Intel Broadwell Thread

[TreVader]

Yeah, Bay Trail promised us this ultra-ultrabook market, but didn't quite come through. Core M seems to have achieved this though, pending real-world benchmarks, etc.

In the absence of real 3rd party benches, I'm trying to translate those slides:

---

The Intel Core M processor enables up to 50 percent faster compute performance and 40 percent faster graphics performance versus the comparable, previous 4th generation Intel Core processor.3 Consumers with older PCs will notice a more significant performance improvement. The Intel Core M processor delivers up to two times the compute performance and up to seven times better graphics compared to a 4- year-old PC, for example.2

---

This 4 year-old PC has an Intel Core i5-420UM, and the test is SYSmark 2014, which I guess makes sense since SYSmark is geared toward business applications.

---

2Intel® Core™ M-5Y70 Processor (up to 2.60GHz, 4T/2C, 4M Cache) vs. Normalized to a 4-year-old PC with Intel® Core™ i5-520UM. Performance based on SYSmark* 2014. Intel® Core™ M 5Y70 compared to Intel® Core™ i5-520UM. Weight based on Intel® Core™ M processor-based 2 in 1 based on Intel® FFRD Llama Mountain. Old PC is OEM laptop with Intel® Core™ i5-520UM and 62WHr battery, 3 lbs weight, 1.1-inch thick.

3Up to 50 percent faster vs. 4th generation Intel Core processors based on: Specfp_rate_base 2006 comparing Intel® Core™ M-5Y70 Processor compared to previous-generation Intel® Core™ i5-4302Y at 4.5W. Up to 40 percent faster graphics vs. 4th generation Intel Core processors based on: 3D Mark Ice Storm comparing Intel® Core™ M-5Y70 Processor with Intel HD graphics 5300 vs. Previous Generation Intel® Core™ i5-4302Y at 4.5W with HD Graphics 4200.

---

Also, to repeat the obvious, 4.5W is the SDP value for Core i5-4302Y. TDP is much higher at 11.5 W.


So, let's look at Intel Core i5-520UM:

http://ark.intel.com/products/47554/Intel-Core-i5-520UM-Processor-3M-Cache-1_06-GHz

1.066 GHz with turbo up to 1.866 GHz and 3 MB cache. Dual-core 4-thread. DDR3-800 dual-channel.

It gets 1336 in PassMark (or 615 single thread):

http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i5+520UM+@+1.07GHz


My 5-year old 2.26 GHz Core 2 Duo P7550 13" MacBook Pro gets 1468 (849 single) in PassMark:

http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core2+Duo+P7550+@+2.26GHz


In comparison, the 4.3 W TDP 1.46 GHz N2805 Bay Trail CPU has a PassMark score of 497 (307 single thread). Ouch!

http://www.cpubenchmark.net/cpu.php?cpu=Intel+Celeron+N2805+@+1.46GHz


The 2.41 GHz J2850 Bay Trail CPU has a PassMark score of 1831 but it has a TDP of 10 W.

http://www.cpubenchmark.net/cpu.php?cpu=Intel+Pentium+J2850+@+2.41GHz

---

If Core M 5Y70 gets twice as much as Core i5-520UM in PassMark, that'd put it in the 2670 range, which would correspond roughly to a 1.7 GHz Core i5-2557M.

http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i5-2557M+@+1.70GHz&id=814

That's a mid-range mobile CPU from 2011, with a max turbo of 2.7 GHz and a 17 W TDP.

In fact, that's the very CPU in the mid-2011 13" MacBook Air.

http://www.everymac.com/systems/app...acbook-air-core-i5-1.7-13-mid-2011-specs.html

---

I know that's a lot of internet handwaving, and PassMark isn't SYSmark, but if even close to being in the right ballpark then that's not too bad, esp. if has vastly improved GPU performance.

I would gladly sacrifice some modern mid-range fan-endowed laptop CPU oomph to get an 1.1 kg 11.9" Retina MacBook with super long battery life.

With my light business-type usage, even my Core 2 Duo 2.26 P7550 is OK, except for the fact that it doesn't have Retina, among other things. If a 1.1 GHz Core M gave me an 80% boost in real-world CPU speed, that's a big jump for me. Even if it is "only" a 50% boost in real-world CPU-speed, but gives me Retina and USB 3, I'll still gladly take the plunge, provided the rest of the machine is to my liking.

Will there be a specific Apple-only SKU released tomorrow or will the top Core M be the 5Y10? I hope so. 1.1-2.6 seems like it's being held back core clocks much higher than that. I wonder how much available thermal headroom is lost when you take the fan out?

 

[CHADBOGA]

Do we have any benchmarks comparing the A7 to Intel's offerings on something other than GeekBench?

 

[Eug]

Will there be a specific Apple-only SKU released tomorrow or will the top Core M be the 5Y10? I hope so. 1.1-2.6 seems like it's being held back core clocks much higher than that. I wonder how much available thermal headroom is lost when you take the fan out?

I see from your sig that you have the 1.7 GHz i5-2557M. My hocus pocus estimate puts the Broadwell Core M 5Y70 at around its CPU performance. Would you be willing to take that level of CPU performance in a 2014 machine if it meant getting an even lighter machine but that has a Retina screen?

If not, and if you think consumers in general wouldn't be happy with it either, then that could mean that unless there is some sort of custom Broadwell Y SKU for Apple, we'll be waiting until 2015 for a Broadwell U based machine.

 

[Qwertilot]

Couldn't they just fit a fan to it and have it run at turbo very much more often or something? A bit like they did with that iMac using a mac book air processor.

The pricing probably doesn't make logical sense a priori - it certainly didn't with the iMac! - but they're big enough to do that sort of thing

 

[Eug]

Couldn't they just fit a fan to it and have it run at turbo very much more often or something? A bit like they did with that iMac using a mac book air processor.

The pricing probably doesn't make logical sense a priori - it certainly didn't with the iMac! - but they're big enough to do that sort of thing

It makes sense, but kind of defeats the purpose.

For the iMac, the desktop chips were hotter than desirable for silence IIRC, but I don't remember all the specs. Since then, desktop chips have dropped in TDP.

Do we have any benchmarks comparing the A7 to Intel's offerings on something other than GeekBench?

Speaking of Geekbench, dunno if this is faked or not, but one guy has posted benches of the iPhone 6, which gets 2920 on a dual-core 1.4 GHz A8 (with 1 GB RAM).

http://9to5mac.com/2014/09/09/iphon...ch-once-again-suggests-1-gb-ram-dual-core-a8/

 

[Idontcare]

I wonder how much available thermal headroom is lost when you take the fan out?

A considerable amount of headroom would be lost. For a low heat-capacity and low thermal-capacity substance the likes of air, the thermal conductivity of conduction is ridiculously less than thermal conduction of convection.

 

[liahos1]

It makes sense, but kind of defeats the purpose.

For the iMac, the desktop chips were hotter than desirable for silence IIRC, but I don't remember all the specs. Since then, desktop chips have dropped in TDP.


Speaking of Geekbench, dunno if this is faked or not, but one guy has posted benches of the iPhone 6, which gets 2920 on a dual-core 1.4 GHz A8 (with 1 GB RAM).

http://9to5mac.com/2014/09/09/iphon...ch-once-again-suggests-1-gb-ram-dual-core-a8/

marginal increase compared to a7. i wonder if this means nvidia is going to be the fastest arm soc player out there.

 

[NTMBK]

A considerable amount of headroom would be lost. For a low heat-capacity and low thermal-capacity substance the likes of air, the thermal conductivity of conduction is ridiculously less than thermal conduction of convection.

They could do something like what NVidia did- stick a honking big metal heat spreader throughout the entire body of the tablet, and basically use the whole thing as a radiator.

 

[Sweepr]

marginal increase compared to a7. i wonder if this means nvidia is going to be the fastest arm soc player out there.

Reality check for those expeting 2x performance every year. If Broadwell-Y is considerably faster than this chip @ 4.5W TDP then kudos to Intel. Can't wait to see a 10'' premium tablet/convertible based on it.

 

[liahos1]

Reality check for those expeting 2x performance every year. If Broadwell-Y is considerably faster than this chip @ 4.5W TDP then kudos to Intel. Can't wait to see a 10'' premium tablet/convertible based on it.

by the time this is out consoleOS should be on its way (Dec 2014). so should be able to get access to android. i wonder if there is a market for premium android tablets with core m in it. LG recently put out a core android based product, the LG TAB book, so shouldnt be out of the question.

 

[Haserath]

Will there be a specific Apple-only SKU released tomorrow or will the top Core M be the 5Y10? I hope so. 1.1-2.6 seems like it's being held back core clocks much higher than that. I wonder how much available thermal headroom is lost when you take the fan out?

To get an idea of conduction vs convection:
http://www.tomshardware.com/reviews/geforce-gtx-750-ti-passive-cooling,3757.html

The card also throttles at 80C.

Core M wouldn't be too bad in the air if it was able to hold its boost. I don't think it has GT3 though while the current Airs do. They might be able to get away with it if the extra thermal headroom and Broadwell tweaks give it enough of a boost.

 

[witeken]

Here are the presentation slides for the Technology Insight: Intel's Next Generation 14nm Microarchitecture for Client and Server: https://intel.activeevents.com/sf14/connect/fileDownload/session/F7EBB2F85164E31DF41CDD863B86286F/SF14_SPCS002_101f.pdf

Enjoy parsing through this.

Awesome, slide 22, anyone? DCC ends at 0.5GHz, according to that slide. So my guess that DCC would also be useful to reduce power consumption in gaming turned out to be correct (there was a whole discussion about DCC and gaming, for those who don't remember).

Reduce power by duty cycling: 80% at 500MHz instead of 100% at 400MHz. Same frequency with
less leakage.

I also like their quote from AnandTech in that slide . ( “both a bit brute force and a bit genius&#8221

 

[TreVader]

To get an idea of conduction vs convection:

http://www.tomshardware.com/reviews/geforce-gtx-750-ti-passive-cooling,3757.html



The card also throttles at 80C.



Core M wouldn't be too bad in the air if it was able to hold its boost. I don't think it has GT3 though while the current Airs do. They might be able to get away with it if the extra thermal headroom and Broadwell tweaks give it enough of a boost.

So why is apple bothering to handicap this MacBook so bad when a tiny, silent fan is a much more efficient setup from an engineering perspective?

Bragging rights?

 

[witeken]

There is also more information about Broadwell Gen8 GPU architecture in the presentation: https://intel.activeevents.com/sf14...164E31DF41CDD863B86286F/SF14_SPCS002_101f.pdf. Broadwell also support HSA.

There are also lots of other details that would be worth for Anand... never mind. Ryan Smith should do a follow-up.

 

[Arachnotronic]

There is also more information about Broadwell Gen8 GPU architecture in the presentation: https://intel.activeevents.com/sf14...164E31DF41CDD863B86286F/SF14_SPCS002_101f.pdf. Broadwell also support HSA.

There are also lots of other details that would be worth for Anand... never mind. Ryan Smith should do a follow-up.

How much do you want to bet that Anand ends up explaining the A8 at today's Apple keynote?

 

[witeken]

How much do you want to bet that Anand ends up explaining the A8 at today's Apple keynote?

He probably joined Apple because they don't give a lot of details which he now has access to. He should take the data and post in on AT.

 

[Khato]

There is also more information about Broadwell Gen8 GPU architecture in the presentation: https://intel.activeevents.com/sf14...164E31DF41CDD863B86286F/SF14_SPCS002_101f.pdf. Broadwell also support HSA.

And in an amusing change of pace, Intel's software support for such is actually ahead of the game with OpenCL 2.0.

 

[TreVader]

I see from your sig that you have the 1.7 GHz i5-2557M. My hocus pocus estimate puts the Broadwell Core M 5Y70 at around its CPU performance. Would you be willing to take that level of CPU performance in a 2014 machine if it meant getting an even lighter machine but that has a Retina screen?

.

I would, but it wouldn't be my only machine.


The 2557M isn't the bottleneck on that machine, it's the 4GB of ram.

 

[IntelUser2000]

The 2557M isn't the bottleneck on that machine, it's the 4GB of ram.

BTW, Passmark is terrible.

I am expecting 5Y70 to be about on par with Core i5 15W.

Speaking of Geekbench, dunno if this is faked or not, but one guy has posted benches of the iPhone 6, which gets 2920 on a dual-core 1.4 GHz A8 (with 1 GB RAM).

http://9to5mac.com/2014/09/09/iphone...-dual-core-a8/

I think Apple themselves just said GeekBench sucks. 16% increase in Geekbench, 25% in real-world.

 

[mikk]

There is also more information about Broadwell Gen8 GPU architecture in the presentation: https://intel.activeevents.com/sf14...164E31DF41CDD863B86286F/SF14_SPCS002_101f.pdf. Broadwell also support HSA.

GPU accelerated HEVC encoding support is nice.

Significant improvements in performance for 2x/4x/8x MSAA sounds nice as well. They improved almost everything by the looks of it.

 

[liahos1]

BTW, Passmark is terrible.

I am expecting 5Y70 to be about on par with Core i5 15W.



I think Apple themselves just said GeekBench sucks. 16% increase in Geekbench, 25% in real-world.

likely select benchmarking

 

[Eug]

BTW, Passmark is terrible.

I am expecting 5Y70 to be about on par with Core i5 15W.

Which one?

 

[liahos1]

http://www.pcper.com/reviews/Processors/IDF-2014-Intel-Shows-Core-M-5Y70-Performance-Numbers

Its glorious

 

[Eug]

http://www.pcper.com/reviews/Processors/IDF-2014-Intel-Shows-Core-M-5Y70-Performance-Numbers

Its glorious

15 Watts?!?

 

[liahos1]

Probe a misprint. This was tested on llama mountain which is 7.2-7.3mm and fabless