Intel first to demonstrate working 45nm chip's..

[dexvx]

Originally posted by: PetNorth
Intel TDP and AMD TDP is measured differently. Intel = tipical power consumption; AMD = max power consumption.

Oh god, not that myth again.

 

[PetNorth]

mith? what myth?

ftp://download.intel.com/design/Pentiumd/datashts/30750602.pdf

p. 76. "NOTES:
1. [...]The TDP is not the maximum power that the processor can dissipate"

 

[dmens]

Nice conspiracy theory. The fact that the datasheet states that thermal solutions should set the stated TDP as the maximum commercial dissipation means that it is the max power in commercial operation at the specified voltage and frequency. AMD should state the same in their info to be honest. As I said before, I can make presler or yonah draw much higher than the TDP by messing with it a bit, and I'm sure I can do the same with any AMD uproc if I knew the procedure.

Oh by the way, I'm sure all your overclocked uprocs are drawing more than the stated TDP at full load... oh no... LOL!

 

[PetNorth]

http://www.lostcircuits.com/cpu/amd_fx60/7.shtml

Look at cpu's power consumption with a rendering task. Now, look at Intel & AMD TDP's.

 

[dmens]

That data is suspect because presler is drawing more than 840, and 30W at load for the venice... sure. No explanation of how the measurement is obtained. Not good enough.

 

[ahock]

Honestly, I dont believe AMD will release 45nm processor at the same time with Intel. They haven't had any 65nm until second half this year. I dont think they will only use their 65nm in less than a year. If so, I dont think they will ramp it at similar rate intel does.

Question did AMD demonstrate a working 65nm silicon?

 

[coldpower27]

Kudos to those who know what they are talking about, right now Intel 65nm process not PROCESSORS, is superior to AMD's non existant 65nm process since AMD has yet to come out with items based on it simply becase it isn't here yet.

You can't compare Intel and AMD processes directly, Intel 90nm process has produced 2 products, Pentium M based processors and Pentium 4, the Pentium M based processor ran quite cool despite being based on Intel's 90nm process, this already proves the heat emitted by a processor, is not as directly linked as you think to the process it's built on, architecture plays a huge role too.

In order to compare fairly, AMD would need to build NetBurst processor on their 90nm SOI process and see how it runs. That way you eliminate architecture from the equation which itself plays a huge role. Or Intel would need to build K8's using their 90nm SS process.

You also have to rememeber Intel's 90nm Pentium M processor have being nice ~ the same wattage of the 90nm Turion's despite Intel not having the SOI advantage to curb leakage.

 

[coldpower27]

Originally posted by: ahock
Honestly, I dont believe AMD will release 45nm processor at the same time with Intel. They haven't had any 65nm until second half this year. I dont think they will only use their 65nm in less than a year. If so, I dont think they will ramp it at similar rate intel does.

Question did AMD demonstrate a working 65nm silicon?

Yeah, I think they have it behind closed door,s but they haven't done a public demonstration yet to my knowledge...

 

[dexvx]

Originally posted by: ahock
Question did AMD demonstrate a working 65nm silicon?

They probably already have it, theres no need to demonstrate it.

There was a huge thread involving LaptopLogic.com's definition of Intel's definition of TDP. Nevertheless, a recent Intel general document stated, in laymen's terms that their TDP is the maximum power dissipated for a product line using default voltage, drawing default current, in a case less than T_max, and using openly available (non-viral) software.

AMD's TDP being maximum is a bunch of BS made by fanboi's, because all you have to do is raise the core voltage and you have power dissipation greater than that of its TDP.

 

[FelixDeCat]

Originally posted by: coldpower27

Kudos to those who know what they are talking about, right now Intel 65nm process not PROCESSORS, is superior to AMD's non existant 65nm process since AMD has yet to come out with items based on it simply becase it isn't here yet.

You can't compare Intel and AMD processes directly, Intel 90nm process has produced 2 products, Pentium M based processors and Pentium 4, the Pentium M based processor ran quite cool despite being based on Intel's 90nm process, this already proves the heat emitted by a processor, is not as directly linked as you think to the process it's built on, architecture plays a huge role too.

In order to compare fairly, AMD would need to build NetBurst processor on their 90nm SOI process and see how it runs. That way you eliminate architecture from the equation which itself plays a huge role. Or Intel would need to build K8's using their 90nm SS process.

You also have to rememeber Intel's 90nm Pentium M processor have being nice ~ the same wattage of the 90nm Turion's despite Intel not having the SOI advantage to curb leakage.

Well said.

Once again, Intel "leaps ahead" by leading the way with new technology. My hope is that they give up the "if we didnt develop it we wont use it" mentality and incorporate a memory controller asap. This is a major advantage for amd and it must be overcome to regain its rightful position.

 

[ahock]

I think Intel will eventually implement the on-die memory controller thru their CSI.... Its just that they are pouring so much resources to it to make it succesful....

But for now from the looks of it, with their FSB architecture, it seems though that they can catch AMD's performance without ondie mem controller

 

[Viditor]

Originally posted by: dexvx
Dont you just love people who have no idea what they're talking about spew random crap?

For people who think AMD > Intel when it comes to power consumption to process technology ratio (which is a totally absurd ratio):

TDP's:
Banias, 130 nm: 29W
Northwood, 130nm: ~90W
Clawhammer, 130nm: ~80W

Dothan, 90nm: 27W
Prescott, 90nm: ~100W
San Diego, 90nm: ~60W

Yonah (DC), 65nm: ~30W
Pressler (DC), 65nm: ~130W
Toledo (DC), 90nm: ~110W

Yes, I do...
TDP IS NOT POWER CONSUMPTION!!!
TDP (or Thermal Design Point) is ONLY a guideline for the thermal design of systems. It in no way indicates the power consumption of a CPU! (I keep pointing this out, but people see a measurement in Watts and make the wrong assumption)

1. Neither Intel nor AMD do actual measurements of working systems to arrive at the TDP, they use simulation models
2. AMD's simulation model is for absolute Max power at Max temp environment, Intel's is for Max power under normal conditions (which they have stated is about 75% of absolute Max power).
3. BOTH are valid guidelines, but AMD's chips will never even hit their TDP, while Intel's chips will often exceed their TDP for brief bursts (but not for sustained periods unless you really try to do just that). While AMD's TDP is guaranteed to be safe under any conditions, Intel's TDP is much closer to realistic conditions.

 

[Viditor]

Originally posted by: ahock
Honestly, I dont believe AMD will release 45nm processor at the same time with Intel. They haven't had any 65nm until second half this year. I dont think they will only use their 65nm in less than a year. If so, I dont think they will ramp it at similar rate intel does.

Question did AMD demonstrate a working 65nm silicon?

AMD demonstrated working and shippable 65nm A64s and X2s in Sept/October last year...
At the conference call for AMD, they stated that instead of a steadily increasing ramp of 65nm, AMD will begin it's 65nm release already at a fully mature yield (later this year).
The main reasons that they haven't begun full 65nm production are:
1. The yields on 90nm are well ahead of expectations (85%+ by most analysts calculations). When they convert to 65nm, they will want the yields to at least 70% first for it to be cost effective.
2. From a power standpoint, there is no need. Netburst just isn't competitive enough to push AMD into 65nm right away.
3. The longer AMD delays, the more tweaks they can add to the new design...

 

[dexvx]

Originally posted by: Viditor
TDP IS NOT POWER CONSUMPTION!!!
TDP (or Thermal Design Point) is ONLY a guideline for the thermal design of systems. It in no way indicates the power consumption of a CPU! (I keep pointing this out, but people see a measurement in Watts and make the wrong assumption)

The above was the TDP for all the cpu's.

 

[Viditor]

Originally posted by: dexvx

The above was the TDP for all the cpu's.

I understand...my point is that it really can't even be used as a point of comparison for power consumption. TDP isn't even a rough guideline for that...

Edit: I just reread what I wrote, and it did sound a bit harsh (sorry m8)...
It really is understandable that so many people (even very well educated people in the field) get this wrong...I would say that 90% of the reporters and supposed "expert" web sites ALSO get it wrong!
The simple truth is that either company could say anything they want for a TDP, and they wouldn't be wrong...though they might lose a lot of sales if their chips overheated.

 

[coldpower27]

Originally posted by: FelixDeKat

Originally posted by: coldpower27

Kudos to those who know what they are talking about, right now Intel 65nm process not PROCESSORS, is superior to AMD's non existant 65nm process since AMD has yet to come out with items based on it simply becase it isn't here yet.

You can't compare Intel and AMD processes directly, Intel 90nm process has produced 2 products, Pentium M based processors and Pentium 4, the Pentium M based processor ran quite cool despite being based on Intel's 90nm process, this already proves the heat emitted by a processor, is not as directly linked as you think to the process it's built on, architecture plays a huge role too.

In order to compare fairly, AMD would need to build NetBurst processor on their 90nm SOI process and see how it runs. That way you eliminate architecture from the equation which itself plays a huge role. Or Intel would need to build K8's using their 90nm SS process.

You also have to rememeber Intel's 90nm Pentium M processor have being nice ~ the same wattage of the 90nm Turion's despite Intel not having the SOI advantage to curb leakage.

Well said.

Once again, Intel "leaps ahead" by leading the way with new technology. My hope is that they give up the "if we didnt develop it we wont use it" mentality and incorporate a memory controller asap. This is a major advantage for amd and it must be overcome to regain its rightful position.

Nah, I think what Intel will do is that they will just use a different fancy name in order to make it look like they made it.

There are other issues to address besides the performance advantage of using an ODMC, for instance AMD doesn't have a chipset buisness to worry about.

Intel right now has a pretty simplistic inventory management due to having LGA775, nad the newer chipsets on LGA775 will support everything from the bottom up, Intel also likes having memory flexibility and upgrades, they would be able to keep upping the speed of support of DDR2 as the Memory Controller is still tied to the chipset.

Intel will eventually move to CSI I believe but not till we reach 45nm NGMA processors at the earliest. They probably want to make the change when NGMA is more a mature and they got it down already.

 

[coldpower27]

Originally posted by: AkumaX
just a question: was Intel the first to have 65nm? and 90nm?

because man, that 90nm tech called Prescott sure turned heads

Intel is always first in transistioning to smaller optical nodes with AMD typically 3-4 quarters behind them. It takes typically 2 years to transistion.

Even though Prescott wasn't very competitive from a power standpoint, it was still cheaper to make then the Athlon 64's of that time thanks to reduce die size and lack of SOI as AMD had yet to go to 90nm.

Now that you think of it, Intel Pentium 4 line up is going to be very simple for them to management as it's all based on Cedar Mill, and that 1 core can produce Pentium D, Pentium 4's and Celeron's in various configurations.

 

[Viditor]

Originally posted by: coldpower27

Intel is always first in transistioning to smaller optical nodes with AMD typically 3-4 quarters behind them. It takes typically 2 years to transistion.

Even though Prescott wasn't very competitive from a power standpoint, it was still cheaper to make then the Athlon 64's of that time thanks to reduce die size and lack of SOI as AMD had yet to go to 90nm.

Now that you think of it, Intel Pentium 4 line up is going to be very simple for them to management as it's all based on Cedar Mill, and that 1 core can produce Pentium D, Pentium 4's and Celeron's in various configurations.

I wouldn't be so sure...just because Intel came out with the process sooner, doesn't mean they had those cost advantages from the smaller die right away. It's most likely that initial yields (at least the first 2 turns...6 months) had very poor yields (this is SOP). Remember that Intel absolutely HAD to convert when they did because they were about to hit a speed wall with Northwood.
Your point on SOI is quite valid though...
All in all, I'd bet (don't know for sure) that Intel's 90nm transition (and probably their 65nm transition) will create chips that are more expensive than their previos nodes...at least for the first 6-9 months.

 

[FelixDeCat]

Originally posted by: coldpower27

Originally posted by: FelixDeKat

Originally posted by: coldpower27

Kudos to those who know what they are talking about, right now Intel 65nm process not PROCESSORS, is superior to AMD's non existant 65nm process since AMD has yet to come out with items based on it simply becase it isn't here yet.

You can't compare Intel and AMD processes directly, Intel 90nm process has produced 2 products, Pentium M based processors and Pentium 4, the Pentium M based processor ran quite cool despite being based on Intel's 90nm process, this already proves the heat emitted by a processor, is not as directly linked as you think to the process it's built on, architecture plays a huge role too.

In order to compare fairly, AMD would need to build NetBurst processor on their 90nm SOI process and see how it runs. That way you eliminate architecture from the equation which itself plays a huge role. Or Intel would need to build K8's using their 90nm SS process.

You also have to rememeber Intel's 90nm Pentium M processor have being nice ~ the same wattage of the 90nm Turion's despite Intel not having the SOI advantage to curb leakage.

Well said.

Once again, Intel "leaps ahead" by leading the way with new technology. My hope is that they give up the "if we didnt develop it we wont use it" mentality and incorporate a memory controller asap. This is a major advantage for amd and it must be overcome to regain its rightful position.

Nah, I think what Intel will do is that they will just use a different fancy name in order to make it look like they made it.

There are other issues to address besides the performance advantage of using an ODMC, for instance AMD doesn't have a chipset buisness to worry about.

Intel right now has a pretty simplistic inventory management due to having LGA775, nad the newer chipsets on LGA775 will support everything from the bottom up, Intel also likes having memory flexibility and upgrades, they would be able to keep upping the speed of support of DDR2 as the Memory Controller is still tied to the chipset.

Intel will eventually move to CSI I believe but not till we reach 45nm NGMA processors at the earliest. They probably want to make the change when NGMA is more a mature and they got it down already.

I sure hope its implemented sooner rather t
an later. Im so ready to upgrade!

 

[coldpower27]

Originally posted by: Viditor

Originally posted by: coldpower27

Intel is always first in transistioning to smaller optical nodes with AMD typically 3-4 quarters behind them. It takes typically 2 years to transistion.

Even though Prescott wasn't very competitive from a power standpoint, it was still cheaper to make then the Athlon 64's of that time thanks to reduce die size and lack of SOI as AMD had yet to go to 90nm.

Now that you think of it, Intel Pentium 4 line up is going to be very simple for them to management as it's all based on Cedar Mill, and that 1 core can produce Pentium D, Pentium 4's and Celeron's in various configurations.

I wouldn't be so sure...just because Intel came out with the process sooner, doesn't mean they had those cost advantages from the smaller die right away. It's most likely that initial yields (at least the first 2 turns...6 months) had very poor yields (this is SOP). Remember that Intel absolutely HAD to convert when they did because they were about to hit a speed wall with Northwood.
Your point on SOI is quite valid though...
All in all, I'd bet (don't know for sure) that Intel's 90nm transition (and probably their 65nm transition) will create chips that are more expensive than their previous nodes...at least for the first 6-9 months.

No Intel's 65nm process is entering production at yield rates about the level of what current 90nm products now according to Intel, the yield rates are doubtfully as low as 50% on 65nm probably at least were looking 70% yield now at the very worse, considering 90nm from Intel is very very mature now. I would bet 80% plus at the current time.

Since your very quick to believe AMD is going to enter 65nm production with mature yields because an AMD representative said so, you have to accept the statement that Intel is entering 65nm with yeild levels of existing 90nm products because Intel representatives said so.

http://www.tgdaily.com/2005/12/16/intel_65nm_production/

No 65nm chips are already cheaper for Pentium D's and Pentium 4 6xx, the only thing you can possibly debate is Yonah, as that is larger then Dothan. Pentium D is smaller and is 2x81mm2 dies from potnetial 2 different wafers compared to a single 206mm2 no contest here. Pentium 6xx is 135mm2 vs 81mm2 also no contest here either.

Prescott is hardly a conversion, there was alot of changes done to the core in the Prescott from Northwood, 90nm is only 1 piece of the issue here.

Prescott I think represents what happens when you try do too much at once. Unlike the AMD play it safe take thier time method. Intel has learned it lesson hence we have simple 65nm like Pentium 6x1 and 9xx, and moderately complex ones like Yonah.

 

[Cooler]

Originally posted by: coldpower27

Originally posted by: Viditor

Originally posted by: coldpower27

Intel is always first in transistioning to smaller optical nodes with AMD typically 3-4 quarters behind them. It takes typically 2 years to transistion.

Even though Prescott wasn't very competitive from a power standpoint, it was still cheaper to make then the Athlon 64's of that time thanks to reduce die size and lack of SOI as AMD had yet to go to 90nm.

Now that you think of it, Intel Pentium 4 line up is going to be very simple for them to management as it's all based on Cedar Mill, and that 1 core can produce Pentium D, Pentium 4's and Celeron's in various configurations.

I wouldn't be so sure...just because Intel came out with the process sooner, doesn't mean they had those cost advantages from the smaller die right away. It's most likely that initial yields (at least the first 2 turns...6 months) had very poor yields (this is SOP). Remember that Intel absolutely HAD to convert when they did because they were about to hit a speed wall with Northwood.
Your point on SOI is quite valid though...
All in all, I'd bet (don't know for sure) that Intel's 90nm transition (and probably their 65nm transition) will create chips that are more expensive than their previous nodes...at least for the first 6-9 months.

No Intel's 65nm process is entering production at yield rates about the level of what current 90nm products now according to Intel, the yield rates are doubtfully as low as 50% on 65nm probably at least were looking 70% yield now at the very worse, considering 90nm from Intel is very very mature now. I would bet 80% plus at the current time.

Since your very quick to believe AMD is going to enter 65nm production with mature yields because an AMD representative said so, you have to accept the statement that Intel is entering 65nm with yeild levels of existing 90nm products because Intel representatives said so.

http://www.tgdaily.com/2005/12/16/intel_65nm_production/

No 65nm chips are already cheaper for Pentium D's and Pentium 4 6xx, the only thing you can possibly debate is Yonah, as that is larger then Dothan. Pentium D is smaller and is 2x81mm2 dies from potnetial 2 different wafers compared to a single 206mm2 no contest here. Pentium 6xx is 135mm2 vs 81mm2 also no contest here either.

Prescott is hardly a conversion, there was alot of changes done to the core in the Prescott from Northwood, 90nm is only 1 piece of the issue here.

Prescott I think represents what happens when you try do too much at once. Unlike the AMD play it safe take thier time method. Intel has learned it lesson hence we have simple 65nm like Pentium 6x1 and 9xx, and moderately complex ones like Yonah.

If they had just did a die shrink with northwood and add 64 bit support it would have been much better.

 

[pm]

Originally posted by: BrownTown
From what I've read about Intel's process for the 45nm node it sounds like its gonna be very different than their last few revisions. They claim its the most revolutionary process since the 1960's. Anyways, whether or not it works could really make or break Intel in 2008.

I have been working with the 45nm Intel process and it certainly is a departure in a number of signifcant ways from previous process technologies although I'm not sure if I'd characterize it as the biggest change since the 60's. I also don't know if it will "make or break Intel" - it seems unlikely that anyone would let the company be put in a situation where this is the case - and it also remains to be seen what everyone else is going to do on 45nm. But, yeah, I agree with you fundamentally... 45nm is going to be very different.

Originally posted by: Viditor
I wouldn't be so sure...just because Intel came out with the process sooner, doesn't mean they had those cost advantages from the smaller die right away. It's most likely that initial yields (at least the first 2 turns...6 months) had very poor yields (this is SOP). Remember that Intel absolutely HAD to convert when they did because they were about to hit a speed wall with Northwood.
Your point on SOI is quite valid though...
All in all, I'd bet (don't know for sure) that Intel's 90nm transition (and probably their 65nm transition) will create chips that are more expensive than their previos nodes...at least for the first 6-9 months.

I'm not sure that I would say that "very poor yields" initially is SOP - at least not at Intel, nor any other high volume manufacturer. Typically, a process is not "deployed" until the yields hit a certain internal goal and knowing this number, I don't think that anyone would ever declare it to be "very poor". Often yield is usually a function of speed or leakage... so by pulling back on speed, you can improve yield. So if your newly deployed process is yielding poorly, the first thing that takes the hit is speed bin. This is not always the case, but it's very common.

The process is usually pretty thoroughly checked out well in advance of shipping - bear in mind that parts sample on a new process years before that process is relied upon for HVM - for example, we are seeing 45nm working IC's today years ahead of their release. Back-end validation (all the stuff that happens between A0 tape-out and final release) takes many quarters... there is plenty of time to tweak yields. A product will not release unless yields are above a set goal - release will be delayed if this is a concern. Normally a part that doesn't yield well is also far more likely to have reliability problems in the field - no one in high-volume manufacturing wants to take a hit with customers (either OEM's or end customers) for high failure rates in the field. So, to summarize, I disagree that "very poor yields" is standard operating procedure on products shipping on a new process.



As far as the discussion of who's process is lower power... you need to, as several people pointed out - compare apples to apples. Comparing microprocessor TDP is not the way to do this. The way that people should be doing this comparison is to pull the trade journal articles (IEEE IEDM articles) for the years when companies are disclosing process information and then look and see what Intel and AMD have for Ion/Ioff ratios for NMOS and PMOS FETs. To my recollection AMD does not usually disclose numbers at this conference, but with the collaboration between IBM and AMD, the IBM numbers should work as a general estimate. I don't have the time, or the motivation to do this, but arguing TDP definitions vs. microarchitectures is not the way to do it.

Ion/Ioff - the ratio of how much current the transistor can source when it's supposed to be on, vs. the amount of current a transistor leaks when it's supposed to be off - is a very good benchmark to compare because leakage and transistor speed usually go hand in hand on a given process at a given company (so you can get great power numbers if you are willing to sacrifice speed bins), but the ratio usually has a "sweet spot" that can be used as a benchmark against other companies.

 

[Viditor]

Originally posted by: coldpower27


Since your very quick to believe AMD is going to enter 65nm production with mature yields because an AMD representative said so, you have to accept the statement that Intel is entering 65nm with yeild levels of existing 90nm products because Intel representatives said so.

http://www.tgdaily.com/2005/12/16/intel_65nm_production/

No 65nm chips are already cheaper for Pentium D's and Pentium 4 6xx, the only thing you can possibly debate is Yonah, as that is larger then Dothan. Pentium D is smaller and is 2x81mm2 dies from potnetial 2 different wafers compared to a single 206mm2 no contest here. Pentium 6xx is 135mm2 vs 81mm2 also no contest here either.

Prescott is hardly a conversion, there was alot of changes done to the core in the Prescott from Northwood, 90nm is only 1 piece of the issue here.

Prescott I think represents what happens when you try do too much at once. Unlike the AMD play it safe take thier time method. Intel has learned it lesson hence we have simple 65nm like Pentium 6x1 and 9xx, and moderately complex ones like Yonah.

Well, Hector Ruiz stated those facts at a Conference call...which means if he's lying he can go to prison. The article you linked isn't even quoting anyone specific about the Intel yields, and certainly nobody in senior management (who could be held liable).
So, please forgive me if I remain dubious as to the articles claims...

 

[AMDrulZ]

I honestly think that AMD get to 65nm sooner than some think. AMD kind of reminds me of Scotty from star trek they always manage to get up and running faster and better than even AMD expected... there like a miracle work. LOL

 

[mxzrider2]

Originally posted by: dmens
Again, process comparisons are totally meaningless if you baseline it with different designs. For example, I can make yonah draw 3-4x more power than in commercial operation by messing with some debug features. Going by your reasoning, I can take yonah-in-debug and claim 65nm sucks ass compared to dothan at 90nm.... but that comparison is obviously garbage.

dude that doesnt matter what is being said is that intel processors at 65nm stil take more power than amd procs at 90 nm. learn to read even though this is a generazation as you are correct that dothan ( SOME) does consume less power than almost all amd procs( current) cept maybe those geo ones or what ever.