WARNING: DON'T BUY CYRIX. READ THIS NEWS!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Butz Yung
Jeff Terrill <jt...@Intrex.net> wrote in article
<3290F4...@Intrex.net>...
> Belmont wrote:
> >
> > In article <Pine.SOL.3.91.961113194256.3537E-100000-100000@sirius>,
> > gp...@herts.ac.uk says...
> > >
> > >> >We are talking abt QUAKE. Get it... QUAKE! Not Duke3D, not
Crusader:
> > No
> > >> >Regret, not DOOM...get it QUAKE QUAKE QUAKE QUAKE! I have a Cyrix
> > and
> > >> its
> > >> >fast, but with QUAKE IT SUCKS! GET IT???!!!
> > >> >
> > >> >Nuff said back.
> > >> >
> > >> How old do you think this Puerto Rican is? He sounds like a big boy
> > of 6
> > >> or 7??
> > >>
> > >>
> > >>
> > >Quake sucks. If you really want to play it why didn't you buy a
> > pentium
> > >in the first place.
> > Interesting statement..i dont know..he's an idiot so that might explain
> > it..
>
> Cyrix sux. It fries. Maybe with enough Zinc Oxide (thermal jelly) and a
> 4 inch heatsink, it'll run for a couple hours.
>
Yes Cyrix CPU do a bad job in QUAKE because QUAKE is a
FPU-intensive(very!) game. As you know, Cyrix FPU is worse than Pentium's
and so the performance will be poor. However, for other games it works fine
but still it cant work with its P-rating if the game use FPU much, and
thats what most 3D games are doing. However if you are using a 3D card the
situation will be much better as much rendering work is offloaded to the
video hardware instead of doing by the slow FPU. Without the FPU, Cyrix is
better than Pentium.
future
Actually, that's bullshit. A Cyrix P166+ performs about like a
Pentium 133 as far as floating point goes... and that is plenty fast
to run Quake at reasonable speed, even without a 3-D video card.
>However, for other games it works fine
>but still it cant work with its P-rating if the game use FPU much, and
>thats what most 3D games are doing.
So what? Even if doing floating point at the speed of a P133, that is
fast enough for any game on the market, as far as I'm concerned.
Note: I'm not a Cyrix user. I've just read up on it thoroughly.
Pete Olson
future (some...@internet.com) wrote:
: On 19 Nov 1996 16:46:24 GMT, "Butz Yung" <bu...@hkstar.com> wrote:
:
: Actually, that's bullshit. A Cyrix P166+ performs about like a
: Pentium 133 as far as floating point goes... and that is plenty fast
: to run Quake at reasonable speed, even without a 3-D video card.
: So what? Even if doing floating point at the speed of a P133, that is
: fast enough for any game on the market, as far as I'm concerned.
:
: Note: I'm not a Cyrix user. I've just read up on it thoroughly.
You are wrong. A Cyrix P166+ does not run Quake at speeds comparable
to a Pentium 133. It falls between the performance of a Pentium 75
and pentium 90, chips that Intel doesn't even sell anymore.
Carl Lang
Actually a Cyrix P166+ does 53 Mflops which is slightly better than a
Intel P90 doing 52 Mflops. (From Wintune 95). The Intger was better
than Intel P166.
Regards, Carl.
had to include extra padding to send mail
Jay Garcia
Pete Olson wrote:
>
> future (some...@internet.com) wrote:
> : On 19 Nov 1996 16:46:24 GMT, "Butz Yung" <bu...@hkstar.com> wrote:
> :
> : Actually, that's bullshit. A Cyrix P166+ performs about like a
> : Pentium 133 as far as floating point goes... and that is plenty fast
> : to run Quake at reasonable speed, even without a 3-D video card.
> :
> : So what? Even if doing floating point at the speed of a P133, that is
> : fast enough for any game on the market, as far as I'm concerned.
> :
> : Note: I'm not a Cyrix user. I've just read up on it thoroughly.
>
> You are wrong. A Cyrix P166+ does not run Quake at speeds comparable
> to a Pentium 133. It falls between the performance of a Pentium 75
> and pentium 90, chips that Intel doesn't even sell anymore.
Just bought 24 P75 chips and 12 P90's from my distributor this morning
and they know nothing about Intel stopping the sale of same .. What
gives ?? .. And YES, it is a MAJOR distributor ..
-Jay-
Pete Olson
Carl Lang (bd...@bedford.waii.com) wrote:
: On 20 Nov 1996, Pete Olson wrote:
:
: > A Cyrix P166+ does not run Quake at speeds comparable
: > to a Pentium 133. It falls between the performance of a Pentium 75
: > and pentium 90, chips that Intel doesn't even sell anymore.
:
: Actually a Cyrix P166+ does 53 Mflops which is slightly better than a
: than Intel P166.
Greg Furness
> So what? Even if doing floating point at the speed of a P133, that is
> fast enough for any game on the market, as far as I'm concerned.
>
> Note: I'm not a Cyrix user. I've just read up on it thoroughly.
You're wrong. I went from 13 to 20 FPS (640x480 - verite 1000) when I
traded in my lame Cyrix P-166+ for an Intel Pentium 133.
Check out: http://sysdoc.pair.com/cpu.html
dang...@sound.net
Maurizio Mansueto <maur...@ee.usyd.edu.au> wrote:
>snip
.
>Aside: all the P5 clones now use RISC architectures and emulate the CISC
>intructions, they foound it easier to copy (ops emulate) the P5 operation
>than to go CISC.
>snip
Cyrix and Pentium = CISC
AMD = RISC (Sort of)
Cyrix proudly proclaims they are pure old fashion Intel compatible
CISC tuned for better performance.
It don't matter what you call a dog with no legs.
He won't come.
Paul R. Erickson
On Wed, 20 Nov 1996 16:55:53 -0600, Jay Garcia <jga...@bigeasy.com>
wrote:ore.
>
>Just bought 24 P75 chips and 12 P90's from my distributor this morning
>and they know nothing about Intel stopping the sale of same .. What
>gives ?? .. And YES, it is a MAJOR distributor ..
Of course they're not gonna tell you buddy! They want you to buy!
They're not even MADE anymore and havent been for what, 6 months? So
either its old, old inventory or they are recycling.
________________________
____Paul R. Erickson____
__ARCLIGHT ENTERPRISES INC(DE)__
++++++++++++++++++++++++++++++++
Univ. of Texas/Austin
Graduate School of Business
MBA 1998 Marketing
"The meek shall inherit the earth only when
we're good and done with it, and with them."
-Fred Rexer, G.I.
Pete Olson
Jay Garcia (jga...@bigeasy.com) wrote:
: Pete Olson wrote:
: >
: > You are wrong. A Cyrix P166+ does not run Quake at speeds comparable
: > to a Pentium 133. It falls between the performance of a Pentium 75
: > and pentium 90, chips that Intel doesn't even sell anymore.
:
: and they know nothing about Intel stopping the sale of same .. What
: gives ?? .. And YES, it is a MAJOR distributor ..
channel. If you have access to Intel's price lists (min 1000 lots)
you'll see that the 75 and 90 do not appear and have not for months.
Intel does not sell Pentiums slower than 100 MHz.
Maurizio Mansueto
If they actually FULL support the Cyrix and AMD chips then these fellas
really fly... faster than their equivalent intel products.
Aside: all the P5 clones now use RISC architectures and emulate the CISC
intructions, they foound it easier to copy (ops emulate) the P5 operation
than to go CISC.
u'll get a reduced performance, and thus the bitching about why intel
chips are better/faster etc...
I figure go with the flow and by what u fell comfortable with:
i.e: do the cost/ reliability/ hassle/ equation and figure whatever rocks
your boat.
sTiMPy
Josh
Pete Olson (nor...@bigdipper.umd.edu) wrote:
: You are wrong. A Cyrix P166+ does not run Quake at speeds comparable
: to a Pentium 133. It falls between the performance of a Pentium 75
: and pentium 90, chips that Intel doesn't even sell anymore.
To put it politely, you're full of shit. In this room sits a Cyrix 6x86
P166+, 32 megs EDO, Trident 9440 2mb PCI. In the other room sits a
Pentium 90, 16 megs EDO, Trident 9440 2mb PCI.
I'm not much of a Quaker, but I do get a lot of enjoyment out of it and I
played it for a week or so on the P90 before I had the Cyrix.
_The_Cyrix_runs_it_faster_. Period. At any resolution. YES, I'm sure it
runs it slower than a Pentium 166 does, and maybe even a Pentium 133,
at higher resolutions. But it does -not- perform like a P90, end of
story.
Yes, FPU performance on the 6x86 blows. If I ran 3DStudio for a living
I'd have a PPro. But until A) Klamath proves itself worthy (likely) B)
M2 proves itself better than Klamath (marginally likely) or C) Dual PPro
200 motherboards become unpopular with the "gotta have it's" and get
cheaper (most likely), it's -quite- adaquete for me, and I bet the same
goes for a -lot- of other people.
And how many demanding FPU applications do you encounter, besides Quake?
3DS, Photoshop plug-ins, Painter plug-ins, etc. I run Photoshop and
Painter on a non-FPU 68040/25 Macintosh and they're plenty fast for
everything short of professional quality image manipulation. The Cyrix
would blow my poor little Mac away, retarded FPU or not.
One game is a silly reason to bash an entire processor, especially when
half the bashing isn't even the whole truth.
BTW - I've got a Matrox Mystique 4mb on the way for the Cyrix. Once the
D3D quake comes out, the FPU might not get taxed so heavily..
--
=====================================================================
Joshua J. Lowe - / - bo...@newworld.bridge.net/bo...@shadow.net
=====================================================================
Pete Olson
Josh (bo...@hyper.shadow.net) wrote:
: Pete Olson (nor...@bigdipper.umd.edu) wrote:
:
: : You are wrong. A Cyrix P166+ does not run Quake at speeds comparable
: : to a Pentium 133. It falls between the performance of a Pentium 75
: : and pentium 90, chips that Intel doesn't even sell anymore.
:
: To put it politely, you're full of shit. In this room sits a Cyrix 6x86
: P166+, 32 megs EDO, Trident 9440 2mb PCI. In the other room sits a
: Pentium 90, 16 megs EDO, Trident 9440 2mb PCI.
:
: I'm not much of a Quaker, but I do get a lot of enjoyment out of it and I
: played it for a week or so on the P90 before I had the Cyrix.
: _The_Cyrix_runs_it_faster_. Period. At any resolution. YES, I'm sure it
: runs it slower than a Pentium 166 does, and maybe even a Pentium 133,
: at higher resolutions. But it does -not- perform like a P90, end of
: story.
Guess again. Your impressions of how Quake runs don't exactly
impress me. Try running some Quake benchmarks on machines
that are identically configured except for the processor. Oh,
gee, what do you know, I've got exactly that right here:
Quake Benchmarks: Cyrix vs. Intel vs. AMD
All benchmarks are from the TIMEREFRESH console command. All I did to get
these frame rates was launch a new single player game, then execute
TIMEREFRESH from the starting position without moving at all.
HARDWARE:
Giga-Byte GA586ATS Motherboard
256K Pipelined-Burst Cache, 6ns
32MB 50ns EDO DRAM
Diamond Stealth Video VRAM PCI 3240 + MVP 2200, 4MB VRAM Installed
SOFTWARE:
HIMEM
Microsoft Mouse v10.0
ODI/IPX/SPX from NetWare v3.12
UNIVBE v5.1 Loaded for VESA 2.0 Support
Cyrix Intel Intel AMD
Vid Mode P166+ P5-133 P5-166 5k86-90
0 14.907466 23.825384 28.590713 16.504581
11 15.207925 26.763811 29.806556 16.613145
12 14.774399 25.238764 27.925444 15.981558
13 14.072603 24.387792 27.065876 14.793053
14 13.620988 22.867025 25.156868 14.208636
15 10.570709 17.461170 19.160210 10.083480
[courtesy H. Cardwell]
Also know that the 5k86 series has been changed in that the speeds
greater than 90 now have an even faster FPU, from what I have been
told (no personal experience on that one).
And yes, that is a motherboard specifically tested and approved
by Cyrix for the 6X86.
DOUGLAS HOFFMAN
On 20 Nov 1996 13:29:59 GMT, nor...@bigdipper.umd.edu (Pete Olson)
wrote:
>future (some...@internet.com) wrote:
>: On 19 Nov 1996 16:46:24 GMT, "Butz Yung" <bu...@hkstar.com> wrote:
>:
>: Actually, that's bullshit. A Cyrix P166+ performs about like a
>: Pentium 133 as far as floating point goes... and that is plenty fast
>: to run Quake at reasonable speed, even without a 3-D video card.
>:
>: So what? Even if doing floating point at the speed of a P133, that is
>: fast enough for any game on the market, as far as I'm concerned.
>:
>: Note: I'm not a Cyrix user. I've just read up on it thoroughly.
>
>You are wrong. A Cyrix P166+ does not run Quake at speeds comparable
>to a Pentium 133. It falls between the performance of a Pentium 75
>and pentium 90, chips that Intel doesn't even sell anymore.
>
Actually you are BOTH wrong! I have a P90 and a Cyrix 166+.The Cyrix
is faster with the same hardware and configuration.According to my
tests,it is equal to a P120 with Quake and much faster than a P133
in Duke 3D!!!
GSF
> >You're wrong. I went from 13 to 20 FPS (640x480 - vQuake) when I
> >traded in my lame Cyrix P-166+ for an Intel Pentium 133.
> person do ... Bottom line is that you've traded in a fast CPU for a Slower one.
Currently, I am using my computer mainly for Direct3D programming, so
actually I'm quite happy with the trade-in. Besides, I'm overclocking
it to 200Mhz ;). And yes, the FPS I gave you was when I was still
running it at 133.
Greg
Guillermo Gonzalez
On 25 Nov 1996 13:45:20 GMT, n...@ee.ed.ac.uk (Nick W Walton) wrote:
>
>We had a Cx166+, a iP133 and a iP166 all running Quake at the same time. The Cx166
>performed fairly well (with an 2Mb ET6000 video card). It got around the same perf-
>ormance as my iP133 with a 2Mb Diamond Stealth 64 DRAM. Surprisingly the iP166 with
>DStealth64 VRAM didn't blow our machines away.
Why is this surprising?? 166 Mhz is only about 25% faster than 133 Mhz...
> It did only marginally better, getting a 18% higher performance.
That sounds about right...
>The Cx166 however varied from game to game. It
>performed admirably, rivalling the iP166 in Duke and Descent, but did worse.
>when run with EarthSeige 2 and Mechwarrior 3.
The more FP a game uses, the slower the Cyrix will be in comparison to an Intel.
Aue
On Wed, 20 Nov 1996 17:30:08 -0600, Greg Furness <gfur...@mtco.com>
wrote:
>> So what? Even if doing floating point at the speed of a P133, that is
>> fast enough for any game on the market, as far as I'm concerned.
>>
>> Note: I'm not a Cyrix user. I've just read up on it thoroughly.
>
>You're wrong. I went from 13 to 20 FPS (640x480 - verite 1000) when I
>traded in my lame Cyrix P-166+ for an Intel Pentium 133.
>
>Check out: http://sysdoc.pair.com/cpu.html
with all due respect, buddy, that's the dumbest thing I've seen a
person do. I've had both cyrix p166+ and p166 and they are about
equal in most aspects. When i was running cyrix p166+. it was kicking
my friend's computer with p133 all over the gameland. That was when
cyrix first came out and they were selling 166+'s for $599 and intels
for $654. Had they lowered the price to today's price which is around
$200 for cyrix 166+ then, i would've kept cyrix no matter what.
I now have P166 however because It turned out my Mb couldn't not fully
support Cyrix.
Bottom line is that you've traded in a fast CPU for a Slower one.
MC
Josh
Paul R. Erickson (p...@mail.utexas.edu) wrote:
: Of course they're not gonna tell you buddy! They want you to buy!
: They're not even MADE anymore and havent been for what, 6 months? So
: either its old, old inventory or they are recycling.
I thought Intel doesn't stop making chips.
Tom LeTourneau
<SNIP>
> And how many demanding FPU applications do you encounter, besides Quake?
>
Try running ACAD 13 for Win95 on a Cyrix for a while! But then again maybe
it's not that FPU intensive :-)!
--
Later...
Tom L.
Warning: E-MAIL ADDRESS HAS BEEN ALTERED TO PREVENT UNWANTED MAIL!
REMOVE ASTERK FROM BEGINING AND END OF ADDRESS TO REPLY BY E-MAIL.
Tim Williams
GSF <gfur...@mtco.com> wrote:
>> >You're wrong. I went from 13 to 20 FPS (640x480 - vQuake) when I
>> >traded in my lame Cyrix P-166+ for an Intel Pentium 133.
>> with all due respect, buddy, that's the dumbest thing I've seen a
>> person do ... Bottom line is that you've traded in a fast CPU for a Slower one.
>
>Currently, I am using my computer mainly for Direct3D programming, so
>actually I'm quite happy with the trade-in. Besides, I'm overclocking
>it to 200Mhz ;). And yes, the FPS I gave you was when I was still
>running it at 133.
BS: I've yet to see a Pent. 133 that would oc to 200. Give it up;
Intel, Cyrix and AMD have NO love for any of you. They are out there
to market their products and make money.
Intel has always offered the "safe bet" (disegarding the buggy
Pentiums which they originally baulked on), AMD has been a little late
to the market with a cheaper/faster/better product than Intel, and
Cyrix.... well I have always thought their stuff acted flaky. But if
you are willing to risk heat problems (lockups and spontaneous
reboots) for a chance at perhaps a MUCH faster machine for the money,
then GOAT HEAD :)
Hell their just chips... not like a REAL argument (Chevy vs. Ford :)
Start this one:
Intel=Chevy
Cyrix=Ford
AMD=Dodge
(I'm an AMD/Dodge fan.)
Scott Lyon
DOUGLAS HOFFMAN wrote:
>
> On 20 Nov 1996 13:29:59 GMT, nor...@bigdipper.umd.edu (Pete Olson)
> wrote:
>
> >future (some...@internet.com) wrote:
> >: On 19 Nov 1996 16:46:24 GMT, "Butz Yung" <bu...@hkstar.com> wrote:
> >:
> >: Actually, that's bullshit. A Cyrix P166+ performs about like a
> >: Pentium 133 as far as floating point goes... and that is plenty fast
> >: to run Quake at reasonable speed, even without a 3-D video card.
> >:
> >: fast enough for any game on the market, as far as I'm concerned.
> >:
> >: Note: I'm not a Cyrix user. I've just read up on it thoroughly.
> >
> >to a Pentium 133. It falls between the performance of a Pentium 75
> >and pentium 90, chips that Intel doesn't even sell anymore.
> >
> Actually you are BOTH wrong! I have a P90 and a Cyrix 166+.The Cyrix
> is faster with the same hardware and configuration.According to my
> tests,it is equal to a P120 with Quake and much faster than a P133
> in Duke 3D!!!
Thats a Crock. I have a P100 and a Cyrix 133+....while the 133+ is about
40% faster in CPU speed, it's about 40% slower in FPU. Plugging them
both into an identical system (SuperMicro M/B, 32Meg, Diamond Stealth64
PCI) yielded the following results in Quake: P100 25.7 FPS, P133+ 15.2
FPS. The P133+ is impressive in windows and most other games, but it
reeks for Quake. So while its a great second computer and plenty fast
for my kids, you need to be honest about it's Quake preformance!
--
********************************************
sco...@cyberlynk.com
GAMERS CORNER -> http://www.cyberlynk.com/~scottl
Handle -> MUGGER. NET QUAKE Rules !!!!!
********************************************
Josh
Tom LeTourneau (*sta...@rangenet.com*) wrote:
: <SNIP>
: > And how many demanding FPU applications do you encounter, besides Quake?
: >
: <SNIP>
: Try running ACAD 13 for Win95 on a Cyrix for a while! But then again maybe
: it's not that FPU intensive :-)!
Thanks for taking my entire quote out of context. I notice you excluded
the REST of what I'd said, about the fact if you used 3DS or other FPU
intensive applications for work and productivity purposes, stay away from
the 6x86. I was referring to the silliness of all the GaMeZ KiDz who
bitch at an entire processor line because it can't run a bloated, resource
hogging game as well due to FPU limitations.
: --
: Later...
: Tom L.
--
Peter Tong
Greg Furness <gfur...@mtco.com wrote:
> You're wrong. I went from 13 to 20 FPS (640x480 - verite 1000) when I
> traded in my lame Cyrix P-166+ for an Intel Pentium 133.
Hard to believe that your 166+ did only 13. My cx5x86-120 does 12.3 on
the timerefresh test with as many cyrix advanced features turned on as
possible.
Peter Tong
Biostar 1433UIV/cx5x86-120
ROMAN TOLEDO
....Does anyone have REAL FPU Benchmarks on the Cyrix 166+ and the
....Intel P166??? The same goes for the 200+ vs the P200 I'm looking to
....buy a new MB and CPU but I'd like to have FACTUAL NUMBERS :-) not
....my cpu is better than yours!@#$%^&*. If anyone can point to
....some???(Winstone96???)on the Web please Email me Thanks....
....
.... Email to 72427...@compuserve.com thanks......
Harry Lee
Josh,
I can vouch for you on the Cyrix speed. I have both an Intel 100 and
a Cyrix P166+ as well. The Cyrix runs Quake a little faster, but I
attribute that to better components (The Pentium has FPM dram, the
Cyrix has EDO dram). Either way, it is definitely faster than an
Intel Pentium 90, and definitely very playable. I've passed it at all
levels and am trying to finish it in the secret nightmare level now.
Harry
On 21 Nov 1996 10:34:40 GMT, bo...@hyper.shadow.net (Josh) wrote:
>Pete Olson (nor...@bigdipper.umd.edu) wrote:
>
>: You are wrong. A Cyrix P166+ does not run Quake at speeds comparable
>: to a Pentium 133. It falls between the performance of a Pentium 75
>: and pentium 90, chips that Intel doesn't even sell anymore.
>
>To put it politely, you're full of shit. In this room sits a Cyrix 6x86
>P166+, 32 megs EDO, Trident 9440 2mb PCI. In the other room sits a
>Pentium 90, 16 megs EDO, Trident 9440 2mb PCI.
>
>I'm not much of a Quaker, but I do get a lot of enjoyment out of it and I
>played it for a week or so on the P90 before I had the Cyrix.
>_The_Cyrix_runs_it_faster_. Period. At any resolution. YES, I'm sure it
>runs it slower than a Pentium 166 does, and maybe even a Pentium 133,
>at higher resolutions. But it does -not- perform like a P90, end of
>story.
>
>Yes, FPU performance on the 6x86 blows. If I ran 3DStudio for a living
>I'd have a PPro. But until A) Klamath proves itself worthy (likely) B)
>M2 proves itself better than Klamath (marginally likely) or C) Dual PPro
>200 motherboards become unpopular with the "gotta have it's" and get
>cheaper (most likely), it's -quite- adaquete for me, and I bet the same
>goes for a -lot- of other people.
>
>And how many demanding FPU applications do you encounter, besides Quake?
>3DS, Photoshop plug-ins, Painter plug-ins, etc. I run Photoshop and
>Painter on a non-FPU 68040/25 Macintosh and they're plenty fast for
>everything short of professional quality image manipulation. The Cyrix
>would blow my poor little Mac away, retarded FPU or not.
>
>One game is a silly reason to bash an entire processor, especially when
>half the bashing isn't even the whole truth.
>
>BTW - I've got a Matrox Mystique 4mb on the way for the Cyrix. Once the
> D3D quake comes out, the FPU might not get taxed so heavily..
>
>
Nick W Walton
We had a Cx166+, a iP133 and a iP166 all running Quake at the same time. The Cx166
performed fairly well (with an 2Mb ET6000 video card). It got around the same perf-
ormance as my iP133 with a 2Mb Diamond Stealth 64 DRAM. Surprisingly the iP166 with
DStealth64 VRAM didn't blow our machines away. It did only marginally better,
getting a 18% higher performance. The Cx166 however varied from game to game. It
performed admirably, rivalling the iP166 in Duke and Descent, but did worse.
when run with EarthSeige 2 and Mechwarrior 3.
Honest punter opinion: + Good chip, excellent for Multimedia, kids games,
"family" computing, 90% of games, Word Processing.
- Intensive games, heavy number crunching, ray tracing,
rendering, high capacity apps. etc.
--
Nick Walton 1st year Postgraduate Nick....@ee.ed.ac.uk (work)
http://www.tardis.ed.ac.uk/~aardvark
Tom Connelly
Well said,
these little Quakers are jerks.
Cheers.
TC
In article <571b80$g...@viper.shadow.net>, bo...@hyper.shadow.net says...
GSF
Nick W Walton wrote:
>
> We had a Cx166+, a iP133 and a iP166 all running Quake at the same time. The Cx166
> performed fairly well (with an 2Mb ET6000 video card). It got around the same perf-
> ormance as my iP133 with a 2Mb Diamond Stealth 64 DRAM.
This sounds highly anecdotal. Did you perform any benchmarks, measure
FPS etc. The Cyrix has a poor FPU.
GSF
Shawn Albright
Earthsiege 2 and Mechwarrior 3(?) or 2 whatever anyway neither use FPU so
that shouldn't even be a factor.
--
Shawn Albright
sbr...@teleport.com
Shawn_A...@Planar.com
Guillermo Gonzalez <gonz...@bridge.net> wrote in article
<32946ef9...@news.bridge.net>...
> On 25 Nov 1996 13:45:20 GMT, n...@ee.ed.ac.uk (Nick W Walton) wrote:
>
> >
> >We had a Cx166+, a iP133 and a iP166 all running Quake at the same time.
The Cx166
> >performed fairly well (with an 2Mb ET6000 video card). It got around
the same perf-
> >ormance as my iP133 with a 2Mb Diamond Stealth 64 DRAM. Surprisingly
the iP166 with
> >DStealth64 VRAM didn't blow our machines away.
>
> Why is this surprising?? 166 Mhz is only about 25% faster than 133
Mhz...
>
> > It did only marginally better, getting a 18% higher performance.
>
> That sounds about right...
>
> >The Cx166 however varied from game to game. It
> >performed admirably, rivalling the iP166 in Duke and Descent, but did
worse.
> >when run with EarthSeige 2 and Mechwarrior 3.
>
Steven Brown
On Mon, 25 Nov 1996 19:32:40 -0600, GSF <gfur...@mtco.com> wrote:
>Nick W Walton wrote:
>>
>> We had a Cx166+, a iP133 and a iP166 all running Quake at the same time. The Cx166
>> performed fairly well (with an 2Mb ET6000 video card). It got around the same perf-
>> ormance as my iP133 with a 2Mb Diamond Stealth 64 DRAM.
>
>This sounds highly anecdotal. Did you perform any benchmarks, measure
>FPS etc. The Cyrix has a poor FPU.
>
>
>GSF
I'll take this sort of honest anecdote over most of the "my piece of
silicon's better than yours" sort of crap that usually occurs in here.
Steven Brown
Edinburgh
Scotland
Email: ste...@netcomuk.co.uk
sbr...@bigfoot.com
Scott Mathers
On Sat, 23 Nov 1996 13:03:59 GMT, ti...@evansville.net (Tim Williams)
wrote:
>
>Intel has always offered the "safe bet" (disegarding the buggy
>Pentiums which they originally baulked on), AMD has been a little late
>to the market with a cheaper/faster/better product than Intel, and
>Cyrix.... well I have always thought their stuff acted flaky. But if
>you are willing to risk heat problems (lockups and spontaneous
>reboots) for a chance at perhaps a MUCH faster machine for the money,
>then GOAT HEAD :)
>
>Hell their just chips... not like a REAL argument (Chevy vs. Ford :)
>
>Start this one:
> Intel=Chevy
> Cyrix=Ford
> AMD=Dodge
>(I'm an AMD/Dodge fan.)
Actually, the 6x86 is very stable, providing you have the right
motherboard.....
R. G.
smat...@vivanet.com (Scott Mathers) wrote:
>On Sat, 23 Nov 1996 13:03:59 GMT, ti...@evansville.net (Tim Williams)
>wrote:
>
>>
>>Intel has always offered the "safe bet" (disegarding the buggy
>>Pentiums which they originally baulked on), AMD has been a little late
>>to the market with a cheaper/faster/better product than Intel, and
>>Cyrix.... well I have always thought their stuff acted flaky. But if
>>you are willing to risk heat problems (lockups and spontaneous
>>reboots) for a chance at perhaps a MUCH faster machine for the money,
>>then GOAT HEAD :)
>>
>>Hell their just chips... not like a REAL argument (Chevy vs. Ford :)
>>
>>Start this one:
>> Intel=Chevy
>> Cyrix=Ford
>> AMD=Dodge
>>(I'm an AMD/Dodge fan.)
>
Yup, I have an overclocked Dodge Shadow - took the model with one
engine size larger than standard, runs a little hotter, but
doesn't leave me stuck beside the transport trucks.
Lee Rhodes
Has anybody had any experience with AMD K5 chips? I know that Microsoft
have given it their seal of approval but has anyone experienced any
problems with them Windows or otherwise? I can find precious little
information about them or how they perform against other manufacturers.
TIA,
Lee Rhodes
Regional Technical College Carlow
Ireland
daman
AMEN!!!
Thomas Mosher
and requires adequate cooling. One user had cooling problems and lockups
when operating the 133 at its 100MHz speed. The problems disappeared
when clocked down to 90 MHz. Using a good heatsink with heatsink grease
cleared the problem when running back at 100 MHz.
--
Thomas Mosher
tmo...@isource.net
Tim Williams
Thomas Mosher <tmo...@isource.net> wrote:
>Lee Rhodes wrote:
>>
>> Has anybody had any experience with AMD K5 chips? I know that Microsoft
>From what I've heard from my clients/associates, the AMD K5 133 runs hot
>and requires adequate cooling. One user had cooling problems and lockups
>when operating the 133 at its 100MHz speed. The problems disappeared
>when clocked down to 90 MHz. Using a good heatsink with heatsink grease
>cleared the problem when running back at 100 MHz.
K5's (at least the PR100 and PR133 -- which both run at 100 MHz clock
speed) run very cool. Each chip says "heatsink and fan required."
But I put those on ANY chip 50 MHz (486) and up... Fans are cheap
insurance.
I can pop the fan off my K5-133 now and the chip is barely warm.
Weird thing: I opened a Compaq server last week -- Pentium 166 with
only a heat sink, no fan. Stupid Compaq, real stupid.
Neer Deth
Wrong! You take the fan off that chip and it'll run for about 20 minutes
then lock up! Personal experience.
--
- May The Remains Of -
- The Dead Rise Once -
- Again To Walk In -
- Your Footsteps -
Cornelis Zijlstra
> K5's (at least the PR100 and PR133 -- which both run at 100 MHz clock
> speed) run very cool. Each chip says "heatsink and fan required."
> But I put those on ANY chip 50 MHz (486) and up... Fans are cheap
> insurance.
Last time i checked the amd site i took a look at the new pentium like
chips of AMD
They just announced the PR120 and the PR133
the jumpersettings for the PR133 were as follows..
clock multiplier at 2x
bus speed 66 mhz
this makes me believe the amd PR-133 actually runs at 133 MHz (I am a
multiplying genius)
I really dont know why people think the new AMD chips are on a different
clock speed..
sigh...
BTW. please correct me if i'm wrong..
Tom
The K5-PR133 doesn't care if you set the multiplier at 1.5x or 2x it still
runs at 100mhz internal clock. The original settings for the K5-PR133 was
66mhz bus 1.5x multiplier 3.52 volts. To try to help resellers using
motherboards that don't show AMD K5-PR133 in the manual AMD just released a
statement saying to set the bus and multiplier just like an intel pentium
133, either way the cpu runs the same , and although I tried to OC it by
going to a 3x multiplier the bios gave AMD K5-PR150 at 120mhz but the
system locked up after the pnp init completed on an asus p55tvp4 with 0103
bios.
tom
Jason Dean Malone
Cornelis Zijlstra wrote:
>
> > K5's (at least the PR100 and PR133 -- which both run at 100 MHz clock
> > speed) run very cool. Each chip says "heatsink and fan required."
> > But I put those on ANY chip 50 MHz (486) and up... Fans are cheap
> > insurance.
>
> chips of AMD
> They just announced the PR120 and the PR133
> the jumpersettings for the PR133 were as follows..
> clock multiplier at 2x
> bus speed 66 mhz
> this makes me believe the amd PR-133 actually runs at 133 MHz (I am a
> multiplying genius)
> I really dont know why people think the new AMD chips are on a different
> clock speed..
>
> sigh...
>
Okay, you're wrong. The PR-120 and 133 run at 90 and 100 MHz,
respectively. See the AMD 5k86 FAQ's.
bd
In article <32A0A5...@exis.net>, da...@exis.net wrote:
>Steven Brown wrote:
>>
>> On Mon, 25 Nov 1996 19:32:40 -0600, GSF <gfur...@mtco.com> wrote:
>Nick W Walton wrote:
>
>> We had a Cx166+, a iP133 and a iP166 all running Quake at the same time.
> The Cx166
>>performed fairly well (with an 2Mb ET6000 video card). It got around the same perf-
>>ormance as my iP133 with a 2Mb Diamond Stealth 64 DRAM.
>This sounds highly anecdotal. Did you perform any benchmarks, measure
>FPS etc. The Cyrix has a poor FPU.
I wouldn't call it poor.. Does a Pentium 100 have poor FPU performance? The
P166+ is about the same as P5/100 FPU yet for $170 the P166+ is signficantly
faster than many P5/166 stock systems in integer applications - a chip that
costs about $400. Cyrix also performs extremely well in cross FPU/integer
apps such as spreadsheets.
If you're into real benchmarks, then consider Winstone which uses real-world
apps.. Here are my Winstone marks that I posted in the past..
______________________________
Gigabyte 586HX (512kPB)
Cyrix P166+ Rev 2.6
32MB EDO RAM
Diamond Edge 3D 2000 video (a fast but not high end card)
Quantum LPS525S 1.6gb drive
Windows 95 (Service Pack 1)
USE MONOSPACED FONT
Results Winstone32 Winstone96
Gigabyte 586HX/512kPB Cyrix P166+ 112 79
Gigabyte 586HX/512kPB Pentium 133 93 67
Dell Dimension XPS P166s 98 80
Gateway 2000 P5-166XL 94 77
HP Vectra 500 Model 515MCx 5/166 85 65
Micron Millennia P166 97 78
Dell Dimension XPS P133c 90 68
DirectWave P200 MVP ProS 109 86
Gateway 2000 P5-200 XL 101 83
Micron Millennia P200 101 86
Zenon Z-Presentation P200 106 85
(32-bit) (16-bit)
The Gigabyte/P166+ kicked Pentium 166 butt in Winstone32 and even beat Pentium
200 stock systems and this is with a relatively slow video card. It did well
against Pentium in Winstone 96 (which is mostly 16bit apps). I don't have any
other Cyrix scores. Anyone else have Cyrix scores? This test is SYSTEM
tests, however, rather than CHIP tests.
The scores of my Gigabyte system would be much higher with a Matrox video
card. The Diamond Edge 2000 is not exactly one of the fastest boards yet the
P166+/Gigabyte system did extremely well. Based on scores I've seen of the
Edge card vs Matrox, the Winstone 32 could be over 120 and the Winstone 96
could be over 85 with a Matrox card on my system. Gotta get one...
If you're not familiar with ZD's Winstone benchmarks, they use common Windows
applications as a benchmark. The advantage is that you get a true application
test. The disadvantage is that it heavily depends on video speed so a
somewhat slow video card will drastically effect the results of an otherwise
fast system (as in my case). The only way to compare different components is
to replace them one by one on the same system an re-run the test with each.
Comparing components from many systems will not work with this type of test -
it can only compare an entire system.
As far as Cyrix and FPU goes, Cyrix does not have any FPU compatibility
problems and is not particularly slow. A P166+ at a 133mhz core speed
performs at about the level of a P5/90-133 depending on the FPU based app.
For under $180, the P166+ fares quite well against a P5/166 at about $400.
And yes, Quake runs fine on a P166+.. about the same feel as a P5/133 and
faster feel than a P5/100. The complaints are unwarranted as far as I'm
concerned.
The apps Winstone 32 uses are:
Business Graphics/DTP:
Adobe(R) PageMaker(R) 6.0
Corel Systems CorelDRAW!(TM) 6.0
Microsoft(R) PowerPoint(R) 7.0
Database:
Borland(R) Paradox(R) 7.0
Microsoft Access 7.0
Word Processing/Spreadsheet:
Microsoft Excel
Lotus(R) Word Pro(TM) 96
Microsoft Word 7.0
Winstone 96 uses:
Business Graphics/DTP: Adobe(TM) PageMaker(TM) 5.0a for Windows
Corel Corporation CorelDRAW!(R) 5.0E2
Microsoft PowerPoint(R) 4.0c for Windows
Database: Borland(R) dBASE(R) 5.0 for Windows
Borland Paradox(R) 5.0 for Windows
Microsoft Access(R) 2.0c for Windows
Microsoft Works 3.0b for Windows
Spreadsheet: Lotus(R) 1-2-3(R) Release 5 for Windows
Microsoft Excel 5.0c for Windows
Microsoft Works 3.0b for Windows
Novell(R) Quattro(R) Pro 6.01 for Windows
Word Processing: Lotus Ami Pro(R) Release 3.1 for Windows
Microsoft Word 6.0c for Windows
Microsoft Works 3.0b for Windows
Novell WordPerfect(R) 6.1 for Windows
I highly recommend you get this CD if you are curious how your SYSTEM
compares.
Bryon Wasserman
On Fri, 29 Nov 1996 11:42:04 +0000, Lee Rhodes <rho...@rtc-carlow.ie>
wrote:
>Has anybody had any experience with AMD K5 chips? I know that Microsoft
>have given it their seal of approval but has anyone experienced any
>problems with them Windows or otherwise? I can find precious little
>information about them or how they perform against other manufacturers.
>
>TIA,
>Lee Rhodes
>Regional Technical College Carlow
>Ireland
Sorry to add, not answer questions, but does anybody have any info on
the FPU speed of the K5 chips?
Thomas Mosher
Paul van de Graaf wrote:
>
> ti...@evansville.net (Tim Williams) wrote:
>
> >I can pop the fan off my K5-133 now and the chip is barely warm.
> >
> >Weird thing: I opened a Compaq server last week -- Pentium 166 with
> >only a heat sink, no fan. Stupid Compaq, real stupid.
>
> massive heat sink and decent cases with quality fans that blow air
> over the heat sink. This is the whole concept behind the ATX case.
>
> You can't make a decent fan the size of a CPU, and the space used to
> house the fan only makes the heat sink smaller. I've seen plenty of
> them fail... mainly on cheapo home-built/garage shop Pentium clones.
>
> I seem to recall that Intel has not advocated the use of CPU fans on
> Pentiums since the 3 volt parts came out. Not sure about the 60 Mhz
> and 66 Mhz 5 volt parts though. The only part I think they've put a
> fan on is the Pentium Overdrive chip, but that's such a dog it isn't
> even worth mentioning.
>
> Paul van de Graaf
That's interesting considering the retail (i.e., boxed) versions of the
Pentiums come with a heatsink/fan permanently bonded to the CPU. Now if
you want to see a heatsink, take a look at some of the older Macs, the
heatsink is the size of a calculator.
--
Thomas Mosher
tmo...@isource.net
Shawn Albright
Id say thats the problem then. Gotta have a special motherboard because
they just arent completely compatible. I see people defending these chips
(Cyrix) just because they are cheap. Thats a load of crap. Intel is the
safest most compatible bet.
Scott Mathers <smat...@vivanet.com> wrote in article
<329e7bd2...@news.idt.net>...
> On Sat, 23 Nov 1996 13:03:59 GMT, ti...@evansville.net (Tim Williams)
> wrote:
>
> >
> >Intel has always offered the "safe bet" (disegarding the buggy
> >Pentiums which they originally baulked on), AMD has been a little late
> >to the market with a cheaper/faster/better product than Intel, and
> >Cyrix.... well I have always thought their stuff acted flaky. But if
> >you are willing to risk heat problems (lockups and spontaneous
> >reboots) for a chance at perhaps a MUCH faster machine for the money,
> >then GOAT HEAD :)
> >
> >Hell their just chips... not like a REAL argument (Chevy vs. Ford :)
> >
> >Start this one:
> > Intel=Chevy
> > Cyrix=Ford
> > AMD=Dodge
> >(I'm an AMD/Dodge fan.)
>
Pete Olson
bd (r...@tyrell.net) wrote:
: In article <32A0A5...@exis.net>, da...@exis.net wrote:
: >Steven Brown wrote:
: >> On Mon, 25 Nov 1996 19:32:40 -0600, GSF <gfur...@mtco.com> wrote:
: >Nick W Walton wrote:
: >> We had a Cx166+, a iP133 and a iP166 all running Quake at the same time.
: > The Cx166
: >>performed fairly well (with an 2Mb ET6000 video card). It got around the same perf-
: >>ormance as my iP133 with a 2Mb Diamond Stealth 64 DRAM.
: >This sounds highly anecdotal. Did you perform any benchmarks, measure
: >FPS etc. The Cyrix has a poor FPU.
:
: If you're into real benchmarks, then consider Winstone which uses real-world
performance of the Cyrix 6X86 for Quake, a detail your
benchmarks have conveniently ignored. See the relevant
bechmarks below:
Quake Benchmarks: Cyrix vs. Intel vs. AMD
All benchmarks are from the TIMEREFRESH console command. All I did to get
these frame rates was launch a new single player game, then execute
TIMEREFRESH from the starting position without moving at all.
HARDWARE:
Giga-Byte GA586ATS Motherboard
256K Pipelined-Burst Cache, 6ns
32MB 50ns EDO DRAM
Diamond Stealth Video VRAM PCI 3240 + MVP 2200, 4MB VRAM Installed
SOFTWARE:
HIMEM
Microsoft Mouse v10.0
ODI/IPX/SPX from NetWare v3.12
UNIVBE v5.1 Loaded for VESA 2.0 Support
Cyrix Intel Intel AMD
Vid Mode P166+ P5-133 P5-166 5k86-90
0 14.907466 23.825384 28.590713 16.504581
11 15.207925 26.763811 29.806556 16.613145
12 14.774399 25.238764 27.925444 15.981558
13 14.072603 24.387792 27.065876 14.793053
14 13.620988 22.867025 25.156868 14.208636
15 10.570709 17.461170 19.160210 10.083480
[courtesy H. Cardwell]
: The scores of my Gigabyte system would be much higher with a Matrox video
^^^^
Hmm, sounds like someone has something to justify. I owned both a P150+
6X86 and a Pentium 166. In general the 6X86 is zippy, but it is simply not
competitive with Pentiums for Quake.
: And yes, Quake runs fine on a P166+.. about the same feel as a P5/133 and
: faster feel than a P5/100. The complaints are unwarranted as far as I'm
: concerned.
Interesting that as soon as you touch upon the software that actually
forms the subject of this thread, we here not the results of benchmarks,
but rather how it feels to you. That doesn't quite cut it. Nice try.
Paul van de Graaf
Pete Olson
Paul van de Graaf (pau...@mindspring.com) wrote:
:
: No. The majority of decent Pentium machines have no CPU fan, but a
: massive heat sink and decent cases with quality fans that blow air
: over the heat sink. This is the whole concept behind the ATX case.
: I seem to recall that Intel has not advocated the use of CPU fans on
: Pentiums since the 3 volt parts came out. Not sure about the 60 Mhz
: and 66 Mhz 5 volt parts though. The only part I think they've put a
: fan on is the Pentium Overdrive chip, but that's such a dog it isn't
: even worth mentioning.
Guess again. About half of time taken up by IPD (Intel Processor
Dealer) seminars and videos is dedicated to heat control, and
a good chunk of that goes to describing the features and abilities
of the CPU fan that Intel includes in every boxed processor for OEMs.
The fans aren't included for snicks.
Bryon Wasserman
On Mon, 25 Nov 1996 19:32:40 -0600, GSF <gfur...@mtco.com> wrote:
>Nick W Walton wrote:
>>
>> We had a Cx166+, a iP133 and a iP166 all running Quake at the same time. The Cx166
>> performed fairly well (with an 2Mb ET6000 video card). It got around the same perf-
>> ormance as my iP133 with a 2Mb Diamond Stealth 64 DRAM.
>
>This sounds highly anecdotal. Did you perform any benchmarks, measure
>FPS etc. The Cyrix has a poor FPU.
>
>
>GSF
Keep in mind that the Stealth 64 DRAM is probably quite a bit slower
than whatever ET6000 card they were using with the Cyrix.
Boris T. Dog
> >>
> >> Has anybody had any experience with AMD K5 chips? I know that Microsoft
> >and requires adequate cooling. One user had cooling problems and lockups
> >when operating the 133 at its 100MHz speed. The problems disappeared
> >when clocked down to 90 MHz. Using a good heatsink with heatsink grease
> >cleared the problem when running back at 100 MHz.
>
> speed) run very cool. Each chip says "heatsink and fan required."
> But I put those on ANY chip 50 MHz (486) and up... Fans are cheap
> insurance.
>
>
> Weird thing: I opened a Compaq server last week -- Pentium 166 with
> only a heat sink, no fan. Stupid Compaq, real stupid.
Compaq isn't the only company too cheap to use fans. Gateway is famous
for it. And strangely enough, we have a Gateway PPro 200 that locks up
constantly and has to cool down. Gateway has replaced EVERY component,
but they still won't admit to a cooling problem. Now we've got a P-133
that's doing the same thing. I replace the heatsink with a fan and the
problem goes away. Funny, huh Gateway?
--
Boris T. Dog
El Rancho Boris
Dripping Springs, TX
Thomas Mosher
has the biggest heatsink I've seen since the Powermacs. Plus it being an
ATX motherboard the heatsink sits behind a large case-mounted fan.
Pentiums not needing heat control? I don't think they put that heatsink
on just for kicks.
--
Thomas Mosher
tmo...@isource.net
Paul van de Graaf
Thomas Mosher <tmo...@isource.net> wrote:
>That's interesting considering the retail (i.e., boxed) versions of the
>Pentiums come with a heatsink/fan permanently bonded to the CPU. Now if
>you want to see a heatsink, take a look at some of the older Macs, the
>heatsink is the size of a calculator.
>--
>Thomas Mosher
>tmo...@isource.net
I assume you are talking about the "Real" Pentium Overdrive units
here. I would guess that this is because Intel can control the
quality of the fans and put some kind of thermal shutdown into the
unit if the fan fails. They can't control the thermal configuration
of an aftermarket consumer's case/machine.
Also this is an end-user warranty situation. Do they really want to
trust the average user to remove the old heat sink which was designed
for the system from the previous CPU, apply heat-sink compound to the
new CPU and put it in the socket? Not to mention the possibility that
it might be a 5 Volt MB and the used would have to put the chip in a
3 Volt adapter? No, They want plug and play, hopefully into a ZIF
socket.
I'm not saying that CPU fans are a bad thing, just that they are not
common in mainstream name-brand Pentium machines.
Paul van de Graaf
Tim Williams
Neer Deth <neer...@quancon.com> wrote:
>> K5's (at least the PR100 and PR133 -- which both run at 100 MHz clock
>> speed) run very cool. Each chip says "heatsink and fan required."
>> But I put those on ANY chip 50 MHz (486) and up... Fans are cheap
>> insurance.
>>
>> I can pop the fan off my K5-133 now and the chip is barely warm.
>>
>> Weird thing: I opened a Compaq server last week -- Pentium 166 with
>> only a heat sink, no fan. Stupid Compaq, real stupid.
>Wrong! You take the fan off that chip and it'll run for about 20 minutes
>then lock up! Personal experience.
No, not wrong...I didn't say I was running it. I just said I could
pop off the fan and it would barely be warm [to the touch of your
finger, etc.]
And I do have one running with only an aluminum heatsink (ripped from
a dead Dell 486/66) w/no fan, but I would ALWAYS reccomend using a
fan.
Please, lose the "Wrong!"
Welbrey A. Hill
C'mon Albright, pull your head out. Only the P200+ needs a special
mbd. due to 75MHz bus capability. Do need proper cooling w/ heat sink
compound at >=P150+. IBM branded seem to run cooler (from reports).
Other problems (NT shutting off cache) due to MS not really doing what
it said it does re. NT internals--nothing unusual there <g>. Not that
anyone would run that resource hog anyway...
.Bill.
--
********************************************
* *
* Bill Hill *
* hi...@freenet.tlh.fl.us *
* *
* Don't blame me, I voted for Harry Browne *
* Charter member, Microsoft-free Society *
********************************************
Mats Andtbacka
bd, in <57stbq$8...@news.metrobbs.com>:
>>This sounds highly anecdotal. Did you perform any benchmarks, measure
>>FPS etc. The Cyrix has a poor FPU.
>I wouldn't call it poor.. Does a Pentium 100 have poor FPU performance?
compared to good FPU performance, yes. not unbelievably, miserably
poor, like the 486, but still poor. for *good* FPU muscle, look to the
DEC Alpha.
(and yes, you _can_ run Quake on an Alpha, provided it's running
Linux. Dave Taylor gave the sources to Linus himself, who did the
port.)
>If you're into real benchmarks, then consider Winstone which uses real-world
>apps..
i'd rather not. that sort of benchmarks are vulnerable not only to
application program versions, but even dates of purchase, as some
manufacturers are in nasty habits of folding bugfixes and
"improvements" in during the product's lifecycle without changing the
version number, or otherwise notifying anybody at all. not to even
mention how the applications are installed and set up, unless the
benchmark specs that out in detail, in which case the detail soon
starts getting intimate enough you might as well write a proper
benchmark.
when i feel crazy enough to trust benchmarks (lies, damned lies,
statistics, benchmarks) i look to SPEC95.
note followups, incidentally.
--
"...Everybody got this broken feeling
like their father or their dog just died..."
- Leonard Cohen
Mark Vorenkamp
The "boxed CPUs" are just that, CPUs. Intel has no way of knowing what case
the chips are going to be used in. Therefore, they have to put a fan on the
chip just to make sure there is "some cooling". As a previous poster
mentioned, the small fans are not reliable. Most of the high end systems
which are designed from the board up, use no fan on the CPU.
Mark
In article <57tng9$k...@hecate.umd.edu>, nor...@bigdipper.umd.edu (Pete Olson)
Stan
Maybe not so wierd - the CPU will probably survive the warranty
period...eh?
Stan -
ti...@evansville.net (Tim Williams) wrote:
>Thomas Mosher <tmo...@isource.net> wrote:
>>Lee Rhodes wrote:
>>>
>>> Has anybody had any experience with AMD K5 chips? I know that Microsoft
>>From what I've heard from my clients/associates, the AMD K5 133 runs hot
>>and requires adequate cooling. One user had cooling problems and lockups
>>when operating the 133 at its 100MHz speed. The problems disappeared
>>when clocked down to 90 MHz. Using a good heatsink with heatsink grease
>>cleared the problem when running back at 100 MHz.
>K5's (at least the PR100 and PR133 -- which both run at 100 MHz clock
jni...@ibm.net
In <32A37C...@isource.net>, Thomas Mosher <tmo...@isource.net> writes:
>Pete Olson wrote:
>>
>> Paul van de Graaf (pau...@mindspring.com) wrote:
>> :
>> : No. The majority of decent Pentium machines have no CPU fan, but a
>> : massive heat sink and decent cases with quality fans that blow air
>> : over the heat sink. This is the whole concept behind the ATX case.
>>
>> Dealer) seminars and videos is dedicated to heat control, and
>> a good chunk of that goes to describing the features and abilities
>> of the CPU fan that Intel includes in every boxed processor for OEMs.
>>
>> The fans aren't included for snicks.
>has the biggest heatsink I've seen since the Powermacs. Plus it being an
>ATX motherboard the heatsink sits behind a large case-mounted fan.
>Pentiums not needing heat control? I don't think they put that heatsink
>on just for kicks.
>Thomas Mosher
>tmo...@isource.net
I think the orginal statement was commenting not having a fan on top
of the heatsink.
Anyway, getting down to it, the thermal compound's (sorry the proper name
escapes me) job is to increase the surface contact of the CPU with the
heatsink by filling in the unsmooth surfaces between the two. (That is why
only put on a very think coat of the stuff on the CPU you don't want the
compound to become and insulator for the CPU.)
The heatsink's job is to absorb the heat from the CPU and then dispurce (sp?)
it. This can be done with either large fins to hopefully use normal air flow
in the enviroment, or with the assistance of a fan(s) pushing air across the
fins other wise place a fan right on top of the heatsink to absorb the heat.
You will see some of the high price servers with this design. Of course they
are usin good muffin type fans that can realy push the air.
In either soultion, as long as a proper design was involved, the end result is
the same. If you have any questions please feel free to ask.
Joe Niffen
bs
Welbrey A. Hill wrote:
>
> C'mon Albright, pull your head out. Only the P200+ needs a special
> mbd. due to 75MHz bus capability. Do need proper cooling w/ heat sink
> compound at >=P150+. IBM branded seem to run cooler (from reports).
> Other problems (NT shutting off cache) due to MS not really doing what
> it said it does re. NT internals--nothing unusual there <g>. Not that
> anyone would run that resource hog anyway...
> .Bill.
the cyrix otherwise it runs alot slower (arent they slow enough
already?)
wayspooled
Someone wrote:
>
> >K5's (at least the PR100 and PR133 -- which both run at 100 MHz clock
> >speed) run very cool. Each chip says "heatsink and fan required."
> >But I put those on ANY chip 50 MHz (486) and up... Fans are cheap
> >insurance.
>
So if the K5 100 and 133 both run at 100, what's the difference between
them other than cost. Did I get ripped?
-From owner of a no problem K5 133.
--
regards, wayspooled
Caution: Due to bug in my mailer, unsolicited commercial advertisements
may be returned to the sender several thousand times. Thanks.
_\|/_
(o o)
---oOO-(_)-OOo----------------------------------------------------
This message is written with recycled letters.
Waco
bs <bsc...@psd.k12.co.us> wrote in article
<32A698...@psd.k12.co.us>...
Not true. Although some MB's don't optimize the Cyrix chips (they should
if they say they support Cyrix processors in the manual) any user that
has a MB that doesn't have the optimizing in bios can put a copy of many
Cyrix Optimizer programs into their Startup group and have the same
functionality.
You can get a copy of some optimizers to try this out at
http://www.ionet.net/~rbdavis/files
Scott Hess
In article <57teal$3...@hecate.umd.edu>,
Just because I thinks it gives a little more perspective, I'm going to
add a couple rows (marked with ***) to this table:
Cyrix Intel Intel AMD
Vid Mode P166+ P5-133 P5-166 5k86-90
0 14.907466 23.825384 28.590713 16.504581
11 15.207925 26.763811 29.806556 16.613145
12 14.774399 25.238764 27.925444 15.981558
13 14.072603 24.387792 27.065876 14.793053
14 13.620988 22.867025 25.156868 14.208636
15 10.570709 17.461170 19.160210 10.083480
***Price $164 $207 $368 $69*
***Price/fps $ 15.51 $ 11.85 $ 19.21 $ 6.84
[courtesy H. Cardwell]
* I got these prices from www.thechipmerchant.com/prices.htm,
12/06/96, and there is no AMD K5-90. $69 is for the AMD K5-100. I
assume that it will perform even better, if not the K5-133 is $108.
Now, I don't play Quake, so I have no idea what the performance
numbers really mean, I'm assuming it reports fps. It looks like the
AMD chips might be slow in an absolute sense, but they also look to be
_cheap_ in an absolute sense. They may not be as fast as the Pentium
of the same "P-rating", but they only cost half as much - but they are
apparently more than half as fast!
To put the debate on a more-or-less even footing, it would be
interesting if someone collected a bigger set of numbers. What about
AMD K5-133? Beyond that, which would be better, getting a real Intel
and a cheap video card, or a Cyrix or AMD and use the savings for a
better video card? [That is, if you can't get the PPro200 plus the
best video card :-).]
Later,
--
scott hess <sh...@one.net> (606) 578-0412 http://w3.one.net/~shess/
<I plan to become so famous that people buy tapes of me reading source code>
Guillermo Gonzalez
On Sat, 07 Dec 1996 16:49:35 GMT, Samauri@alias.#cyberpass.net (Samauri) wrote:
>
>The K5-133 uses a 100 MHz clock but has a faster internal structure. It is
>touted to run the system at Pentium 133 overall speeds.
>
>FWIW, I overclocked my K5-133 to 120 MHz (using a 75 MHz bus speed and the
>default 1.5 multiplier) and it runs fine. Performance increase of around
>12-15% over the usual 100 MHz.
So what do your Quake benchmark numbers look like?
Michael Chan
gonz...@bridge.net (Guillermo Gonzalez) writes:
>On Sat, 07 Dec 1996 16:49:35 GMT, Samauri@alias.#cyberpass.net (Samauri) wrote:
>>
>>The K5-133 uses a 100 MHz clock but has a faster internal structure. It is
>>touted to run the system at Pentium 133 overall speeds.
>>
>>FWIW, I overclocked my K5-133 to 120 MHz (using a 75 MHz bus speed and the
>>default 1.5 multiplier) and it runs fine. Performance increase of around
>>12-15% over the usual 100 MHz.
>
there's no such thing as an amd running at 120Mhz, 75 x 1.5 == 112.5,
not 120Mhz. Personally, i can't say that i noticed the performance difference,
since it runs so fast anyways...
Michael C.
>So what do your Quake benchmark numbers look like?
--
Michael Chan
m-c...@uiuc.edu
http://www.students.uiuc.edu/~m-chan
Leho Kraav aka -= Macman =-
One beautiful day, mand...@news.abo.fi (Mats Andtbacka) decided that |4 Dec
1996 00:04:16 GMT| was the right time to write about "Re: Quake performance with
Cx166+":
>compared to good FPU performance, yes. not unbelievably, miserably
>poor, like the 486, but still poor. for *good* FPU muscle, look to the
>DEC Alpha. (and yes, you _can_ run Quake on an Alpha, provided it's running
>Linux. Dave Taylor gave the sources to Linus himself, who did the
>port.)
So... how fast is Quake on a DEC Alpha machine <:]?
E-mail: mac...@online.ee
WWW: http://www.online.ee/~macman1/
Fidonet: 2:490/14.11
Nick Savoiu
In a previous article Welbrey A. Hill (hi...@fn5.freenet.tlh.fl.us) thus spoke:
> C'mon Albright, pull your head out. Only the P200+ needs a special
> mbd. due to 75MHz bus capability. Do need proper cooling w/ heat sink
> compound at >=P150+. IBM branded seem to run cooler (from reports).
> Other problems (NT shutting off cache) due to MS not really doing what
> it said it does re. NT internals--nothing unusual there <g>. Not that
> anyone would run that resource hog anyway...
And if you do there are software patches out there that flip back the WB
cache bit in the control registers and, voila, your 6x86 runs as
expected...
--
Nick @msn.com | Check out my homepage at:
Savoiu @paul.rutgers.edu | http://paul.rutgers.edu/~savoiu
Tim Roberts
Doesn't overclocking CPU's burn um up!
--
Tim Roberts
In article <58ig5g$j...@vixen.cso.uiuc.edu>, Michael Chan <m-
ch...@ehsn15.cen.uiuc.edu> writes
Welbrey A. Hill
The AMD 5KR-133 has internal architecture improvements much like the
Cyrix 6x86 series. The internals of the original 5K are nearly ident-
ical to the Intel, therefore performance is one for one.
Nick Savoiu
In a previous article Scott Hess (sh...@one.net) thus spoke:
> In article <57teal$3...@hecate.umd.edu>,
> nor...@bigdipper.umd.edu (Pete Olson) writes:
> Quake Benchmarks: Cyrix vs. Intel vs. AMD
> Just because I thinks it gives a little more perspective, I'm going to
> add a couple rows (marked with ***) to this table:
> Cyrix Intel Intel AMD
> Vid Mode P166+ P5-133 P5-166 5k86-90
> 0 14.907466 23.825384 28.590713 16.504581
> 11 15.207925 26.763811 29.806556 16.613145
> 12 14.774399 25.238764 27.925444 15.981558
> 13 14.072603 24.387792 27.065876 14.793053
> 14 13.620988 22.867025 25.156868 14.208636
> 15 10.570709 17.461170 19.160210 10.083480
> ***Price $164 $207 $368 $69*
> ***Price/fps $ 15.51 $ 11.85 $ 19.21 $ 6.84
> [courtesy H. Cardwell]
Funny, using HOST_SPEEDS, I get 20ms/frame = 50fps at 320x200. I have a
6x86 P166+.
Pete Olson
Nick Savoiu (sav...@paul.rutgers.edu) wrote:
:
: Funny, using HOST_SPEEDS, I get 20ms/frame = 50fps at 320x200. I have a
: 6x86 P166+.
Host speed is an amazingly poor benchmark, as it returns the result
of single frames and is very sensitive to changes in environment.
I have gotten numbers as low as 10 at 320x200, full screen.
The best built-in benchmark in 1.06 is timedemo.
From the command line in DOS, type
quake -safe +timedemo demo2
to get the best possible benchmark.
Roderick Castillo
sav...@paul.rutgers.edu (Nick Savoiu) wrote:
>In a previous article Scott Hess (sh...@one.net) thus spoke:
>> In article <57teal$3...@hecate.umd.edu>,
>> nor...@bigdipper.umd.edu (Pete Olson) writes:
>> Quake Benchmarks: Cyrix vs. Intel vs. AMD
>> Just because I thinks it gives a little more perspective, I'm going to
>> add a couple rows (marked with ***) to this table:
>> Cyrix Intel Intel AMD
>> Vid Mode P166+ P5-133 P5-166 5k86-90
>> 0 14.907466 23.825384 28.590713 16.504581
>> 11 15.207925 26.763811 29.806556 16.613145
>> 12 14.774399 25.238764 27.925444 15.981558
>> 13 14.072603 24.387792 27.065876 14.793053
>> 14 13.620988 22.867025 25.156868 14.208636
>> 15 10.570709 17.461170 19.160210 10.083480
>> ***Price $164 $207 $368 $69*
>> ***Price/fps $ 15.51 $ 11.85 $ 19.21 $ 6.84
>> [courtesy H. Cardwell]
>Funny, using HOST_SPEEDS, I get 20ms/frame = 50fps at 320x200. I have a
>6x86 P166+.
What is HOST_SPEEDS and where is it available?
Roderick
Pete Olson
Roderick Castillo (100....@germanynet.de) wrote:
:
: What is HOST_SPEEDS and where is it available?
It is console command that returns the time required
to draw a single frame. As it uses tiny 1 frame
samples, it is a very poor measure of general Quake
performance.
A good way to test Quake performance under standardized
conditions is to enter the following at the DOS prompt:
quake -safe +timedemo demo2
You need version 1.06 for this, and you still have to
make sure that the play area is consistent between the
various machines you wish to compare (such as full
screen or in a teeny window in Quake).
No salutes for your surrender, nothing noble in your fate...Christ, what have you done?
In article <59hslb$h...@hecate.umd.edu>, nor...@bigdipper.umd.edu (Pete Olson) writes:
> Nick Savoiu (sav...@paul.rutgers.edu) wrote:
> :
> : Funny, using HOST_SPEEDS, I get 20ms/frame = 50fps at 320x200. I have a
> : 6x86 P166+.
>
> of single frames and is very sensitive to changes in environment.
> I have gotten numbers as low as 10 at 320x200, full screen.
>
> The best built-in benchmark in 1.06 is timedemo.
> From the command line in DOS, type
>
> quake -safe +timedemo demo2
>
> to get the best possible benchmark.
The best benchmark I've seen was posted at www.stomped.com, it
recommends going to a NEW GAME, then, without moving, hit the ~
(tilde) to bring up the command line.. type TIMEREFRESH and hit enter.
This will spin you in a complete circle, and give you an average frame
rate. While this method may not reflect dynamic gameplay framerate,
it will give you an accurate method of gauging your lowest FPS, IMO.
Try it as described above, then try just walking around the main
entrance room then doing TIMEREFRESH, note the difference.
I installed a P-166+ (Cyrix) last night and benched Quake .. I took
a 20% decrease in framerate than that posted by my Intel P-100.
>
--
Robert Spelman | "Does it bother me at all, my rival is
00rbs...@virgo.bsuvc.bsu.edu | Neanderthal? ... IQ is no problem
Ball State University | Here, we won't be playing Scrabble(C)
Muncie, IN | For her hand I fear ..." -Sting
G.H. Stoltz
nor...@bigdipper.umd.edu (Pete Olson) wrote:
>Nick Savoiu (sav...@paul.rutgers.edu) wrote:
>:
>: Funny, using HOST_SPEEDS, I get 20ms/frame = 50fps at 320x200. I have a
>: 6x86 P166+.
>Host speed is an amazingly poor benchmark, as it returns the result
>of single frames and is very sensitive to changes in environment.
>I have gotten numbers as low as 10 at 320x200, full screen.
>The best built-in benchmark in 1.06 is timedemo.
>From the command line in DOS, type
>quake -safe +timedemo demo2
>to get the best possible benchmark.
I've run that in the past on my system, a P166+ and got numbers
ranging from 25-26 frames per second, while not as fast as a true
pentium chip from intel, it is more than acceptable to me.
JackMan
you can also try and just run timedemo demo2 on it....that will give
you and accurate fps also....let you see what you average through an
episode during action...
On 28 Dec 96 14:53:04 -0500, 00rbs...@bsuvc.bsu.edu (No salutes for
your surrender, nothing noble in your fate...Christ, what have you
done?) wrote:
>In article <59hslb$h...@hecate.umd.edu>, nor...@bigdipper.umd.edu (Pete Olson) writes:
>> Nick Savoiu (sav...@paul.rutgers.edu) wrote:
>> :
>> : Funny, using HOST_SPEEDS, I get 20ms/frame = 50fps at 320x200. I have a
>> : 6x86 P166+.
>>
>> Host speed is an amazingly poor benchmark, as it returns the result
>> of single frames and is very sensitive to changes in environment.
>> I have gotten numbers as low as 10 at 320x200, full screen.
>>
>> The best built-in benchmark in 1.06 is timedemo.
>> From the command line in DOS, type
>>
>> quake -safe +timedemo demo2
>>
>> to get the best possible benchmark.
>
James Dean
Ok first lets look at the chart...>>> In article <57teal$3...@hecate.umd.edu>,
>>> nor...@bigdipper.umd.edu (Pete Olson) writes:
>>> Quake Benchmarks: Cyrix vs. Intel vs. AMD
>
>>> Just because I thinks it gives a little more perspective, I'm going to
>>> add a couple rows (marked with ***) to this table:
>
>>> Cyrix Intel Intel AMD
>>> Vid Mode P166+ P5-133 P5-166 5k86-90
>>> 0 14.907466 23.825384 28.590713 16.504581
>>> 11 15.207925 26.763811 29.806556 16.613145
>>> 12 14.774399 25.238764 27.925444 15.981558
>>> 13 14.072603 24.387792 27.065876 14.793053
>>> 14 13.620988 22.867025 25.156868 14.208636
>>> 15 10.570709 17.461170 19.160210 10.083480
>>> ***Price $164 $207 $368 $69*
>>> ***Price/fps $ 15.51 $ 11.85 $ 19.21 $ 6.84
If this is so first what is the clock on the motherboards of these CPU's? and what
video cards are they using cause you see on my machine (crappy ISA 512k board) with a
50MHz bus and p100 oc'd to 150MHz I can get about 14-16 FPS in vid mode 0 but with a
66MHz bus and oc'd to 133 I get 18+ FPS so I think you need to revise these or get
more details for them to be accurate.
Chris
magicb...@bort.mv.net (James Dean) wrote:
Who cares anyway Quake is the crapiest games ever made. It's been so
poorly writen that it uses FPU calculation even for whole numbers so
dont wonder why its so chitty.
Quake is not only a shitty game to play but it's so stupidly writen
that as a matter of fact its the only games writen this way (where all
calculation are relying on the FPU even for the whole number) and the
only good thing about this is that we will never see any other games
so badly writen and this is a real relief and peoples using this bunch
of shitt as a so called benchmark to compare Intel CPUs against Cyrix
CPU are putting one of their finger in their ass and another one in
their eye.
Quake sucks as hell please lets forget it and talk about serious
matter.
John Hill
Chris wrote:
> Who cares anyway Quake is the crapiest games ever made. It's been so
> poorly writen that it uses FPU calculation even for whole numbers so
> dont wonder why its so chitty.
>
> Quake is not only a shitty game to play but it's so stupidly writen
> that as a matter of fact its the only games writen this way (where all
> calculation are relying on the FPU even for the whole number) and the
> only good thing about this is that we will never see any other games
> so badly writen and this is a real relief and peoples using this bunch
> of shitt as a so called benchmark to compare Intel CPUs against Cyrix
> CPU are putting one of their finger in their ass and another one in
> their eye.
>
> Quake sucks as hell please lets forget it and talk about serious
> matter.
Are we just a little bit testy Duke-boy? Quake may not be the best
written game in your opinion, but it IS the future of gaming. The days
of the stand-alone-game are numbered (as well as the 2D pixel-drawn
graphics engine of much lesser games). Just look at the number of
people on the Internet playing Multiplayer Quake. And the comparison of
the different cpu's in Quake has a valid point. The FPU performance of
the Cyrix P166+ is so poor that an AMD 5k85-90 that costs 50% less gives
comparable numbers. I had to make a decision on how to invest $200 into
my PC, and after seeing the overall performance of the CP166+ vs. IP133,
I was sold on Intel.
John
Steve Krems
> >
>
> Who cares anyway Quake is the crapiest games ever made. It's been so
> poorly writen that it uses FPU calculation even for whole numbers so
> dont wonder why its so chitty.
>
While not an avid gamer I strongly disagree with your statements. Quake
seems to be a great real world benchmark. It uses CPU, FPU, Video and
memory subsystems. The fact that it uses the FPU even for whole numbers
is a great idea....if its there use it!!!
Steve
Pete Olson
Chris (Qu...@sucks.shitt) wrote:
:
: Who cares anyway Quake is the crapiest games ever made. It's been so
: poorly writen that it uses FPU calculation even for whole numbers so
: dont wonder why its so chitty.
Trolling is not cool.
houser
On Fri, 03 Jan 1997 23:09:11 GMT, Qu...@sucks.shitt (Chris) wrote:
>magicb...@bort.mv.net (James Dean) wrote:
>
>>Ok first lets look at the chart...
>>>>In a previous article Scott Hess (sh...@one.net) thus spoke:
>>>>> In article <57teal$3...@hecate.umd.edu>,
>>>>> nor...@bigdipper.umd.edu (Pete Olson) writes:
>>>>> Quake Benchmarks: Cyrix vs. Intel vs. AMD
>>>
>>>>> Just because I thinks it gives a little more perspective, I'm going to
>>>>> add a couple rows (marked with ***) to this table:
>>>
>>>>> Cyrix Intel Intel AMD
>>>>> Vid Mode P166+ P5-133 P5-166 5k86-90
[clip]
>Who cares anyway Quake is the crapiest games ever made. It's been so
>poorly writen that it uses FPU calculation even for whole numbers so
>dont wonder why its so chitty.
>
>Quake is not only a shitty game to play but it's so stupidly writen
>that as a matter of fact its the only games writen this way (where all
>calculation are relying on the FPU even for the whole number) and the
>only good thing about this is that we will never see any other games
>so badly writen and this is a real relief and peoples using this bunch
Actually, an upcoming game called MDK will probably use the same FPU
tricks as Quake.
Jak Crow
John Hill (john...@connect.net) wrote:
: Are we just a little bit testy Duke-boy? Quake may not be the best
: written game in your opinion, but it IS the future of gaming. The days
: of the stand-alone-game are numbered (as well as the 2D pixel-drawn
: graphics engine of much lesser games).
Terminator: Future Shock and Descent 1/2 use polygon graphic
engines and were out before Quake. Does that mean Quake just copied them?
Multiplayer is all well and fine, but for something to be a truly
good game, it should have just as strong one-player value as multiplayer.
Single player gaming should always come before the network play.
WILLIAM MICHAEL KIRBY
Chris wrote:
>
> magicb...@bort.mv.net (James Dean) wrote:
>
> >Ok first lets look at the chart...
> >>>In a previous article Scott Hess (sh...@one.net) thus spoke:
> >>>> In article <57teal$3...@hecate.umd.edu>,
> >>>> nor...@bigdipper.umd.edu (Pete Olson) writes:
> >>>> Quake Benchmarks: Cyrix vs. Intel vs. AMD
> >>
> >>>> Just because I thinks it gives a little more perspective, I'm going to
> >>>> add a couple rows (marked with ***) to this table:
> >>
> >>>> Cyrix Intel Intel AMD
> >>>> Vid Mode P166+ P5-133 P5-166 5k86-90
> >>>> 11 15.207925 26.763811 29.806556 16.613145
> >>>> 12 14.774399 25.238764 27.925444 15.981558
> >>>> 13 14.072603 24.387792 27.065876 14.793053
> >>>> 14 13.620988 22.867025 25.156868 14.208636
> >>>> 15 10.570709 17.461170 19.160210 10.083480
> >>>> ***Price $164 $207 $368 $69*
> >>>> ***Price/fps $ 15.51 $ 11.85 $ 19.21 $ 6.84
> >
> >If this is so first what is the clock on the motherboards of these CPU's? and what
> >video cards are they using cause you see on my machine (crappy ISA 512k board) with a
> >50MHz bus and p100 oc'd to 150MHz I can get about 14-16 FPS in vid mode 0 but with a
> >66MHz bus and oc'd to 133 I get 18+ FPS so I think you need to revise these or get
> >more details for them to be accurate.
> >
>
> poorly writen that it uses FPU calculation even for whole numbers so
> dont wonder why its so chitty.
>
> Quake is not only a shitty game to play but it's so stupidly writen
> that as a matter of fact its the only games writen this way (where all
> calculation are relying on the FPU even for the whole number) and the
> only good thing about this is that we will never see any other games
> so badly writen and this is a real relief and peoples using this bunch
> CPU are putting one of their finger in their ass and another one in
> their eye.
>
> Quake sucks as hell please lets forget it and talk about serious
> matter.
John Hill
I do agree with you about Descent and T:FS, Id knew that the engines in
these games would give greater flexibility, and a much better gaming
experience. I don't think Id 'copied' them (I believe Descent came out
about 8-10 months before Quake did, and 8-10 months would be a very
short period of time to reverse engineer a gaming engine and rewrite
it), but I do believe they made a better engine. The multi-player
capabilities of Quake and it's engine design make it one of the best
games I've played. Destroying a computer-based enemy, exploiting the
weaknesses in the AI, and being victorious over silicon just isn't as
fun as doing the same to a carbon-based enemy. Network play, IMHO, is
becoming more important and will soon be the primary concern of game
designers.
John
graehl
> > Terminator: Future Shock and Descent 1/2 use polygon graphic
> > engines and were out before Quake. Does that mean Quake just copied them?
> > Multiplayer is all well and fine, but for something to be a truly
> > good game, it should have just as strong one-player value as multiplayer.
> > Single player gaming should always come before the network play.
>
> I do agree with you about Descent and T:FS, Id knew that the engines in
> these games would give greater flexibility, and a much better gaming
> experience. I don't think Id 'copied' them (I believe Descent came out
> about 8-10 months before Quake did, and 8-10 months would be a very
> short period of time to reverse engineer a gaming engine and rewrite
> it), but I do believe they made a better engine.
Descent was out a LOOONG time ago while T:FS was out maybe a few months
before the final version. I'm not sure about T:FS, but Descent/Descent2
have a special-purpose engine which is able to get decent frame rates by
limiting the maps to the kind you can explore the whole of by "touching
your arm to the wall" and following it (I'm not sure exactly what the
restriction is but I've heard there is one and the levels are definitely
simpler, even though there is perspective-correct texture mapping and
polygon enemies) Quake currently has the most advanced/accurate/fast 3d
engine around ... no other game that I know of has texture level light
mapping (as opposed to per-polygon lighting) and unrestricted 3d
objects/levels
Pan...@chatlink.com
>> Who cares anyway Quake is the crapiest games ever made. It's been so
>> poorly writen that it uses FPU calculation even for whole numbers so
>> dont wonder why its so chitty.
>>
>
>seems to be a great real world benchmark. It uses CPU, FPU, Video and
>memory subsystems. The fact that it uses the FPU even for whole numbers
>is a great idea....if its there use it!!!
Actually using FPU for whole numbers is slower then using Interger instructions
for the same thing.
Guillermo Gonzalez
Not when you can use the FPU for whole numbers while the integer unit is
busy processing other things...
Kevin Miller
>> Who cares anyway Quake is the crapiest games ever made. It's been so
>> poorly writen that it uses FPU calculation even for whole numbers so
>> dont wonder why its so chitty.
>>
>> Quake is not only a shitty game to play but it's so stupidly writen
>> that as a matter of fact its the only games writen this way (where all
>> calculation are relying on the FPU even for the whole number) and the
>> only good thing about this is that we will never see any other games
>> so badly writen and this is a real relief and peoples using this bunch
>> of shitt as a so called benchmark to compare Intel CPUs against Cyrix
>> CPU are putting one of their finger in their ass and another one in
>> their eye.
>>
>> Quake sucks as hell please lets forget it and talk about serious
>> matter.
>Your obviously a penis brain. Thank You.
Penis brain?! Aren't you being just a little cocky? :)
Kevin Miller
"Jazz is not dead; it just smells funny" -- Frank Zappa
No 6
>Who cares anyway Quake is the crapiest games ever made. It's been so
>poorly writen that it uses FPU calculation even for whole numbers so
>dont wonder why its so chitty.
>
>Quake is not only a shitty game to play but it's so stupidly writen
>that as a matter of fact its the only games writen this way (where all
>calculation are relying on the FPU even for the whole number) and the
>only good thing about this is that we will never see any other games
>so badly writen and this is a real relief and peoples using this bunch
>of shitt as a so called benchmark to compare Intel CPUs against Cyrix
>CPU are putting one of their finger in their ass and another one in
>their eye.
>
>Quake sucks as hell please lets forget it and talk about serious
>matter.
This sounds like a user who bought a Cyrix and wished he's spent a bit
extra and got a real processor.
No 6
>Terminator: Future Shock and Descent 1/2 use polygon graphic
>engines and were out before Quake. Does that mean Quake just copied them?
like Descent is a copy of Star Wars which was released in the arcades
in the early to mid 1980s, but it clearly a pointless argument.
>Multiplayer is all well and fine, but for something to be a truly
>good game, it should have just as strong one-player value as multiplayer.
>Single player gaming should always come before the network play.
And who are you to make this decision? There have been some games
written purely for multi-player environment. Take MUDs for example.
There are some good single player games, and some equally enjoyable
mutli-player games. I can't see how anyone can say that all games
should be judged on their solo play capabilities.
Steve.
No 6
>Who cares anyway Quake is the crapiest games ever made. It's been so
>poorly writen that it uses FPU calculation even for whole numbers so
>dont wonder why its so chitty.
>
>Quake is not only a shitty game to play but it's so stupidly writen
>that as a matter of fact its the only games writen this way (where all
>calculation are relying on the FPU even for the whole number) and the
>only good thing about this is that we will never see any other games
>so badly writen and this is a real relief and peoples using this bunch
>of shitt as a so called benchmark to compare Intel CPUs against Cyrix
>CPU are putting one of their finger in their ass and another one in
>their eye.
>
>Quake sucks as hell please lets forget it and talk about serious
>matter.
There a sinple solution to all this. Don't play it, and even better,
stop ranting about it.
Jak Crow
On Mon, 06 Jan 1997 00:07:12 GMT, n...@the.villiage (No 6) wrote:
>>Multiplayer is all well and fine, but for something to be a truly
>>good game, it should have just as strong one-player value as multiplayer.
>>Single player gaming should always come before the network play.
>And who are you to make this decision? There have been some games
>written purely for multi-player environment. Take MUDs for example.
>There are some good single player games, and some equally enjoyable
>mutli-player games. I can't see how anyone can say that all games
>should be judged on their solo play capabilities.
So why bother including a one-player option at all then?
Albert W. Dorrington
In article <32CDD8...@adnc.com>, Steve Krems <ste...@adnc.com> writes:
> > >
> >
> > Who cares anyway Quake is the crapiest games ever made. It's been so
> > poorly writen that it uses FPU calculation even for whole numbers so
> > dont wonder why its so chitty.
> >
>
> While not an avid gamer I strongly disagree with your statements. Quake
> seems to be a great real world benchmark. It uses CPU, FPU, Video and
> memory subsystems. The fact that it uses the FPU even for whole numbers
> is a great idea....if its there use it!!!
>
What people keep failing to realize here was that Quake was written
to take specific advantage of the Intel Pentium's FPU architecture.
(pipelining of FP operations, etc.)
The Cyrix 6x86 processors use an entirely different architecture which
does not pipeline FP operations, yet allows them to run in parallel
with non-FP operations.
I would be willing to bet, that if Id Software could be convinced to
recode Quake to take specific advantage fo the Cyrix 6x86 architecture
that the Cyrix version would be a bit faster than the Intel version.
(not that it will ever happen though.)
- Al
--
Al Dorrington
awdo...@ictest.delcoelect.com Database Admin
Delco Electronics - IC CIM Unix Sysadmin
Kokomo, Indiana, USA Phone: 317.451.9655
jason
James Dean wrote:
>
> Ok first lets look at the chart...
> >>In a previous article Scott Hess (sh...@one.net) thus spoke:
> >>> In article <57teal$3...@hecate.umd.edu>,
> >>> nor...@bigdipper.umd.edu (Pete Olson) writes:
> >>> Quake Benchmarks: Cyrix vs. Intel vs. AMD
> >
> >>> Just because I thinks it gives a little more perspective, I'm going to
> >>> add a couple rows (marked with ***) to this table:
Has anyone had experience with Techmedia Monitors? (particularly the
1500N or 1548G) Or could you recommend another 15" monitor for a tight
buget? Thank you:)
Paul Beer
On 4 Jan 1997 04:37:19 GMT, jak...@nntp.best.com (Jak Crow) wrote:
>John Hill (john...@connect.net) wrote:
>
>: Are we just a little bit testy Duke-boy? Quake may not be the best
>: written game in your opinion, but it IS the future of gaming. The days
>: of the stand-alone-game are numbered (as well as the 2D pixel-drawn
>: graphics engine of much lesser games).
>
> Terminator: Future Shock and Descent 1/2 use polygon graphic
>engines and were out before Quake. Does that mean Quake just copied them?
> Multiplayer is all well and fine, but for something to be a truly
>good game, it should have just as strong one-player value as multiplayer.
>Single player gaming should always come before the network play.
>
Numbskull. Heard of baseball? Soccer? Rugby league? These are all
team games of course, and all the more fun for it. And this is where
Quake excells, as it becomes so much more than graphics or sound or
playing against some computer program. I accept that there are
networks in perhaps a few countries where you can play other pc games
against other people... but you were emphasising single player value.
Quake may be best viewed as a ticket to multiplayer gaming which is a
times awesome and unforgettable in a way which no computer generated
opponent will ever be. The single player aspect of Quake may leave a
little to be desired in some ppl's eyes, but then again many people
only play it single-player, so it must be worth it in their eyes. Of
course there are the copious numbers of (free) new levels that may be
downloaded for single player, besides all the multiplayer levels.
Some of these levels outdo the id originals in workmanship and ideas!
In a sense Quake may have 'copied' other gaming technology, but I
would see it as a major evolution, not just a copy. Well I hope you
will forgive me for calling you a numbskull now... but you did kind of
ask for it by confining your argument to 'computer' games; That's like
saying a single player golf game should always take priority over
playing against others. Multiplayer Quake is as much more than a
computer game, as people are more than computers (given the
limitations of internet play. Of course some set up LAN games ... woo
hoo!)
Okay Jak, you are now licensed to call me a numbskull, or whatever you
like!
Paul
BTW how the hell did this thread end up in Windows 95 ng? I have
removed half of the groups it was crossposted to from my address
window.
Stephen Moore
Albert W. Dorrington wrote:
>
> In article <32CDD8...@adnc.com>, Steve Krems <ste...@adnc.com> writes:
> > > >
> > >
> > > Who cares anyway Quake is the crapiest games ever made. It's been so
> > > poorly writen that it uses FPU calculation even for whole numbers so
> > > dont wonder why its so chitty.
As in Chitty Chitty Bang Bang?
Is Dick Van Dyke a Quake fan now?
What's this guy talking about?
;)
Steve.
--
+-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+
| Stephen Moore - aka Granny Killer |
| Punishment Squad Quake Clan |
| mailto:*st...@shpcorp.dnet.co.uk* |
| WWW: http://www.niweb.com/dnet/dnetKhbw |
+-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+
| Remove *'s before replying by email!! |
+-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+
Jonathan Williamson
>I do agree with you about Descent and T:FS, Id knew that the engines in
>these games would give greater flexibility, and a much better gaming
>experience. I don't think Id 'copied' them (I believe Descent came out
>about 8-10 months before Quake did, and 8-10 months would be a very
>short period of time to reverse engineer a gaming engine and rewrite
>it)
A game like descent takes significantly less processing for each frame
than a game like quake, descent uses a clipping method called portals
which consist of pre-generated tables of what is visible from each are
in the map. This is possible because descent is a very linear game
compared to something like quake, especcialy because of the ability for
the player in quake to get to pretty well any part of the map. As for
future shock I believe it is just pre-rendered animation (but don't
quote me on that). As for terminator I have never seen it so I can't
comment...
cya,
Jonathan=-
--
,o888o, Jonathan Williamson AKA ShRoOm=- ,o888o,
8888888 ~~~~~~~~ ~~~~~~~~~~ ^^^ ~~~~~~~~ 8888888
| | Email: Jona...@chipping.demon.co.uk | |
.\\|//. Fax: (0114) 281 9435 (between 4-6pm) .\\|//.
A.M.P. Barcellos
Albert W. Dorrington wrote:
>
> In article <32CDD8...@adnc.com>, Steve Krems <ste...@adnc.com> writes:
> to take specific advantage of the Intel Pentium's FPU architecture.
> (pipelining of FP operations, etc.)
>
> The Cyrix 6x86 processors use an entirely different architecture which
> does not pipeline FP operations, yet allows them to run in parallel
> with non-FP operations.
>
> I would be willing to bet, that if Id Software could be convinced to
> recode Quake to take specific advantage fo the Cyrix 6x86 architecture
> that the Cyrix version would be a bit faster than the Intel version.
> (not that it will ever happen though.)
Some practical experience:
I just have upgraded my PC from
Pentium 60 256kb cache 16Mb RAM SVGA PCI Trident 9440-PA 1Mb
to
Cyrix P166+ 512kb cache 32Mb EDO RAM Matrox Mistique 2Mb
All games became notably faster, including EF2000, Duke, Descent I/II,and
GP2, allowing virtually always the highest resolution and detail.
All BUT Quake.
To my despair, the performance remained similar, and
it does not allow me to run 640x480 smoothly. Even considering that
VESA 2.0 is implemented on Mistique's hardware.
So, is the problem an exagerated floating point arithmetics used
in Quake?
Regards,
Marinho.
--
==== Antonio Marinho Pilla Barcellos ==================================
Ph.D. Student phone: +44 191 222 8007
Department of Computing Science fax: +44 191 222 8232
University of Newcastle upon Tyne WWW: http://www.ncl.ac.uk/~n4521529
The United Kingdom NE1 7RU e-mail: A.M.P.B...@newcastle.ac.uk
=======================================================================
Dave Chernetz
aul...@mindspring.com wrote:
>
> On Tue, 07 Jan 1997 18:05:44 +0000,"A.M.P. Barcellos"
> >So, is the problem an exagerated floating point arithmetics used
> >in Quake?
>
yeah. today I switched my cyrix which I was running at 166+ (133MHz)
for an intel 133MHz. quake fps went from 15 to 28 just like that.
Everything else feels the same.
--
Dave Chernetz "I'd give my right arm to be ambidextrous."
aka Pizpot --AOLer Kowman
cher...@mts.net
http://www.mts.net/~chernetz/
R. Luke
*jak...@best.com* (Jak Crow) wrote:
>On Mon, 06 Jan 1997 00:07:12 GMT, n...@the.villiage (No 6) wrote:
>>>Multiplayer is all well and fine, but for something to be a truly
>>>good game, it should have just as strong one-player value as multiplayer.
>>>Single player gaming should always come before the network play.
>>And who are you to make this decision? There have been some games
>>written purely for multi-player environment. Take MUDs for example.
>>There are some good single player games, and some equally enjoyable
>>mutli-player games. I can't see how anyone can say that all games
>>should be judged on their solo play capabilities.
> So why bother including a one-player option at all then?
... So people can learn the levels and develop strategy before going
multi-player. Just think of a "green" Quake player in a game with
a bunch of deathmatch veterans.... FRAG CITY!!!
Rob
cam macintosh
As for
future shock I believe it is just pre-rendered animation (but don't
quote me on that). As for terminator I have never seen it so I can't
> comment...
ok, i won't quote you, but:
i think you are reffering to creature shock (all pre-rendered
animations) instead of future shock (a true 3d game).
by the way, terminator is terminator:future shock