Will Intel's 45nm send AMD to the grave?

[dmens]

At least Intel is honest enough to say the TDP is a recommendation for thermal solutions, because that is what it is.

Tell me, how do you think "maximum current" to the CPU can be measured? With a power virus. That will always be a good estimation of maximum power draw.

 

[Keysplayr]

Boy OP, you sure chose the "right" forum to ask this question. This is the AMD fanboy capitol of the internet!! . I like AMD, but people can get pretty crazy about it around these parts.

 

[mikeford]

Many technical advances can't be protected, (held away from competitors), so Intel can't out tech Amd or vice versa for more than a few months. A better question would be to look at the profitablity of various semiconductor sectors.

 

[Hacp]

Originally posted by: keysplayr2003
Boy OP, you sure chose the "right" forum to ask this question. This is the AMD fanboy capitol of the internet!! . I like AMD, but people can get pretty crazy about it around these parts.

wow I agree. AMD fanboys run amuck here, and you knew you were going to get flamed for posting that.

 

[dexvx]

I see the TDP myth is pretty prevalent on AT.

 

[Maximilian]

Only one person can answer that, AMD's CEO.

 

[dmens]

Originally posted by: Viditor
BTW, on the original point of the post...
AMD and IBM have also been developing both 45nm and 32nm for over a year now. The only difference is that they have not seen fit to advertise their step by step progress.

What's your point? I recall IBM making a big deal out of their 90nm SOI, even as horrific yield problems was crippling its push to production. Remember early power5? You figure if they had a breakthrough on 45nm like intel just did, they'd make a press release to boost investor confidence. And if there's anything that counts in this field, it's a new process.

And since when did intel advertise their step-by-step on 45nm? It's been around for months and the first time the public has heard anything on its progress was just the other week.

 

[TraumaRN]

Originally posted by: dmens

Originally posted by: Viditor
BTW, on the original point of the post...
AMD and IBM have also been developing both 45nm and 32nm for over a year now. The only difference is that they have not seen fit to advertise their step by step progress.

What's your point? I recall IBM making a big deal out of their 90nm SOI, even as horrific yield problems was crippling its push to production. Remember early power5? You figure if they had a breakthrough on 45nm like intel just did, they'd make a press release to boost investor confidence. And if there's anything that counts in this field, it's a new process.

And since when did intel advertise their step-by-step on 45nm? It's been around for months and the first time the public has heard anything on its progress was just the other week.

Maybe cuz IBM learned their lesson to make sure it works first....seems SOI works great now. I figure once it's ready IBM and AMD will show off working samples as well.

 

[dmens]

Right, and that big 45nm SRAM intel showed works pretty well too.

I'll never understand the some people here with the unlimited cynicism towards intel and fervent optimism towards amd. Seems almost religious.

 

[TraumaRN]

Originally posted by: dmens
Right, and that big 45nm SRAM intel showed works pretty well too.

I'll never understand the some people here with the unlimited cynicism towards intel and fervent optimism towards amd. Seems almost religious.

I never meant to come off as a cynic, personally I'm not surprised Intel came out with this first. But to be honest once it's ready it's ready for both AMD and Intel, I'll pass my judgement when working chips and benchmarks come out. I may have devotion to AMD at the moment but if Intel comes out with a better product at a good price, you can bet I'll be switching right back.

And I think people are cynical towards Intel because Intel has done nothing to impress some of the people here, it seems to me the expectation here(at AT) is lower power, lower heat, etc, and most people dont think they can get that from Intel at the moment.

 

[dmens]

I can understand legitimate criticism, but a lot of commentary here is based on false information or outright lies. It doesn't help the forum because it reduces threads with potential for good debate to retarded snippy one-liners. I can also understand confusion regarding the straight facts. But if I'm going to write a huge vehement rant, I'd at least try to make sure my info is correct.

I read AT because it balances out the everything-is-fine chant from work, and I'd rather not have to wade through a load of crap to get to the useful commentary.

 

[isp]

i would love for intel to take back the crown...if for no other reason than to thin out the amd fanboys...it's really getting disgusting these days

 

[themusgrat]

We AMD fanboys will be around until Intel does somehing good. Duh. But they have made a good start with their newest P4s, the ones that OC really well.

EDT: And when AMD's CEO was asked the same question as the OP, he said that people assume that AMD will never make a better product than the current ones. Intel has a way to go until hey are ready for mass production.

 

[Markfw]

Originally posted by: isp
i would love for intel to take back the crown...if for no other reason than to thin out the amd fanboys...it's really getting disgusting these days

And I am sick of the Intel fanboys speculating (like how AMD will die when 45nm comes out) and how good the new processors will be that nobody has benched, but they will rule, and kick AMD's butt.

I want something better out of Intel, and they CAN do it, they just haven;t yet.

 

[n7]

Originally posted by: keysplayr2003
Boy OP, you sure chose the "right" forum to ask this question. This is the AMD fanboy capitol of the internet!! . I like AMD, but people can get pretty crazy about it around these parts.

ROFL...this coming from a crazy nVidia fanboi :laugh::roll:

I do agree that this forum is filled with fanbois though.

I wouldn't say the current mods don't do their job, but if i was one, there'd be some major house cleaning, starting with the video forum, & then here...

 

[Markfw]

Originally posted by: DeathBUA

Originally posted by: dmens
Right, and that big 45nm SRAM intel showed works pretty well too.

I'll never understand the some people here with the unlimited cynicism towards intel and fervent optimism towards amd. Seems almost religious.

I never meant to come off as a cynic, personally I'm not surprised Intel came out with this first. But to be honest once it's ready it's ready for both AMD and Intel, I'll pass my judgement when working chips and benchmarks come out. I may have devotion to AMD at the moment but if Intel comes out with a better product at a good price, you can bet I'll be switching right back.

And I think people are cynical towards Intel because Intel has done nothing to impress some of the people here, it seems to me the expectation here(at AT) is lower power, lower heat, etc, and most people dont think they can get that from Intel at the moment.

Almost exactly my point. Add to that performance, and price/performance, and that is missing with Intel right now also.

 

[isp]

Originally posted by: Markfw900
snip

I agree we should just wait and see. IMO, speculation is what breeds fanboyism. They speculate so much that sooner or later it becomes the bonafide truth inside their head. Then at that point it's impossible to get through to them.

 

[Praxis1452]

This thread is utterly useless.

 

[Viditor]

Originally posted by: dmens
At least Intel is honest enough to say the TDP is a recommendation for thermal solutions, because that is what it is.

Tell me, how do you think "maximum current" to the CPU can be measured? With a power virus. That will always be a good estimation of maximum power draw.

Ummm...they both say that. In fact, that's exactly what TDP means! The reason for the difference in the determination of values is because Intel relies on throttling much more than AMD does, and this is certainly fair as Intel is better at it than AMD.
Sadly, 90% of the people who read reviews don't understand TDP at all...even the well educated ones. It's almost like the old MHz myth.
As to determination of "maximum current", Intel does indeed use a "power virus" which generates another number that they DON'T publish called "Power Delivery Design".
From Intel's whitepaper:

Thermal Design Targets: Power delivery designs are typically based on the theoretical maximum power that is drawn by the processor, which is based on a synthetic (power virus) code. The theoretical maximum power drawn is based on a synthetic code that is designed to use resident data from the on-chip caches (L1 or L2). The pipelines and queues are maintained full to the best possible extent. Given the superscalar and superpipelined architectures of the microprocessor, such activity could conceivably occur over brief bursts of time, but would not likely be sustainable over long periods. Furthermore, if the thermal designs are done to a lower power target (Ptdp for example) than the maximum power, the thermal capacity of the system may be able to support temporary bursts of power consumption over short durations, without violating the CPU thermal specifications

Some points on this...
1. In general, the Ptdp is ~75% of the Pdd, though this is not a hard and fast rule...
2. The Pdd is measured using a highly confidential synthetic software suit (as is AMD's) so replication or verification is not possible.
3. Intel's Pdd is calculated at normal case temperatures, AMD uses maximum case temperatures

Just so we're clear...my whole point here isn't that AMD is greater than Intel or vice versa. My point is USING TDP AS A POWER GUIDELINE IS JUST PLAIN WRONG (sorry for the shouting...).

BTW...For those of you decrying things like "myth" and "Fanbois", could you at least post some argument rather than just throwing rant remarks? While dmens and I disagree, he has at least had the courage to post arguments and facts to back up what he is saying...and I very much respect that!

 

[dmens]

Good points. But in regards to point 3, in a lower environment temperature, I'd argue the power draw from the CPU would drop due to slower switching and decreased current drive. But given the small difference between "normal" and "maximum" case temps, I doubt it'd make too much of a difference.

But at least we agree that TDP (or whatever other names it has) is just a recommendation. I know it as TDP because that's what everyone calls the tests run to stress the chip and look for powerdown ideas.

 

[Viditor]

Originally posted by: dmens

Originally posted by: Viditor
BTW, on the original point of the post...
AMD and IBM have also been developing both 45nm and 32nm for over a year now. The only difference is that they have not seen fit to advertise their step by step progress.

What's your point? I recall IBM making a big deal out of their 90nm SOI, even as horrific yield problems was crippling its push to production. Remember early power5? You figure if they had a breakthrough on 45nm like intel just did, they'd make a press release to boost investor confidence. And if there's anything that counts in this field, it's a new process.

And since when did intel advertise their step-by-step on 45nm? It's been around for months and the first time the public has heard anything on its progress was just the other week.

This is an excellent illustration of my previous point...
While IBM and AMD didn't collabarate on the 90nm SOI process, they are very similar. However, IBM released early and had absolutely horrible yields...AMD delayed release and had amazingly GOOD yields.
At the moment, AMD really has no need to boost investor confidence (their stock is up ~40% in the last 3 months...), while Intel certainly does (down ~20% in the same time).
Add to that what Ruiz said at the last CC about AMD going forward
"we intend to underpromise and overperform"
Any announcement about the 45nm and 32nm process would have to be a joint one...
I think your last point was the key question...why did Intel feel the need for a press release last week?

 

[Viditor]

Originally posted by: dmens
Good points. But in regards to point 3, in a lower environment temperature, I'd argue the power draw from the CPU would drop due to slower switching and decreased current drive. But given the small difference between "normal" and "maximum" case temps, I doubt it'd make too much of a difference.

But at least we agree that TDP (or whatever other names it has) is just a recommendation. I know it as TDP because that's what everyone calls the tests run to stress the chip and look for powerdown ideas.

Ooops...that's because I screwed up and wrote Pdd instead of Ptdp. Must be that age thing again. You are absolutely right though...TDP is ONLY useful for thermal design as a guideline for OEMs, and really has nothing to do with comparing Intel and AMD on power/thermal usage.

 

[dmens]

Yeah, the timing of the intel release is related to the Q4 earnings. But enough with press releases...

This brings back a point I made in another topic, that process is king. Just like how intel needs to deliver a new uarch, AMD needs to catch up on process. Given their current 65nm gap, I'd have to say the 45nm is just as lagged, if not worse. The issue of high volume manufacturing becomes more critical at smaller features, and that is something IBM/AMD never had a good track record with.

As process size decreases and the trend towards multicore continues, I'd argue that process generations will have even more impact than before based on the number of chips a single wafer can produce. It is amplifying the effect of economies of scale. I am looking forward to see the results of AMD's 65nm ramp and how soon they can pull it off, because it will be a good indicator of future trends, IMO.

 

[Viditor]

Originally posted by: dmens
Yeah, the timing of the intel release is related to the Q4 earnings. But enough with press releases...

This brings back a point I made in another topic, that process is king. Just like how intel needs to deliver a new uarch, AMD needs to catch up on process. Given their current 65nm gap, I'd have to say the 45nm is just as lagged, if not worse. The issue of high volume manufacturing becomes more critical at smaller features, and that is something IBM/AMD never had a good track record with.

As process size decreases and the trend towards multicore continues, I'd argue that process generations will have even more impact than before based on the number of chips a single wafer can produce. It is amplifying the effect of economies of scale. I am looking forward to see the results of AMD's 65nm ramp and how soon they can pull it off, because it will be a good indicator of future trends, IMO.

Fair enough...just keep in mind that there is a difference between when it's POSSIBLE to release and when it's actually released. Let me break it down into pros and cons:

PROS
There are 3 main advantages to converting to a smaller node...
1. Reducing power requirements by decreasing lengths
2. Increasing overall yield/wafer
3. Adding things (i.e. cache or 64 bit extensions) to existing designs without increasing the footprint

CONS

1. Existing equipment must be discarded creating a Capital Loss
2. All new nodes must be ramped over time, reducing production


1. In the current situation, AMD has no need for "pros" 1 and 3 to remain highly competitive...
2. AMD uses 10% of their lines for R&D which allows them to improve a ramp before they initiate volume production
3. There is a mathematical point where the increased yield of conversion to 65nm outweighs the costs of the "cons". For AMD, this point is calculated by their APM 3.0 software.

I certainly have neither the math nor the data to know where that point is, but judging by the brilliant success of their "late" release of 90nm, I do have a great deal of confidence that AMD does. Also, keep in mind that the longer they delay, the greater their initial yield will be when they DO convert to volume on 65nm...

 

[Dadofamunky]

Originally posted by: dmens
Didn't an overclocked yonah do well against FX-60? yonah @ 2.7ghz comes close on power consumption to a stock FX-60.

http://www.oc.com.tw/article/0601/readocarticle.asp?id=4895

Yeah, that's a great link, doesn't display in Explorer without crashing and probably installs a malware program on your PC. Sorry, I'll wait until the pros wade in with their assessments.