"APM's performance estimates put a 3GHz X-Gene at roughly half the integer performance of a 2.4GHz Sandy Bridge."
Yet the chart at the top says X-Gene vastly outperforms Sandy Bridge per core in SpecINT? I can't read the tiny text in the image, but I must be missing something big.
the tiny text reads the core count :) so at the top you have 32 cores, then 16 and so on... so if 32 core ARM outperforms 4 core sandy bridge 2x, then per core it still doesn't add up, but then I can't read everything that's written in tiny text either...
Since they are talking about single socket performance, I'm assuming they mean you can fit so many ARM cores in one socket that eventually you will pass a Sandy Bridge socket in perf/watt.
Its chips like this that make the MacBook Air on ARM rumor seem like it can and will become a reality. It wont be soon, but by 2014-15, its almost a "why wouldn't they?" question, opposed to a why would they...
As an owner of both a 2010 MBA and a 2011 and having seen the drop in battery life and the increase in heat in the new model, I am looking forward to a fast ARM SOC for laptops.
Considering Apple bought 2 chip design companies in the past 3 years, I can only assume that they think they can design a better chip than Intel which is low power and has great battery life.
And when you think about what people actually do on the web most of the time, browse the web, check e-mail, chat on IM, most of that stuff is not CPU-demanding. In fact, arguably, CPUs from 5 years ago would be fine for most day to day work. But getting more battery life out of our machines is really key.
Intel Atom and competitive in the same sentence? Really? I mean, if their development speeds up, Atom could be very competitive. But with current products they are quite a ways off versus AMD's Brazos, let alone ARM, for power/performance.
This sounds exactly like something that is going to have a derivative in a future gaming console. Good buy red ring of death. Hello 16 core armv8 soc + hd6000 graphics under a single lid.
I'll believe it when they ship product and have the implied claims verified by 3rd parties. I know the performance numbers are for a picked benchmark (that is apparently very friendly to large numbers of RISC cores), but the implication is "X-Gene is going to wipe the floor with x86 (specifically Intel)".
I have no problem believing they can build high performance ARM cores (they just have to architect for it, and they did that).
I do have a problem believing they can do 3 GHz clock speed and vastly improved IPC for that small of a power penalty, with interconnects at 400 Gbps (that is 50 GB/s, or 50 lanes of PCIe 3.0). Interfaces that fast aren't cheap to run, power wise. Something smells fishy. The ARM ISA may be nice and efficient, but it's not magic, it doesn't obviate physics.
Plus by 2013 when these come out, Haswell will integrate more onto the chip (more power savings), and Atom (the real competition here) will be at 22 nm, and probably much improved architecturally.
I have just read the space can be shrink in more than 90% The energy costs cut by almost half And we do not know the SoC prices but I bet it will be far below intel ones.
As you can put 2 SoCs or 3 to beat Intel performance, even at a desktop cpmputer will make it cheaper.
But we do not know the price yet
i will not have any problem in having a desktop system with 3 or 4 of this - cheap SoCs - and as they say the space and the energy cost is lower than x86 solutions perhaps for servers will be a good product too.
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.
Facebook
Twitter
RSS
13 Comments
Back to Article
Andypro - Monday, November 14, 2011 - link
"APM's performance estimates put a 3GHz X-Gene at roughly half the integer performance of a 2.4GHz Sandy Bridge."Yet the chart at the top says X-Gene vastly outperforms Sandy Bridge per core in SpecINT? I can't read the tiny text in the image, but I must be missing something big.
MantasPakenas - Monday, November 14, 2011 - link
the tiny text reads the core count :) so at the top you have 32 cores, then 16 and so on... so if 32 core ARM outperforms 4 core sandy bridge 2x, then per core it still doesn't add up, but then I can't read everything that's written in tiny text either...Kristian Vättö - Monday, November 14, 2011 - link
Full size pic is here: http://images.anandtech.com/galleries/1532/Screen%...shiznit - Monday, November 14, 2011 - link
Since they are talking about single socket performance, I'm assuming they mean you can fit so many ARM cores in one socket that eventually you will pass a Sandy Bridge socket in perf/watt.Doormat - Monday, November 14, 2011 - link
Its chips like this that make the MacBook Air on ARM rumor seem like it can and will become a reality. It wont be soon, but by 2014-15, its almost a "why wouldn't they?" question, opposed to a why would they...shiznit - Monday, November 14, 2011 - link
As an owner of both a 2010 MBA and a 2011 and having seen the drop in battery life and the increase in heat in the new model, I am looking forward to a fast ARM SOC for laptops.dagamer34 - Monday, November 14, 2011 - link
Considering Apple bought 2 chip design companies in the past 3 years, I can only assume that they think they can design a better chip than Intel which is low power and has great battery life.And when you think about what people actually do on the web most of the time, browse the web, check e-mail, chat on IM, most of that stuff is not CPU-demanding. In fact, arguably, CPUs from 5 years ago would be fine for most day to day work. But getting more battery life out of our machines is really key.
eio - Monday, November 14, 2011 - link
Larabee in ARM flavor, on a new process nodedew111 - Monday, November 14, 2011 - link
Intel Atom and competitive in the same sentence? Really? I mean, if their development speeds up, Atom could be very competitive. But with current products they are quite a ways off versus AMD's Brazos, let alone ARM, for power/performance.hooflung - Monday, November 14, 2011 - link
This sounds exactly like something that is going to have a derivative in a future gaming console. Good buy red ring of death. Hello 16 core armv8 soc + hd6000 graphics under a single lid.bjacobson - Friday, November 18, 2011 - link
what would keep them from moving to ARM? The internet runs Linux, so I don't see why it would be a big deal to recompile and jump.dertechie - Saturday, November 19, 2011 - link
I'll believe it when they ship product and have the implied claims verified by 3rd parties. I know the performance numbers are for a picked benchmark (that is apparently very friendly to large numbers of RISC cores), but the implication is "X-Gene is going to wipe the floor with x86 (specifically Intel)".I have no problem believing they can build high performance ARM cores (they just have to architect for it, and they did that).
I do have a problem believing they can do 3 GHz clock speed and vastly improved IPC for that small of a power penalty, with interconnects at 400 Gbps (that is 50 GB/s, or 50 lanes of PCIe 3.0). Interfaces that fast aren't cheap to run, power wise. Something smells fishy. The ARM ISA may be nice and efficient, but it's not magic, it doesn't obviate physics.
Plus by 2013 when these come out, Haswell will integrate more onto the chip (more power savings), and Atom (the real competition here) will be at 22 nm, and probably much improved architecturally.
mitcoes - Monday, October 22, 2012 - link
I have just read the space can be shrink in more than 90%The energy costs cut by almost half
And we do not know the SoC prices but I bet it will be far below intel ones.
As you can put 2 SoCs or 3 to beat Intel performance, even at a desktop cpmputer will make it cheaper.
But we do not know the price yet
i will not have any problem in having a desktop system with 3 or 4 of this - cheap SoCs - and as they say the space and the energy cost is lower than x86 solutions perhaps for servers will be a good product too.