Moving An Embedded Project From x86 to PowerPC

An anonymous reader writes "This series on embedded development shows you how to migrate a project prototype from x86 to PowerPC. This initial installment explains the realities and rationale of the project: it introduces the robotic submarines that were the start of the project (and where they came from), and describes the Linux/GCC development environment and the bare-bones Kuro Box PowerPC development board."

Re:Other way around

[RAMMS+EIN]

There are definitely good reasons to not go with x86. At the core, x86 is a crufty architecture, which will always have a negative inpact on things like power efficiency.

You mean...

[2nd Post!]

Develop on PowerPC and then migrate to the x86?

But I quote from the article:

These processors are large and extremely energy-hungry; active cooling is almost universally required in x86 designs, and mains power is preferable. The dominant design factor steering x86-based designs is the baggage required by backwards compatibility (both in the CPU core and other support circuitry on the motherboard), and many people are doubtless aware of the history behind this. Vast effort has been invested to modernize the CPU and overall system architecture, but even the current 64-bit architectures are constrained to some degree by legacy considerations. There are relatively few true SoC offerings based around an x86-compatible core, and thus significant external circuitry is always required in x86 systems. It should also be noted that, unlike the vast majority of embedded devices, x86 chips lack on-chip JTAG or other hardware debugging interface support.

The problem is how to hook up mains, active cooling (liquid maybe?) and designing the external support circuitry for x86.

The PowerPC doesn't have that problem because:

one tends to find ARM in low-system-cost, high-volume applications, particularly where power consumption is the prime criterion; x86 is typically found in applications where PC-compatibility is the prime selection criterion, and PPC is often found in applications where performance is the prime criterion. The specific advantages I was hoping to realize in moving to PPC were reduced power consumption, elimination of active cooling (required in my Geode design because of the combined power dissipation of the CPU and its mandatory companion IC), and enhanced performance in some digital image processing (machine vision) code.

So the answer? PPC has lower thermal constraints, higher performance given fixed energy requirements, and lower part count.

insane really use a Mips or ARM SOC

[johnjones]

why not use a intel Xscale (ARM) or a quad core 1GHz BCM1250 (mips) SOC from broadcom OR run a AMD alchemy au1200 500MHz (mips)

and run netbsd or linux

the whole thing is skewed he does mention ARM but discounts it because not powerful enough and opts for a 200MHz PPC yet a Xscale can clock at 500MHz so can the MIPS and trust me the broadcom has 4 CPU's to do the processing so its going to be outgunning ANYTHING the PPC camp has

I would have gone with a Xscale (intel) or Alchemy (AMD) solution

regards

John

Re:insane really use a Mips or ARM SOC

[2nd Post!]

Sure, but these guys went from x86->PPC successfully, so should and did are two different things.

I'm sure there are reasons for all these architectures. The PPC has the benefit of being reasonably scalable too: There are PPC cores from 200MHz up to 2.5GHz so you can develop on one and deploy on another.

Re:Other way around

[LWATCDR]

No.
x86 is not the end all and be all of computers.
It fails at things like power to mips. Just remember the fastest computer Blue/Gene does not run x86.
Even Microsoft is dropping the X86 for the new XBox.
x86 is not dead yet thanks to AMD but it will be interesting to see what the future holds.

McCraigor wiggler

[bhima]

I've used a McCraigor "wiggler" in the past and was very underwhelmed. It's sort of the winmodem of JTAG adapters.

It's a shame he hosted his project on Windows, I would have like to have seen it on OS X or Linux. I suppose he will wind up using Cygwin which I never really got on with... I always found easier just be in a *NIX environment to begin with.

Here's hoping after they finish jettisoning the PC business we quit seeing so many Windows hosted articles on IBM Developerworks

Re:McCraigor wiggler

[Amiga Trombone]

It's a shame he hosted his project on Windows, I would have like to have seen it on OS X or Linux.

Where do you see that he hosted it on Windows? He hosted it on a Kuro Box [revogear.com], which is a Linux-based PPC device. The only role Windows plays is running the setup program to load the Kuro system image.

Actually, you can even get around that. Someone on the Kuro forum [revogear.com] put together instructions for loading the system image from a Mac [revogear.com].

The Kuro Box is a pretty neat toy in it's own right. If you want to play around w

Re:McCraigor wiggler

[bhima]

The wiggler is a windows only thing, which is why I don't like it.

I already do PPC development so I'm not really in the market for a cheap device.

Re:McCraigor wiggler

[kju]

> The wiggler is a windows only thing, which is why I don't like it.

Not true. It is just another lowcost jtag interface and can easily be accessed with linux. Having said that, its worth to note, that you can build an equivalent device yourself for under $3. Digilent (www.digilent.us) also offers some cheap jtag cable, even for usb (and some nice cpld and fpga devboards)

Homepage of the author

[sdcmk]

The homepage for Lewin Edwards [larwe.com] is interesting. He is doing other neat projects in addition to his submarine project mentioned in the article.