Atom vs ARM9

Hello,

We are considering using an upcoming version of the Intel Atom SoC, I
believe the current code name is tunnel creek. Anyway, my boss asked me the
other day how we could expect the Intel Atom to perform in comparison to a
comparable ARM9(same amount of cache, clock, bus to RAM, amount of RAM). I
was a bit stumped, another engineer said he would expect the Atom to blow
away a comparable ARM9, he said he would expect it to be two to three times
faster. I'm not familiar with the architecture of the Atom, or what tricks
Intel has come up with recently. It's been about five years since my
computer architecture courses in school. Can anyone done any benchmarks, or
could anyone comment from a theoretical point of view?

Regards,
Eddie

"dawydiuk" <dawydiuk@gmail.com> wrote in message news:29ednbUHM7VP58rXnZ2dnUVZ_j-dnZ2d@giganews.com...
> Hello,
>
> We are considering using an upcoming version of the Intel Atom SoC, I
> believe the current code name is tunnel creek. Anyway, my boss asked me the
> other day how we could expect the Intel Atom to perform in comparison to a
> comparable ARM9(same amount of cache, clock, bus to RAM, amount of RAM). I
> was a bit stumped, another engineer said he would expect the Atom to blow
> away a comparable ARM9, he said he would expect it to be two to three times
> faster. I'm not familiar with the architecture of the Atom, or what tricks
> Intel has come up with recently. It's been about five years since my
> computer architecture courses in school. Can anyone done any benchmarks, or
> could anyone comment from a theoretical point of view?

Atom and ARM9 are two completely different CPUs aimed at different markets.
It's a bit like comparing a 386 with a Pentium. Atom would indeed be about twice as
fast at the same clockspeed, but the ARM9 would use significantly less power even
if you clocked it faster to get the same performance. ARM9's typically run at around
200MHz (the fastest are 600MHz), so it's not in the same league as a 1.6GHz Atom.

A better comparison would be with the 1GHz Cortex-A8, which is like Atom
an in-order dual issue superscalar. There are few publicly available benchmarks,
but ARM published benchmarks showing Cortex-A8 easily beat Atom in web
browsing while using just a fraction of the power. I tried a Javascript benchmark
which the 600MHz Cortex-A8 in the iPhone runs faster than my 1.8GHz Athlon64.
It's funny Intel originally promoted Atom talking about it's supposedly superior web
browsing capabilities compared to ARM...

Power consumption is Atoms main issue, not only the core but the whole chipset.
There are people who claim that Intel can overcome these issues, but I believe that
the extra cost of all the x86 compatibility is large enough that it cannot make an
x86 CPU which would be suitable for a mobile phone.

Wilco 

Wilco Dijkstra wrote:

> Power consumption is Atoms main issue, not only the core but the whole chipset.
> There are people who claim that Intel can overcome these issues, but I believe that
> the extra cost of all the x86 compatibility is large enough that it cannot make an
> x86 CPU which would be suitable for a mobile phone.

Roger Wilco,

In 1943, it was reported that the world would only ever need a maximum 
of 5 personal computers.

Almost anything is possible. :-)

Cheers Don...




-- 
Don McKenzie

Site Map:            http://www.dontronics.com/sitemap
E-Mail Contact Page: http://www.dontronics.com/email
Web Camera Page:     http://www.dontronics.com/webcam
No More Damn Spam:   http://www.dontronics.com/spam

Breakout, Prototype, Development, & Robotics Boards.
http://www.dontronics-shop.com/sparkfun-electronics.html

Don McKenzie schreef:
> Wilco Dijkstra wrote:
> 
>> Power consumption is Atoms main issue, not only the core but the whole 
>> chipset.
>> There are people who claim that Intel can overcome these issues, but I 
>> believe that
>> the extra cost of all the x86 compatibility is large enough that it 
>> cannot make an
>> x86 CPU which would be suitable for a mobile phone.
> 
> Roger Wilco,
> 
> In 1943, it was reported that the world would only ever need a maximum 
> of 5 personal computers.

I doubt that in 1943 they were even considering the concept of 
*personal* computers.

On Sat, 11 Jul 2009 10:40:08 +0200, Dombo wrote:

>> In 1943, it was reported that the world would only ever need a maximum 
>> of 5 personal computers.
> 
> I doubt that in 1943 they were even considering the concept of 
> *personal* computers.

Yep; it was "5 computers", period.

On Fri, 10 Jul 2009 12:30:58 -0500, "dawydiuk" <dawydiuk@gmail.com>
wrote:

>Hello,
>
>We are considering using an upcoming version of the Intel Atom SoC, I
>believe the current code name is tunnel creek. Anyway, my boss asked me the
>other day how we could expect the Intel Atom to perform in comparison to a
>comparable ARM9(same amount of cache, clock, bus to RAM, amount of RAM). I
>was a bit stumped, another engineer said he would expect the Atom to blow
>away a comparable ARM9, he said he would expect it to be two to three times
>faster. I'm not familiar with the architecture of the Atom, or what tricks
>Intel has come up with recently. It's been about five years since my
>computer architecture courses in school. Can anyone done any benchmarks, or
>could anyone comment from a theoretical point of view?
>
>Regards,
>Eddie

Another metric for your boss: how long does Intel usually keep a chip
in production?

John

Nobody wrote:
> On Sat, 11 Jul 2009 10:40:08 +0200, Dombo wrote:
> 
>>> In 1943, it was reported that the world would only ever need a maximum 
>>> of 5 personal computers.
>> I doubt that in 1943 they were even considering the concept of 
>> *personal* computers.
> 
> Yep; it was "5 computers", period.

ohhps, correct, take that back.

I didn't check the words that are coming out of my mouth (or keyboard 
for that matter) :-)

Don...




-- 
Don McKenzie

Site Map:            http://www.dontronics.com/sitemap
E-Mail Contact Page: http://www.dontronics.com/email
Web Camera Page:     http://www.dontronics.com/webcam
No More Damn Spam:   http://www.dontronics.com/spam

Breakout, Prototype, Development, & Robotics Boards.
http://www.dontronics-shop.com/sparkfun-electronics.html

John Larkin wrote:
> On Fri, 10 Jul 2009 12:30:58 -0500, "dawydiuk" <dawydiuk@gmail.com> wrote:
> Another metric for your boss: how long does Intel usually keep a chip
> in production?
another hint :
why is he the boss and asks lower ranking people ?
that should ring the bell ;-)
Go ARM, since Atom is needed only when legacy SW is required.
And today, PC legacy CAN be overcome, and with it, soooo many troubles.

> John
yg
-- 
http://ygdes.com / http://yasep.org

>Atom and ARM9 are two completely different CPUs aimed at different
markets.
>It's a bit like comparing a 386 with a Pentium. Atom would indeed be
about twice as
>fast at the same clockspeed, but the ARM9 would use significantly less
power even
>if you clocked it faster to get the same performance. 

Interesting, this is exactly what I was looking for.

>ARM9's typically run at around
>200MHz (the fastest are 600MHz), so it's not in the same league as a
1.6GHz Atom.

There are faster chips available. We are currently designing in a 1.2 GHz
ARM9 from Marvell. Although, Marvell owns the rights to modify the
instruction set so to be fair it's a modified ARM9 core.

>Power consumption is Atoms main issue, not only the core but the whole
chipset.
>There are people who claim that Intel can overcome these issues, but I
believe that
>the extra cost of all the x86 compatibility is large enough that it
cannot make an
>x86 CPU which would be suitable for a mobile phone.

We build industrial SBCs... I believe we will be connecting a touch panel
display to the Atom(assuming we design in this part). I really wasn't sure
how the 1.2GHz ARM9 from Marvell would compare to a similiar Intel Atom
chip... Thanks for the info.

Regards,
Eddie

>Another metric for your boss: how long does Intel usually keep a chip
>in production?

Intel is trying to break into the embedded industry again, so they have
been doing a few things differently to try to appeal embedded designers.
For instance several versions of the Atom will be guarenteed to be
available for at least five years.

Regards,
Eddie