Freescale announced MPC8610 integrated host processor today. It's got PCIe x8 & x4, LCD controller, I2S/AC97, IrDA and so on. Let's check it out!
http://www.freescale.com/webapp/sps/sit ... de=MPC8610

Cool, SoC + AltiVec support!

EFIKA 3? :D

_________________
Czech PowerPC User Group - http://powerpc.lukysoft.cz
MorphOS Software database - http://morphos.lukysoft.cz/en

Maybe more like a 2nd chance for a shrinked PegasosIII.

Edit: A bit more L2 cache would have been nice, but anyway, if the price tag is okay this excuses a lower L2 cache.

* e600 core built on Power Architecture technology, with 256 KB backside L2 cache with ECC
* Up to 1333 MHz performance and 3060 MIPS
* AltiVec® 128-bit vector processing engine to accelerate image recognition and encoding/decoding
* DDR/DDR2 SDRAM Memory Controller with ECC (up to 533 MHz)
* Integrated Display Controller supports up to SXGA 1280 x 1024 resolution and 24 bits per pixel
* Two synchronous serial interface (SSI) controllers for I2S or AC97 audio inputs/outputs
* Two PCI Express® Interfaces, one with 1x/2x/4x/8x lanes for connecting graphics processors
* PCI 2.2 Interface at 32-bits and 66 MHz
* Two four-channel DMA controllers
* Enhanced Local Bus with 32-bit multiplexed address/data for ROM, NAND or NOR Flash
* Two Fast/Serial Infra-Red Interfaces (FIRI/SIRI)
* Serial Peripheral Interface (SPI)
* Two dual universal asynchronous receiver/transmitters (DUARTs)
* Two I2C Controllers
* Up to 32 general-purpose input/output (GPIO) ports


yummy...

Indeed, gimme gimme... But does the chip actually exist? Can a development board be ordered? How much does it cost? Remember, freescale is not Intel, they can not deliver on their promises as easily...

WoW

This thing sounds awesome!
It seems to be the perfect CPU to build a real nice PowerEFIKA / AMIGA / LAPTOP/ MacMini.

Is this CPU available already?

From the brochure ( http://www.freescale.com/files/32bit/do ... pdf?fpsp=1 ):
"
Availability
• Samples and evaluation boards now
• Production mid-2008
"

_________________
CzP
http://czanik.blogs.balabit.com/

..which usually means:

Availability
• Samples and evaluation boards next year
• Production mid-2009

The pre-announcement dates back 1+ year, though.

I actually wouldn't mind getting my hands on one of these.

With the inclusion of altivec, isn't this essentially a G4 SoC? Couple that with the ability to use graphics cards that outrank the ones available for my desktop >_<

This is a little more power than you usually see for embedded computing..

Freescale's has ideas for this one, as I'm sure everyone here does as well. :p

_________________
I <3 tiny computers.

Yes, but everyone else (including Freescale) does nothing with it.

James, pass this on to the others at Directron...

The overall electronics industry has a food chain that sets expectations on who funds development based on who benefits. If we are talking about basic enablement, then Freescale should be funding items such as Flash, Java, OpenGL, etc. and whatever else is needed for the markets they want to pursue. That being said, Freescale has been quite creative over the years "using other peoples money" or leveraging their strength in key markets. Automotive and Networking are two examples of where Freescale invests much effort in getting complete enablement for the markets, but actually spends very little compared to the output. Third party companies recognize the dominance of Freescale and fund much of the effort.

This has set an expectation at Freescale to fund very little of such development. For markets they don't dominate, they have invested large amounts money into the enablement to try and gain a foothold. Wireless, infrastructure DSP, and telematics are some examples. It is interesting to note that they have actually gained little share with all of those efforts. The reason these investments were made is that those markets were identified as high priority areas (we should keep that in mind as we go ahead).

Now, let's look at the media devices we have been discussing on our blog lately. It's a new area that today is dominated by just a few players like Apple and it is still not clear exactly what the timeline and a truly successful package will look like (beyond the iPod/BTW, the new iPod-video-Nano is a very good device). Recall the Newton was an Apple product too and that was some time ago (it can take time when you are a market leader). Although we are personally convinced think the market is ripe, Freescale still views the EFIKA and anything like it and the market it makes as a risk (whatever is extolled this week) and therefore Freescale executives will invest very little until the market forces become more clear. Also, remember that parts of Freescale will still view this as a PC like device ("a computer") and therefore want nothing to do with it post Apple.

So from the Freescale side, we can expect very little other than some support within whatever budget they have already. We may find some "enablement" funding as the overall and future Freescale plan comes together. That said, in terms of funding Freescale's development effort on end products, we just don't see that happening other than maybe for a few key software enablement activities. Freescale is a component supplier. Even though the industry has started to migrate up the food chain with semiconductor companies taking on more and more, Freescale is somewhat stuck in the past due to the reasons and history noted in the first paragraph above.

Now, let's look at the ODM's -- they are at the top of the food chain. They stand to gain the most from end unit sales and ironically it seems they invest the least (Dell is a good example!).

BTW, we see a tremendous opportunity to link the EFIKA Pad for the distribution of Olympics 2008 content. MorphOS makes for a perfect partner for this broad and narrowcast opportunity. We can now confirm that the 5121e can handle the video required and the target screens we have discussed. Imagine all the EFIKA Pads traveling back to their new homes after the Olympics. It is hard to imagine a better way to establish an international reputation.

LET'S FINALLY HAVE SOME FUN!

:D

R&B

P.S. And, yes, there will be a Pegasos III, but it will be somewhat later and in a manner no one expects...;-)

_________________
http://bbrv.blogspot.com

After reading this long, quality post from BBRV, I first thought about an equally long, quality answer (as if I was able!).
Then, an immediate idea came to mind:

freescale does not help.
What if Genesi swithes chip provider?

Life is soooooo much easier in Intel land...

(me ducks - again!)

Flames accepted.

More cut-throat, you mean. Intel is a dead-end alley we'd get smashed into the ground by 100 other companies for using.

Outside of Intel, name another chip supplier - IBM, AMCC, PA Semi? They're all operating on the same principle as Freescale. For instance, none of them would do something like sponsor the OLPC project.

Use ARM or MIPS or maybe something curious and new like Blackfin? Well, Marvell are a good supplier *AND* customer of ours already. Maybe we should move to PXA270's and Marvell wireless and take advantage of their newfound Linux-friendliness?

What do you suggest, flames aside?

_________________
Matt Sealey

snip!

So, think of the missing parts and find us a southbridge that includes them. This is the major problem using PowerPC right now is there is no interconnect we could rely on to connect - at least the 970MP had Hypertransport in the northbridge.

What do we have here, PCI Express? PCI? Nobody makes bridges with those interconnects anymore but.. ULi? That means convincing nVidia.. and they haven't been nice in the past.. then, think of the LCD controller blurb. Why would you have an LCD controller supporting high resolutions on the chip then specify the 8x PCI Express lane as being "for connecting graphics controllers"? Doesn't that make the LCD controller sound kinda useless?

As for "Efika 3", this is more like a Pegasos 3 idea - a large, performance desktop computer or a full laptop. I wouldn't slap the Efika brand on it, personally :)

_________________
Matt Sealey

Marvell as a customer? What have you built for them? Perhaps some kind of service? I'm curious.
That sounds consistent. Yet the problem persists: You can build hardware when you have a signed contract from a buyer, for ten thousand units. Or you build such a consumer device, that it's given for granted that you will sell it in the thousands. An I sounding pedantic, or like a nine year old? 8^)
The Marvell move you mention makes sense. ARM processors are nice, and getting more powerful each day. This is possible because they already have a market (mobile gizmoz, nonetheless). Market for PowerPC is... let's say... dubious?
Well, stop being ridiculous myself. It's clear that I was suggesting you things that you had knowledge of, since ages.
I have one last suggestion, but it is in fact a change of business model. Something too big to discuss lightly, without being "in the know".

Now, going back to that new MPC8610 chip:
Ouch! It's true! It has LCD controller, but no ethernet or USB... A plain MPC5200 looks more useful by itself. Also, like you say, it might be impossible to find a southbridge to add those functions. Imagine a new Pegasos III without USB and ethernet (and a weak LCD output)... Sure, with a lot of PCI expansion, to add cards (for which drivers don't exist).
Reminds me of Commodore launching the Amiga 2000: Even though it's model number was twice as big as the original 1000, it had the same functions and speed as the 500 model (and costed four times more). Oh, yeah, it had plenty of empty slots. Everybody knows which was the most important Amiga model.

Hey, Genesi, why don't you make the new Amiga 500? Is there a market for it now?