• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Intel Increases Pipeline stages towards 30 w/ Precott design?

Yeah, it's on the front page of Anandtech too. Doesn't make sense. If it enables higher clock speeds with the cost of higher latency and sacrifice of IPC's , then what's the difference between a 3.2C P4 running with a shorter pipe then a Prescott running at 3.6 with more pipes?
 
The point is the future, and marketing. Being able to scale far higher than 3.6, and making people feel warm and fuzzy for buying something that sounds fast.
 
Ouch. They better make sure they have pretty impressive branch misprediction software built into that. Otherwise it's gonna suffer.
 
Originally posted by: Cat
The point is the future, and marketing. Being able to scale far higher than 3.6, and making people feel warm and fuzzy for buying something that sounds fast.
Click to expand...

A 3.2 only sounds fast? As far as I could tell, It's still one of the fastest chips on the market.

It is indeed marketing, but that doesn't mean it doesn't perform.
 
I never implied that the Northwood 3.2 wasn't fast. I'm merely confirming the fact that a lengthened pipeline normally drops effective IPC, but is probably a sound engineering practice if Intel wants to further escalate clock speed. I don't really care how long a processor's pipelines are, as long as it gets the job done for a reasonable price. If this is what it takes to get to 4 and 5 GHz, and do it cheaply, I'm all for it.
 
I'm going to play devil's advocate here...

People made the same comments when the P4 came out and found that the P3 was faster at a given clockspeed. We were subjected to arguments from both sides of the fence...on the Intel side, it was "more pipelines mean higher frequency" and on the anti-Intel side, it was "more pipelines mean more latency and a bigger penalty for branch prediction misses so this design is going to flop."

Fast forward to today...the P4 design is currently chugging along at 3+ ghz and by almost any benchmark is far faster than the fastest P3 it supplanted. Likewise, in terms of actual performance, the deeply-pipelined P4 design and the shallow-pipelined Athlon design have largely kept pace with one another. If at any point in time one design were found to offer, say, 5x higher throughput than the other design, I think one could confidently claim that design is superior. But in fact that hasn't been the case. At no time over the past 2-3 years has the deeply-pipelined design had a significant performance advantage over the shallow pipeline design. Or visa versa. So these arguments are largely moot.

From a marketing standpoint, though, you can't argue with Intel's logic. And let's be honest, Intel's in business to make money so marketing advantages matter.

 
I'm not impressed. Oh well, AMD has been in the trunk of the car for the past couple years, so it would be nice for Intel to sit in the back seat for a while.
 
Originally posted by: Rectalfier
I'm not impressed. Oh well, AMD has been in the trunk of the car for the past couple years, so it would be nice for Intel to sit in the back seat for a while.
Click to expand...

OK, I am an AMD man as you can see, but I have built P4's. BUT, there is no way I am going to comment on the Prescott until I see at least ONE benchmark article from even a semi-reputable site, preferably 3 or more... I would love to see Intel in the back seat for a few months and give AMD time to make a profit (if that happens), but it is WAY too early to even suggest that....
 
Originally posted by: arcas

Fast forward to today...the P4 design is currently chugging along at 3+ ghz and by almost any benchmark is far faster than the fastest P3 it supplanted. Likewise, in terms of actual performance, the deeply-pipelined P4 design and the shallow-pipelined Athlon design have largely kept pace with one another. If at any point in time one design were found to offer, say, 5x higher throughput than the other design, I think one could confidently claim that design is superior. But in fact that hasn't been the case. At no time over the past 2-3 years has the deeply-pipelined design had a significant performance advantage over the shallow pipeline design. Or visa versa. So these arguments are largely moot.
Click to expand...
One thing to keep in mind though, Arcas, is that Intel may not be so lucky this time around. With the P4, they could seed compilers with information on how to handle the longer pipeline, and more importantly, how to handle SSE2. Prescott however has just the longer pipeline, so there's a decent chance Intel won't be able to get back as much effeciency out of the Prescott, since the only thing going for it this time are more pipeline compiler optimizations, along with the larger cache and the better BPU.
 
I for one hope the Prescotts come through and pull ahead of the Athlon64's. I like the Athlon64's but their chipsets are quite underwhelming. Intel chipsets are almost always rock solid. If AMD would make their own I would be more willing to give them a chance.
 
Here's the way I see it. Intel has already delayed the Prescot due to heat issues. I think they are trying to work around this by increasing the pipeline. It just seems that Intel has done this all of a sudden, since they couldn't produce their original Prescott design, and have thus delayed it. Could this be a massive trip for Intel? The release of the Pentium 4EE sure seemed like an act of desperation. Rest assured, Intel will have something good for us again, but it looks like 2004 will be AMD's biatch.
 
Originally posted by: Rectalfier
Here's the way I see it. Intel has already delayed the Prescot due to heat issues. I think they are trying to work around this by increasing the pipeline. It just seems that Intel has done this all of a sudden, since they couldn't produce their original Prescott design, and have thus delayed it. Could this be a massive trip for Intel? The release of the Pentium 4EE sure seemed like an act of desperation. Rest assured, Intel will have something good for us again, but it looks like 2004 will be AMD's biatch.
Click to expand...

Extending the pipeline isn't something that can be done in 3 months. They would basically have to go back to the drawing board to extend the pipeline. This was planned well before Prescott was delayed. They couldn't do that in 3 months or even 6 months. Extending the pipeline would not help it deal with heat either.

I still don't think it will be extended to 30. We know it will go up but I see it being closer to 22 or 24 up from 20. Extending to 30 is HUGE jump.

 
One thing to keep in mind though, Arcas, is that Intel may not be so lucky this time around. With the P4, they could seed compilers with information on how to handle the longer pipeline, and more importantly, how to handle SSE2. Prescott however has just the longer pipeline, so there's a decent chance Intel won't be able to get back as much effeciency out of the Prescott, since the only thing going for it this time are more pipeline compiler optimizations, along with the larger cache and the better BPU.
Click to expand...

They also have SSE3 now to seed compilers with.
 
Originally posted by: arcas
I'm going to play devil's advocate here...

People made the same comments when the P4 came out and found that the P3 was faster at a given clockspeed. We were subjected to arguments from both sides of the fence...on the Intel side, it was "more pipelines mean higher frequency" and on the anti-Intel side, it was "more pipelines mean more latency and a bigger penalty for branch prediction misses so this design is going to flop."

Fast forward to today...the P4 design is currently chugging along at 3+ ghz and by almost any benchmark is far faster than the fastest P3 it supplanted. Likewise, in terms of actual performance, the deeply-pipelined P4 design and the shallow-pipelined Athlon design have largely kept pace with one another. If at any point in time one design were found to offer, say, 5x higher throughput than the other design, I think one could confidently claim that design is superior. But in fact that hasn't been the case. At no time over the past 2-3 years has the deeply-pipelined design had a significant performance advantage over the shallow pipeline design. Or visa versa. So these arguments are largely moot.

From a marketing standpoint, though, you can't argue with Intel's logic. And let's be honest, Intel's in business to make money so marketing advantages matter.
Click to expand...

I'm gonna play the logical advocate ^^

In theory, I believe AMD has much more room to work w/ in terms of performance.. They're chugging along at what, 2.2Ghz as their flagship... Intel has a 3.2 as their flagship... for the most part, they run on par w/ each other, correct? Now, what if AMD kept the current pipeline length, but added another 1ghz to the Frequency... We'd have an Intel Killer. It would be more expensive, yeah... but I doubt it would still be anywhere near Intel's EE.

Bill

 
Originally posted by: Rectalfier
Here's the way I see it. Intel has already delayed the Prescot due to heat issues. I think they are trying to work around this by increasing the pipeline. It just seems that Intel has done this all of a sudden, since they couldn't produce their original Prescott design, and have thus delayed it. Could this be a massive trip for Intel? The release of the Pentium 4EE sure seemed like an act of desperation. Rest assured, Intel will have something good for us again, but it looks like 2004 will be AMD's biatch.
Click to expand...

Increasing the pipeline tends to increase power dissipation, not reduce it. And the P4EE is no more an act of desperation than the Athlon FX.

 
I'm gonna play the logical advocate ^^

In theory, I believe AMD has much more room to work w/ in terms of performance.. They're chugging along at what, 2.2Ghz as their flagship... Intel has a 3.2 as their flagship... for the most part, they run on par w/ each other, correct? Now, what if AMD kept the current pipeline length, but added another 1ghz to the Frequency... We'd have an Intel Killer. It would be more expensive, yeah... but I doubt it would still be anywhere near Intel's EE.

Bill
Click to expand...

The current AMD design may not scalable to 3.2 Ghz. Even the 90nm parts are not being shown to scale to 3+ Ghz. I'm willing to be the current AMD design is good to 2.6 to 2.8Ghz before they need to redesign it again.

 
In defence of Intel/Prescott(ugh), along with this are some improvements to Prescotts HT and other features. I suspect these improvements negate the longer pipeline situation.
 
Originally posted by: mikecel79
Originally posted by: Rectalfier
Here's the way I see it. Intel has already delayed the Prescot due to heat issues. I think they are trying to work around this by increasing the pipeline. It just seems that Intel has done this all of a sudden, since they couldn't produce their original Prescott design, and have thus delayed it. Could this be a massive trip for Intel? The release of the Pentium 4EE sure seemed like an act of desperation. Rest assured, Intel will have something good for us again, but it looks like 2004 will be AMD's biatch.
Click to expand...

Extending the pipeline isn't something that can be done in 3 months. They would basically have to go back to the drawing board to extend the pipeline. This was planned well before Prescott was delayed. They couldn't do that in 3 months or even 6 months. Extending the pipeline would not help it deal with heat either.
Click to expand...
Best post so far. 😉

 
Originally posted by: Wingznut
Originally posted by: mikecel79
Originally posted by: Rectalfier
Here's the way I see it. Intel has already delayed the Prescot due to heat issues. I think they are trying to work around this by increasing the pipeline. It just seems that Intel has done this all of a sudden, since they couldn't produce their original Prescott design, and have thus delayed it. Could this be a massive trip for Intel? The release of the Pentium 4EE sure seemed like an act of desperation. Rest assured, Intel will have something good for us again, but it looks like 2004 will be AMD's biatch.
Click to expand...

Extending the pipeline isn't something that can be done in 3 months. They would basically have to go back to the drawing board to extend the pipeline. This was planned well before Prescott was delayed. They couldn't do that in 3 months or even 6 months. Extending the pipeline would not help it deal with heat either.
Click to expand...
Best post so far. 😉
Click to expand...

Thanks. I don't think people have any idea how much work and engineering goes into making a processor. I don't think even I know how much work goes into it. So many people talk about just extend the pipeline or just ramp it up 1Ghz. I don't think they realize how much validation and testing goes into a CPU before it's released.
 
Delays aren't very good, but they're not super horrible either. Just like how AMD delayed K8 for a long time, but they filled in with chips like the Barton, then did a solid K8 release in terms of performance, Intel may have some other design up their sleeve; (a Northwood shrink without new features?) that may keep performance decent until this "P4 with SSE3" comes along.
 
Hmm, the News.com.com article seems to come from who I think of as a credible source, but they say the XP has 15 stages? I thought K7's had 10 stages! 😕
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top