64-bit desktop computing unnecessary, says Intel CTO

Pandaren

Golden Member
Sep 13, 2003
1,029
0
0
didn't expect him to say anything different

Of course not. I am sure that Yamhill is real and inactive on Prescott processors. Intel won't enable it unless it becomes clear that Itanium is sinking for good.
 

BoberFett

Lifer
Oct 9, 1999
37,562
9
81
He can say that, but he knows deep down they're going to need 64 bit soon.

Intel loves to play the numbers game. If they can sacrifice a few IPCs for a higher clockspeed, they're more than happy to because they know that Joe Schmoe can tell the Intel is faster. I mean, the Pentium's got 3 giggihertz and the Athalon only has 2. That's half again as fast!

Marketed correctly, the Athlon 64 will fly off the shelves, if to nobody other than ill informed users. And while we techno-elites may frown at AMD for underhanded marketing, they won't be the first to play the bit-count game, and they won't be the last. At least it might help their bottom line a bit so we have some competition to Intel in the coming years.
 

dexvx

Diamond Member
Feb 2, 2000
3,899
0
0
Originally posted by: BoberFett
Intel loves to play the numbers game. If they can sacrifice a few IPCs for a higher clockspeed, they're more than happy to because they know that Joe Schmoe can tell the Intel is faster. I mean, the Pentium's got 3 giggihertz and the Athalon only has 2. That's half again as fast!

Then explain to me why Centrino Pentium-M laptops are so damn popular and why the Itanium II 1.5/6MB is the highest single processor in the world.

Your logic is flawed.
 

HokieESM

Senior member
Jun 10, 2002
798
0
0
Well, he's right in a sense. I don't think we NEED 64-bit computing. Especially for Joe-Schmoe, average consumer. Look at the Pentium-M. Would you love to see it on a 90nm process running at 2.4GHz? (Look at some of the PIII-S benchmarks when overclocked.... and that's with a 160MHz FSB or so).

BUT I'm not an Intel fanboy.... and I really DO think they should be roasted (pardon the pun) for planning on putting out Prescott pumping out ~105W of heat. I don't want a space-heater in my room--which is why I went from the Tbird 1.4 (MAN that thing was hot) to a P4 Northwood. Not to mention, I built my first "quiet" computer about a month ago--and I LOVE it (its a PIII-S, btw). Its nice not to have a whiney hard drive in your ear.... not to mention, its funny to see your friends turn it "on" to play games (when its actually already on). :)

I look forward to the arrival of the Athlon64. I don't have the cash to buy one (which might be good--forcing me to wait on 64-bit compilers/apps could be a good thing)..... but I like the competition that it will bring.
 

BoberFett

Lifer
Oct 9, 1999
37,562
9
81
Originally posted by: dexvx
Originally posted by: BoberFett
Intel loves to play the numbers game. If they can sacrifice a few IPCs for a higher clockspeed, they're more than happy to because they know that Joe Schmoe can tell the Intel is faster. I mean, the Pentium's got 3 giggihertz and the Athalon only has 2. That's half again as fast!

Then explain to me why Centrino Pentium-M laptops are so damn popular and why the Itanium II 1.5/6MB is the highest single processor in the world.

Your logic is flawed.

Uh oh, little boy wants to play with the big kids. Sigh.

Yes you're right. My logic is flawed, because the shelves at Best Buy are full of Itanium based desktops. Damn near every computer I see stuffed into the back of an SUV filled with screaming kids has a big Itanium sticker on it.
rolleye.gif


And what exactly do you mean by "the Itanium II 1.5/6MB is the highest single processor in the world"? The highest what? It returns the highest benchmarks in TPC tests? Highest framerate in Half Life 2? It can smoke more ganja than any other chip? Please, quantify your statement.

As far as the Centrino, what are some sales figures compared to laptops based on other CPUs? Please compare them to other laptops in terms of clock speed and units sold to home users.

Come back with some ammo, fanboy.
 

dexvx

Diamond Member
Feb 2, 2000
3,899
0
0
Originally posted by: BoberFett
Originally posted by: dexvx
Originally posted by: BoberFett
Intel loves to play the numbers game. If they can sacrifice a few IPCs for a higher clockspeed, they're more than happy to because they know that Joe Schmoe can tell the Intel is faster. I mean, the Pentium's got 3 giggihertz and the Athalon only has 2. That's half again as fast!

Then explain to me why Centrino Pentium-M laptops are so damn popular and why the Itanium II 1.5/6MB is the highest single processor in the world.

Your logic is flawed.

Uh oh, little boy wants to play with the big kids. Sigh.

... blah blah blah....

Come back with some ammo, fanboy.

Lets get to the point of the arguement instead of straying, little kid.

You said Intel would sacrifice high IPC for high clock speed. I pointed out 2 products in Intel's line that has low clockspeed but high IPC. I have disproven your lovely theory about Intel loving high clock speeds.

So come back with some ammo, fanboi and prove me wrong.
 

Excelsior

Lifer
May 30, 2002
19,047
18
81
Originally posted by: dexvx
Originally posted by: BoberFett
Originally posted by: dexvx
Originally posted by: BoberFett
Intel loves to play the numbers game. If they can sacrifice a few IPCs for a higher clockspeed, they're more than happy to because they know that Joe Schmoe can tell the Intel is faster. I mean, the Pentium's got 3 giggihertz and the Athalon only has 2. That's half again as fast!

Then explain to me why Centrino Pentium-M laptops are so damn popular and why the Itanium II 1.5/6MB is the highest single processor in the world.

Your logic is flawed.

Uh oh, little boy wants to play with the big kids. Sigh.

... blah blah blah....

Come back with some ammo, fanboy.

Lets get to the point of the arguement instead of straying, little kid.

You said Intel would sacrifice high IPC for high clock speed. I pointed out 2 products in Intel's line that has low clockspeed but high IPC. I have disproven your lovely theory about Intel loving high clock speeds.

So come back with some ammo, fanboi and prove me wrong.

No you haven't disproven the theory that Intel loves high clock speeds. Its a fact. We are talking about desktop chips here, not the Itanium. Now, in desktop chips, does Intel favor high mhz and low IPC or high IPC and lower mhz? You lose.
 

jbond04

Senior member
Oct 18, 2000
505
0
71
Originally posted by: BoberFett

And what exactly do you mean by "the Itanium II 1.5/6MB is the highest single processor in the world"? The highest what? It returns the highest benchmarks in TPC tests? Highest framerate in Half Life 2? It can smoke more ganja than any other chip? Please, quantify your statement.

...

Come back with some ammo, fanboy.

Here are the SPEC2000 results for the Intel Itanium II 1.5GHz:
FP Base/Peak: 2119/2119
Int Base/Peak: 1322/1322

Here are the SPEC2000 results for the AMD Opteron 246:
FP Base/Peak: 1209/1293
Int Base/Peak: 1248/1317

These are both for single processor systems. The Intel Itanium benchmarks come from an HP Integrity RX2600 server and the AMD Opteron benchmarks come from an AMD configured (Rioworks motherboard) system.

And Excelsior, I don't think the high MHz, low IPC strategy is necessarily a bad one. I don't know if you've taken a look at the P4 lineup lately, but the 3.2GHz P4 is definitely faster than the low MHz, high IPC Athlon XP 3200+. Granted, the Athlon 64 is rumored to be pretty impressive, but Prescott is waiting just around the corner (early November), running at 3.4GHz with several architectural advantages over the Northwood. I doubt the "high MHz, low IPC" Prescott will be an easy processor to beat...
 

SexyK

Golden Member
Jul 30, 2001
1,343
4
76
I never understood why people slammed Intel for the NetBurst architecture. Sure, the P4 gets less done per cycle than the Athlon XP. So what? It was designed to ramp in clock speed which it's done beautifully... it's just a different approach to design, and I'd say its been pretty succesful looking at the performance of the 3.2GHz P4C. People who think it's lower IPC is a design flaw are seriously confused.
 

miniMUNCH

Diamond Member
Nov 16, 2000
4,159
0
0
Scientific and entainment apps are driving the 64 bit computing platform onto the desktop...with hardware advances in the desktop come new apps for the masses... for example: direct real-time rendering of animated movies (imagine being able to download a animated moive in 10 Mb of compressed openGL, etc. scripts and watch final fantasy or something rendered real time).

Also, the big iron concept or the huge cluster is becoming somewhat passe. Instead of buying a 1 million dollar linux cluster composed of X number of processors...why don't we use all our desktops (which in total are 10X processors) at night to run the computations? There is software available today that does this and the systems can be a variety of different operating systems.


Mr. CTO is talking about how windows and word wont consume 4GB...no duh...word better not consume 4GB of space in 4 years or people will burn Microsoft top brass at the stake.
 

HokieESM

Senior member
Jun 10, 2002
798
0
0
Originally posted by: miniMUNCH
Scientific and entainment apps are driving the 64 bit computing platform onto the desktop...with hardware advances in the desktop come new apps for the masses... for example: direct real-time rendering of animated movies (imagine being able to download a animated moive in 10 Mb of compressed openGL, etc. scripts and watch final fantasy or something rendered real time).

Also, the big iron concept or the huge cluster is becoming somewhat passe. Instead of buying a 1 million dollar linux cluster composed of X number of processors...why don't we use all our desktops (which in total are 10X processors) at night to run the computations? There is software available today that does this and the systems can be a variety of different operating systems.


Mr. CTO is talking about how windows and word wont consume 4GB...no duh...word better not consume 4GB of space in 4 years or people will burn Microsoft top brass at the stake.

Actually, one of the BIG talks is that clusters are beginning to be passe for scientific computing--because we need faster SERIAL execution of code. :) Clusters are great.... as are fast desktops with distributed computing. But nothing can make up for pure speed. Of course, we're talking research-level computing for people with deep pockets. I agree that 64-bit computing will let more people do research at home. Its amazing how fast a computer is today... compared to 10 years ago. (which isn't that long ago in ANYTHING but technology)

You make a good point about video rendering. I think DVD ripping/editing will become more the norm--especially when digital camcorders become more popular (among the masses).

 

mfs378

Senior member
May 19, 2003
505
0
0
Low IPC is not neccesarily a flaw, but it is a compromise. In this case, Intel sacrificed IPC for clock speed. I feel that their goals in processor design should be delivering the best performance at a given price point. Clock speed is irrelevant to me. But I guess that is too much to expect of the majority of people who buy computers nowadays.
 

Alkali

Senior member
Aug 14, 2002
483
0
0
What else is an INTEL cto going to say?!!!!??


Of course they say that, otherwise they would say, "Go ahead, buy our direct competitors 64-bit processors you need them!"
 

imgod2u

Senior member
Sep 16, 2000
993
0
0
Scientific and entainment apps are driving the 64 bit computing platform onto the desktop...with hardware advances in the desktop come new apps for the masses... for example: direct real-time rendering of animated movies (imagine being able to download a animated moive in 10 Mb of compressed openGL, etc. scripts and watch final fantasy or something rendered real time).

I can see how scientific applications (particularly ones dealing with big integers) can benefit but how does 64-bit computing drive entertainment apps which are mainly FP-intensive?
 

JavaMomma

Senior member
Oct 19, 2000
701
0
71
Originally posted by: dexvx
Originally posted by: BoberFett
Intel loves to play the numbers game. If they can sacrifice a few IPCs for a higher clockspeed, they're more than happy to because they know that Joe Schmoe can tell the Intel is faster. I mean, the Pentium's got 3 giggihertz and the Athalon only has 2. That's half again as fast!

Then explain to me why Centrino Pentium-M laptops are so damn popular and why the Itanium II 1.5/6MB is the highest single processor in the world.

Your logic is flawed.


While I agree with you, I have also found many stores do not list the clock speed of the Centrino...makes it very annoying when shopping around because even the store clerks dont know.
 

dexvx

Diamond Member
Feb 2, 2000
3,899
0
0
Originally posted by: Excelsior
Originally posted by: dexvx
Originally posted by: BoberFett
Originally posted by: dexvx
Originally posted by: BoberFett
Intel loves to play the numbers game. If they can sacrifice a few IPCs for a higher clockspeed, they're more than happy to because they know that Joe Schmoe can tell the Intel is faster. I mean, the Pentium's got 3 giggihertz and the Athalon only has 2. That's half again as fast!

Then explain to me why Centrino Pentium-M laptops are so damn popular and why the Itanium II 1.5/6MB is the highest single processor in the world.

Your logic is flawed.

Uh oh, little boy wants to play with the big kids. Sigh.

... blah blah blah....

Come back with some ammo, fanboy.

Lets get to the point of the arguement instead of straying, little kid.

You said Intel would sacrifice high IPC for high clock speed. I pointed out 2 products in Intel's line that has low clockspeed but high IPC. I have disproven your lovely theory about Intel loving high clock speeds.

So come back with some ammo, fanboi and prove me wrong.

No you haven't disproven the theory that Intel loves high clock speeds. Its a fact. We are talking about desktop chips here, not the Itanium. Now, in desktop chips, does Intel favor high mhz and low IPC or high IPC and lower mhz? You lose.

Yes I have disproven it. Pentium-M and Itanium are low clock high IPC. Do you understand that?

Note the original poster said "Intel loves to play the numbers game. If they can sacrifice a few IPCs for a higher clockspeed, they're more than happy to..." He did not say Intel loves to play blah blah blah with desktops. So next time how about taking some reading classes before you stumble again?
 

ElFenix

Elite Member
Super Moderator
Mar 20, 2000
102,374
8,499
126
Originally posted by: dexvx
Yes I have disproven it. Pentium-M and Itanium are low clock high IPC. Do you understand that?

Note the original poster said "Intel loves to play the numbers game. If they can sacrifice a few IPCs for a higher clockspeed, they're more than happy to..." He did not say Intel loves to play blah blah blah with desktops. So next time how about taking some reading classes before you stumble again?

you must have quit reading right there and not read the part where the processors are marketed to joe schmo who knows damn near nothing about computers.
 

ElFenix

Elite Member
Super Moderator
Mar 20, 2000
102,374
8,499
126
Originally posted by: jbond04

And Excelsior, I don't think the high MHz, low IPC strategy is necessarily a bad one. I don't know if you've taken a look at the P4 lineup lately, but the 3.2GHz P4 is definitely faster than the low MHz, high IPC Athlon XP 3200+. Granted, the Athlon 64 is rumored to be pretty impressive, but Prescott is waiting just around the corner (early November), running at 3.4GHz with several architectural advantages over the Northwood. I doubt the "high MHz, low IPC" Prescott will be an easy processor to beat...

i think with similar engineering budgets both techniques could work equally well, but in different areas (if that makes sense)
 

BoberFett

Lifer
Oct 9, 1999
37,562
9
81
Thanks for the backup guys. Apparently this guy needs to work on his reading comprehension. This WHOLE conversation is about desktops. Not 32 way servers running Itanium.

The title of the damn thread is "64-bit desktop computing unnecessary, says Intel CTO"