Intel's Broadwell Core M-5Y70: The First Benchmarks

[burkhartmj]

I couldn't care less about fanless, so comparing to Atom is nearly useless. For those wondering about how it compares to an i5, the i5-4300U [SP3] got 2.82 in cinebench, vs 2.67 for this guy.

Fanless means its going against iPads, Galaxy tabs and phones and iPhones, etc. If you don't care about those devices and are looking for a laptop replacement, SP2/SP3, then you should stop reading this article because you're not getting it here.

The price heavily disagrees with you. This is an ultrabook class price, and the comparable haswell processors are in ultrabooks. The price-comparable intel processors to ARM and Nvidia designs are Baytrail Atom, and they are what is competing against iPads and Galaxy Tabs. Fanless isn't likely to be limited to tablets anyways, it'll show up in many ultrabooks, anything to make them thinner and lighter.

 

[Doug Lord]

Atom is good enough CPU wise for MS Office. What we need to know is how the GPU compares to Apple or Nvidia.

 

[ingtar33]

they'll only keep that price if apple picks it up for their ipads... if apply doesn't buy this chip, they'll give it away for free to OEMs (like they currently are giving away bay trail) to push ARM out of the low end market, and finish off AMDs tablet/laptop market share.

 

[InvalidError]

Atom is good enough CPU wise for MS Office. What we need to know is how the GPU compares to Apple or Nvidia.

There probably won't be anything "comparing" to Nvidia's K1 any time soon since all other mobile SoCs are far more concerned with low-power than peak graphics performance. Apple only claims a 50% improvement for the A8's GPU and that does not put it anywhere near the K1. The Core-M only has slightly beefier IGP than Haswell so I would not expect miracles there either.

If you want significantly higher performance mobile graphics out of Intel's CPUs, you will need to wait for the 25-45W Haswell-M with GT5e graphics. Those should be about twice as fast as GT3e and finally bring Intel's IGP semi-usable levels.

 

[abitoms]

@Don

Fanless I believe

thank you. i then wonder why the author did not consider mullins chips

Mostly because Mullins doesn't yet have a fanless tablet design win. You can't actually buy it.
As far as I know anyway, and I have looked. I can only find it in an HP 10z laptop, but if you're aware of a fanless tablet/convertible with Mullins, please let me know!

Which is quite sad, really. It has potential but manufacturers won't pick it up. I hear the 10z has poor battery life, but haven't looked closely at it.

@don / @cleeve
yes, it is unfortunate that Mullins could have had more/better design wins with better batteries - it would have been fair competition. thank you for the clarification.

 

[hannibal]

Hmmm... I was expecting this between 800 to 899$... so it will be 1000$... Outch...
But, when there is no competition, no need to cut prices... Have to look for the next gen Atom... more palatable to my purse.

 

[Joao611]

I know why you didn't include Haswell-Y results in there, but can you get some, please? I'd really like to see the comparison.

 

[cleeve]

I know why you didn't include Haswell-Y results in there, but can you get some, please? I'd really like to see the comparison.

I tried! I put a lot of time into it, actually.

I have a Lenovo Miix2 with Core i5-4202Y. But the damn thing gives me worse benchmark results than the Bay Trail tablet, so I'm trying to figure out what is wrong before publishing.

I expect we'll have an in-depth Core M review in the near future, and Haswell-Y will be included, to be sure.

 

[sonofliberty08]

looks like this chips are definitely for intel fanboi, why do we need an chips cost $300 for mobile? there are bunch of better option on the market, lets hope Apple will acquire this chips for the icrap

 

[InvalidError]

looks like this chips are definitely for intel fanboi, why do we need an chips cost $300 for mobile? there are bunch of better option on the market

If you want a reasonably fast x86 chip under 9000... mW, the Core-M is pretty much the only ticket in town.

Yes, the price is ridiculously high for what basically amounts to an under-clocked, under-volted desktop i3 with 4MB cache but there is nothing else anywhere close to that performance and power level so Intel can afford to use the take-it-or-leave-it approach.

 

[patrickjp93]

Maybe they will release less expensive broadwell Ys as well...

The performance specs on Broadwell-Y are already low enough that it should not need any lower-end parts. What it could really use is a $100 price drop. The $150-300 markup on Intel's mobile CPUs compared to their nearest equivalent desktop counterparts really bugs me..

Then get Qualcomm and AMD off their asses and have them design a competitive performance product.

@Don

Fanless I believe

thank you. i then wonder why the author did not consider mullins chips

Because Mullins is shaping up to be an utter failure with almost no improvements on the previous generation.

 

[sonofliberty08]

looks like this chips are definitely for intel fanboi, why do we need an chips cost $300 for mobile? there are bunch of better option on the market

If you want a reasonably fast x86 chip under 9000... mW, the Core-M is pretty much the only ticket in town.

Yes, the price is ridiculously high for what basically amounts to an under-clocked, under-volted desktop i3 with 4MB cache but there is nothing else anywhere close to that performance and power level so Intel can afford to use the take-it-or-leave-it approach.

the quad core ARM chips was fast enough and serve me well for my mobile needs, i don't think i need any so call fast x86 mobile chips on the go at the moment, i only rely on my x64 desktop/workstation for the performance needs.

 

[ingtar33]

The problem is mullins out performed bay trail pretty badly yet gained zero traction because intel is giving away baytrail for free. Why should AMD or Qualcomm get off their backside and make a superior product if everytime they do it, intel just gives away the inferior product for free? Its the same as what happened 2003-2007, when intel gave away p4s for free to prevent athlon64 of athlon64 fx cpus from gaining market share.

Then of course instead of making money from Mullins no one uses it because baytrail is cheaper (as in free), and then intel comes out with what looks like x3 the performance of baytrail in this core m chip... and the opportunity vanishes for AMD in less then 4 months. sorta hard to play on an even field when chipzilla can break the rules at will to tilt the field in their favor whenever they need to, heck, chipzilla spends something stupid like x2 in R&D then AMD gross' in a year (note, gross, not net).

The numbers i see in this preview should freak AMD out. They no longer are remotely competitive with intel if this is the case; the only sector of the chip business that they made any money in the past 2 years has been in mobile. this part makes their whole mobile lineup irrelevant.

 

[InvalidError]

the quad core ARM chips was fast enough and serve me well for my mobile needs, i don't think i need any so call fast x86 mobile chips on the go at the moment, i only rely on my x64 desktop/workstation for the performance needs.

ARM chips do not run standard x86 applications and the Core-M is intended for premium convertibles/ultrabooks which will usually be purchased by people who do not want to be ball-and-chained to their desktop or do not want to bother with owning a desktop at all.

 

[jimmysmitty]

So I want the same usecase- which is a form factor of surface pro 2 or 3. I'm happy with the size, weight and battery life. Core M, if used in next surface pro, would be a downgrade in processor speed - while still being just as expensive. That is why I want benchmarks - to compare how much slower it is.

Very different markets. The Surface Pro 3 with the i3, for example, is a 11.5W TDP vs this 4.5W TDP. This is probably meant for more battery life and to compete with the ARM Cortex series, which seems to be doing very well from the presentations.

As well this could be the next step towards a x86 phone CPU which I am very interested in. I have used all the ARM types from single core to 8 core and the performance is not as impressive overall, on the other hand they are great for battery life which is what this CPU is focusing on.

 

[LummusMaximus]

I would love to see a Core M sereis processor combined with something like the 820m in a 10 to 11 inch chassis.

 

[Junit151]

Wait, 300$?

If this is true, then Intel lost my hope on the Core M platform. Then you would rather want a Core i7 than a processor only as fast as the Pentium.

Now show me a snappy fanless i7 tablet. Oh noes, there isn't any?

*cough* Surface pro 3 *cough*

 

[jimmysmitty]

Wait, 300$?

If this is true, then Intel lost my hope on the Core M platform. Then you would rather want a Core i7 than a processor only as fast as the Pentium.
Now show me a snappy fanless i7 tablet. Oh noes, there isn't any?

*cough* Surface pro 3 *cough*

http://winsupersite.com/mobile-devices/surface-pro-3-fan-and-heat

The Surface Pro 3 has a fan. It can run in fanless mode in very low power situations but when being fully utilized, it requires a fan to keep it from overheating. After all it is a 15w TDP part.

 

[Free2play_noobs]

Okay ,Now where is your comparison with MULLINS & BEMA .

 

[sonofliberty08]

the quad core ARM chips was fast enough and serve me well for my mobile needs, i don't think i need any so call fast x86 mobile chips on the go at the moment, i only rely on my x64 desktop/workstation for the performance needs.

ARM chips do not run standard x86 applications and the Core-M is intended for premium convertibles/ultrabooks which will usually be purchased by people who do not want to be ball-and-chained to their desktop or do not want to bother with owning a desktop at all.

if u really need to run traditional x86 APPs on the go, the cheap E/A series APU or i3 ultrathin laptop was the better option, x86 for tablets and smartphones are useless, ARM APPs work better for smartphones and tablets, anyone who get the x86 tablets/smartphones are basically computer noobs who only run facebook and youtube most of the times, the pro will always want to hv an desktop/workstation.

 

[sonofliberty08]

The problem is mullins out performed bay trail pretty badly yet gained zero traction because intel is giving away baytrail for free. Why should AMD or Qualcomm get off their backside and make a superior product if everytime they do it, intel just gives away the inferior product for free? Its the same as what happened 2003-2007, when intel gave away p4s for free to prevent athlon64 of athlon64 fx cpus from gaining market share.

Then of course instead of making money from Mullins no one uses it because baytrail is cheaper (as in free), and then intel comes out with what looks like x3 the performance of baytrail in this core m chip... and the opportunity vanishes for AMD in less then 4 months. sorta hard to play on an even field when chipzilla can break the rules at will to tilt the field in their favor whenever they need to, heck, chipzilla spends something stupid like x2 in R&D then AMD gross' in a year (note, gross, not net).

The numbers i see in this preview should freak AMD out. They no longer are remotely competitive with intel if this is the case; the only sector of the chip business that they made any money in the past 2 years has been in mobile. this part makes their whole mobile lineup irrelevant.

to against this unfair situation, the best way was to boycott all the manufacturer who take intel bribe and go full intel only, like boycott the asus zenfone and fonepad line, to warn them for not to be the puppets of intel giant.

 

[Jaroslav Jandek]

Pretty amazing. Extra $249 (tray price) for triple performance in the same form factor is IMHO worth it!
See the comparison on Intel's Ark: http://ark.intel.com/compare/78416,83611,83612,83610

 

[deppman]

First, this is an impressive package. Great performance, and it beats the Tegra K1 at least in raw performance. But there are some points for caution:

1. Although the graphics scores look great, I wonder what the visual fidelity is on the Intel parts. So far, the K1 was had fantastic, full dp rendering just like on the desktop, and has not only outperformed the competition, but all while providing a higher quality image (see Anand). I wonder if the Intel had to make compromises to reach these performance numbers.

2. The cost is insane! It's about 6-7x the cost of a K1. One can buy a complete K1 tablet (miPad, Shield Tablet, the Nexus 9) for about the same or *less* cost than just the Core M board.

3. With Denver and Erista on the horizon, Core M perf and perf/W dominance might not be as long lived as Intel might hope, especially if the next gen parts can get to 16nm, and Denver turns out as impressive as NV hopes.

4. What's the GPGPU capability of this chip? The K1 can use the same CUDA code as used in supercomputers. What's Intel's answer - OpenCL? Is that well supported and competitive?

5. Perhaps the greatest sign of caution for me is the tablet performance demo video. Is the Intel solution so weak that they have to show it against an off-brand chinese A9 tablet? Any one with have an IT neron knows the 8-core aspect is simply a diversion, since the app almost certainly use at most 4 cores. My suspicion is that if the actually competed with post-2011 technology - like a miPad or Shield tablet (4 x A15r2p3 CPU, Kepler graphics with GPGPU acceleration for image processing) they might actually lose or see indistinguishable results. And then spectators might come to the conclusion that only fools would spend a ~$300 premium for little or no difference.

Again, an impressive package, and I might consider a Core-M for my next Linux laptop or desktop. But those are the areas that concern me.

 

[patrickjp93]

the quad core ARM chips was fast enough and serve me well for my mobile needs, i don't think i need any so call fast x86 mobile chips on the go at the moment, i only rely on my x64 desktop/workstation for the performance needs.

ARM chips do not run standard x86 applications and the Core-M is intended for premium convertibles/ultrabooks which will usually be purchased by people who do not want to be ball-and-chained to their desktop or do not want to bother with owning a desktop at all.

if u really need to run traditional x86 APPs on the go, the cheap E/A series APU or i3 ultrathin laptop was the better option, x86 for tablets and smartphones are useless, ARM APPs work better for smartphones and tablets, anyone who get the x86 tablets/smartphones are basically computer noobs who only run facebook and youtube most of the times, the pro will always want to hv an desktop/workstation.

Um, no.If Apple released a PROFESSIONAL Macbook pro with ECC memory and a mobile Quadro, I'd happily pay $6,000 for it rather than deal with a tower desktop. The downside would of course be only 2 USB ports (screw you apple) and the expense of thunderbolt for connecting to a drive cage or external GPUs for extra horsepower at the workplace or at home.

 

[patrickjp93]

First, this is an impressive package. Great performance, and it beats the Tegra K1 at least in raw performance. But there are some points for caution:

1. Although the graphics scores look great, I wonder what the visual fidelity is on the Intel parts. So far, the K1 was had fantastic, full dp rendering just like on the desktop, and has not only outperformed the competition, but all while providing a higher quality image (see Anand). I wonder if the Intel had to make compromises to reach these performance numbers.

2. The cost is insane! It's about 6-7x the cost of a K1. One can buy a complete K1 tablet (miPad, Shield Tablet, the Nexus 9) for about the same or *less* cost than just the Core M board.

3. With Denver and Erista on the horizon, Core M perf and perf/W dominance might not be as long lived as Intel might hope, especially if the next gen parts can get to 16nm, and Denver turns out as impressive as NV hopes.

4. What's the GPGPU capability of this chip? The K1 can use the same CUDA code as used in supercomputers. What's Intel's answer - OpenCL? Is that well supported and competitive?

5. Perhaps the greatest sign of caution for me is the tablet performance demo video. Is the Intel solution so weak that they have to show it against an off-brand chinese A9 tablet? Any one with have an IT neron knows the 8-core aspect is simply a diversion, since the app almost certainly use at most 4 cores. My suspicion is that if the actually competed with post-2011 technology - like a miPad or Shield tablet (4 x A15r2p3 CPU, Kepler graphics with GPGPU acceleration for image processing) they might actually lose or see indistinguishable results. And then spectators might come to the conclusion that only fools would spend a ~$300 premium for little or no difference.

Again, an impressive package, and I might consider a Core-M for my next Linux laptop or desktop. But those are the areas that concern me.

You have to remember Intel is building iGPUs for the compute power more than for actual graphics outputs. That's just a bonus. Intel sees ARM and HSA as a serious threat to its interests and market share of servers/supercomputers/mainframes. Why do you think the top Broadwell SKU promises a whopping 2 Teraflops of performance? Though you can use a K1 for gpgpu compute, it's only rated slightly above Iris Pro 5200 at I think 1.06 Teraflops vs 832 Gigaflops.

There are probably some fidelity compromises, and of course we have to wait for Skylake before Intel puts a tessellation engine on its integrated GPU (which will make it capable of more serious gaming far beyond Broadwell's Iris Pro 6200).

As for perf/watt, you have big x86 cores in there at only 4.5 watt TDP. Intel is not going to let ARM keep claiming its amazing perf/watt numbers, especially when the thermals now say x86 is king by a huge margin. Intel says Skylake is again mobile-oriented, so I expect 3-watt parts by the end of 2016, and I expect ARM's TDP to continue creeping upwards as it has the last few years. Without better branch prediction they're against a brick wall to get more performance. It's either boost clocks or boost IPC, which requires good branch prediction, and the best ARM offering is 34% accurate on the AMD Opterons bs. Intel's now (I believe) 79% accurate branch predictor. That circuitry is thermally expensive, but boosting clocks is electrically AND thermally expensive. ARM can either make individual instructions take a bit less time (not much room on that given how rudimentary the instruction set is in the first place) or they improve the pipelining and branch prediction.

I don't know what the gigaflop rating is on the K1, but if Broadwell's top SKU has the claimed 2.03 Teraflops (GPU only) at 1GHz on 96 EUs, I suspect Core M, with 24 EUs, has about 440Gigaflops peak given the 850MHz boost clock, plus the 200 Gigaflops from the CPU cores themselves and their numerous SIMD units (4 per core).

OpenCL is used a ton in the supercomputer world. Intel's Xeon Phis outnumber Nvidia's Teslas by a huge margin due to their perf/watt advantage and raw performance advantage.

I'm assuming the Intel tablet had a very good aluminum backplate and heat spreader, but I'd also bet it has the graphics power of mullins easily. I somewhat doubt it can match the K1, but most tablet apps are still built on x86, so Intel has that advantage too.