Intel's Medfield & Atom Z2460 Arrive for Smartphones: It's Finally Here
by Anand Lal Shimpi on January 10, 2012 8:00 PM ESTThe GPU
The PowerVR SGX 540 in Medfield is no different from what you'd get in an OMAP 4460, with the exception that it's clocked a bit higher at 400MHz.
The SGX 540 here is a remnant of Intel's earlier strategy to have Medfield out far sooner than it actually is going to show up on the market. Thankfully Intel has plans to introduce a PowerVR SGX 543MP2 based Medfield successor also before the end of the year.
Video Decode/Encode Support, Silicon Hive ISP
Intel relies on two more IP blocks from Imagination Technologies: the VDX385 and VDE285 for 1080p video decode and encode. Intel claims support for hardware accelerated 1080p30 decode, High Profile. Maximum supported bitrate is apparently up to 50Mbps, although Intel only demonstrated a 20Mbps High Profile stream:
Intel also claims support for 1080p30 video encode.
Medfield's ISP is provided by Intel owned Silicon Hive. The ISP supports cameras ranging from 5MP to 16MP (primary sensor), with the reference design standardizing on an 8MP sensor. Medfield supports burst capture at up to 15 fps (8MP).
The Process
Intel bifurcated its process technology a few years ago, offering both low power and high performance versions of each of its process nodes. Today those process nodes are staggered (45nm LP after high perf 32nm, 32nm LP debuts after high performance 22nm, etc...) however Intel plans on bringing both in lockstep.
Medfield debuts on Intel's 32nm LP process. The only details we have from Intel are that leakage is 10x lower than the lowest on 45nm. Compared to Moorestown, Medfield boasts 43% lower dynamic power or 37% higher frequency at the same power level.
The bigger and more valid comparison is to TSMC's 28nm process, which is what companies like Qualcomm will be using for their next-generation SoCs. It's unclear (and very difficult) to compare different architectures on different processes, but it's likely that Intel's 32nm LP process is more comparable to TSMC's 28nm LP process than it would be to any 4x-nm node.
It is important to note that Intel seems very willing to sacrifice transistor density in order to achieve lower power consumption where possible. I don't believe Intel will have the absolute smallest die sizes in the market, but I also don't believe it's clear what the sweet spot is for mobile SoCs at this point. It's quite likely that Apple's ~120mm^2 target is likely where everyone will eventually end up in the near term.
The Roadmap
Although Medfield is already posting competitive performance numbers, its current competition is roughly a year old. Within the next two quarters we'll see smartphones and tablets shipping based on Qualcomm's Krait. The next-generation Snapdragon platform should be Cortex A15-like in its performance level
Today we have Medfield, a single core Atom paired with a PowerVR SGX 540 built on Intel's 32nm LP process. Before the end of the year we'll see a dual-core Atom based Medfield with some form of a GPU upgrade. I wouldn't be too surprised to see something like a PowerVR SGX 543MP2 at that point either. In tandem Intel will eventually release an entry level SoC designed to go after the more value market. Finally we'll see an Intel Atom based SoC with integrated Intel baseband from its Infineon acquisition - my guess is that'll happen sometime in 2013.
Facebook
Twitter
RSS
164 Comments
View All Comments
extide - Wednesday, January 11, 2012 - link
Why does clockspeed matter? People should stop focusing on the clock so much. The real performance metric is performance per watt. Anyone remember the P4? How about Bulldozer? If intel can get more clockspeed in the same thermal envelope, then good job and they should be able to compare them side by side. I know ARM vendors would clock their chips faster, but then they run into thermal limitations.Wilco1 - Wednesday, January 11, 2012 - link
For single-threaded benchmarks clockspeed is the only thing that matters. I agree performance per watt is far more important in the real world. This is why dual or quad cores give better performance per watt than a single high clocked core. I don't believe ARM cores are thermally limited, Tegra 3 has 4 cores at 1.3GHz, and even faster SoCs are coming soon.milli - Thursday, January 12, 2012 - link
"For single-threaded benchmarks clockspeed is the only thing that matters"Ever heard of issue-width or instruction re-ordering? Ever heard of MIPS/Mhz? If you have, how can you say such a thing?
Wilco1 - Thursday, January 12, 2012 - link
IPC matters of course but only at similar frequencies. And frequency differences are typically much larger than IPC differences. For example 2-way out-of-order execution gives around 25% better IPC than 2-way in-order, however the frequency difference in the article is 33-60%. So given a large enough difference in frequency, you would expect an in-order to beat out-of-order.milli - Thursday, January 12, 2012 - link
You can't just paste performance numbers on a cpu based on it's high level architecture. Your example might be right for cpu A & B but you can't apply it just to every cpu.Next you'll tell me that a Cortex A15 is as fast clock for clock as a Phenom just because they are both 3-wide OoO architectures? Rest assured that a K10.5 core will be more than double as fast as an A15 (and i'm sure, up to 5-6x faster).
french toast - Thursday, January 12, 2012 - link
I can tell you now that cortex a15 wont be a million miles off clock for clock,even if it doesn't beat it.Obviously cache sizes/latency as well as bandwidth will play a part, but cortex a15 will be competitive with phenom, on a tiny fraction of the die space and power consumption.
cortex-a9 is nearly on par with a ULV core 2 duo clock for clock as difficult as that seems.
milli - Thursday, January 12, 2012 - link
Oh french toast, I've seen your comments here before. You just crack me up. Such a fanboi. I didn't even know there was such a thing as an ARM fanboi but you prove me wrong.FYI, an ULV C2D is around 3 to 10x faster than an A9 (clock for clock) and an A15 will get nowhere near a Phenom. Sorry to burst your bubble.
kaiyao - Wednesday, January 11, 2012 - link
While this chip for phones is finally out, does anyone know if Intel going to release any tablet chips anytime soon? Perhaps a dual/quad core version of this chip?Because Intel should really push out a chip competitive with ARM when Windows 8 comes out. I imagine if the chip performs as well as an ARM (in terms of performance and power efficiency), and if Intel matches the pricing of ARM chips, Windows 8 tablet manufacturers would definitely choose x86 over ARM to advertise compatibility with legacy applications.
I remember that the previous "mobile chip" from Intel did not work with Windows 7 due to something along the lines of the lack of PCI bus support, but since Microsoft can port Windows 8 to ARM, clearly this PCI bus is not an issue (if Microsoft modifies Win8 a bit). I presume application code will not be affected by the presence of the PCI bus.
guilmon19 - Wednesday, January 11, 2012 - link
I read somewhere, sorry i don't have a link, that intel was going to release a dual core version by the final quarterMumrik - Wednesday, January 11, 2012 - link
Page 4: "and I wouldn't be surprised if more aren't on the way."Isn't that the opposite of what you meant Anand?