Bay Trail is still seasons away but engineering samples are, and this should be no surprise, already in use at least for research and development purposes. Someone, somewhere down the line, decided to run a benchmark which was posted online. AnTuTu, the benchmark utilized, measures a spread of factors including memory, integer performance, floating point performance, 3D performance, and so forth. Unfortunately it does also include some non-CPU/GPU factors in its score, albeit barely, so best take it with a grain of salt.
The Droid Guy compiled it with a few scores from competing, ARM, devices.
Image Credit: The Droid Guy
The Silvermont-based chip, clocked at an… actually quite modest 1101 MHz, received a synthetic score of 43416. To put that in comparison: arguably the fastest ARM processor on the market, the Qualcomm's Snapdragon 800, tends to find itself with a score around the 30,000-32,000 range which is about 27-31% slower than Intel. The very popular albeit soon deprecated Nexus 7, powered by the Tegra 3, scores 12726.
Personally, I am getting a little flashback of the Intel vs. AMD battle about 8 years ago. We seem to be close to a Conroe (Core 2 Duo) vs. AMD Athlon 64 FX point between Intel and ARM. Intel eclipsed the AMD Athlon 64 FX-57 (update: I meant FX-62) and kept throwing more money at research than AMD could possibly afford. Unless ARM can severely undercut Bay Trail, Intel could follow past trends and simply bury their competitors with tens of billions in capital investment until their products are so far ahead that consumers default to Intel products.
If history repeats itself, this leaves Qualcomm and others in a difficult position. The solution seems to be either to tread water in a price point that Intel ignores or to collectively dump money into ARM and run the "out-research Intel" treadmill. Remember, this is a company who will dump twice AMD's revenue into their Research and Development year-over-year to keep ahead. Unlike Intel's GPU efforts, which did not seem like a problem that cash could solve alone, they know how to make processors.
I would not make business decisions under the assumption x86 will keep Intel hobbled indefinitely.
This is only a big deal if
This is only a big deal if the Silvermont hits a power envelope equivalent to that of the ARM stuff.
These benchmarks are great, but at what cost on battery life?
Yeah to note amd is still
Yeah to note amd is still hurting badly but they are just now starting to recover from past missteps in the x86 arena.
My feeling is amd will go for smartphones after intel does most of the leg work getting all the mobile os’s up and running on x86.
Hah! That’d be cute for AMD.
Hah! That'd be cute for AMD.
AMD has already stated
AMD has already stated they’re not going into that sector.
I don’t know if x86 will ever
I don’t know if x86 will ever be able to effectively compete against ARM in a mobile/small device environment. The performance may be there, but as far as per-watt goes if Intel really wants to compete I think they’ll have to get a little more RISC-y.
That, or battery tech will have to evolve significantly. (That would be nice.)
You evidently didn’t keep up
You evidently didn’t keep up with how Intel made Clover Trail consume less than comparable ARM processors, with the z2760 managing to consume LESS power than dual core Krait;
while I was honestly expecting a bit more on that side, recent leaks show that the dual core mobile 1.46GHz Bay Trail part will have an SDP of 2.5W and a TDP of 4.5W, which is in line with the Snapdragon 600 currently used in cellphones, while higher frequency and quad core parts will reach 4.5W/7.5W, which is perfectly acceptable for a tablet (and the reason why I’m really looking forward to bay trail convertible windows tablets).
Anyway x86 CPUs have been RISC processors with a CISC-to-micro-op converter for almost two decades, that argument doesn’t really stand. Removing some backward-compatible parts could definitely help but don’t think it could be a revolution, Intel surprisingly already managed to reach ARM-like power consumption.
http://www.anandtech.com/show/6536/arm-vs-x86-the-real-showdown
uhh no it doesn’t snapdragon
uhh no it doesn’t snapdragon 600 does 2.5 to 3W TDP. vs 4.5 for Intel as you point out. That’s 50% higher…not even close to get smartphone makers to even think about switching.
But look how long took Intel
But look how long took Intel to get there? Back when ARM poked that bear claiming Intel couldn’t match them in power useage, 3 months later intel has a cpu down to power usage in ARM’s range. On top of it, Intel Atom cpu was still faster then ARM cpu. So since only been like maybe a year since intel started that route they are very close.
Intel has a process node lead
Intel has a process node lead over ARM at the moment, but put the ARM chip on the same process node, and re-test! Also the metrics that some of these Intel tests use, and the lack of an Intel retail version, that can be purchased through a retail channel, and tested indipendently should lead anyone to take these results with a grain of salt!
Its not 3 months later jeez,
Its not 3 months later jeez, this chips has been in the works for allot longer. This is not the same as Intel vs AMD. People who think Intel will just crush the ARM platform are in for a big surprise.
There isn’t really that much
There isn’t really that much more to say that isn’t mentioned in the conclusion of the anandtech article you linked.
My impression, informatize me if I’m wrong, is that Intel maintains a process tech advantage over the ARM camp, however ARM is currently ahead of Intel in terms of architecture. In my opinion I think it will be easier for the ARM camp to move to better/more competitive process techs than it will be for Intel to move to better architectures.
In the end, both those are just bets on the future. (Personally, I prefer a LP i5/i7 in my tablets.)
I think it should also be
I think it should also be pointed out that while it did have great power and performance as a smartphone or tablet SoC it wasn’t very good.
Qualcomm has a much nicer package than Intel. By adding radios alone it’s a much more enticing offer than past and current Intel parts of the time. It’s also my belief that in real use the Krait part would actually have shown better battery life in a phone, maybe not so much in a tablet.
I still think Intel is ARMs biggest threat. Currently, they’re package isn’t as good, and I don’t think they’ll be able to compete on price.
“I don’t think they’ll be
“I don’t think they’ll be able to compete on price”
Huh? High end ARM chips are in the $25-30 range whereas the high end Atom is in the $50-60 range.
There’s already been faked
There’s already been faked mobile benchmark submissions with other benchmarks. I wouldn’t put too much merit into a bench that shows up on AnTuTu either. Nothing to see here, move along.
http://www.chiphazard.com/2012/09/22/galaxy-s3-decimates-iphone-5-in-geekbench-scores-previous-comparison-was-fake/
Yep, like always, we’ll need
Yep, like always, we'll need to wait and see to know for sure.
Is it really necessary to
Is it really necessary to write with such a snarky and arrogant tone?
It looks especially bad when coupled with laughable quotes like “tread water in a price point that Intel ignores”.
In the mobile market it doesn’t matter what Intel ignores because everybody who’s anybody ignores Intel. And as long as Apple, Samsung, LG and HTC keep ignoring Intel and their precious price points, how fast Intel chips are doesn’t matter.
Snarky and arrogant was not
Snarky and arrogant was not the tone.
It feels a lot like how AMD ignored Intel back in 2004/2005. They thought they could beat anything Intel could throw at them. The problem is Intel will spend tens of billions of dollars each and every year in research and development if need be.
There sure is someone being
There sure is someone being snarky and arrogant in this discussion, I’m just not sure it’s the author.
agreed
agreed
It is not simply an Intel vs
It is not simply an Intel vs AMD comparison, One company with wads of cash verses another with less r&d cash! ARM is a whole industry, who just happen to license the ARM holdings IP at different levels, ranging from outright using the ARM supplied refrence CPUs, to licensing the ARM instruction set and designing circuitry the from the ground up, like Apple and a few others do! Apple, for sure, has the money to take on Intel! With Apples hoard of cash, throw Samsung and Qualcomm into the mix, and just these 3, can out spend Intel, and Keep ARM in the forefront for quite some time! Intel’s pricing will keep it out of the market, except for the very High end/High cost market where price does not matter! Intel has very few years with which to leveredge its process technology lead, and in the next 3 to 5 years, most of that process lead will be all but gone!
No mobile OEM wants an Intel Ring through its nose, like Intel had with the x86 and OEMs in the past! ARM, for all points and purposes, provides an Industry Standard Instruction set, backed by ARM Holding’s IP and decades worth of Software IDEs, and applications, and software toolchains, that Many OEMs from washing machines to phones, tablets, and chromebooks use in their products, at a much better price point than Intel will ever be able to provide!
Yes Intel has the most powerful x86 processors, but that comes at a high price, and Intel still needs AMD and Nivida descrete GPUs to win the gaming benchmarks!
Anandtech: “The ARM Diaries, Part 1: How ARM’s Business Model Works”
http://www.anandtech.com/show/7112/the-arm-diaries-part-1-how-arms-business-model-works
Exactly, its Intel vs ARM,
Exactly, its Intel vs ARM, Qualcomm, Apple, Nvidia and all the others invested in the ARM platform. To compare this fight with the AMD/x86 wars is missing so many different factors that its nothing of the like.
That is kind-of the point of
That is kind-of the point of the last paragraph (excluding the single-line one).
If they want to keep Intel down, they will need to "collectively dump money into ARM and run the 'out-research Intel' treadmill"… together.
Sure, an entire industry can often beat a single company; but, that single company is Intel. If the industry doesn't take Intel seriously… well… Intel is going to keep spending (as they said at Computex) $12 Billion USD per year. That is a lot of R&D money… if the companies under the ARM banner don't take them seriously.
They really should… especially if Intel is as close as the benchmark suggests.
In the tablet market is where
In the tablet market is where ARM needs to take Intel seriously, as well as the high end smart phone market!
Intel’s cpu prices are more costly, and for Intel to sale in any other lower cost devices, would require Intel to lower its prices, and this will affect Intel’s ability to spend the large amount of R&D money on lower margin devices! One reason Intel’s chips cost so much money is the large amount of R&D capitol that Intel has to spend, and Intel has to amortize that cost by selling large amounts of expensive chips! If Intel can not produce a chip that can compete with ARM on the performence and price front and quickly gain enough market share to sell millions of units, Intel is going to have a hard time justifying the large R&D budget to its investors! Intel is already looking for clients who need chip fabrication services, beacuse Intel’s chips fabs, cost billions to run, and unless Intel keeps those fabs running at as close to full capacity an possible, those fabs can quickly move from being a cash cow to a cash drain!
Who says they don’t take
Who says they don’t take Intel seriously? There lively hood is based on ARM succeeding. These companies know how to make low power work. They have been doing it for many years. Intel is the new kid in this area and its completely different world than desktop and server world that Intel has been in.
People forget what happened to Intel when they tried to take on Nvidia and AMD a few years ago in GPU’s. They got embarrassed and slinked away.
I referred to GPUs in the
I referred to GPUs in the story…
To me, it seems like GPUs were time-sensitive when it comes to research… as opposed to cash-sensitive. That was not something that just better fabs and a few teams working in parallel can solve… it required a lot of supporting software… and so forth.
Mobile SoCs feel more like something Intel can just throw money at.
As long as Apple uses ARM,
As long as Apple uses ARM, Intel is irrelevant in smartphones and tablets. Performance will only get faster in future ARM architectures.
And do not forget that Nvidia
And do not forget that Nvidia will be licensing some of its GPU IP, and this GPU IP will be used by current and future companys, ARM based and others, to provide graphics and GPGPU acceleraton for Future Mobile platforms! Nivida has a lot of experience and IP built around its HPC/Supercompiter products! I am sure that Apple, Samsung, Qualcomm, etc. will be evaluating and may use the Nivida IP in some of their future products!
It strikes me that ARM stuff
It strikes me that ARM stuff is still cheap, effective logic for such things as calculators, refrigerators, and cars, and they will sell tons into those markets. Serious computing tasks, especially multitasking, in high end mobile devices will be best served with chips that best serve desktops and laptops: x86 chips from intel and maybe amd.
I am a pure desktop guy with no need, knock on wood, for mobile gadgets powered by ARM crap.
Intel is going to win.
All
Intel is going to win.
All this talk of battery life is non-sense.
If people cared about things like battery life, do you think the auto industry would still be selling tucks? Not in a million years.
People will get hooked on performance numbers and not worry about battery life.
Intel managed to kill the RISC processors in the mainframe and workstation market. They’re going to kill ARM as well.
And now ARM based server
And now ARM based server chips are entering the market, to give Intel competition in that market, ARM is far from having its RISC based processors killed, ARM is everywhere, when you by a laptop/PC, chances are very high that the controller for the hard drive or SSD is an ARM based controller, not to mention the Wifi and other devices around your house! RISC is not dead, RISC is on more Devices than x86 CISC could ever be!
Lol people don’t care about
Lol people don’t care about battery life….just so freaking funny…and clueless.
If people only cared about super high performance why are desktops going the way of the dinosaurs? The only people who care are the hardware enthusiasts. A very small niche.
If a server farms were run by
If a server farms were run by your logic, the server services company would be bankrupt on power bills! Do you even know what it costs for the power to run a big server site! Do you think that all the money these multibillion dollar companys make comes from people like you, in your mothers basement playing on you gaming rig, the money to develop your gaming rig CPUs comes from mostly the sales to the server industry and enterprises, these “poeple” you speak for only make up a small fraction of the market, and the main theme of this article, is about mobile, where power usage and cost rule!
Your Mommy’s Basement World View shows, and you are so Butthurt over the reality, Intel no longer will rule!
Bay Trail is nothing special,
Bay Trail is nothing special, Intel will have a slight lead in terms of CPU performance but in GPU it will get oblitirated with ease. Bay Trail specs leaked and AMDs Temash wil keep pressuring Bay Trail.
I looked at specs, Bay Trail does not stand a chance above 10tdp
Bay Trail is 50% faster per clock compared to Clovertrail+ that at 1ghz cinebench 11.5 scores 0.10 pts in single thread performance so Bay Trail will score 0.15 while AMDs Temash at 1ghz scores 0.26 pts single thread.
ARM chips at 20nm will crush Bay Trail and AMDs Beema/Leopard will crush Bay Trail. So, dont get hyped up about Bay Trail, it will be still crappy.
can’t we have intel + nvidia
can’t we have intel + nvidia solution for tablet or high smartphone ?
Another Architectural Holy
Another Architectural Holy War with added willy-waving and bogus logic thrown in for free!
We’ve been here before: How’d that work out for MIPS, SPARc, ALPHA, Silicon Graphics and the AMD 29K?
Seems to me that there are too many ‘offerings’ in the tablet and smartphone markets and its about to get even more crowded.
Joe Public will need some form of product differentiator.
Let the Marketing commence.
See you at the Intel April 2014 earnings conference call folks.
(I’m a long term Intel stockholder – 30 years)
Better diversify your
Better diversify your portfolio, beacuse the PC/Laptop market is saturated, and I would love to have a laptop/PC with one of those ARM server chips in it, Plenty of cores, and not the xeon server processor High price! Yes give me a 16 or higher core ARM server farm on a chip for cheeper than a core i7, and I will be very happy to 86 the x86!
Those ARM based chromebooks are just the start, of getting Intel Outside!
Intel has the best silicon
Intel has the best silicon engineers on the planet. Oh, and they have their own fabs to respin chips with any changes immediately. ARM is simply a designer. Intel does both. Don’t f*ck with Intel.
Really, do you even Know what
Really, do you even Know what it costs to make the masks for those chips, it costs millions, do you Know how long it takes to get a wafer line up and running, with low enough defect rates, so that all your profits are not eaten up by wasted product! Changes can not be implimented immediately in any high volumn chip FAB from any company, just look how long it takes Intel to fix the intel chipset defects! Apple has some damn good silicon chip engineers, Apple licensed the ARM instruction set only, then built their own microprocessor, from the ground up, around the ARM instruction set, to acheive some of the lowest power consuming tablets/smartphones, and Apple sells millions, and has a much bigger bank account than Intel! ARM holdings designs a refrence CPU and other IP that a whole industry uses, and ARM licenses come in different levels, from using the standard ARM refrence design, to licensing just the instruction set, and doing what Apple, Qualcomm, and a few others do, which is design a CPU to run the ARM instruction set! Fabless production will rule the marketplace, and allows many smaller companys to enter the ARM based ecosystem Industry, an Industry without an Intel ring through its nose!
CPU world is INTEL’s
CPU world is INTEL’s bailiwik. They are ALREADY closing in and can afford to throw enormous resources YEAR after YEAR to make products so compelling that manufacturers can’t resist. The GPU argument is a straw man because they were never close to the top graphics companies. To dismiss Intel as a competitor by looking in the rear view mirror is childish. They want market share in the mobile space, and I would bet they will get it.
Not If Apple has any thing to
Not If Apple has any thing to say about it, and Apple has the big bucks to take on Intel! The future does Not have an Intel Ring through its nose! Mobile device makers have knoledge of Intel’s past business practices, and even IBM, at the dawn of the PC era, knew enough to force Intel to second source and cross license the x86 instruction set with AMD! IBM did this to keep its CPU supplies safe from monopolistic market forces! The ARM and Linux products and ecosystem represents the individual’s and entire industry’s collective will, to be free of the unhealthy WINTEL alliance!
Intel FTW…I will be looking
Intel FTW…I will be looking for a “Silvermont” tablet !
Intel has the brains/fabs baby !
Another benchmarking pitfall
Another benchmarking pitfall by Intel and overhype spotted.
This was posted on Nvidia’s tweets.
http://www.bdti.com/InsideDSP/2013/07/11/JeffBierImpulseResponse