 |
 |
|
 |
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Anyone know the difference between a:
AMD Athlon
Pentium III
Intel Celeron
for similar speed?
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Greg M. Johnson" wrote:
>
> Anyone know the difference between a:
>
> AMD Athlon
> Pentium III
> Intel Celeron
>
> for similar speed?
My research has the performance in speed listed in the order you
present above fastest to slowest. My own tests indicate about 14%
better performance of the AMD over the P-III but I admit it was
rather subjective since I don't own a P-III of the same speed
as my AMD and had to scale my results to a slower P-III.
--
Ken Tyler - 1400+ POV-Ray, Graphics, 3D Rendering, and Raytracing Links:
http://home.pacbell.net/tylereng/index.html http://www.povray.org/links/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Greg M. Johnson <gre### [at] my-dejanews com> wrote:
: for similar speed?
Define "speed". (Cpu clock frequency? MIPS? MFLOPS? Something else?)
--
char*i="b[7FK@`3NB6>B:b3O6>:B:b3O6><`3:;8:6f733:>::b?7B>:>^B>C73;S1";
main(_,c,m){for(m=32;c=*i++-49;c&m?puts(""):m)for(_=(
c/4)&7;putchar(m),_--?m:(_=(1<<(c&3))-1,(m^=3)&3););} /*- Warp -*/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Warp" <war### [at] tag povray org> wrote in message
news:3a5f35e7@news.povray.org...
> Define "speed". (Cpu clock frequency? MIPS? MFLOPS? Something else?)
>
I assumed (and I think Ken assumed) that he meant rendering speed (what
other benchmark is there?).
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Tom Melly <tom### [at] tomandlu co uk> wrote:
: I assumed (and I think Ken assumed) that he meant rendering speed (what
: other benchmark is there?).
If rendering speed (with povray) is the same, then it really doesn't
matter which CPU you use :)
--
char*i="b[7FK@`3NB6>B:b3O6>:B:b3O6><`3:;8:6f733:>::b?7B>:>^B>C73;S1";
main(_,c,m){for(m=32;c=*i++-49;c&m?puts(""):m)for(_=(
c/4)&7;putchar(m),_--?m:(_=(1<<(c&3))-1,(m^=3)&3););} /*- Warp -*/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Warp wrote:
> If rendering speed (with povray) is the same, then it really doesn't
> matter which CPU you use :)
Do you really mean that a 450 MHz Pentium II is as fast as a 450 MHz Pentium
III, Celeron, etc., as far as povray rendering time? It goes against my
intuition, or at least my cynical view that it's not that simple.
Post a reply to this message
|
 |
|  |
|  |
|
 |
From: Christoph Hormann
Subject: Re: OT: cpu speed & rendering time
Date: 12 Jan 2001 14:05:27
Message: <3A5F5576.BC54C873@gmx.de>
|
|
 |
|  |
|  |
|
 |
"Greg M. Johnson" wrote:
>
> Anyone know the difference between a:
>
> AMD Athlon
> Pentium III
> Intel Celeron
>
> for similar speed?
If you mean Povray speed (i.e. mainly FPU performance), AFAIK the Athlon
is fastest when compared at the same Frequency. Furthermore I think the
Celeron is slower than the P3.
BTW, the German c't magazine regularly publishes Povray benchmark results
for all CPUs.
Christoph
--
Christoph Hormann <chr### [at] gmx de>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
 |
|  |
|  |
|
 |
From: Thorsten Froehlich
Subject: Re: OT: cpu speed & rendering time
Date: 12 Jan 2001 14:23:58
Message: <3a5f59ce@news.povray.org>
|
|
 |
|  |
|  |
|
 |
In article <3A5F48B9.FBA4E5E0@my-dejanews.com> , "Greg M. Johnson"
<gre### [at] my-dejanews com> wrote:
> Do you really mean that a 450 MHz Pentium II is as fast as a 450 MHz Pentium
> III, Celeron, etc., as far as povray rendering time? It goes against my
> intuition, or at least my cynical view that it's not that simple.
No, you are confusing CPU speed with render time. They are two _different_
things. You asked for a comparison of "CPU speed" vs "render time". Render
time is a measurement of "CPU speed", thus you can't use one to compare the
other...
I think what Warp wants to point out is that CPU clock frequency (which is
what you seem to be up to) has absolutely no value to compare anything.
With this knowledge, the only useful comparison is to give a full system
configuration, both hardware and software, include such obscure factors like
disk fragmentation, and then look at the render time.
I suspect your question is in the direction of what CPU should I buy to
render fastest with POV-Ray? As for this question, find a friend with a
similar configuartion you would like to buy and install POV-Ray. Is it fast
enough for you compared to you current system? If yes, buy exactly the same
system, if no, go to another friend and repeat the exercise.
Thorsten
____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trf de
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
In article <3a5f59ce@news.povray.org> , "Thorsten Froehlich"
<tho### [at] trf de> wrote:
> Render
> time is a measurement of "CPU speed", thus you can't use one to compare the
> other...
Grr, this was supposed to be "isn't a measurement of"...
____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trf de
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Thanks but that's hopelessly impractical advice. What I see are sales on
different PC's. I was sort of waiting for the price of a 1000 MHz PIII to drop
but now I see AMD chips fall into reasonable prices. And I have **already** once
before asked my local Circuit City if I could install povray on each of their
machines. The clerk asked his manager, who asked me if I had ever heard of
something called a "computer virus," the risk being too great.......
Thorsten Froehlich wrote:
> I suspect your question is in the direction of what CPU should I buy to
> render fastest with POV-Ray? As for this question, find a friend with a
> similar configuartion you would like to buy and install POV-Ray. Is it fast
> enough for you compared to you current system? If yes, buy exactly the same
> system, if no, go to another friend and repeat the exercise.
>
> Thorsten
>
> ____________________________________________________
> Thorsten Froehlich, Duisburg, Germany
> e-mail: tho### [at] trf de
>
> Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Greg M. Johnson" wrote:
>
> Warp wrote:
>
> > If rendering speed (with povray) is the same, then it really doesn't
> > matter which CPU you use :)
>
> Do you really mean that a 450 MHz Pentium II is as fast as a 450 MHz Pentium
> III, Celeron, etc., as far as povray rendering time?
Believe or not, the above is usually true for the three processors you named.
In my experience, the main factor in POV performance is the processor's FPU
(Floating Point Unit), the part of the processor which handles math-intensive
computations such as - surprise - raytracing. The three processors you mention
all belong to the same family of processors (which also includes the old Pentium
Pro, but not Pentium IV). As such they have all but identical floating point
units.
I used to have a Celeron myself, and its POV benchmarks were practically equal
to those of Pentium II / Pentium III at the same clock speed.
--
Margus Ramst
Personal e-mail: mar### [at] peak edu ee
TAG (Team Assistance Group) e-mail: mar### [at] tag povray org
Home page http://www.hot.ee/margusrt
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Greg M. Johnson" wrote:
>
> Anyone know the difference between a:
>
> AMD Athlon
> Pentium III
> Intel Celeron
>
> for similar speed?
Athlon (and its little brother, Duron) is probably best of the three, at least
in POV-Ray.
Pentium III should be similar in terms of performance, but costs quite a bit
more than an equivalent Athlon.
Celeron, while probably a reasonable performer in POV, is by now widely regarded
as overpriced and underpowered. It recently experienced a mild facelift
(migrating to a 100MHz bus speed), but in my opinion that's too little too late.
If you want a good machine for POV-Ray, I'd recommend the following:
Athlon "Thunderbird" for best performance (currently goes up to 1.2 GHz)
Athlon "Duron" for best value (currently up to 850 MHz)
The P III is also a solid processor if you've got money to burn. It reputedly
has less compatibility problems than Athlon (although I've had no trouble with
my Duron).
I don't recommend the Celeron; nor do I recommend the new Pentium 4, at least
until system prices drop and optimized software becomes available.
--
Margus Ramst
Personal e-mail: mar### [at] peak edu ee
TAG (Team Assistance Group) e-mail: mar### [at] tag povray org
Home page http://www.hot.ee/margusrt
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Greg M. Johnson" wrote:
> > If rendering speed (with povray) is the same, then it really doesn't
> > matter which CPU you use :)
>
> Do you really mean that a 450 MHz Pentium II is as fast as a 450 MHz Pentium
> III, Celeron, etc., as far as povray rendering time? It goes against my
> intuition, or at least my cynical view that it's not that simple.
Indeed it should. That's not what he means. "Statistics" are easily abused by
people who don't understand the math or do well and can use it to their benefit.
Marketing people know how to do that. Like the Pentium IV; more MHz but same
megaflops, or RDRAM, faster transfer but slower access, or those supposed "40x"
cd-roms that take five seconds to spin up and misread (ie reread) data... people
like numbers. Americans are the worst when it comes to that...
--
David Fontaine <dav### [at] faricy net> ICQ 55354965
My raytracing gallery: http://davidf.faricy.net/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Greg M. Johnson" wrote:
> Anyone know the difference between a:
>
> AMD Athlon
> Pentium III
> Intel Celeron
>
> for similar speed?
You have 'em in order. Any of the new AMDs should give POV a significant
boost over Intel for equivalent MHz, and those all come in the same
range (Athlon and PIII up to around 1.2GHz I believe). The Pentium IV
allows clock speeds much higher but actually does less per cycle IIRC,
so in the end a high-end PIV is probably slower rendering than a
high-end PIII. My advice, go for AMD. (And for equivalent MHz the mac
chips waste any of them, but they don't come nearly as fast in MHz; I'm
pretty sure the fastest PC chips still beat the fastest mac chips. But
macs don't count. ;)
--
David Fontaine <dav### [at] faricy net> ICQ 55354965
My raytracing gallery: http://davidf.faricy.net/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Christoph Hormann wrote:
BTW, the German c't magazine regularly publishes Povray benchmark results
> for all CPUs.
>
> Christoph
>
Is this information posted on the web?
...Chambers
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
David Fontaine wrote:
<snip>
supposed "40x"
> cd-roms that take five seconds to spin up and misread (ie reread) data... people
> like numbers. Americans are the worst when it comes to that...
Oh yeah!?!? Where are your numbers to PROVE that!
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Dearmad wrote:
> > supposed "40x"
> > cd-roms that take five seconds to spin up and misread (ie reread) data... people
> > like numbers. Americans are the worst when it comes to that...
>
> Oh yeah!?!? Where are your numbers to PROVE that!
Ok, change it to "I get the impression Americans are the worst when it comes to
that..."
--
David Fontaine <dav### [at] faricy net> ICQ 55354965
My raytracing gallery: http://davidf.faricy.net/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
oh play coy with me... as if you didn't get it... sure... yeah right...
--
http://www.teleport.com/~dearmad
But why bother? I'm not interesting.
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Margus Ramst wrote:
>
> "Greg M. Johnson" wrote:
> >
> > Warp wrote:
> >
> > > If rendering speed (with povray) is the same, then it really doesn't
> > > matter which CPU you use :)
> >
> > Do you really mean that a 450 MHz Pentium II is as fast as a 450 MHz Pentium
> > III, Celeron, etc., as far as povray rendering time?
>
> Believe or not, the above is usually true for the three processors you named.
> In my experience, the main factor in POV performance is the processor's FPU
> (Floating Point Unit), the part of the processor which handles math-intensive
> computations such as - surprise - raytracing. The three processors you mention
> all belong to the same family of processors (which also includes the old Pentium
> Pro, but not Pentium IV). As such they have all but identical floating point
> units.
Not really. The FPU of the Athlon is a lot better than that of a PII or
PIII at the same clock speed.
I take Ken's side on the issue, go for the Athlon.
> I used to have a Celeron myself, and its POV benchmarks were practically equal
> to those of Pentium II / Pentium III at the same clock speed.
The main difference between a Celeron and a "true" PII or PIII is the
lack of onchip cache, so the speed difference with FPU intensive apps
tends to be less noticeable than with "regular" use, which is moving
gigaglobs of memory in and out of the cache (as is word-processing or
sheet-spreading)
--
Francois Labreque | Unfortunately, there's no such thing as a snooze
flabreque | button on a cat who wants breakfast.
@ | - Unattributed quote from rec.humor.funny
videotron.ca
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Francois Labreque wrote:
>
> Not really. The FPU of the Athlon is a lot better than that of a PII or
> PIII at the same clock speed.
>
I totally agree; I was talking specifically about the 3 processors he named,
Celeron, PII & PIII
> The main difference between a Celeron and a "true" PII or PIII is the
> lack of onchip cache
Current Celerons do have onchip L2 cache, but it's smaller than that of
PII/PIII.
--
Margus Ramst
Personal e-mail: mar### [at] peak edu ee
TAG (Team Assistance Group) e-mail: mar### [at] tag povray org
Home page http://www.hot.ee/margusrt
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
And another one here, just in case...
:o)
--
http://www.teleport.com/~dearmad
But why bother? I'm not interesting.
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
On Fri, 12 Jan 2001 09:34:52 -0500, "Greg M. Johnson"
<gre### [at] my-dejanews com> wrote:
>Anyone know the difference between a:
>
>AMD Athlon
>Pentium III
>Intel Celeron
>
>for similar speed?
Performance at the same CPU clock is in the order you have specified
them, fastest on top. The Pentium III is better suited for games
(better supported AGP on the mainboard).
Check Tom's Hardware Guide (any search engine should be able find it),
it is the number one place for such FAQs.
Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] vip bg
TAG e-mail : pet### [at] tag povray org
Post a reply to this message
|
 |
|  |
|  |
|
 |
From: Christoph Hormann
Subject: Re: OT: cpu speed & rendering time
Date: 13 Jan 2001 04:23:07
Message: <3A601E7B.3EC1149@gmx.de>
|
|
 |
|  |
|  |
|
 |
Ben Chambers wrote:
>
> Is this information posted on the web?
> ...Chambers
I don't think so.
Might check yourself: http://www.heise.de/ct/
Christoph
--
Christoph Hormann <chr### [at] gmx de>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
David Fontaine wrote:
> "Greg M. Johnson" wrote:
>
> > Anyone know the difference between a:
> >
> > AMD Athlon
> > Pentium III
> > Intel Celeron
> >
> > for similar speed?
>
> You have 'em in order. Any of the new AMDs should give POV a significant
> boost over Intel for equivalent MHz, and those all come in the same
> range (Athlon and PIII up to around 1.2GHz I believe). The Pentium IV
> allows clock speeds much higher but actually does less per cycle IIRC,
> so in the end a high-end PIV is probably slower rendering than a
> high-end PIII. My advice, go for AMD. (And for equivalent MHz the mac
> chips waste any of them, but they don't come nearly as fast in MHz; I'm
> pretty sure the fastest PC chips still beat the fastest mac chips. But
> macs don't count. ;)
>
> --
> David Fontaine
Motorola makes the PowerPC G4 733mhz processor. Why are you calling it a
mac chip Apple has little to do with its development from my understanding.
Anyway I recommend the AMD processor despite the local economy's near
complete dependence on Intel. Intel has just made so many mistakes in the
last year. They recalled a million motherboards. I had one of them, and I
can't tell you how many times I thought I figured out why my computer was
behaving so badly. Oh the frustration. Well lets hope their 64 bit
processors will be better.
Dan Johnson
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
In article <3A5FC650.E20A9960@faricy.net>, David Fontaine
<dav### [at] faricy net> wrote:
> (And for equivalent MHz the mac chips waste any of them, but they
> don't come nearly as fast in MHz; I'm pretty sure the fastest PC
> chips still beat the fastest mac chips. But macs don't count. ;)
The "Mac chips" are PowerPCs, which are mainly made by Motorola. Apple,
Motorala, and IBM worked together on these chips, but apparently
Motorola makes the specific ones used in Macs. (and they don't seem to
be working together as much any more)
And while there are PowerPC G4 machines out with over 700MHz (and dual
533MHz G4 machines), the different architectures mean you can't compare
MHz directly to the Pentium compatible chips. There are some things the
PPC chips are *much* faster at, and other things they are a bit slower
at, with the same MHz rating. The biggest advantage is integer math, or
single-precision float math using the Altivec (Velocity Engine) SIMD
acceleration, which means these processors aren't that great for POV,
which very heavily uses double-precision floats and can't take advantage
of SIMD instructions. Not surprisingly, Macs are best at things like
PhotoShop.
--
Christopher James Huff
Personal: chr### [at] mac com, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tag povray org, http://tag.povray.org/
<><
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Margus Ramst <mar### [at] peak edu ee> wrote:
: Believe or not, the above is usually true for the three processors you named.
: In my experience, the main factor in POV performance is the processor's FPU
: (Floating Point Unit)
This is a totally wild guess, but examining the povray code, I would say
that far less than 50% of it (when compiled to machine code) is FPU code.
Perhaps 10% or even less.
Of course the FPU code is concentrated on places which are called
frequently, which raises the % of executed FPU code compared to executed
CPU code. Still, I would say that less than 50% of the code executed by
povray is FPU code.
So the CPU execution speed would play an important role in rendering times.
My point in my (sarcastic) answer was totally misunderstood.
What I said was: "If povray rendering speed is the same in all three
processors, then it doesn't matter which one you use."
I didn't say that the rendering speed is the same in the three processors.
--
char*i="b[7FK@`3NB6>B:b3O6>:B:b3O6><`3:;8:6f733:>::b?7B>:>^B>C73;S1";
main(_,c,m){for(m=32;c=*i++-49;c&m?puts(""):m)for(_=(
c/4)&7;putchar(m),_--?m:(_=(1<<(c&3))-1,(m^=3)&3););} /*- Warp -*/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Dan Johnson wrote:
> Anyway I recommend the AMD processor despite the local economy's near
> complete dependence on Intel. Intel has just made so many mistakes in the
> last year. They recalled a million motherboards. I had one of them, and I
> can't tell you how many times I thought I figured out why my computer was
> behaving so badly. Oh the frustration. Well lets hope their 64 bit
> processors will be better.
Ooh, that sucks. Intel does seem to be going downhill.
--
David Fontaine <dav### [at] faricy net> ICQ 55354965
My raytracing gallery: http://davidf.faricy.net/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
At least AMD is going uphill. I like to see a little change of scenery.
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
I found this reference, too:
�http://www.tabsnet.com/benchmark/pbench.php3
�
Post a reply to this message
|
 |
|  |
|  |
|
 |
From: Christoph Hormann
Subject: Re: OT: cpu speed & rendering time
Date: 15 Jan 2001 14:34:39
Message: <3A6350CF.7C9F6945@gmx.de>
|
|
 |
|  |
|  |
|
 |
Ben Chambers wrote:
>
> Christoph Hormann wrote:
> BTW, the German c't magazine regularly publishes Povray benchmark results
>
> > for all CPUs.
> >
> > Christoph
> >
>
> Is this information posted on the web?
> ...Chambers
I just looked up some values from previous tests, they used the chess2
scene coming with povray 3.1 with 320x240 -aa, but did not specify any
further details, especially not the focal blur quality used. I tested it
with FB_Quality_Default on my K6/500 while some other apps where running
and came to 253 PPS which seems quite reasonable. :-(
Even though chess2.pov is one of the larger sample scenes from povray 3.1
it nearly only uses the CPU (< 1Mb memory use) and therefore i did not
include any other information on the test systems used.
The last issue of the c't (02/2001) contained a test of Pentium 4 systems,
therefore there are quite a lot of values for this processor.
Celeron 533 MHz 333 PPS
Celeron 600 MHz 339 PPS
Pentium 3 733 MHz 512 PPS
Pentium 3 750 MHz 522 PPS
Pentium 3 866 MHz 604 PPS
Pentium 3 1.0 GHz 691/704/691 PPS
Pentium 4 1.4 GHz 738/724/745/745/791/731 PPS
Pentium 4 1.5 GHz 791/783 PPS
Pentium 4 1.6 GHz 834 PPS
Duron 600 MHz 471/474 PPS
Duron 650 MHz 515/515 PPS
Duron 700 MHz 552/556/560 PPS
Athlon 1.2 GHz 960/948 PPS
Christoph
--
Christoph Hormann <chr### [at] gmx de>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
On Fri, 12 Jan 2001 09:34:52 -0500, "Greg M. Johnson"
<gre### [at] my-dejanews com> wrote:
>Anyone know the difference between a:
>
>AMD Athlon
>Pentium III
>Intel Celeron
>
>for similar speed?
Comparisons for similar speed (mhz) have *no sense*
You should do the only reasonable comparison: similar price
Btw the athlon is the best in both fields
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Angelo 'kENpEX' Pesce wrote:
>
> >Anyone know the difference between a:
> >
> >AMD Athlon
> >Pentium III
> >Intel Celeron
> >
> >for similar speed?
>
> Comparisons for similar speed (mhz) have *no sense*
> You should do the only reasonable comparison: similar price
> Btw the athlon is the best in both fields
And in reliability? And in floating point?
(Note: I'm not sarcastic, I really ask)
--
Adrien Beau adr### [at] free fr http://adrien.beau.free.fr/
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |