So what is XMOS good for?

persia · 02-21-2010, 06:21 PM

What does having it on board do for the AmigaONe machine? So far all I know is that it's integer only and you could fry your motherboard if you hook power up to it wrong. While those maybe considered selling points, what else does it do?

Hell Labs · 02-21-2010, 06:34 PM

Drives up price.

Makes it look like they are making unique hardware that would be worth the cost.

Makes them look busy.

Gulliver · 02-21-2010, 06:45 PM

I couldnt agree more!

Hans_ · 02-22-2010, 02:20 PM

Quote:

Originally Posted by Hell Labs

Drives up price.

Makes it look like they are making unique hardware that would be worth the cost.

Makes them look busy.

Uh, those XCore chips cost less than $10 US. I know that it's fun to be negative and piss on someone else's work, but do try to keep the criticism sane.

XMOS call their chips "software defined silicon." What I think that they mean with this** is that there are no built in peripherals. Instead, its low latency event-driven multi-threaded architecture means that you can create any peripheral I/O that you like using software. This is something that "normal" micro-controllers can't do (or would struggle to do reliably).

What would it be used for on the X1000? No idea exactly. It's greatest use is likely to be with the Xena slot. Forget about using it for graphics or emulation, because it's no processing powerhouse, unless you string a whole set of these cores together (via the Xena slot) and figure out how to use them as stream processors. Given its strength with low latency event driven processing, this would be useful for hardware hackers, people doing robotics stuff, etc.

"But that's not desktop," I hear some people say. Who cares? People call the Amiga a "hobby" computer, so why not put stuff in it that caters to computer/electronics hobbyists?

Why not just put it on a PCI card or on a USB dongle? Well, what's the point of having a low latency device if you then put it at the end of a high latency bus? Sure, you could still develop XCore stuff via a USB connected Xcore chip, but you would immediately take anything that would work better being tightly integrated off the table.

Why put the XCore chip on all machines instead of just the hobbyists who want to use it? Well, let us say that one of these hobbyists makes some cool gizmo that you like, and want to have. With standardized hardware you could ask the hobbyist to make you one, and plug it straight in.

All of this is completely hypothetical. We're not going to know what can be done with it until the machine is available, and in the hands of those who can develop stuff. If you don't see the point in the XCore chip, then maybe you should focus on the rest of the machine; multi-core >1.6 GHz, PCI-Express, etc. Surely that's interesting enough in its own right?

Hans

** I'm not familiar with their architecture, so I could be completely wrong

**Piru** · 02-22-2010, 03:37 PM

Quote:

Originally Posted by Hans_

We're not going to know what can be done with it until the machine is available

We know what can't be done with it, however.

Quote:

If you don't see the point in the XCore chip, then maybe you should focus on the rest of the machine; multi-core >1.6 GHz, PCI-Express, etc.

Somehow the whole marketing drive of this "X1000" seems to be built around this chip. If it's not so important, then why make such a big deal about it? I find this odd.

**koaftder** · 02-22-2010, 04:26 PM

Quote:

Originally Posted by Piru

We know what can't be done with it, however.

Somehow the whole marketing drive of this "X1000" seems to be built around this chip. If it's not so important, then why make such a big deal about it? I find this odd.

My guess is that they are hoping to sell "expansion" modules that utilize the x-link switching feature of these devices. SO basically you can pay 100 bucks or something for each "device" that will do something lame. They're hoping to get developer support for it, which ain't gonna happen because it's lame and uninteresting and unprofitable in this context.

persia · 02-22-2010, 08:14 PM

So now we know it's slower than USB, has only 64K ram and has no no direct access to system ram... Curiouser, and curiouser. Unless these boards have some sort of other use and they were able to piggy back the Amiga One production onto a much larger production order...

**Matt_H** · 02-21-2010, 06:48 PM

C'mon guys, at least give some developers a chance to play with it before writing it off as useless.

dammy · 02-21-2010, 07:41 PM

Quote:

Originally Posted by Matt_H

C'mon guys, at least give some developers a chance to play with it before writing it off as useless.

How much RAM does it have? 64K?

**koaftder** · 02-21-2010, 08:58 PM

I'm thinking of porting the compiler over to AROS, just to be a dick, just so AROS users can brag that they got to play with the cheap XMOS dev kits first without paying 2,000 for the privilege of having the uber l33t new hyperion OS/mobo combo and still have to compile their junk on a mac or pc.

https://www.xmos.com/technology/design-tools-source

dammy · 02-22-2010, 07:36 AM

Quote:

Originally Posted by koaftder

I'm thinking of porting the compiler over to AROS, just to be a dick, just so AROS users can brag that they got to play with the cheap XMOS dev kits first without paying 2,000 for the privilege of having the uber l33t new hyperion OS/mobo combo and still have to compile their junk on a mac or pc.

https://www.xmos.com/technology/design-tools-source

I thought it was all closed sourced tools, nice to see I'm wrong. How big of a job would it be to do the port?

**SamuraiCrow** · 02-22-2010, 10:35 AM

Quote:

Originally Posted by koaftder

I'm thinking of porting the compiler over to AROS, just to be a dick, just so AROS users can brag that they got to play with the cheap XMOS dev kits first without paying 2,000 for the privilege of having the uber l33t new hyperion OS/mobo combo and still have to compile their junk on a mac or pc.

https://www.xmos.com/technology/design-tools-source

It's an LLVM-based toolchain! I hope you do port it because I want to make Mattathias BASIC use that same toolchain! The trickiest part should be that their IDE is Eclipse and requires Java. Other than that, it's just a matter of making the calling conventions for AROS libraries using TableGen and setting the alignment for AROS' structure alignment policies. There's already an x86 and PowerPC backend for LLVM but x86 seems to get more attention.

haywirepc · 02-22-2010, 10:49 AM

Tripitaka

Okay, except for embedded devices and video game consoles... which use 2+year old slower chips...Power pc is dead dead dead.

From wikipedia :

"In 2004, Motorola exited the chip manufacturing business by spinning off its semiconductor business as an independent company called Freescale Semiconductor. Around the same time, IBM exited the embedded processor market by selling its line of PowerPC products to Applied Micro Circuits Corporation (AMCC) "

What do you think about the future of power pc without the millions and millions of R+D by IBM and Motorola GONE? Intel/AMD wins hands down and thats that my friend. Freescale and applied micro circuits corporation don't have 5% of the budget of motorola or ibm.

They can not and will never again compete with intel/amd, no way.

I for one, can't understand anyone gambling the future of their company
on a dying processor. Its dead or dying. It will NEVER be the competitor it once was, NEVER. You can be ignorant of this and put your blinders on
but that fact will NEVER change.

I for one would buy AOS in a hearbeat if it ran on x86 and so would so many others, yet they still ignorantly think thats not a good idea.

Also, I love "Our most ambitious project to date" If the most ambition they can muster is attaching an off the shelf microcontroller to a substandard overpriced power pc motherboard... I don't know thats kind of sad really.

Somehow I hoped for more ambition than that...

Steven