ARM, TSMC Tape-out 16nm Cortex-A57 Processor

[exfileme]

It's the fastest, most energy efficient design from ARM yet.

ARM, TSMC Tape-out 16nm Cortex-A57 Processor : Read more

 

[chicofehr]

Does this mean 16nm GPU soon? my 7950 is struggling with all the 8K textures and 30" monitor on Skyrim

 

[axarm]

This is tapeout. That is a design on paper not silicon. Arm will not have full productions of their A15 finfet till sometime in 2014 at the earliest. The A57 will NOT be produced before the A15. So full production is at least 2 years if not 3 years off.

 

[hero1]

If only they could jump from 28nm GPU chips to this and increase performance and efficiency. I would like to see the benchmarks for this. Intel has got to get going and bring something to the game fast if they are to take a chunk out of ARM shares in mobile sector.

 

[axarm]

Intel will be going to 22nm finfet this fall from 32nm nonfinfet now. Look at the 32nm Lenovo K900 benchmarks and they faster than the Samsung S4.With the move to 22nm finfet and other changes, Intel's Baytrail will have twice the performance. Arm will never have the performance or efficiency lead in mobile SoC again (that includes servers).

 

[tmk221]

[citation][nom]axarm[/nom] Arm will never have the performance or efficiency lead in mobile SoC again (that includes servers).[/citation]

Let's see Intel taking the lead in efficiency area first. Then we can discuss whether ARM will be able to catch up again or not...

 

[g00fysmiley]

looking forward to the increased processing power and efficiency, but still screens burn the most power, need mroe efficiency there to but still looking forward to benchmarks and seeing if the claimed performance increases will be as much as claimed

 

[-Fran-]

Charlie said something about "read between the lines and cut the BS".

There's more info in the Steamroller CPU thread in the forums.

Cheers!

 

[A Bad Day]

[citation][nom]tmk221[/nom]Let's see Intel taking the lead in efficiency area first. Then we can discuss whether ARM will be able to catch up again or not...[/citation]

They already do, according to Anandtech: http://www.anandtech.com/show/6529/busting-the-x86-power-myth-indepth-clover-trail-power-analysis

Now the question is, can Intel get manufacturers interested to switch from ARM?

 

[Bricktop]

If I had to guess, A57 devices will hit the market in mid-to-late 2014 at the earliest. TSMC hasn't announced that the 16nm FinFet fab is production ready yet. Once that happens, Nvidia and Qualcomm have to get some samples and determine how to implement/modify these cores into their SoC design. Additionally, while the A57 will be produced using the 16nm Fab, the the transistors on the chip will be 20nm according to Engadget.

By the time these chips make it to the consumer, Intel will have already pushed out 14nm products. Judging by the fact that Intel is opening up it's Fab to outside businesses (possibly even Apple), I would assume they will have additional production space to create 14nm SoC's soon after the release of "Broadwell". Intel just got it's feet wet in the smartphone space, but I think it will be a heated battle; Intel vs ARM by 2015.

 

[InvalidError]

[citation][nom]axarm[/nom]This is tapeout. That is a design on paper not silicon.[/citation]
Tape-out is one of the last steps before before lithography masks are made and first silicon can be produced. Depending on how much of a hurry ARM and TSMC are to have their first test silicon out, they could have their first (potentially) working chips in 3-4 months. The likelihood of nailing it on the first spin is low but it can happen... but even if they do, they will likely need a few more months to tune yields.

If there aren't any major setbacks, it could turn up in consumer devices in late 2014.

 

[danwat1234]

[citation][nom]axarm[/nom]Intel will be going to 22nm finfet this fall from 32nm nonfinfet now. Look at the 32nm Lenovo K900 benchmarks and they faster than the Samsung S4.With the move to 22nm finfet and other changes, Intel's Baytrail will have twice the performance. Arm will never have the performance or efficiency lead in mobile SoC again (that includes servers).[/citation]
They've been at 22nm and will continue with 22nm Haswell this year and 14nm Broadwell in mid/late 2014 or so

 

[hector2]

April Fools, right ? Tapeout means nothing. It's a software operation. They haven't even made the masks. They could just as easily have taped out 5nm. Wake me up when they actually fab their first silicon. It'll take another year after that to get to production-worthy yields

 

[InvalidError]

[citation][nom]hector2[/nom]April Fools, right ? Tapeout means nothing. It's a software operation.[/citation]
In a real production environment, the tape-out is the last step before the design gets shipped out for masks. From there to actual silicon, it takes 2-3 months if all goes well. Since TSMC needs to produce actual silicon to test their process yield, I would bet that TSMC is in just as much of a hurry as ARM is to find out how well the chips will turn out.

There is every reason to believe ARM's tape-out will be made into silicon as quickly as TSMC can accommodate since they both need it: ARM needs to prove their power-efficiency on a process that rivals Intel's and TSMC needs a test device to benchmark their new process yield. Intel usually produces SRAM chips for yield testing - simple regular structure that is very easy to test.

 

[jkflipflop98]

If they're actually producing c2c 16nm finfets, it's going to take more than 3 months to tune the silicon. It's no easy task.

 

[InvalidError]

[citation][nom]jkflipflop98[/nom]If they're actually producing c2c 16nm finfets, it's going to take more than 3 months to tune the silicon. It's no easy task.[/citation]
ARM likely isn't expecting fully working silicon on the first spin either. But both companies get something to work from with that... TSMC gets a design to practice/benchmark their yield with and ARM gets silicon to see if back-annotated post-PAR simulation matches silicon results.

Being the first guinea pig on TSMC's 16nm process likely comes with substantial discounts on those first masks and wafers, which is great for early silicon that most likely won't work (or at least not perfectly) on the first try anyway.

For TSMC, it does not matter too much whether or not ARM's design is faulty or not since TSMC can still determine how good their process is by examining the dies with a microscope between metallization layers as they fab the chips.