FP
performance, at least in the eyes of Joe User.
The ideal low cost chip would have been to pack the application
performance of a K6-266 or 686MX-PR266 and the FP performance of a
Celeron 266.
Having said this, if a person wants to make a *primarily* game machine
for the lowest cost, I will not call him a dumbass if he is to use
Celeron. I may do it myself (and keep my K6 and Cyrix systems in my
office). A Celeron system (if you are not using that 440EX mobo) with a
440LX or BX chipset can still at least upgrade to a faster PII later on
when these becomes more affordable. Celeron pretty much sits where a
486SX used to be.
Rgds,
Chris
>
>=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>Kasey Chang DCL, Fremont, CA k a s e y c @ d i s c o p y . c o m
>PDOXWIN, Star Trek, Computer Games, Science Fiction, Writing, && more
>---------------------------------------------------------------------
>Commercial use of this e-mail address implies your consent to pay me
>amounts of up to US$100.00 per e-mail message from you received by me.
Daniel Schwartz
Check out Tom's hardware site for some info on Celeron. Apparently this
sucker is the most overclockeable PII yet.
dan
d...@ucla.edu
Jason Petersen
Kasey Chang (fix address before replying to me!) wrote:
>
> On Fri, 17 Apr 1998 09:08:44 GMT, "Victor Healey"
> <fourphun....@bellsouth.net> wrote:
>
> >Run - don't walk to
> >http://www.tomshardware.com/cpubench.html
> >Look for section under updated chip family review.
>
> Absolutely BS, according to the PC World preview. The folowing points
> where offered...
>
> 1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
> into a M/B designed for Celeron.
From what the latest issue of Boot says about Celeron, this is true.
However, they say that it's possible to fit a standard PII mobo with the
Celeron.
--
Jason
http://members.aol.com/pete6306
The system is down.
Type any key to continue
Type any other key to quit
Jason Petersen
Andrew wrote:
>
> Why would you want a P-II without its level 2 cache.??????
>
> Its like buying a Porsche and putting a Ford engine in it.
So you think that the absence of L2 cache will make it faster?? :p
T. Bull
Question:
If the Celeron has no cache to slow down its overclocking......does a
Pentium 2 with the L2 cache disabled in the bios overclock more easily????
HMMMM.......
Bull
Barry Betambeau wrote in message
<892858208.6434.0...@news.demon.co.uk>...
>
>Foolio wrote in message <6h8f4e$89e$1...@news.ccit.arizona.edu>...
>>
>>>
>>>But the Celeron is a BAD technical idea. Its just a marketing >scheme
...
>>
>>
>>agree totaly. no cache ==> go to memory _every_ _single_
>>_instruction_. Lame
>
>Wrongo! sunshine, you still have the L1 cache.
>In test I've done disabling L2 only made 20-25% difference in speed, slower
>obviously.
>
>
>
>
steevo
celeron has no case, its like a naked p2..saves manufacturing costs
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
steevo
"It's all a bunch of tree-huggin' hippie crap!" --E.Cartman
Ga'ash Soffer wrote in message <35380BB4...@home.com>...
Joe B
My sentiments exactly.....Joe b
Kendall Helmstetter Gelner
-Y...@Like.I.Want.Spam.Read.My.Sig.Please (The Bombigator's Grandkid)
wrote:
>I'm not an Intel basher. I use Intel products at work and am very happy with
>them. I own all three different colors of the Intel Bunny Suit dolls.
There are five colors:
Yellow, Green, Blue, Pink, Violet
But I pretty much agree with the rest of your post.
This important topical post brought to you by:
O \'\Kgelner@ \'\
You know, for kids! \'\ @bigfoot. \'\
\'\ .com\'\
Kenneth Lee
Luis Constantino wrote in message <3537644D...@cornell.edu>...
>Hmm,
> That article is interesting. But look at it again... If you will then
you
>will see that in almost every benchmark, the old P2 333 and 300 are around
3 or
>4th. and that's with them NOT overclocked. So don't throw away your
memory,
A P2-333 OLD? I'm about to get my P2-333 on Sunday, and you're saying its
OLD?
They say computers get obsolete once you get it...but this is ridiculous..!
somebody's saying its obsolete before I even lay my hands on it....!
who cares about a P2-400, they're for the richy rich...
Barry Betambeau
Antek S. Baranski wrote in message <35376935...@cern.ch>...
>
>I don't think so because the L2 cache-ships won't work on 208 MHz!
(
Some do ; )
As you said
>have a look at the article again and you will say that Tom's says that the problem
>with overclocking the PII is not the CPU core but the chips used for the L2 cache.
>:-((
Sometimes the system starts with the L2 cache disabled, but once it DOES start its quite reliable. If I could work out how to jack the supply voltage to the cache up by 20-25% I'm sure it would be solid. And dont fret about over heating, the 333 Deschute runs cooler than the 233 Klamath!
>Exactly the only problem with the BX/Celeron combo is the price you have to pay for
>memory that *will* actually work in the system, which is quite important, I think
>:-))
Yep, the call you have to make "is a Celeron/BX/PC100ram better value than a Deschutes 333/LX/SDRAM ?"
Barry
Barry Betambeau
Andrew wrote in message <6h7bm6$5...@news5-gui.server.cableol.net>...
>Why would you want a P-II without its level 2 cache.??????
>
>
Somebody who was interested in SAVING enough money that he could then get a
Voodoo2 3d card?
A combination that doesn't care a shit about L2 presence/performance when
you play Quake2
But I suppose the problem is, who the hell plays Quake? it never did catch
on, did it?(sarcasm :)
>
Barry Betambeau
Bonker
>Hmm,
> That article is interesting. But look at it again... If you will then you
>will see that in almost every benchmark, the old P2 333 and 300 are around 3 or
>4th. and that's with them NOT overclocked. So don't throw away your memory,
>CPU's and LX6's yet... What I'd Love to see would be a descutes 333
>overclocked to(83*5) 416 in there.... I've got a funny feeling it'd top
>everything there no?
> As far as the BX/Celeron combo, it has the potential to be a cheap
>alternative, IF the PC100 ram prices drop....
>Just my .02
>
>Lue
OK, Win95 finally stopped buttfucking me and I now have my new Abit
LX6 rev 'K' mobo + new PII-333 running (apparently) stable at 416MHZ
(5x83). Unfortunately, I haven't got the Voodoo 2 card yet so I can't
run the Quake test etc to see what kinda scores this bitch will pump
out. I'd expect it to beat a PII-400/100 on a BX mobo though ...
BTW, for those who care, my system is:
Abit LX6 rev 'K' mobo (just purchased)
Intel Boxed PII-333 (just purchased. Not clock-locked)
1x32MB 10ns SDRAM (just purchased. Cheapo, they look like Hyundai
chips on a no-name stick)
Quantum Fireball ST 6.4GB hard drive (6 months old)
Matrox Mystique 220 4MB (6 months old)
SB32 (1 year old)
Adaptec 1510 SCSI card (18 months old)
I tried the 92Mhz bus speed but funnily enough, the Mystique didn't
want to play at all, just got a black screen. I have a friend with a
Millenium II 4MB, I'll borrow that card in a couple of days to see if
that can hack it ;)
I'm currently running the drive at PIO 3, just in case. Will check out
the full whack if and when this has been running solid for a couple of
days or so.
BTW, I only have 1x32MB cos' that's all the SDRAM the supplier had at
the time, will be picking up 2x64MB next week, hopefully they'll work
just as well as the 1x32MB seems to ...
Bonker.
W.J. Ernst
Maybe chicks in 'alt.binaries.erotica.chicks'.
Foolio <Q...@r.s.com> wrote in article <6h8f9k$tqo$1...@news.ccit.arizona.edu>...
> So are you going to pick up chicks with your cool new Celeron P.O.S
> machine???
>
>
> Zorak wrote in message <6h7sl5$nli$1...@winter.news.erols.com>...
> >But when your objective is to pick up chicks, the Porsche with the Ford
> >engine will still work as good!
> >
> >Andrew wrote in message <6h7bm6$5...@news5-gui.server.cableol.net>...
> >>Why would you want a P-II without its level 2 cache.??????
> >>
> >>Its like buying a Porsche and putting a Ford engine in it.
> >>
Kim Wilson
I may be wrong but I thought that after the last instance of creeps
remasking chips Intel hard coded the chip ID into the chip itself.
Regards
Kim
kwi...@esi.co.nz
SS396
On Fri, 17 Apr 1998 15:28:48 -0400, "John Smith" <jsm...@company.com>
wrote:
>Sorry, I have to take issue with this. I have built some reliable, screaming
>Ford engines putting out 1 hp/cu.in or more (though I am at heart a Chevy
>man, not a Fordnatic. Had to clarify this ;) )
Good, then you're aware that GM did that with *production* engines
some 30/40 odd years ago with a Fuel Injected 283 cu.in in the
Corvette, it had 283 cu.in and 283 HP .
> How does a 500 hp. 460 with
>over 550 lbs/ft of torque sound, all within a useable RPM range? Try getting
>that level of everyday reliable performance out of a Porsche for even 2x the
>price! Or a 320 hp. 5.0 (not even close to its potential)? It could even be
>made to fit in a Porsche, but why go thru the trouble and devaluation of
>that overpriced sports car, except if you want a more reliable and
>potentially as if not more powerful engine? There is a 351C (yep! Ford!) in
>the Pantera. BTW, with turbocharging, even the lowly Pinto engine can put
>out 300+ *reliable* horsepower. (still a Pinto, though...)
How 'bout 425 HP out of my solid lifter, factory stock (unmodified)
396 SS Chevelle and a 50,000 mile warranty in 1969 ?? with a 4 speed
and 4.11 rear gearing ? :)
>That L2 cache
>seems not to affect rendering performance quite so much as having a
>single-issue FPU. The P2---> Celeron is sort of like dropping the
>compression ratio and retarding the timing for 87 octane gasoline. By
>contrast, the P2 FPU---> AMD/Cyrix FPU is like removing the spark plug wires
>on four of those 8 cylinders. Maybe if AMD and Cyrix would get on the ball
>before Intel flattens them with its ponderous bulk...
Ahhhh that's apples and Oranges , Intel sets the standards and it's
gotta be a Bixxx and doubly difficult working with technology that
some magazines have compared to the Manhattan Project (1st atomic bomb
for gen X'ers) chip details the size of the wavelength of green light
etc, chip yields, backwards/Windows/Intel compatibility all the while
making sure you're *NOT* trampling over one of Intel's designs knowing
full well they'll *sue* at the drop of a hat, and *RACING* to make
sure that by the time you hit the market with one generation of chip
say at 266Mhz someone else hasn't eaten your lunch by being at the
market with the *NEXT* generation of chip at 400Mhz.
In comparing CPU technology to automobiles the latest Corvette engine
isn't much different at all (sans electronics) than it's great-great
grand-daddy the 265 cu.in OHV V8 introduced in 1955 .... And I too
think Porsche's suck :)
cqu...@iafrica.com
On Fri, 17 Apr 1998 14:20:06 +0200, "Antek S. Baranski"
<Antek.Baranski@"euro.lab.for.part.phys"cern.ch> wrote:
>> Why would you want a P-II without its level 2 cache.??????
>How about the price and speed? Have a look at
>http://sysdoc.pair.com/cpubench.html and look at the GAMES results with
>a Celeron running at 400 MHz (OVER CLOCKED!!!!)
OK; makes sense. Instead of paying $$$$ for a PII-100MHz, drop in a
Celeron, beat it to death, and just buy more as they burn out
(consider it a consumable overhead, like toner cartridges) until the
PII you wanted gets cheap enough to buy. Chances are good that you
might get that on a single Celeron, if it lasts a year.
As long as the associated non-financial costs (data loss and downtime)
are acceptable, it's as good an end-user-level strategy as any. In
fact, as it is only the disposable CPU that is overclocked, risk of
data loss is smaller than the johnnys who overclock the PCI.
Niggling doubt; how long do you think it will be before Intel puts
clock-limiting circuitry into Celeron? It would be typical Intel to
do that even if it cost as much as adding the L2 back.
Amaury
Even if it did and was faster, the whole point of the Celeron is that
it only costs $180 whereas a P2 400mhz CPU costs $800. It has the FPU
performance of a Pentium 2 400, and because it doesn't have a L2 cache,
you can overclock it to 400mhz with a BX 100 mhz motherboard. Alot more
bang for the buck. Now, how stable it runs at that speed from its 266
mhz setting is another question.
T. Bull wrote:
>
> Question:
>
> If the Celeron has no cache to slow down its overclocking......does a
> Pentium 2 with the L2 cache disabled in the bios overclock more easily????
>
> HMMMM.......
>
> Bull
>
> Barry Betambeau wrote in message
> <892858208.6434.0...@news.demon.co.uk>...
> >
cqu...@iafrica.com
On Fri, 17 Apr 1998 15:54:28 +0200, "Antek S. Baranski"
<Antek.Baranski@"euro.lab.for.part.phys"cern.ch> wrote:
>2a) An EX-chipset board **SUCKS MAJOR** but very cheap! (Castrated BX chipset)
Nope, it's a castrated LX chipset. Not the best place to try 100MHz
base speed. If you are going to try Celeron at 100MHz, budget for a
real 100MHz BX mobo, pref not Intel (risk of overclock limit)
>2b) A BX-chipset board (the only choice I think) and not that much more
>expensive than an EX-board.
I'll believe that when I see it. LX still costs more than Socket 7,
and I don't expect BX to be cheaper. Even with incrimental upgrades
such as VX over FX, and TX over VX, Intel always priced higher for at
least 3-6 months... which takes us at least halfway out of Celeron's
window of relevance (Mendocino cometh?)
With Intel, there's no such thing as a free lunch. Intel has replaced
IBM as the godfathers of "functional pricing" and its R&D is geared to
enforcing this rather than developing better solutions for the user.
Peter McMillan
jmwi...@unity.ncsu.edu (John Michael Wilson) wrote:
>A little relevant information. Intel makes P-II 266's in both 0.35u
>and 0.25u.
For an uninitiated knucklehead like me, how does one tell the
difference when purchasing?
Cheers,
Peter.
To email me, change .com in my address to .au
Bonker
On Sat, 18 Apr 1998 08:34:12 GMT, cqu...@iafrica.com wrote:
>On Fri, 17 Apr 1998 15:54:28 +0200, "Antek S. Baranski"
><Antek.Baranski@"euro.lab.for.part.phys"cern.ch> wrote:
>
>>2a) An EX-chipset board **SUCKS MAJOR** but very cheap! (Castrated BX chipset)
>
>Nope, it's a castrated LX chipset. Not the best place to try 100MHz
>base speed. If you are going to try Celeron at 100MHz, budget for a
>real 100MHz BX mobo, pref not Intel (risk of overclock limit)
Hold on dude. For all we know, the cut-down EX could be the BEST place
to go at 100+ ...
Bonker.
Andreas Grossfurtner
the count <eglam...@angelfire.com> in comp.sys.ibm.pc.games.action,
Fri, 17 Apr 1998 12:47:08 -0400:
>I doubt it would top a 600 MHz Alpha-PC ;)
>Of course, you can't run Win95 on an AlphaPC, but you can run NT
>(a 64 bit native port of NT, that is :)
600 Mhz... I wonder whether WinNT compatible games would run on such a
machine...? ;-)
--
URs,
Andreas Grossfurtner / yc...@usa.net
ICQ: 3462412; PGP public key available
news.epix.net
Don't you think Intel will have thought of all these things? They are not
as dumb as many think (just a little slow! it did take them YEARS to put any
overclocking prodoction on chip!
Overclocking:
well, hmmmm, let's see, from what I've heard recently no new P2 chips
are without the O/C protection.
Board:
According to ALL reliable sources (Tom is ok, but he does not have the
connections with the inside that most mags do, he does good reviews, be he
is far from being right about everything) sources I have read, the chip will
not be able to fit in a standard P2 board, mainly so you can't stick it in a
BX board and pass it off as a true P2! All the test versions are running on
standard P2 boards for the simple reason that there are NO EX boards
available yet!!!! They are the same chip in all but cache, so they will
run on a standard board, but all it takes to change is key a part of the
chip, an, wam, you can't fit it anymore, no big deal to change before the
production run! All of the chips that have been reviewed are pre-production,
remember that!
Mendocino:
If Intel is contimplating a next generation chip the SAME YEAR they
must know that something is wrong!!
The real reason for the Celeron is so they can shut down socket 7 production
forever, no other, it is not so you can get a quick fast PC, they just want
to end support for the socket 7 as quick as possible, we all know that Intel
wants to kill 7 as soon as they can!
I would love to hear any responses you might have (from the number of posts
this is obviously a hot topic) I enjoy hearing what people think the market
will do. but I won't be on the group for a couple of days, could you please
also forward any responses to vont...@usa.net
Tom von Trott
-J wrote in message <3537b123....@news.whro.org>...
>On Fri, 17 Apr 1998 17:13:34 GMT, Kas...@spamsucker.discopy.com (Kasey
>Chang (fix address before replying to me!)) wrote:
>
>>On Fri, 17 Apr 1998 09:08:44 GMT, "Victor Healey"
>><fourphun....@bellsouth.net> wrote:
>>
>>>Run - don't walk to
>>>http://www.tomshardware.com/cpubench.html
>>>Look for section under updated chip family review.
>>
>>Absolutely BS, according to the PC World preview. The folowing points
>>where offered...
>>
>>1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
>>into a M/B designed for Celeron.
>
>I didnt know this... could be an interesting point. Is this true?
>
>>
>>2) Celeron is SLOWER than all its "cheap" clone competitors such as
>>the AMD K6-233 or the Cyrix 6X86MX. It has NO onboard secondary cache.
>
>So overlcock it to 400MHz and blow them all away.
>
>>
>>
>>3) In business apps testing, Celeron is SLOWER THAN MOST Pentium-200
>>MMX's, despite its 266 MHz clock speed.
>
>see above
>
>>
>>4) Intel is already designing Celeron's follow-up, codename Mendocino,
>>which should debut before end of the year.
>>
>>Conclusion (mine, not PC World): any one who buys a Celeron-based PC
>>is a dumba**.
>>
>
>I will trust Toms Hardware Guide over PC World *any* day of the week.
>Not to mention that PC Week had an early version of the chip, *PLUS*
>PC Week did not overclock it to 400MHz with a 100MHz bus.
>
>Enjoy,
>
>-J
Antek S. Baranski
cqu...@iafrica.com wrote:
> On Fri, 17 Apr 1998 15:54:28 +0200, "Antek S. Baranski"
> <Antek.Baranski@"euro.lab.for.part.phys"cern.ch> wrote:
>
> >2a) An EX-chipset board **SUCKS MAJOR** but very cheap! (Castrated BX chipset)
>
> Nope, it's a castrated LX chipset. Not the best place to try 100MHz
> base speed. If you are going to try Celeron at 100MHz, budget for a
> real 100MHz BX mobo, pref not Intel (risk of overclock limit)
Uhm...... LX cannot run at 100 MHz! The only chipsets in the market at present (for
PII & co.) that can run 100 MHz are BX and EX and nothing else.
> >2b) A BX-chipset board (the only choice I think) and not that much more
> >expensive than an EX-board.
>
> I'll believe that when I see it. LX still costs more than Socket 7,
> and I don't expect BX to be cheaper. Even with incrimental upgrades
> such as VX over FX, and TX over VX, Intel always priced higher for at
> least 3-6 months... which takes us at least halfway out of Celeron's
> window of relevance (Mendocino cometh?)
Have a look at: http://www.pricewatch.com/ CPU -> motherboard -> 440 BX $ 165.- for
a BRAND NEW motherboard is not a lot of money!!!
> With Intel, there's no such thing as a free lunch. Intel has replaced
> IBM as the godfathers of "functional pricing" and its R&D is geared to
> enforcing this rather than developing better solutions for the user.
I agree.... people should be warned! I think that INTeL must have some plan
(eliminate the competition or at least have them licence Slot 1).
Cheers,
--
ing. Antek S. Baranski
European Laboratory for Particle Physics (CERN) - IT division - ASD group
Tel: +41 - (0)22 - 76 742 47 Fax: +41 - (0)22 - 76 786 30 ICQ: 1127269
REMOVE THE "bla bla" to-reply
Antek S. Baranski
steevo wrote:
> celeron has no case, its like a naked p2..saves manufacturing costs
Exactly the Celeron has no fancy case around it like the *real* PII has.
> --
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> steevo
Antek S. Baranski
Robato Yao wrote:
> In <MPG.fa157bff...@news.atl.bellsouth.net>, fourphun....@bellsouth.net (Victor Healey) writes:
>
> <SNIP>
>
> Does the 440EX chipset support 100MHz? That's another question, when
> you consider that the socket 7 MBs now have them.
Yes the 440EX is a castrated 440BX thus 100MHz functional.
> Rgds,
>
> Chris
Antek S. Baranski
Jason Petersen wrote:
> Andrew wrote:
> >
> > Why would you want a P-II without its level 2 cache.??????
> >
> > Its like buying a Porsche and putting a Ford engine in it.
>
> So you think that the absence of L2 cache will make it faster?? :p
AOL here we come :-))
Look and wheep: http://sysdoc.pair.com/cpubench.html
> -
> Jason
> http://members.aol.com/pete6306
Barry Betambeau
T. Bull wrote in message <35380...@news.snowhill.com>...
>Question:
>
>If the Celeron has no cache to slow down its overclocking......does a
>Pentium 2 with the L2 cache disabled in the bios overclock more easily????
>
>HMMMM.......
Dunno, problem here is that its already going as fast as my m/b(LX) can be
set!
With or without L2 enabled.
Other reports I've read suggest that systems are clock-locking at 400 Mhz on
BX chip sets,
but I believe that can be diasabled.
Barry
Antek S. Baranski
Antek S. Baranski wrote:
> cqu...@iafrica.com wrote:
>
> > On Fri, 17 Apr 1998 15:54:28 +0200, "Antek S. Baranski"
> > <Antek.Baranski@"euro.lab.for.part.phys"cern.ch> wrote:
> >
> > >2a) An EX-chipset board **SUCKS MAJOR** but very cheap! (Castrated BX chipset)
> >
> > Nope, it's a castrated LX chipset. Not the best place to try 100MHz
> > base speed. If you are going to try Celeron at 100MHz, budget for a
> > real 100MHz BX mobo, pref not Intel (risk of overclock limit)
>
> Uhm...... LX cannot run at 100 MHz! The only chipsets in the market at present (for
> PII & co.) that can run 100 MHz are BX and EX and nothing else.
Oops...... Seems like I made a mistake up there!!!! SORRY!!!! The EX is a castrated LX
board offcourse because the Celeron is designed for 266 / 66 running and nothing
else... SORRY for this confusion!!!!
Cheers and sorry again for the mistake!!! :-))
Tony Skinner
On Fri, 17 Apr 1998 19:47:06 GMT, jho...@removemewhro.org (-J) wrote:
>>Absolutely BS, according to the PC World preview. The folowing points
>>where offered...
>>
>>1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
>>into a M/B designed for Celeron.
>
>I didnt know this... could be an interesting point. Is this true?
I read this as well. It seems valid to me. If they allowed the Celeron
to work in all Pentium II boards, it would seem that someone could
easily build a system and deem it a Pentium II - 266MHz system rather
than a Celeron system. Windows 95 would probably report it as a
Pentium II or Pentium Pro. The only way they could distinguish the
chip is if they recognized the poor performance characteristics of the
chip or used a CPU identification program which showed it to be a
Celeron or listed the processor has having no level 2 cache.
>
>>
>>2) Celeron is SLOWER than all its "cheap" clone competitors such as
>>the AMD K6-233 or the Cyrix 6X86MX. It has NO onboard secondary cache.
>
>So overlcock it to 400MHz and blow them all away.
>
Overclocking to 400MHz will not show a significant increase in Office
applications. The main memory will still be completely uncached. The
only reason Quake and Quake II still benefit without the level 2 cache
is due to way in which Quake uses the FPU (addressing Tom's results
with Quake II and the Celeron).
>>
>>
>>3) In business apps testing, Celeron is SLOWER THAN MOST Pentium-200
>>MMX's, despite its 266 MHz clock speed.
>
>see above
* see above *
>
>>
>>4) Intel is already designing Celeron's follow-up, codename Mendocino,
>>which should debut before end of the year.
>>
>>Conclusion (mine, not PC World): any one who buys a Celeron-based PC
>>is a dumba**.
>>
>
>I will trust Toms Hardware Guide over PC World *any* day of the week.
>Not to mention that PC Week had an early version of the chip, *PLUS*
>PC Week did not overclock it to 400MHz with a 100MHz bus.
Well, go ahead and buy a Celeron. It's the consumers choice, I just
hope your completely happy with your purchase. Personally I wouldn't
want one because I use my PC for a wide variation of tasks from gaming
to CAD work.
Tony Skinner
Herbavor
In article <6h8ahh$j4g$1...@news1-alterdial.uu.net>, jsm...@company.com says...
>Sorry, I have to take issue with this. I have built some reliable, screaming
>Ford engines putting out 1 hp/cu.in or more (though I am at heart a Chevy
>man, not a Fordnatic. Had to clarify this ;) ) How does a 500 hp. 460 with
>over 550 lbs/ft of torque sound, all within a useable RPM range? Try getting
>that level of everyday reliable performance out of a Porsche for even 2x the
>price!
Ever hear of what Powerhaus, Turbo Performance Center, or any of the other
Porsche aftermarket shops do to Porsches? How about DAILY DRIVEN 600-750
horsepower 911's? that's with air conditioning on them too. Best of all
they don't get overheated in stop and go traffic. And don't talk about
reliability.
As far as long term reliability goes, Porsche leads any US or Asian auto
manufacturer in long term reliability. Over 85% of all Porsches that were
ever made that haven't been totaled in crashes or stolen are still in good
enough condition to drive on a daily basis and a lot of them are still driven
on a regular basis. That info is from auto registrations in the US and
Europe. Look at endurance racing. Who has the most wins in the major
endurance races? Porsche of course. Over 1/2 of the 24 hrs of LeMans that
Porsche has entered since Porsche first entered have been won by Porsche.
Since the early 1990's several stock 911's and 993's have been driven to the
24 hours of Daytona and competed in the race and finished in the top 13
overall and the top 10 in their class. The only mods they have had were the
ones required to meet the safety rules of the race, roll cages, racing
approved fuel tanks, approved size wheels and tires, and a fire extinguisher
mounted in the car. The only cars that finished in higher places were cars
that weren't street legal and weren't even close to their street going
counterparts.
And another thing about horsepower, what good is horsepower if you have to
use a lot of it to move around a heavy car? I have owned 2 Fords and 1
Porsche. I am a big fan of Chavy and Ford, but in everything but drag
racing, horsepower numbers don't impress me too much. I can do laps on
Watking Glen in my modified 914-4 (1900 lbs and 130-140 HP) that are within a
few seconds of the newest Corvettes. I know many 914-6 owners that have
daily driven 914-6 2.7's that weigh 2100 lbs and the engines produce 250-270
horsepower. They out accelerate Corvettes, Mustangs, and Camaros from 0-60
and in the 1/4 mile and also have no trouble embarassing them in an autoX
race or on a race track. A 914-6 with a 2.7 costs a little more than a
Camaro, but a lot less than a Camaro.
>Or a 320 hp. 5.0 (not even close to its potential)?
Yea but much more horsepower than that form a 5.0 requires internal mods that
can get costly if you want it in a daily driver. I know that moding Porsche
engines is expensive, but when you compare it to the price of the car, its
cheap and the ratio of the price of the mods compared to the price of the car
is the same as a Mustang or Camaro, or even less in a lot of cases.
>It could even be
>made to fit in a Porsche, but why go thru the trouble and devaluation of
>that overpriced sports car, except if you want a more reliable
see the comments about reliability above.
>and
>potentially as if not more powerful engine? There is a 351C (yep! Ford!) in
>the Pantera. BTW, with turbocharging, even the lowly Pinto engine can put
>out 300+ *reliable* horsepower. (still a Pinto, though...) T
I do like the kits to put a Chevy 350 or 383 in a 914 and use it for track
racing and drag racing. Its cheap horsepower, but throws off the balance of
the car.
In closing I'd like to say why compare Porsches anf Fords or Chevys? They
are aimed at different markets and have different purposes most of the time.
I drive a ford as my daily driver, but race with my Porsche. I'd love to buy
a 993 turbo for a daily driver, but the insurance would kill me and be more
than the car payments.
'bavor
>Andrew wrote in message <6h7bm6$5...@news5-gui.server.cableol.net>...
>>Why would you want a P-II without its level 2 cache.??????
x...@x.org
> Why would you want a P-II without its level 2 cache.??????
>
> Its like buying a Porsche and putting a Ford engine in it.
>
> Senior Technical Specialist
> Fujitsu (ICL) Ltd
>
>
>
because you can clock it to 400/100, giving pentium2 300-350 speed.
shows how much you know mr technical specialist ;)
--
40% of Usenet traffic is spam
40% spam cancels
20% .....discussion
http://spam.abuse.net
x...@x.org
> On 17 Apr 1998 08:44:03 -0400, ssto...@freenet.columbus.oh.us (Samuel
> Stoddard) wrote:
>
> >: Why would you want a P-II without its level 2 cache.??????
> >
> >The big reason is the overclockablility. 400mhz w/ no problems! Hell yea
>
> And the P2 isn't overclockable???
its more overclockable because it has no cache no hold it
back. 266-400. try clocking a p2-266 at 400.
x...@x.org
> When I've tried this, I've seen actual speed of application performance drop
> by 60-75%, depending on the app. Games tend to take a bigger hit than
> business apps. So your overclocked Celeron running at 450 Mhz would
> probably run at about half the speed of a normally clocked full-fledged P-II
> at 233 Mhz.
'probably'. in other words this is a pure speculation.
>
> You'd be better of getting a cheap socket 7 motherboard and a Cyrix
6x86M2
> 200Mhz chip. It would run as fast, wouldn't burn out, and would
cost about
> 60% as much as your Celeron/Slot I motherboard combo.
are you fucking stupid? read the topic. look at the benchmarks. cyrix
is shit for fpu, celeron better than mmx and k6. celeron gives 80fps
in quake 2.
> Thats probably what the OEM's will decide, too. They'll put out
Celeron
> models for the rubes unwilling to shell out the dough for a real PII
machine,
> but willing to pay an extra $100-200 for a machine endorsed by
breakdancing
> idiots in bunny suits. But the Celeron still can't beat a socket 7
chip/MB
> combo from AMD or Cyrix. And the AMD K6 really is perfectly
adequate for most
it can and does, empatically. look at the goddamn benchmarks before
you ponificate any further on this subject.
x...@x.org
> On Fri, 17 Apr 1998 09:08:44 GMT, "Victor Healey"
> <fourphun....@bellsouth.net> wrote:
>
> >Run - don't walk to
> >http://www.tomshardware.com/cpubench.html
> >Look for section under updated chip family review.
>
> where offered...
>
> 1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
> into a M/B designed for Celeron.
>
> the AMD K6-233 or the Cyrix 6X86MX. It has NO onboard secondary cache.
>
>
> MMX's, despite its 266 MHz clock speed.
>
> which should debut before end of the year.
>
> Conclusion (mine, not PC World): any one who buys a Celeron-based PC
> is a dumba**.
conclusion 2: people who dont read the topic are dumbasses.
Nanker Phelge
Peter McMillan wrote:
>
> jmwi...@unity.ncsu.edu (John Michael Wilson) wrote:
>
> >A little relevant information. Intel makes P-II 266's in both 0.35u
> >and 0.25u.
>
> For an uninitiated knucklehead like me, how does one tell the
> difference when purchasing?
>
> Cheers,
> Peter.
I think that poster is wrong, unless he's talking laptop computers.
First Deschutes PII (0.25) chip for desktop was the 333. Intel started
0.25 process with the Tillamook notebook chips, which where PentiumMMX
chips at 233/266. Pentium II notebook CPU's, which are just out, are
0.25 also, so there's a PII 266 at 0.25 for notebooks, but not the
desktop chips, those are 0.35 .
Wolfe
> >>
> >>1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
> >>into a M/B designed for Celeron.
> >
>
> I read this as well. It seems valid to me. If they allowed the Celeron
> to work in all Pentium II boards, it would seem that someone could
> easily build a system and deem it a Pentium II - 266MHz system rather
> than a Celeron system. Windows 95 would probably report it as a
> Pentium II or Pentium Pro. The only way they could distinguish the
> chip is if they recognized the poor performance characteristics of the
> chip or used a CPU identification program which showed it to be a
> Celeron or listed the processor has having no level 2 cache.
>
According to owners of the Intel AL440LX motherboard, BIOS 8 adds proper
identification for the Celeron processor. So it's atleast compatible
with some boards.
Although there may be physical compatibility issues with some boards.
From the Intel datasheet:
"The Intel Celeron processor requires a new retention mechanism.
This retention mechanism may require motherboard hole dimensions
to be increased to 0.159" diameter holes if low cost plastic
fasteners are used to secure the retention mechanisms in place.
The larger diameter holes are necessary to provide a robust
structural design that can guarantee stringent shock and vibe
testing. If captive nuts are used in place of the plastic
fasteners, then either the 0.159" or the 0.140" diameter holes
will suffice as long as the corresponding sized RMAM is used."
- Chris
--
To email, replace 'nospam' with 'erols'. (mailto:ch...@erols.com)
http://www.erols.com/chare/ - PC Hardware Links. CPU Upgrade FAQ.
MB
Peter McMillan wrote in message <353874f...@news.uq.edu.au>...
>jmwi...@unity.ncsu.edu (John Michael Wilson) wrote:
>
>>A little relevant information. Intel makes P-II 266's in both 0.35u
>>and 0.25u.
>
>For an uninitiated knucklehead like me, how does one tell the
>difference when purchasing?
>
>Cheers,
>Peter.
>
I think the above poster is mistaken. The newest PII300 (100mhz front side)
and higher CPUs are based on .25. The 266 is still on .35. The only way
you will know is because thats all they are made with.
max...@ix.netcom.com
On Sat, 18 Apr 1998 20:41:05 GMT, jea...@NOSPAMaccessus.net (Tony
Skinner) wrote:
>On Fri, 17 Apr 1998 19:47:06 GMT, jho...@removemewhro.org (-J) wrote:
>
>>>Absolutely BS, according to the PC World preview. The folowing points
>>>where offered...
>>>
>>>1) While Celeron is a Slot 1 CPU, you CANNOT fit a regular Pentium II
>>>into a M/B designed for Celeron.
>>
>>I didnt know this... could be an interesting point. Is this true?
>
>I read this as well. It seems valid to me. If they allowed the Celeron
>to work in all Pentium II boards, it would seem that someone could
>easily build a system and deem it a Pentium II - 266MHz system rather
>than a Celeron system. Windows 95 would probably report it as a
>Pentium II or Pentium Pro. The only way they could distinguish the
>chip is if they recognized the poor performance characteristics of the
>chip or used a CPU identification program which showed it to be a
>Celeron or listed the processor has having no level 2 cache.
>>
>>>
>>>2) Celeron is SLOWER than all its "cheap" clone competitors such as
>>>the AMD K6-233 or the Cyrix 6X86MX. It has NO onboard secondary cache.
>>
>>So overlcock it to 400MHz and blow them all away.
>>
>Overclocking to 400MHz will not show a significant increase in Office
>applications. The main memory will still be completely uncached. The
>only reason Quake and Quake II still benefit without the level 2 cache
>is due to way in which Quake uses the FPU (addressing Tom's results
>with Quake II and the Celeron).
>
>>>
>>>
>>>3) In business apps testing, Celeron is SLOWER THAN MOST Pentium-200
>>>MMX's, despite its 266 MHz clock speed.
>>
>>see above
>
>* see above *
>
>>
>>>
>>>4) Intel is already designing Celeron's follow-up, codename Mendocino,
>>>which should debut before end of the year.
>>>
>>>Conclusion (mine, not PC World): any one who buys a Celeron-based PC
>>>is a dumba**.
>>>
>>
>>I will trust Toms Hardware Guide over PC World *any* day of the week.
>>Not to mention that PC Week had an early version of the chip, *PLUS*
>>PC Week did not overclock it to 400MHz with a 100MHz bus.
>
>Well, go ahead and buy a Celeron. It's the consumers choice, I just
>hope your completely happy with your purchase. Personally I wouldn't
>want one because I use my PC for a wide variation of tasks from gaming
>to CAD work.
>
>Tony Skinner
Quite frankly , I don't see why anyone would want a 266 period.
my 2 cents,
max
Peter McMillan
Nanker Phelge <ruki...@somewhere.in.the.world> wrote:
>I think that poster is wrong, unless he's talking laptop computers.
>First Deschutes PII (0.25) chip for desktop was the 333. Intel started
>0.25 process with the Tillamook notebook chips, which where PentiumMMX
>chips at 233/266. Pentium II notebook CPU's, which are just out, are
>0.25 also, so there's a PII 266 at 0.25 for notebooks, but not the
>desktop chips, those are 0.35 .
Thanks for the clarification. :)
Hartman
On Sat, 18 Apr 1998 08:33:49 GMT, cqu...@iafrica.com wrote:
>On Fri, 17 Apr 1998 14:20:06 +0200, "Antek S. Baranski"
><Antek.Baranski@"euro.lab.for.part.phys"cern.ch> wrote:
>
>>> Why would you want a P-II without its level 2 cache.??????
>
>>How about the price and speed? Have a look at
>>http://sysdoc.pair.com/cpubench.html and look at the GAMES results with
>>a Celeron running at 400 MHz (OVER CLOCKED!!!!)
>
>OK; makes sense. Instead of paying $$$$ for a PII-100MHz, drop in a
>Celeron, beat it to death, and just buy more as they burn out
>(consider it a consumable overhead, like toner cartridges) until the
>PII you wanted gets cheap enough to buy. Chances are good that you
>might get that on a single Celeron, if it lasts a year.
There are things called 'THERMOMETERS'
A lot of new motherboards will automatically throttle clock-cycles if
your fan malfunctions or the CPU starts getting too hot (and too hot
is hot). Then of course there are the ones you own right now, touch
the cpu after a few hours running, if too hot to touch, get bigger
fan, don't OC by as much.
Hartman
>Its like buying a Porsche and putting a Ford engine in it.
Porsche vs Cobra... GO
Roger Wong
jea...@NOSPAMaccessus.net (Tony Skinner) wrote:
>Overclocking to 400MHz will not show a significant increase in Office
>applications.
A 400mhz FPU will really cut my 3D rendering times.
Is it possible to fit 4 overclocked Celeron's on a single
motherboard? I have a vision of cheap rendering farms
dancing in my head.
Dirk Hufnagel
news.epix.net wrote:
>
> Don't you think Intel will have thought of all these things? They are not
> as dumb as many think (just a little slow! it did take them YEARS to put any
> overclocking prodoction on chip!
>
> Overclocking:
>
> well, hmmmm, let's see, from what I've heard recently no new P2 chips
> are without the O/C protection.
>
> Board:
> According to ALL reliable sources (Tom is ok, but he does not have the
> connections with the inside that most mags do, he does good reviews, be he
> is far from being right about everything) sources I have read, the chip will
> not be able to fit in a standard P2 board, mainly so you can't stick it in a
> BX board and pass it off as a true P2! All the test versions are running on
> standard P2 boards for the simple reason that there are NO EX boards
> available yet!!!! They are the same chip in all but cache, so they will
> run on a standard board, but all it takes to change is key a part of the
> chip, an, wam, you can't fit it anymore, no big deal to change before the
> production run! All of the chips that have been reviewed are pre-production,
> remember that!
>
According to the c't test in issue 08/98, the Celeron won't fit in a normal P2 board.
translated from German:
"
The Celeron fits in the Slot 1 of the Pentium II, finds
however no more stop in the so far usual guide rails. Also
the normal Pentium II heat sinks are not suitable for
the new processor. Thus the cheap CPU needs both a new
heat sink and a new guide rail - these then certainly no
more Pentium II fits in.
"
If you check Tyans and Abits website, they claim P2 and Celeron support
for their new BX motherboards. We have to wait and see if you can
really put a normal Celeron CPU on these boards.
dirk
cycho
A. Tom does not do good reviews. Tom is biased and narrow minded, But
then so am I.
B. an overclocked piece of shit is still a piece of shit.
c. Intel IS as dumb as you think!
d. A cheap slot 1 cpu is still more costly than a Socket 7, the next
generation of socket 7 chips will fit on current MB's and unless they
have on chip L2 cache a simple bios flash will suffice to upgrade, to go
from Celeron 1 to Celeron 2(Mendocino, which will be no match for
Cayenne and K6 3d) you will definately have to upgrade the MB, this is
Intel we are talking about.
e. Not buying Celeron will send the message to Intel that they can't
just sell the easiest thing for them to make, the question of they're
profit margin slipping has arisen and I find it amusing that people
think this chip is somehow less profitable for Intel, if they sell one
they make money because they sold none in that makret prior to Celeron,
and it is simply a crippled PII, no research, minimal manufaccturing
chamngees and viola, a marketing gimmick ids born. I wouldn't use one in
a Tamagachi(sic).
Antek S. Baranski
cycho wrote:
>
> <SNIP>
>
> A. Tom does not do good reviews. Tom is biased and narrow minded, But
> then so am I.
Huh? Would you care to explain why you find Tom's reviews biased and narrow minded?
> B. an overclocked piece of shit is still a piece of shit.
True.
> c. Intel IS as dumb as you think!
INTeL and dumb? I think they are *THE* marketing machine in the world... after Microsoft
perhaps. :-)
> d. A cheap slot 1 cpu is still more costly than a Socket 7, the next
> generation of socket 7 chips will fit on current MB's and unless they
> have on chip L2 cache a simple bios flash will suffice to upgrade, to go
> from Celeron 1 to Celeron 2(Mendocino, which will be no match for
> Cayenne and K6 3d) you will definately have to upgrade the MB, this is
> Intel we are talking about.
Correct INTeL won't do anything un less they can get money ouf of it.
> e. Not buying Celeron will send the message to Intel that they can't
> just sell the easiest thing for them to make, the question of they're
> profit margin slipping has arisen and I find it amusing that people
> think this chip is somehow less profitable for Intel, if they sell one
> they make money because they sold none in that makret prior to Celeron,
> and it is simply a crippled PII, no research, minimal manufaccturing
> chamngees and viola, a marketing gimmick ids born. I wouldn't use one in
> a Tamagachi(sic).
Why not use it for your gaming machine? Nothing wrong with the CPU for the market INTeL
released it into, the gaming market. If INTeL would have sold the Celeron as *THE* office
CPU then nobody would buy them but now for games, its the perfect thing. And I think that
AMD, IBM etc. should get their act together.....
just my $0.02
Antek S. Baranski
Anthony Hoelscher wrote:
> I see Tom has this newsgroup creaming in its pants at the prospect of
> running the Celeron at 400MHz. But here's something to think about:
>
> A BX motherboard costs about $60 more than an LX motherboard. A BX
> motherbaord will need more expensive RAM than an LX motherboard. If you
> were to use that $60 to buy a better CPU rather than buying a new
> motherboard, you could get a real Pentium II 266 instead of the lobotomized
> Celeron 266. (A Celeron is $187, a real P II 266 is $250.) And then you
> can overclock that P2 266 up to 333MHz by setting your bus at 83MHz.
> According to Tom's own benchmarks, a P2 333MHz with cache is only 2.5 fps
> difference from a 400MHz Celeron.
Funny.. just a small question have you tried overclocking a PII-266 lately? :-)
I have 7 PII-266 system, 5 were bought in July 1997 and run 300 or more.....
flawlessly on the other hand the latest (March 1998) addition of 2 PII-266 will
either not boot at all or give screen corruption when clocked at 300 MHz. All 7
systems are identical ASUS KN97X (rev 1.02 or 1.03), 160MB EDO, Matrox Mil II
4MB.
So your above statement is only true when some has an _old_ PII-266 (Please
don't think that an PII-266 is OLD, its just that there are two versions out on
the market nowadays).
> And I'll bet the fact that Tom's benchmark is with a 66MHz bus means
> that the P II 266 overclocked to 333MHz is actually faster than a 400MHz
> Celeron.
>
> Not only that...When You're Not Playing Quake Your Other Applications Will
> Be Running at Full P2 333 Speeds, Where a Celeron Will Choke Big Time.
Do you really need the P2-333 performance for MS Word? For MSVC++ YES and other
real time applications where YOU the user has to wait for the PC, but in MS
Word the PC is waiting for you to type.
> So take a deep breath guys, step back, and re-analyze this situation.
But I do agree with you people shouldn't jump the gun.... ;-)
Cheers,
Antek S. Baranski
x...@x.org wrote:
>
> <SNIP>
>
>
> conclusion 2: people who dont read the topic are dumbasses.
If you don't have anything substantial to add to the thread would you
mind..... I saw one fruitfull reply from you and the rest were all kind
of....... #$#%@$#%^^*(#&*$#
> 40% of Usenet traffic is spam
> 40% spam cancels
> 20% .....discussion
In which of the above categories does your posting fit? ;-)