Actually, the very definition of threshold voltage is different from fab to fab based on how they calculate it. Current flow through the transistor is not ON and OFF; it ramps up with the applied gate voltage from zero to saturated along a curve. The threshold voltage (Vt) value is based upon the interpretation of that graph, not an absolute value, and there are different beliefs on how it should be calculated. Sounds like Intel has characterized and perfected their process to such a high state that they can repeadtedly hit the same Vt target with very tight control limits.
Yeah very interesting indeed and the fact that processors won't require heatsinks is absolutely madness, I can already see how much more thinner they will be developed if the need for a coolant is not required.
it is fun how they always come back to pentium architecture on this kind of experiment. repeated times. seems like someone there really likes that cpu!
It is (A) relatively simple (low element count) so it is easier to debug and (B) contains just about complete instruction set that is used by applications nowadays, minus the extensions (MMX, SSE* etc.). Why not go for it when making proof-of-concept chips ?
Wonder what fab node they are doing this on. I can't imagine how small an original Pentium would be fabbed at 32nm or 22nm. Original Pentium had 3.1 million transistors up to 3.3M (w/MMX) fabbed with 800nm and die size of 294mm^2. Just taking the simplified approach: 800nm/32nm = 25, 294mm^s/25 = 58.8mm^2inch.
Clarkdale is 384M transistors and 81mm^2 on 32nm. Doing the simplified 384M/3.1M = 123.87. 81mm^2/123.87 = 7.28mm^2.
Pretty sure that neither of these is the correct way to do this but the 2nd is probably closer.
You forgot the 800nm or 32nm are in 1 dimension only. To get the area of the die, you must square the dimensions: (800nm/32nm)^2 = (25)^2 = 25 x 25 = 625. Using your figures, die size of (294mm^2)/625 = 0.47mm^2 die size for original Pentium if @ 32nm.
Clarkdale @ 384M transistors and 81mm^2 on 32nm gives: 384M/3.1M = 123.87. 81mm^2/123.87 = 0.654mm^2 die size for original Pentium if @ 32nm.
0.47mm^2 is very close to 0.654mm^2 and the difference can be accounted for by differences in transistor sizes and the area occupied by components other than transistors.
btw 294/25 = 11.76 but units are wrong.. don't even know where the inches came from. Don't forget you are dealing in area (mm^2) not linear distance so it would be 294/(25*25) = 0.4704 mm^2 Now that's fricken tiny... but given that even if you can make transistors that small you still might need to do extra magic to get it to work so I'd say the Clarkdale calculation would be closer... wait, that calc isn't right either... 81/123.87 = 0.65 mm^2 So that agrees well, a pentium today would be less than mm on a side..... wowsers.
My guess is a P54C core, since those have already been moved to a modern process for the SCC. This research very likely came out of the SCC project. The P54C was built on a 600 nm process. If moved to a 40 nm process, the scale would be (600/32)^2 = around 350 cores in the same area as one 600 nm process core. Of course, a better approach would be to use perhaps 48 such cores and use the rest of the area for an on-die network switch, shared message passing RAM, and DRAM controllers. Oh wait...that IS the SCC.
The current conducted by a transistor is an exponential function of the input voltage. The amount of current you consider to be the "on" state determines the associated "on" input voltage. Thus the Treshold voltage is somewhat arbitrary; it depends upon the current chosen to represent the "on" state.
The "on" current, and thus the threshold voltage, can be chosen over a wide range. For low power operation, the current, and associated threshold voltage, would be chosen to be a very low value.
<i>"NVT Pentium won't be <b>productized</b> but the"</i>
ahhhhh
my pet hate
cant you just say won't be produced. or wont be manufactured.
maybe its an acceptable americanism. in which case the americans can ignore the post whilst i bask in the fellow hatred of this word with my compatriots......
Since you and your countrymen seem to be a little dull, I'll help you out.
"In general terms, "productization" means simply taking an otherwise generic type of service or support offering, and redefining and packaging it more as a "product" offering. For example, if your organization provides a variety of computer-related customer services to both the business and consumer markets, you may wish to "productize" these offerings to appeal to each market directly. By differentiating and packaging your offerings as either "professional computer services for the business community" and "professional-style computer services for the home or school PC user" may be one way of doing so."
Yes, but it's stupid, is the point. It's a classic example of corporate douche-speak. If this is what you're trying to get across, say "it will not be put into production". If you're "packaging" your offerings, say that, not that you're "productizing" your offerings. I think if somebody said that to me in a sales pitch, that would be the end of the meeting right there. At the very least, I'd push a shedload harder for better terms than I might otherwise. I once started steering business away from a major vendor because the sales rep said "irregardless" in conversation. It only went for a few days, but it was a few grand they missed out on because their college-educated (MBA, if I had to take a guess) rep speaks worse English than most of the high-school dropouts I've known (and there have been a lot.)
Assuming by the way you look down on Bubbacup and his compatriots that you're an American, JKflipflop98, please stop. You're making the rest of us look like assholes.
Uhh, there were many CPUs with no heat sink in the original Pentium era. In fact, I seem to recall seeing some Pentium systems that managed it. Is This really the accomplishment they are making it out to be?
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
21 Comments
Back to Article
DEEPAYAN - Thursday, September 15, 2011 - link
that is some impressive r&d. pointing out the very volt when switching happens is a enormous taskQuikstyle - Thursday, September 15, 2011 - link
Actually, the very definition of threshold voltage is different from fab to fab based on how they calculate it. Current flow through the transistor is not ON and OFF; it ramps up with the applied gate voltage from zero to saturated along a curve. The threshold voltage (Vt) value is based upon the interpretation of that graph, not an absolute value, and there are different beliefs on how it should be calculated. Sounds like Intel has characterized and perfected their process to such a high state that they can repeadtedly hit the same Vt target with very tight control limits.soliozuz - Saturday, September 17, 2011 - link
Yeah very interesting indeed and the fact that processors won't require heatsinks is absolutely madness, I can already see how much more thinner they will be developed if the need for a coolant is not required.marc1000 - Thursday, September 15, 2011 - link
it is fun how they always come back to pentium architecture on this kind of experiment. repeated times. seems like someone there really likes that cpu!Arnulf - Friday, September 16, 2011 - link
It is (A) relatively simple (low element count) so it is easier to debug and (B) contains just about complete instruction set that is used by applications nowadays, minus the extensions (MMX, SSE* etc.). Why not go for it when making proof-of-concept chips ?fic2 - Thursday, September 15, 2011 - link
Wonder what fab node they are doing this on. I can't imagine how small an original Pentium would be fabbed at 32nm or 22nm. Original Pentium had 3.1 million transistors up to 3.3M (w/MMX) fabbed with 800nm and die size of 294mm^2.Just taking the simplified approach: 800nm/32nm = 25, 294mm^s/25 = 58.8mm^2inch.
Clarkdale is 384M transistors and 81mm^2 on 32nm.
Doing the simplified 384M/3.1M = 123.87. 81mm^2/123.87 = 7.28mm^2.
Pretty sure that neither of these is the correct way to do this but the 2nd is probably closer.
doylecc - Thursday, September 15, 2011 - link
You forgot the 800nm or 32nm are in 1 dimension only. To get the area of the die, you must square the dimensions: (800nm/32nm)^2 = (25)^2 = 25 x 25 = 625.Using your figures, die size of (294mm^2)/625 = 0.47mm^2 die size for original Pentium if @ 32nm.
Clarkdale @ 384M transistors and 81mm^2 on 32nm gives:
384M/3.1M = 123.87.
81mm^2/123.87 = 0.654mm^2 die size for original Pentium if @ 32nm.
0.47mm^2 is very close to 0.654mm^2 and the difference can be accounted for by differences in transistor sizes and the area occupied by components other than transistors.
marraco - Saturday, September 17, 2011 - link
I'm sure that 625 Pentium processors can do far more work than any 32 nm processor (for highly parallelizable tasks).futrtrubl - Thursday, September 15, 2011 - link
btw 294/25 = 11.76 but units are wrong.. don't even know where the inches came from.Don't forget you are dealing in area (mm^2) not linear distance so it would be 294/(25*25) = 0.4704 mm^2
Now that's fricken tiny... but given that even if you can make transistors that small you still might need to do extra magic to get it to work so I'd say the Clarkdale calculation would be closer... wait, that calc isn't right either... 81/123.87 = 0.65 mm^2
So that agrees well, a pentium today would be less than mm on a side..... wowsers.
Jaybus - Friday, September 16, 2011 - link
My guess is a P54C core, since those have already been moved to a modern process for the SCC. This research very likely came out of the SCC project. The P54C was built on a 600 nm process. If moved to a 40 nm process, the scale would be (600/32)^2 = around 350 cores in the same area as one 600 nm process core. Of course, a better approach would be to use perhaps 48 such cores and use the rest of the area for an on-die network switch, shared message passing RAM, and DRAM controllers. Oh wait...that IS the SCC.doylecc - Thursday, September 15, 2011 - link
The current conducted by a transistor is an exponential function of the input voltage. The amount of current you consider to be the "on" state determines the associated "on" input voltage. Thus the Treshold voltage is somewhat arbitrary; it depends upon the current chosen to represent the "on" state.The "on" current, and thus the threshold voltage, can be chosen over a wide range. For low power operation, the current, and associated threshold voltage, would be chosen to be a very low value.
toyotabedzrock - Friday, September 16, 2011 - link
The 3d transistor gate may have improved made it an even steeper curve.Isn't the atom based on the Pentium?
Bubbacub - Friday, September 16, 2011 - link
<i>"NVT Pentium won't be <b>productized</b> but the"</i>ahhhhh
my pet hate
cant you just say won't be produced. or wont be manufactured.
maybe its an acceptable americanism. in which case the americans can ignore the post whilst i bask in the fellow hatred of this word with my compatriots......
JKflipflop98 - Friday, September 16, 2011 - link
Since you and your countrymen seem to be a little dull, I'll help you out."In general terms, "productization" means simply taking an otherwise generic type of service or support offering, and redefining and packaging it more as a "product" offering. For example, if your organization provides a variety of computer-related customer services to both the business and consumer markets, you may wish to "productize" these offerings to appeal to each market directly. By differentiating and packaging your offerings as either "professional computer services for the business community" and "professional-style computer services for the home or school PC user" may be one way of doing so."
fluxtatic - Saturday, September 17, 2011 - link
Yes, but it's stupid, is the point. It's a classic example of corporate douche-speak. If this is what you're trying to get across, say "it will not be put into production". If you're "packaging" your offerings, say that, not that you're "productizing" your offerings. I think if somebody said that to me in a sales pitch, that would be the end of the meeting right there. At the very least, I'd push a shedload harder for better terms than I might otherwise. I once started steering business away from a major vendor because the sales rep said "irregardless" in conversation. It only went for a few days, but it was a few grand they missed out on because their college-educated (MBA, if I had to take a guess) rep speaks worse English than most of the high-school dropouts I've known (and there have been a lot.)Assuming by the way you look down on Bubbacup and his compatriots that you're an American, JKflipflop98, please stop. You're making the rest of us look like assholes.
Bubbacub - Friday, September 16, 2011 - link
is there a way of getting italics in comments? html in text doesnt work as in my last postCamikazi - Friday, September 16, 2011 - link
Anandtech uses BBCode I believe, let's see if I am right...Camikazi - Friday, September 16, 2011 - link
[noparse]Yep BBCode, so it would be for Italics for Bold.[/noparse]Camikazi - Friday, September 16, 2011 - link
OK so not all BBCode :P anyway [ i ], [ /i ] and [ b ], [ /b ].CZroe - Tuesday, September 20, 2011 - link
Uhh, there were many CPUs with no heat sink in the original Pentium era. In fact, I seem to recall seeing some Pentium systems that managed it. Is This really the accomplishment they are making it out to be?stephenbrooks - Sunday, September 25, 2011 - link
Solar panels, yes, but can it run off a POTATO BATTERY?