Thought that would turn up here.

...and there's this: Freescale-adopts-ARM-cores-in-QorIQ-line

Inevitable...

> Inevitable...

I like what's also written in the article:

"The architecture will sport ARM, Power and accelerator cores [...]. [...] Freescale aims to differentiate itself by offering a menu of both general purpose ARM and Power cores along with a new set of packet-processing accelerators [...]. [...] The company must create a software environment that straddles both the big endian Power and little endian ARM architectures. [...] ARM will not replace Power cores for Freescale, at least in the foreseeable future. The company said it has long-term plans to continue designing and supporting Power cores. [...] Power-based chips will provide higher-end performance than the ARM cores are likely to reach in the near term [...]. “You will see the Power architecture for a long time--it will co-exist for a decade-plus,” he said. The company has made no change in the resources it puts behind Power cores, and “we are still going to refine existing Power cores [because] we see Power has a strong base in this sector,” he added. [...] In the short term, Freescale has a deep and broad investment in Power."

And from the Freescale press release:

"The Layerscape architecture is the foundation of a broad array of forthcoming QorIQ multicore processors, [...] leveraging Power Architecture and ARM technologies as appropriate. [...] Future QorIQ processors based on the Layerscape architecture will leverage Power Architecture technology. Freescale continues to enhance and invest in its industry-leading Power Architecture e6500 core, which will be among the highest-performance cores featured in the new portfolio."
http://media.freescale.com/phoenix.zhtml?c=196520&p=irol-newsArticle&ID=1706486

And from the FTF interview:

"Power Architecture continues to be a very significant aspect of our IP, of our offering [...]."
http://www.youtube.com/watch?v=HD8FYGIZ2Nw

So regarding the QorIQ product line, Power Architecture stays in and will be further developed, and ARM will be added. Nice :-)

I bet this is something that some people thought would never happen; ARM breaking into PPC stronghold. Personally, I was sure it would happen sooner or later, but didn't think it would happen this soon, my guess was that it would have to take some ARMv8 CPU's for them to do this, but it seems they are going for the *low-end* now already (undoubtly to be followed by high end with the new generations of ARM CPU's to come). PPC is dead for the desktop market since 2007, and now a process has started to corner it in its new domains. ARM won't reach the high-end segments in these markets until the next generation of CPU's of course, but I am convinced this marks the beginning of a new way of reasoning over at Freescale, the start of a process towards a new era. Freescale isn't immune to costs, and why keep developing two architectures, costing twice as much, if you could manage all your needs with just one?

[ Edited by takemehomegrandma 20.06.2012 - 12:03 ]

Quote:

They cant drop Power because customers depend on it. They cant switch to another platform in short term.

> now a process has started to corner it in its new domains.

The networking/communications domain is far from a "new domain" for PPC. It's been going strong in that domain since the mid-1990s:

http://en.wikipedia.org/wiki/PowerQUICC#PowerQUICC_I (1995)
http://en.wikipedia.org/wiki/PowerPC_400#PowerPC_403 (1994)

The PPC had its debut in the desktop computing domain in March 1994, so it started at about the same time in both domains.

> Freescale isn't immune to costs, and why keep developing two
> architectures, costing twice as much, if you could manage all your
> needs with just one?

Because Freescale's need is to get money from its customers in exchange for products, and (some?, most? of) its networking/communications customers' current and foreseeable-future need is being offered (existing and new/better/faster) Power Architecture products for several technical reasons (see also what itix said).

Nobody said it would happen overnight, but it has now begun...

Quote:

Cortex-A15 is due this year...

PPC did have its own high-end embedded niche but A8 and A9 have now got a big chunk of that level of performance. A15 looks like it'll finish the job.
Freescale in particular pretty much owned automotive and had a big chunk of networking but a lot of the ARM guys are chasing these now.

Quote:

It's nothing new. They did have several different ISAs (ARM, Coldfire, 68K, PPC, 8-bit) until fairly recently but they have been replacing their designs with ARM based designs for some time now.

It's got to the point where there's really no advantage in designing your own CPU any more unless it's something really exotic or special (like console CPUs). It's far easier and cheaper just to license a design. That way you get a chip that you can still tweak but you don't have to worry about developing a compiler or apps or whatever.

Quote:

Yes, I meant the various upcoming ARMv8 incarnations for "high-end"...

Quote:

It's clearly the trend!

Quote:

Yes, huge advantages on the SW side, which of course is essential. But even if you license an ARM design (and other on-chip controllers) you can very much design your own CPU (as in "chip") around those, to target different markets and applications. IMHO there are only benefits in this, and that's why it will happen. PPC is now *deader* than dead, but it might just not know it yet...

Addendum:

>> what I would like to know is whether IBM's stock A2 core (as used in
>> the PowerEN and Power BQC chips) has it. I ask because I've
>> read conflicting third party statements on that matter (and no statement
>> by IBM itself).

> http://www.theregister.co.uk/2011/08/22/ibm_bluegene_q_chip/
> As that detailed article doesn't mention AltiVec/VMX for the A2 core
> I assume that it's not there.

Found an IBM source:

"The A2 processor does not support vector single instruction, multiple data (VMX) instructions."
http://www.redbooks.ibm.com/redbooks/pdfs/sg247948.pdf (page 77 ("61"))

> Hard to believe they have the resources to create this.

According to the Applied Micro CEO, not only have they created their X-Gene core from the ARMv8 ISA spec but they also helped ARM Ltd. to outline this spec to begin with:

"When we did the deal with ARM, we licensed the micro architecture. We did not get ARM’s cores... we got 300 pieces of paper, out of which half were blank, which we helped fill up and thus got a discount (laughs)."
http://www.businessworld.in/businessworld/businessworld/content/Open-Source-Ultimate-Leveller.html

Update:

>> You were probably not expecting this :-D

> Absolutely. This comes most unexpected. I'd have never thought someone
> would supersede nVidia regarding ARM64/ARMv8

Cavium trailing behind Applied Micro:

http://www.cavium.com/newsevents_Cavium_Licenses_ARM_Architecture.html
http://www.cavium.com/newsevents_Cavium_Unveils_Project_Thunder.html
http://www.cavium.com/processor_Project_Thunder.html
http://www.arm.com/about/newsroom/arm-and-cavium-extend-relationship-with-armv8-architecture-license.php
http://blogs.arm.com/smart-connected-devices/773-arm-in-servers-cavium-welcome-to-the-party/

...and nVidia seems to be slacking.

Andreas_Wolf,
Quote:

Don't think Cavium are quite in the same market but but it looks like they're headed in the same direction.
BTW I read an interview with Applied Micro a while back where they said they expect to start sampling soon and see shipping products by the end of next year. 4 - 32 cores @ 2.5GHz.

Quote:

They might just not have anything to talk about yet.

BTW I was digging around eBay and found the cheapo tablets are even cheaper and there's now ARM laptops appearing. Tablets are £30 and Laptops (i.e. netbooks) £50. A 10" 1GHz A9 laptop is £80 ...with free postage.

You can now buy a new laptop for less than a second hand Mac mini. That's just mind boggling.

We are now into the era of seriously cheap computing.

> Don't think Cavium are quite in the same market

True. There's a nice blog article by HP summarizing the different target applications for Applied Micro's X-Gene and Cavium's Project Thunder:

http://h30507.www3.hp.com/t5/Hyperscale-Computing-Blog/Pathfinders-Making-it-Matter-Welcome-to-Applied-Micro-and-Cavium/ba-p/119167

More on Applied Micro's X-Gene:

Novel-server-CPUs-glow-at-Hot-Chips?pageNumber=1

First chip has 8 cores @ 2.5GHz, second chip gets 16 @ 3GHz.

> I've heard the 3 core IBM Wii2 a few times now. Sounds perfectly plausible.

More info from June:

"Three cores (fully coherent)."
http://www.vgleaks.com/world-premiere-wii-u-specs/

And some more recent info:

"The Wii U's IBM-made CPU is made up of three Power PC cores. We've been unable to ascertain the clock speed of the CPU (more on this later), but we know out of order execution is supported. [...] The mention of the Wii U's CPU raised our eyebrow - as it has done for many developers we've spoken to. While its clock speed remains private, most developers agree it is lower than the clock speed offered by the PS3 and Xbox 360's CPUs - disappointing for many."
http://www.eurogamer.net/articles/2012-08-30-how-powerful-is-the-wii-u-really

Update:

>> "the CPU is very likely to be a PowerPC of one sort or other too.
>> Then again, [...] there are strong and credible rumors of it being an
>> AMD x86 core ala Trinity. In any case more SemiAccurate moles are
>> still saying PowerPC, that architecture makes much more sense here."
>> http://semiaccurate.com/2011/12/05/exclusive-xbox-next-chip-just-taped-out/

> "Sources close to Xbox World magazine have revealed that Durango’s
> devkit features a “monster” 16-core IBM PowerPC CPU"
> http://sillegamer.com/2012/04/06/xbox-720-devkit-specs-detailed-includes-16-core-processor/

"SemiAccurate has been saying for a while that all signs were pointing toward a PowerPC, specifically an IBM Power-EN variant. The dark horse was an x86 CPU, but it was a long shot. It looks like the long shot came through, moles are now openly talking about AMD x86 CPU cores"
http://semiaccurate.com/2012/09/04/microsoft-xbox-next-delay-rumors-abound/

If the rumour of Oban being an x86 chip holds true, then I think the Durango devkit doesn't have PPC either (just wouldn't make sense).

> Cortex-A15 is due this year...

I really didn't expect it so soon:

http://en.wikipedia.org/wiki/Apple_A6

Apple being first is especially funny as two months ago Samsung claimed their Exynos 5 chip
to be "the first chip in the market to integrate Cortex-A15 dual core" (page 4).

Apple A6 having Cortex-A15 cores has not been officially confirmed yet, but there are strong pro arguments:
http://munsie.tumblr.com/post/31483017459/iphone-5-processor

Correction:

> Apple A6 having Cortex-A15 cores has not been officially confirmed yet

As it has been revealed now, it's not Cortex-A15. It's not Cortex at all but rather Apple's very own implementation of the ARMv7-A ISA. With this, Apple joins the ranks of Marvell (Sheeva PJ4/PJ4B) and Qualcomm (Scorpion, Krait) in developing their own ARMv7 cores rather than just licensing ready-made Cortex-A cores from ARM Ltd.

http://www.anandtech.com/show/6292/iphone-5-a6-not-a15-custom-core

I'm anxious to know how Apple's A6 core compares to Cortex-A15 and Krait performance-wise. And I guess that Samsung still can manage to have "the first chip in the market to integrate Cortex-A15 dual core" with their Exynos 5 :-)

Andreas_Wolf,
Quote:

When they bought they bought PA-Semi I suspected they'd do their own CPU.
I've been wondering if it was something of their own in the A6.

> When they bought they bought PA-Semi I suspected they'd do their own CPU.

Let's not forget Intrinsity here :-)

Andreas_Wolf,
Quote:

Think that they will use any of Intrinsity's gating technology?

> Think that they will use any of Intrinsity's gating technology?

It's well known that the Cortex-A8 core in Apple's A4 chip uses Fast14, and in the linked posting minator said:

"Before they were bought, there were rumours of Intrinsity working on a 2Ghz Cortex-A9."

From this I presume that also the Cortex-A9 core in Apple's A5 and A5X chips uses the technology. Thus I find it highly plausible that Fast14 technology has been applied to the custom core in Apple's A6 chip as well.

That could lead to a really frightening prcessor.