I'm still planning my future computer. (Calling all geeks)

[faded glory]

"You really have no idea at all what you're talking about, cut your losses..."

Oh boy I could soooo prove you wrong asher and your lucky my abilities to argue these things have always been, rather lacking. A multiproccesser technology is just a SHARING of processing. IT doesnt mean there are 2 processers working to make the world a better place.

CPU 1: Hi, I will do the Graphics app's, Ballistics, and physics

CPU 2: Ok, I guess Ill cover the rest.

And that is what it boils down to. It really ISNT that much faster.


"What HyperThreading (otherwise known as Simultaneous MultiThreading) does is allow two threads to run on the same processor, rather than one. If one thread is doing AI work on the ALU, another thread can be doing physics calculations using the SIMD and/or FPU units at the same time"

Asher that is a load of marketing horse****! An AMD XP 3000 beats a P4 with 'HT' technology in high end software applications......what does that tell you? HT is horse****.

Oh and dont talk down to me... I know what these things are. I dont take down notes from mega-games like other nerds. Who think they know how it works, but have no clue.

[faded glory]

Is this some kind of joke?

You didnt know?

Somebody doesnt have his A+ Certification...tsk..tsk..

[Asher]

Originally posted by faded glory
Oh boy I could soooo prove you wrong asher and your lucky my abilities to argue these things have always been, rather lacking.

That's because you constantly argue about things you have no knowledge about.

A multiproccesser technology is just a SHARING of processing. IT doesnt mean there are 2 processers working to make the world a better place.

I'm not sure what you're trying to say. If you look at the System Properties dialog with a system with HyperThreading enabled, you'll see two processors listed. It means the CPU acts as two processors, so it can run two threads which interact with different components of the CPU...

Asher that is a load of marketing horse****! An AMD XP 3000 beats a P4 with 'HT' technology in high end software applications......what does that tell you? HT is horse****.

Which apps would those be?
It certainly wouldn't be any of these, would it?

Oh and dont talk down to me... I know what these things are.

No, you don't.

I dont take down notes from mega-games like other nerds. Who think they know how it works, but have no clue.

You're not a nerd, and maybe you shouldn't argue with one. Chances are they know what they're talking about more than you will.

[faded glory]

Oh yes Buslines. The little grids etched on everything. The real bottlenecks that all the IRQ and DMA requests go through with the data to the cpu.. IBM owns the patents. They have since the 8 bit ISA

[Asher]

Originally posted by faded glory
Is this some kind of joke?

You didnt know?

Somebody doesnt have his A+ Certification...tsk..tsk..

Got it back in the 10th grade. Any dumbass with 2 hours of sparetime before the test could get his A+ cert.

You're still talking out of your ass about "buslines", and you clearly don't understand why Serial ATA exists, or anyone's future plans for it.

Oh yes Buslines. The little grids etched on everything. The real bottlenecks that all the IRQ and DMA requests go through with the data to the cpu.. IBM owns the patents. They have since the 8 bit ISA

Serial ATA is a replacement for today's current Parallel ATA. IBM, of course, is one of the people who helped develop Serial ATA, by the way.

It's still not a bus. And IBM doesn't have a patent on busses in particular, just some components used in some busses. Those busses, ironically, are being phased out in 2003-2004 by Intel and its new standards...

[Urban Ranger]

Originally posted by Asher
What HyperThreading (otherwise known as Simultaneous MultiThreading) does is allow two threads to run on the same processor, rather than one. If one thread is doing AI work on the ALU, another thread can be doing physics calculations using the SIMD and/or FPU units at the same time.

The very fundamental problem is this: these CPUs are already supposedly "super-scalar," with this executing multiple micro-ops, retiring out of order, branching look ahead, and all that good stuff. So, fundamentally, all the execution units are supposed to be occupied doing something, because of the "smart" controls.

So, when you find some idle execution units in one of them, it can mean one of two things:

1. The control logic isn't smart enough
2. The execution speed outstrips data transfer speed

Neither of these problems can be fixed by "hyperthreading," that's why I said it is a gimmick.

[Asher]

Originally posted by Urban Ranger
The very fundamental problem is this: these CPUs are already supposedly "super-scalar," with this executing multiple micro-ops, retiring out of order, branching look ahead, and all that good stuff. So, fundamentally, all the execution units are supposed to be occupied doing something, because of the "smart" controls.

So, when you find some idle execution units in one of them, it can mean one of two things:

1. The control logic isn't smart enough
2. The execution speed outstrips data transfer speed

Neither of these problems can be fixed by "hyperthreading," that's why I said it is a gimmick.

I don't see why you'd think a superscalar architecture automatically would mean all of the units are being fed?

The problem is the units are all being fed, provided that thread uses the units. A processor today can only use one thread, remember, if that thread is only touching ALU stuff throughout the pipeline, the SIMD/FPU stuff sits idle.

[Asher]

If HyperThreading is a gimick, by the way, why did both Intel and DEC implement it? And why do apps optimized for HyperThreading show such gains, even in a primitive implementation in today's P4s?

[faded glory]

edit brb

[Urban Ranger]

Originally posted by Asher
I don't see why you'd think a superscalar architecture automatically would mean all of the units are being fed?

When I mentioned "execution units," I referred to the ALU's since they are fundamental. FPU is optional.

Originally posted by Asher
The problem is the units are all being fed, provided that thread uses the units. A processor today can only use one thread, remember, if that thread is only touching ALU stuff throughout the pipeline, the SIMD/FPU stuff sits idle.

"Threads" make no difference at that level.

Remember, "superscalar" architecture is supposed to mean multiple micro-ops executed at the same clock cycle because of the hardware parallelism. This should translates to all execution units (ALU's) in use every cycle. If this is not the case, they should fix this problem.

[faded glory]

[Asher]

Originally posted by Urban Ranger
When I mentioned "execution units," I referred to the ALU's since they are fundamental. FPU is optional.

"Threads" make no difference at that level.

Remember, "superscalar" architecture is supposed to mean multiple micro-ops executed at the same clock cycle because of the hardware parallelism. This should translates to all execution units (ALU's) in use every cycle. If this is not the case, they should fix this problem.

I'm still confused why you think that superscalar translates to all execution units in use every cycle. Even if it were the case that all ALUs were in use, it's not the case that the FPU and SIMD units are in use, which HyperThreading helps fix.

HT is one way to increase ILP, that's why it's there. And it does work in the realworld in most cases, will be improved in Prescott (Prescott doubles much of the logic on the chip in key areas, as well as L1/L2 cache), and adds instructions for thread syncs.

[faded glory]

yes so the p4 is about a hair faster. Does it really live up to its hype? Thats the question. No it doesnt...

[Asher]

Originally posted by faded glory
(image)

What's this supposed to show? That the 3GHz P4 is exactly the same speed as a 3000+ Athlon in a specific 3DS benchmark?

So what?

[Asher]

Originally posted by faded glory
yes so the p4 is about a hair faster. Does it really live up to its hype? Thats the question. No it doesnt...

1) The P4's Osweld is looking at is nearly identical in price to the AMD
2) The 3DS benchmark you showed still shows the P4 being faster, and 3DS has not been optimized for HyperThreading, just generic multithreading.

That was never the question, by the way. The question was, what CPU should Osweld get...

And I don't see why you'd judge the "hype" for HyperThreading by showing a benchmark of a program which was made before HyperThreading even existed.