Introducing Intel's 14nm Node and the Broadwell Processor

[cleeve]

Intel finally provides solid information on Haswell's successor, the next-generation Broadwell core. We also learn some detailed info about the new 14nm processing node, a must-read for CPU enthusiasts who are interested in the future of Intel's Core!

Introducing Intel's 14nm Node and the Broadwell Processor : Read more

 

[Mike Stewart]

Nope... still not good enough !
at 4.6Ghz my 2700K is more than a capable CPU .
Bring on the Skylake,, then we'll talk .

 

[InvalidError]

With Intel heavily focusing on power-efficiency, my bet is clocks are not going to get bumped by more than 200MHz, quite possibly less: by collapsing the multiplier pipeline from five cycles to three and supersizing a bunch of other things, that is a whole lot more logic per stage and of those additions chip away at any timing closure margins that may have been gained from the shrink.

I would expect this to also translate into even more unpredictable and voltage/temperature-sensitive overclock outcomes.

 

[Shneiky]

2700K at 4.0/4.4 turbo 1 core for silent operation. Guess this year, I will change the Evo to something beefier and go 4.2 or 4.3 with 4.6 as a turbo to remain silent. Rocking 2011 year gear and still not finding a reason to upgrade. For a person who renders almost all the time, this CPU stagnation is frustrating.

 

[Gaurav Rai]

Its good to see Intel working so hard on their thermal department. Gaming is great, but you cant help feeling guilty about mother earth every time you fire up your pc. Meanwhile Amd innovates with 220W processor XD

 

[ceeblueyonder]

intel need die shrinks to cram in their specializalize units or cpu's. i am not really sure. but, intel has a bunch of specialize instructions built into their chips to beef it up, make it have faster ipc but also relying on software to take advantage of it. things like quicksync comes to mind or the SSE or whatever it is called. in comparison, amnd chips like the fx-series chips and the phenoms before it seems simpler, to me. a more general computing unit. kind of like powerpc chips or arm chips. they compose of execution cores or integer cores and floating point units and thats it. no instruction sets and quicksync decoders to gain software-driven advantage. but i don't really know. just hunches. i tend to make hunches but hopefully, they're educated hunches. so i am rooting for amd! intel's die shrinks seems to be like a monopolistic grip that keeps others with better "Architectures," simpler more general logical integer units to stay behind. since these cores with die shrinks, say an amd at 14nm, too would probably blow the top off of a competing intel that is also at 14nm. again, general computing units which i think AMD, IMB POWERPC, ARM chips have an inherent advantage to intel's inferior x86 architecture. AMD and POWERC both were the first x64 cpu's. but i could be wrong. again, just a hunch. intel blows amd out of the water today b/c intel has chips at 14nm, competing with AMD that has 32nm or 28nm chips. and also the software-driven instruction sets that intel has crammed into their chips that make software developers basically just check a box or intel has supported them to make intel chips run even faster than say AMD, which doesn't have quicksync or SSE or whatever it's called.

thus beat amd. but, to me, amd chips like the fx-series and the phenoms before have a simplicity to them that i admire. although i can't specifically say how or what it is.

 

[qlum]

While better efficiency is nice and all I fear intel won't do enough for gamers to warrant an upgrade of their cpu. When overclocked Haswell doesn't do much above Sandy bridge and while intel may not have the strongest competition from amd on the higher end anymore if people won't upgrade their cpu's it will hurt them on the longer run.

 

[InvalidError]

Meanwhile Amd innovates with 220W processsor XD

Can you really call it innovation when AMD needs a 200W chip to compete with Intel's sub-100W chips? Unless you meant innovation in the high-tech space-heater market.

Intel has gone down the crank-clocks-power-be-damned path with Prescott about a decade ago and that did not work too well. AMD just tried the same thing and "shockingly," that did not work particularly well for them either.

 

[ceeblueyonder]

it's also odd that intel is the die shrink boy. die shrink cpu company doing all the die shrinks. perhaps, the die shrink industry can only support one coompany at a time. as if, supporting two companies that are also doing tick tocks would "saturate" the indsutry too much. it probably has a lot to do with greed. money. but, that doesn't matter. what we all need to know is that just because intel beats amd today doesn't mean intel makes better chips. or that intel has superior technology. we need to take a closer look at what intel is doing and what amd is not doing and then look at their products more closely than just geekbench scores or ceinbench scores. look under the hood.

 

[Shneiky]

@InvalidError,

What Gaurav Rai said:
"Meanwhile Amd innovates with 220W processsor XD"

Was humor. You know, like Ha-ha and stuff?

 

[InvalidError]

intel blows amd out of the water today b/c intel has chips at 14nm, competing with AMD that has 32nm or 28nm chips.

Even if you compare Sandy Bridge (32nm) Intel CPUs with AMD's FX83xx (28nm) which theoretically gives the advantage to AMD, Intel's older chips still win most benchmarks. Intel being one process node ahead has very little to do with their performance lead; their architecture itself is just that far ahead.

 

[xtreme5]

Good work, but unfortunately there will be no huge difference in performance as compare to Haswell, intel is going to make thier value down. Bring up something Stormy, lets wait for IGPU.

 

[mapesdhs]

"... Having said that, Moore's Law appears to continue unabated for the moment. ..."

Hardly. Performance of the current Intel 4-core isn't that much better than the
equivalent model from 18 months ago. I know they've improved power consumption,
etc., but without significant speedups, most potential users really won't care.

Mike Stewart, you should be able to run your 2700K at 5.0. Every 2700K I've
obtained runs at 5 no problem, with good temps, etc.

OTOH, the chipset improvements with Z97 do at least offer a vaguely passable
rationale for upgrading, re the greater number of Intel SATA3 ports, newer
storage tech, etc. If budget was not an issue, I'd build with a 4790K without
hesitation.

Can't help feeling though, with various comments I've seen this past few weeks,
that what may be holding many people back from their ideal build is RAM pricing
which is now completely ridiculous. RAM is just too expensive. Huge step backwards
in system cost. And please I don't want to hear about chip shortages, etc., we all
know why RAM is more expensive now, because it's happened so many times before:
the suppliers don't like the pricing levels, so they restrict supply to raise prices. Well
IMO it's counter productive, because I can't be the only one who thinks no thanks,
I'm not paying that much for an 8GB 1600 kit when for about a 3rd less one could
get an 8GB 2133 kit a year+ ago, so heck with it I'll look for used kits instead, save
a bundle. I've bought four used GSkill 2x4GB 2133 kits this year, saved over 100 UKP
so far.

Price drops & efficiency improvements on CPUs are all fine & lovely, but what's the
point if potential future power savings are being wiped out by an artificially upfront
cost increase via the RAM?

Ian.

 

[RedJaron]

Intel has gone down the crank-clocks-power-be-damned path with Prescott about a decade ago and that did not work too well. AMD just tried the same thing and "shockingly," that did not work particularly well for them either.

Which makes it all the more funny considering the Athlon XPs at the same time were more focused on efficient computing with better IPC instead of insane clock rates. You'd think AMD would have learned enough from that time not to fall into the Netburst trap.

 

[maroon1]

It is not up to 5%
Intel slides says >5% which mean over 5%.

[Edit: Thx, fixed!]

 

[balister]

"... Having said that, Moore's Law appears to continue unabated for the moment. ..."

Hardly. Performance of the current Intel 4-core isn't that much better than the
equivalent model from 18 months ago. I know they've improved power consumption,
etc., but without significant speedups, most potential users really won't care.

Moore's Law states that the number of transistors on the chip will double every 24 months: http://en.wikipedia.org/wiki/Moore's_law

From the Wiki article: Moore's law is the observation that, over the history of computing hardware, the number of transistors in a dense integrated circuit doubles approximately every two years.

Double Transistors <> Double Performance (although early on it seemed that way)

 

[TopLuca]

I am not that impressed

 

[robholden]

"For example, if we compare Intel's 22nm to 14nm nodes, we find that transistor fin pitch (the space between fins) has been reduced from 60nm to 42nm, transistor gate pitch (the space between the edge of adjacent gates)"

Actually pitch means the space from the center of one fin to the center of the adjacent fin... it is not just the space between the 2 fins...

 

[InvalidError]

Double Transistors <> Double Performance (although early on it seemed that way)

Back in those days, newer chips with more transistors were also on a smaller process, significantly higher clocks and usually accompanied with some fundamental performance enhancements/breakthroughs so the performance doubling every ~18 months was a combination of multiple compounding factors.

Today, practically all the fundamental discoveries have been made and all they are doing is refine them so that side of performance scaling is effectively shut down. The clock scaling also appears to have hit a brick wall since the latency hit from making pipelines longer to enable higher clocks causes the execution pipelines to stall on dependencies more often and negate gains from higher clocks. Process wise, they are at a point where they are starting to fight with fundamental laws of physics, which does not help with smooth progress either.

There is little reason to believe things are going to improve much any time soon when all aspects are well into their diminishing return curve.

 

[ceeblueyonder]

intel blows amd out of the water today b/c intel has chips at 14nm, competing with AMD that has 32nm or 28nm chips.

Even if you compare Sandy Bridge (32nm) Intel CPUs with AMD's FX83xx (28nm) which theoretically gives the advantage to AMD, Intel's older chips still win most benchmarks. Intel being one process node ahead has very little to do with their performance lead; their architecture itself is just that far ahead.

FX-83xx series are 32mm, btw/fyi. fx-8350 vs. i7-2600k is probably a fair fight. i bet they'd trade blows. or, an fx-8350 is not far behind if it is behind. and amd has a software/platform/optimization disadvantage, meaning that the programs are not optimized for amd chips since most pc's have intel chips inside them.

 

[ceeblueyonder]

intel blows amd out of the water today b/c intel has chips at 14nm, competing with AMD that has 32nm or 28nm chips.

Even if you compare Sandy Bridge (32nm) Intel CPUs with AMD's FX83xx (28nm) which theoretically gives the advantage to AMD, Intel's older chips still win most benchmarks. Intel being one process node ahead has very little to do with their performance lead; their architecture itself is just that far ahead.

FX-83xx series are 32mm, btw/fyi. fx-8350 vs. i7-2600k is probably a fair fight. i bet they'd trade blows. or, an fx-8350 is not far behind if it is behind. and amd has a software/platform/optimization disadvantage, meaning that the programs are not optimized for amd chips since most pc's have intel chips inside them.

 

[mapesdhs]

Moores law says nothing about performance. It only has to do with the number of transistors on a chip doubling roughly every 18 months. ...

It was never a Law as such, merely an observation that seemed to be conveniently accurate many years
ago, but often it's quoted as being either a performance doubling or a density doubling every year and a
half (not 24 months). And for the other poster, wikipedia is not the word of god.

Either way, my point still stands - neither angle has been true for a long time now.

Ian.

 

[InvalidError]

fyi. fx-8350 vs. i7-2600k is probably a fair fight. i bet they'd trade blows. or, an fx-8350 is not far behind if it is behind.

The i7-2600k wins most benchmarks by a 10-20% margin and quite a few by a more substantial 30-50% lead. The only benches AMD wins by a significant margin (~15%) are 7zip and 2nd-pass h264.
http://www.anandtech.com/bench/product/697?vs=287

To make the FX a more even match for the stock i7, it needs at least an extra 600MHz.

 

[ozicom]

They made the "tick" with getting same performance with less watts and made "tock" with making the architecture better and giving better performance with that watts. So Intel's play seems good for me. You'll get better gaming performance when game developers develops games for that architecture. For example you can get better gaming experience with less performance devices like PS4 or Xbox because developers are making special games for those platforms. Intel's move from 22nm to 14nm is a good choice. I'm waiting for new products.

 

[ceeblueyonder]

fyi. fx-8350 vs. i7-2600k is probably a fair fight. i bet they'd trade blows. or, an fx-8350 is not far behind if it is behind.

The i7-2600k wins most benchmarks by a 10-20% margin and quite a few by a more substantial 30-50% lead. The only benches AMD wins by a significant margin (~15%) are 7zip and 2nd-pass h264.
http://www.anandtech.com/bench/product/697?vs=287

To make the FX a more even match for the stock i7, it needs at least an extra 600MHz.

i did acknowledge that "if the fx-8350 is behind, it isn't behind by much." 10-20% is not much to me. it's not what you considered in your earlier post as intel having an architectural advancement b/c if you wanna speak architecture, AMD patented x64. x64 is better than x86 which intel uses. correct me if i'm wrong.