AMD K6/2 faster than a 933MHz Pentium II?

http://www.cpu-world.com/cgi-bin/Com...ACTION=Compare

Hello I am just curious to know why the 'doom' benchmaks do not show any
improvement when going froom the K6/2 to the Pentium II 933mhz.
The 1100 Duron ain't much better either.


??? Explaination ???

On Sun, 07 Aug 2005 21:48:53 GMT, "Donald McTrevor"
wrote:


http://www.cpu-world.com/cgi-bin/Com...ACTION=Compare

Hello I am just curious to know why the 'doom' benchmaks do not show any
improvement when going froom the K6/2 to the Pentium II 933mhz.
The 1100 Duron ain't much better either.


??? Explaination ???


You are relying far too much on overly simplified
benchmarks. Even so, game benchmarks are often detemined
more by video than CPU.

In the case of the P3, it appears the chosen motherboard
used a particularly poor chipset for the P3 era (Via 693).
I predict that if the chipset on the P3 motherboard had been
i440BX, Via 694, i820 or i815, the P3 would've had a more
even opportunity (to isolate the video as a bottleneck).

Further, insufficient detail is given about the testing
scenario. We could assume they attempted to provide a fair
gaming benchmark by reusing the same video card, which would
be relatively older (slower) than the average card on a P3
box.

You cannot use those benchmarks for comparison of gaming
because the system they used (if the same as mentioned here,
http://www.cpu-world.com/info/bench/...m_P5M4_M_.html
has 100MHz FSB, PC100 memory, and AGP video. As I'd already
mentioned, your system has several limitations in addition
to CPU, it really wasn't worth the time to upgrade except
perhaps for a young child's use.

"kony" wrote in message
...
On Sun, 07 Aug 2005 21:48:53 GMT, "Donald McTrevor"
wrote:



http://www.cpu-world.com/cgi-bin/Com...II-300GP+%2875
+MHz+2.9V%29+%28CPGA%29+-+Tekram+P5M4-M%2B&CPU1=AMD+K6-2+300+-+AMD-K6-2%2F30
0AFR-66+%28SPGA%29+-+Tekram+P5M4-M%2B&CPU2=Intel+Pentium+III+933+-+933%2F256
%2F133%2F1.7V+%28FCPGA%29+%28FCPGA%29+-+Epox+EP-693A&CPU3=AMD+Duron+1100+-+D
HD1100AMT1B+%28PGA%29+-+Abit+KT7-RAID&CPU4=&PAGE_NAME=COMPARE_BENCH&ACTION=C
ompare

Hello I am just curious to know why the 'doom' benchmaks do not show any
improvement when going froom the K6/2 to the Pentium II 933mhz.
The 1100 Duron ain't much better either.


??? Explaination ???


You are relying far too much on overly simplified
benchmarks. Even so, game benchmarks are often detemined
more by video than CPU.

In the case of the P3, it appears the chosen motherboard
used a particularly poor chipset for the P3 era (Via 693).
I predict that if the chipset on the P3 motherboard had been
i440BX, Via 694, i820 or i815, the P3 would've had a more
even opportunity (to isolate the video as a bottleneck).

Further, insufficient detail is given about the testing
scenario. We could assume they attempted to provide a fair
gaming benchmark by reusing the same video card, which would
be relatively older (slower) than the average card on a P3
box.

Well whatever the case it is a bit shocking that it is actually slower!!
I wouldn't be to happy if I had upgraded my CPU to play 'Doom'
(I assume it's a game) only to find it was slower!!


You cannot use those benchmarks for comparison of gaming
because the system they used (if the same as mentioned here,
http://www.cpu-world.com/info/bench/...m_P5M4_M_.html
has 100MHz FSB, PC100 memory, and AGP video. As I'd already
mentioned, your system has several limitations in addition
to CPU, it really wasn't worth the time to upgrade except
perhaps for a young child's use.

I think you will find young children are more demanding of a system than
I am!! Fancy graphics don't impress me much, space invaders is about
as good as a computer game gets.

On Sun, 07 Aug 2005 23:34:27 GMT, "Donald McTrevor"
wrote:

Further, insufficient detail is given about the testing
scenario. We could assume they attempted to provide a fair
gaming benchmark by reusing the same video card, which would
be relatively older (slower) than the average card on a P3
box.

Well whatever the case it is a bit shocking that it is actually slower!!

Not at all, it goes back to what I've alreay written, that
you cannot just install a different CPU and expect the other
deficiencies in a system to no longer be bottlenecks, in
fact the extent to which they're bottlenecks will be even
more significant.


I wouldn't be to happy if I had upgraded my CPU to play 'Doom'
(I assume it's a game) only to find it was slower!!

Your system will be significantly slower than any in the
comparison at Doom no matter what CPU you have in it.



I think you will find young children are more demanding of a system than
I am!! Fancy graphics don't impress me much, space invaders is about
as good as a computer game gets.

I was thinking of very young children, having no idea how to
use a computer and just having something to peck at on a
keyboard and learn to use a mouse.

"Donald McTrevor" wrote in message
...

http://www.cpu-world.com/cgi-bin/Com...ACTION=Compare

Hello I am just curious to know why the 'doom' benchmaks do not show any
improvement when going froom the K6/2 to the Pentium II 933mhz.
The 1100 Duron ain't much better either.


??? Explaination ???

I dont for a second believe those figures are anything like accurate - 3D
games demand good floating point performance from a processor and the Cyrix
was abysmal at that. The idea that a Cyrix 300 could give 60% of the
framerate you'd get from even a PII300 would stretch crudility never mind a
933mhz chip.

On Mon, 08 Aug 2005 07:23:08 GMT, "Sleepy"
wrote:


"Donald McTrevor" wrote in message
...

http://www.cpu-world.com/cgi-bin/Com...ACTION=Compare

Hello I am just curious to know why the 'doom' benchmaks do not show any
improvement when going froom the K6/2 to the Pentium II 933mhz.
The 1100 Duron ain't much better either.


??? Explaination ???

I dont for a second believe those figures are anything like accurate - 3D
games demand good floating point performance from a processor and the Cyrix
was abysmal at that. The idea that a Cyrix 300 could give 60% of the
framerate you'd get from even a PII300 would stretch crudility never mind a
933mhz chip.

The video card they mention is a particularly poor one, Rage
8MB.

"Sleepy" wrote in message
. uk...

"Donald McTrevor" wrote in message
...


http://www.cpu-world.com/cgi-bin/Com...ACTION=Compare

Hello I am just curious to know why the 'doom' benchmaks do not show any
improvement when going froom the K6/2 to the Pentium II 933mhz.
The 1100 Duron ain't much better either.


??? Explaination ???

I dont for a second believe those figures are anything like accurate - 3D
games demand good floating point performance from a processor and the
Cyrix
was abysmal at that. The idea that a Cyrix 300 could give 60% of the
framerate you'd get from even a PII300 would stretch crudility never mind
a
933mhz chip.


I don't see why we should doubt the figures as I a sure they would have
been questioned if the results looked wong.
Anyway the 300mhz K6/2 I bought for a £1 out performed the PIII 933mhz!!
So I am happy,

On Mon, 08 Aug 2005 20:04:37 GMT, "Donald McTrevor"
wrote:


I don't see why we should doubt the figures as I a sure they would have
been questioned if the results looked wong.

They are being questioned, there are far too many benchmarks
to count, that directly contradict the 'site you linked.


Anyway the 300mhz K6/2 I bought for a £1 out performed the PIII 933mhz!!
So I am happy,


On average a properly configured P3 box will be over 50%
faster, sometimes over 100% faster. So long as you're happy
though, I suppose that's what matters in the end.

"kony" wrote in message
...
On Mon, 08 Aug 2005 20:04:37 GMT, "Donald McTrevor"
wrote:


I don't see why we should doubt the figures as I a sure they would have
been questioned if the results looked wong.

They are being questioned, there are far too many benchmarks
to count, that directly contradict the 'site you linked.

Really? Well I struggled to find them with google, I even posted here
earlier, requesting benchmarks (without any replies).
If any one can post benchmarks to contradict the site please do!!

If they were incorrect I imagine it would have been spotted now,
after several years!!

If you still doubt them why not drop them an email?




Anyway the 300mhz K6/2 I bought for a £1 out performed the PIII 933mhz!!
So I am happy,


On average a properly configured P3 box will be over 50%
faster, sometimes over 100% faster. So long as you're happy
though, I suppose that's what matters in the end.

Well yes but you cannot compare CPU's running in boxes with
vastly different configurations.
Well I might get my K6 tomorrow, I have already forked out
£2.50 on some thermal compound, (more than it cost for the CPU!!).

On slighty unhappier note it seems that a an AMD K6-2 will interpretate
a X2 clock as a X6 clock, so I might be able to run a 450MHz AMD.
So maybe I should have bought a faster processor!!
Will I have to spend another £1 to get one?!!

Maybe the graphics card was a bottle neck but it still showed the
PIII running slower.

I don't think the benchmarks are wrong because if they are they
must have got a whole series of them wrong, here the PIII 700
is also slower.
I think the truth is the K6-2 is a great processor which punches
above it weight in some applications.

On Mon, 08 Aug 2005 21:49:55 GMT, "Donald McTrevor"
wrote:


"kony" wrote in message
.. .
On Mon, 08 Aug 2005 20:04:37 GMT, "Donald McTrevor"
wrote:


I don't see why we should doubt the figures as I a sure they would have
been questioned if the results looked wong.

They are being questioned, there are far too many benchmarks
to count, that directly contradict the 'site you linked.

Really? Well I struggled to find them with google, I even posted here
earlier, requesting benchmarks (without any replies).
If any one can post benchmarks to contradict the site please do!!

1) You can't compare two CPUs by running benchmarks that
stress OTHER parts of a system.

2) I had already attempted to take a short-cut regarding
performance, it is NOT worth the time to do all this
research on a minor upgrade to a decade old system.

3) The 'site you linked, did list figures for synthetic CPU
performance, most of the figures prior to the Game entries
would be the more relevant comparison between the CPUs.

4) Try Google searching for K6-2 & Pentium 3 benchmarks


If they were incorrect I imagine it would have been spotted now,
after several years!!

No, consider that quite a few web reviews are wrong or
inconclusive, if not so poorly set up to be worthless
indications. Just take my word for it, the performance
difference is very large between a K6-2 and a (near 1 GHz)
coppermine P3. There is absolutely nothing the K6-2 does
anywhere near as fast, except possibly some VERY rare
application that only had 3DNow optimizations, and offhand I
can't think of any such applications that didn't have at
least MMX if not SSE support if they had 3Dnow by the time a
P3 Coppermine was near 1GHz.



If you still doubt them why not drop them an email?


At this point who cares?
Do you realize how much time it would take to email every
single web reviewer who had inconclusive (if not worse)
reviews, stats, etc, etc ? It's just not worth the time,
and besides that, the average reviewer is a young kid that
gets very defensive when they're corrected, will argue
themselves into a hole and then just ignore all logic and
other reviews. Even so, the page you linked was merely a
set of results, it did not appear to be a direct comparison
of the two CPUs for any useful purpose- you have taken one
small subset of data out of a much larger amount, out of
context.


On average a properly configured P3 box will be over 50%
faster, sometimes over 100% faster. So long as you're happy
though, I suppose that's what matters in the end.

Well yes but you cannot compare CPU's running in boxes with
vastly different configurations.

That's exactly what the review did, because it used
different motherboard chipsets.

Well I might get my K6 tomorrow, I have already forked out
£2.50 on some thermal compound, (more than it cost for the CPU!!).

Did it ever dawn on you that it might've been so cheap
because it doen't have very good performance? I have a
stack of old K6 CPUs "somewhere", I lost them and it
wouldn't be worth the trouble to find them. It's not just
that they had low performance but so many limitations
compared to other slightly newer yet (today) quite cheap if
not free hardware. Just a couple years later the typical
motherboard had vastly better memory performance running
same PC100 memory, actually had a fair amount of
compatibility with AGP video cards, could run ATA66 hard
drives and support far larger drives. As mentioned
previously, there is a certain "hump" to get over for
moderately reasonable performance at modern tasks and it
takes a little bit newer system to meet that goal.



On slighty unhappier note it seems that a an AMD K6-2 will interpretate
a X2 clock as a X6 clock, so I might be able to run a 450MHz AMD.
So maybe I should have bought a faster processor!!
Will I have to spend another £1 to get one?!!

Unhappier? That's exactly what you should be wanting, since
your board doesnt even support 100MHz FSB.


Maybe the graphics card was a bottle neck but it still showed the
PIII running slower.

No, it showed an entire system running slower. It was a bit
of coincidence that it happened to have a P3 in it.
I'm not going to spend any more time on this silliness, it's
a known fact that a K6-2 is the slowest CPU one could buy,
by the time a P3 Coppermine CPU came out. Well on second
thought there were probably some odd IBM/Via/C3 CPUs still
in the market but they never had much interest from anyone
concerned about performance.


I don't think the benchmarks are wrong because if they are they
must have got a whole series of them wrong, here the PIII 700
is also slower.

Ignorance is bliss?

I think the truth is the K6-2 is a great processor which punches
above it weight in some applications.

To find truth, first you would need more than one
ill-configured benchmark. There were tons of benchmarks out
there and it's just silly to dwell on what is known
industry-wide. There is NO K6-2 that is as fast as even the
slowest P3, even the generation before the P3 coppermine
which was slower and only went up to around 600MHz.

However, same things apply that I mentioned about your
system, that there are other bottlenecks that just a CPU.
If you put a P3 (or a K6-2, or whatever) on a horrible or
misconfigured motherboard, or have OS problems, it may
easily effect the performance.