slvr_phoenix
Splendid
You have some very good points. But it shows how you seem to concentrate mostly on just AMD. So I'll give you some thoughts from the Intel side as well so that you don't have to worry about any lack of competition coming up.
Yes, the Hammer may be able to squash the current P4 designs on a clock for clock basis. But the current P4 isn't the end of the line for the P4. Don't you think that when Intel switches to .13micron etching they'll be able to fit that FPU back into the P4 specs (and more)? And to boot it'll run cooler as well AND have a great more potential than just a measily 1.5GHz.
While the P4 seems to be a dinky piece of arguable junk right now, anything based on the P4 in the future won't be. The P4 is only so crappy right now because of it's FPU.
AND, the P4 has a lot more room for growth than just by means of clock speed. Think about how little gain there is using DDR SDRAM in an Athlon or P3. Why is that? You would think that twice the bandwidth should mean twice the speed. But it's because the chips just weren't designed for high memory bandwidth and thus gain next to nothing from it.
But look at the P4. It's completely different. Intel did a VERY smart thing there by making it so bandwidth intensive. Look how drastic of a difference there is between a P4 running with PC600 RDRAM and a P4 running with PC800 RDRAM. And now imagine what the very same clock speed P4 will be like if you were to put memory with twice the bandwidth in. Or even more. Yes, not only is RDRAM ramping up their clock speed, but also their bus. That means it has a lot more potential for future speed improvements.
And that means that the P4 will get quite noticable speed improvements JUST FROM MEMORY BANDWIDTH ALONE. Nifty, eh? I wonder how long it'll take AMD to catch on to that kind of hardware development.
My only hope is that when Intel does design some SDRAM chipsets for the P4 that these chipsets work in the same way and will allow for much higher bandwidth SDRAM than what is currently available. If not, I can actually see a point in the future where RDRAM might actually serve a purpose. (Even if the RAMBUS company still is run by jerks.)
So yes, while AMD does have some good products now and does look to have at the least an interesting product with their Hammers, Intel is FAR from defeated.
In fact, the only reason that AMD gained so much on Intel is that Intel had assumed it was the only real PC CPU power in the computer market and it could take as long as it wanted to give us processor improvements AND bleed us dry in the process.
Then out of the shadows comes AMD with a massive blow using their Athlon core. And Intel had nothing to fight back with because they'd just been lazy about R&D.
But now that Intel KNOWS it isn't the only one trying to be King Of The Hill, they're kicking their R&D engineers into gear and coming out with some stuff that has great potential. Right now I'm more concerned that Intel will blow AMD away because all AMD thinks that they have to compete against is a P4 chip that's missing half of it's FPU to meet die size requirements. Once Intel fixes that, AMD will be in for a big surprise. Of that, I have no doubts. Especially because by that time RDRAM will be even faster, thus making P4 systems just that much better.
I just hope AMD can cope so that we can continue to see all of these nifty developments between the two companies.
So when you look at both sides of the coin, it really doesn't look like either company is going under any time soon. We'll have at the least another two years of battles.
- Sanity is purely based on point-of-view.
Yes, the Hammer may be able to squash the current P4 designs on a clock for clock basis. But the current P4 isn't the end of the line for the P4. Don't you think that when Intel switches to .13micron etching they'll be able to fit that FPU back into the P4 specs (and more)? And to boot it'll run cooler as well AND have a great more potential than just a measily 1.5GHz.
While the P4 seems to be a dinky piece of arguable junk right now, anything based on the P4 in the future won't be. The P4 is only so crappy right now because of it's FPU.
AND, the P4 has a lot more room for growth than just by means of clock speed. Think about how little gain there is using DDR SDRAM in an Athlon or P3. Why is that? You would think that twice the bandwidth should mean twice the speed. But it's because the chips just weren't designed for high memory bandwidth and thus gain next to nothing from it.
But look at the P4. It's completely different. Intel did a VERY smart thing there by making it so bandwidth intensive. Look how drastic of a difference there is between a P4 running with PC600 RDRAM and a P4 running with PC800 RDRAM. And now imagine what the very same clock speed P4 will be like if you were to put memory with twice the bandwidth in. Or even more. Yes, not only is RDRAM ramping up their clock speed, but also their bus. That means it has a lot more potential for future speed improvements.
And that means that the P4 will get quite noticable speed improvements JUST FROM MEMORY BANDWIDTH ALONE. Nifty, eh? I wonder how long it'll take AMD to catch on to that kind of hardware development.
My only hope is that when Intel does design some SDRAM chipsets for the P4 that these chipsets work in the same way and will allow for much higher bandwidth SDRAM than what is currently available. If not, I can actually see a point in the future where RDRAM might actually serve a purpose. (Even if the RAMBUS company still is run by jerks.)
So yes, while AMD does have some good products now and does look to have at the least an interesting product with their Hammers, Intel is FAR from defeated.
In fact, the only reason that AMD gained so much on Intel is that Intel had assumed it was the only real PC CPU power in the computer market and it could take as long as it wanted to give us processor improvements AND bleed us dry in the process.
Then out of the shadows comes AMD with a massive blow using their Athlon core. And Intel had nothing to fight back with because they'd just been lazy about R&D.
But now that Intel KNOWS it isn't the only one trying to be King Of The Hill, they're kicking their R&D engineers into gear and coming out with some stuff that has great potential. Right now I'm more concerned that Intel will blow AMD away because all AMD thinks that they have to compete against is a P4 chip that's missing half of it's FPU to meet die size requirements. Once Intel fixes that, AMD will be in for a big surprise. Of that, I have no doubts. Especially because by that time RDRAM will be even faster, thus making P4 systems just that much better.
I just hope AMD can cope so that we can continue to see all of these nifty developments between the two companies.
So when you look at both sides of the coin, it really doesn't look like either company is going under any time soon. We'll have at the least another two years of battles.
- Sanity is purely based on point-of-view.