@koft

Quote:

koft wrote: programming model isn't similar to LRB. With LRB you can pass threads to and from the processor, you wont be able to with Xena because it's alien. It's just some gimped processor on a bus with 64kb sram. What a friggin joke. It's not even "amiga like".

As for alien, refer to Toshiba Spurs Engine (CELL based quad-SPE).

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

For anyone looking for a problem that this solution can fix, here is a trivial problem that some Classic supporters might find interesting:

With the XMOS chip (or any FPGA, for that matter), you could emulate the floppy drive logic contained within the legacy Paula chips. With a driver running in OS4, programming the XMOS chip to act like a Paula, you could connect any old PC floppy drive to the output pins and have it read/write old Amiga-formatted disks at full speed.

_________________
cheers
tony

Hyperion Support Forum: http://forum.hyperion-entertainment.biz/index.php

I wonder if xena is same as this: http://www.xmos.com/system/files/xs1-l1-64lqfp-ds.pdf

Or does it contain something extra. (& how it will be accessed from PPC side)


Are MorphOS people also interested in X ....
http://www.morphosppc.com/article/do-you-know-what-a-transputer-is/
?


btw. is this company in any way involved in x1000 development: http://www.transtech.co.uk/
(see the CPU)

Last edited by KimmoK on 06-Jan-2010 at 12:43 PM.
Last edited by KimmoK on 06-Jan-2010 at 12:39 PM.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@JonathanMay

Welcome to the forum and thanks for coming clarifying our doubts. I hope this machine Amiga-side will be as interesting for you as the XMOS-side is for us.

the question is how many Xenas will we need?

So basically they are using a Xmos chip to emulate 68K and recompile the code?

@Samurai_Crow

Quote:

The XCore toolchain is based on LLVM. That means for PowerPC it will only be a matter of writing an AmigaOS 4 backend for LLVM and recompiling.

Isn't that what you're doing/using for matthias basic?

_________________
Peter Swallow.
A1XEG3-800 [IBM 750FX PowerPC], running OS4.1FE, using ac97 onboard sound.

"There are 10 types of people in the world: those who understand binary, and those who don't."

@Chuckt

No, they have petunia. But perhaps it would be possible as well.



+
I bet most of these ides would be doable also on xena.
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=27379&forum=33

Last edited by KimmoK on 06-Jan-2010 at 02:33 PM.
Last edited by KimmoK on 06-Jan-2010 at 02:14 PM.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

some xmos users discussing X1K here

http://www.xmoslinkers.org/forum/viewtopic.php?f=3&t=700

Does anyone have any answers for their questions?

_________________
AmigaOne X1000

@MrFancyPants

It is possible to run Java on LLVM using VMKit. Xena might not help bring VMKit to the Amiga but LLVM might.

@Swoop

Quote:

Swoop wrote: @Samurai_Crow Quote: The XCore toolchain is based on LLVM. That means for PowerPC it will only be a matter of writing an AmigaOS 4 backend for LLVM and recompiling. Isn't that what you're doing/using for matthias basic?

Yes, we'll be using LLVM for Mattathias.

might that be that xorro could be utilized to interface with lcd modules or with further xmos chain to control such modules in a manner like in this example?:
https://www.xmos.com/products/development-kits/led-reference-design-kit

@CodeSmith Quote:

If it's only connected to the PCIe bus, even if it's a busmaster, any PCIe graphics card will easily blow it out of the water (when's the last time you saw a card with only 8 shaders?)

AFAIK, sending data to & from a CUDA program running on a gfx-card is a fairly expensive operation, which will have to be "batched". That implies high latency. And one of XCore's main attributes is extremely *low* latency in respoding to I/O, so (if it is wired into the mobo correctly) then it could respond much faster than any CUDA gfx-card.

_________________
Author of the PortablE programming language.
It is pitch black. You are likely to be eaten by a grue...

@wawa

Sure! Why not?

@tonyw Quote:

With the XMOS chip (or any FPGA, for that matter), you could emulate the floppy drive logic contained within the legacy Paula chips. With a driver running in OS4, programming the XMOS chip to act like a Paula, you could connect any old PC floppy drive to the output pins and have it read/write old Amiga-formatted disks at full speed.

I thought of this too - no need for a full-blown Catweasel! Just a simple Xorro card to act as the connector to a floppy drive.

IMHO the *best* thing someone could design for Xorro is a multi-purpose I/O-connector card. So it would provide tons of different connectors, and you would be able to choose which one(s) to use in software.

Last edited by ChrisH on 06-Jan-2010 at 04:46 PM.

_________________
Author of the PortablE programming language.
It is pitch black. You are likely to be eaten by a grue...

Here's a challenging idea for a Xorro plug-in card: A physics-calculation accelerator, the sort of thing that gfc-cards are now being used for on the PC. If you had a large array of XCore chips on one card, then it should be possible to make each of them work on a different bit of the physics calculation, all in parallel.

I suppose ideally someone would make a Xorro card with a lot of XCore chips on, and then we could use it for whatever we liked. Since single XCore chips are supposedly cheap, you could have 10 to 20 of them on a card that cost (say) £100. While the raw processing power of such a card wouldn't be stunning, it's ability to do vast numbers of operations in parallel could be...

_________________
Author of the PortablE programming language.
It is pitch black. You are likely to be eaten by a grue...

Ok... So I connect one or more of these 64Core boards to my Xorro slot. Then what?

http://www.hexus.net/content/item.php?item=17513

So... What is the floating point performance of each core like? Are we going to get a good render farm out of it? Realtime Raytracing?
The slashdot term "Imagine a Bewulf cluster of these things" springs to mind.

So... Can we make each core run as a web server? Does this fulfill a market demand of some type? Low power, heavy load.

If any of these things are possible aren't we filling a pretty exclusive niche.

_________________
Leo Nigro, CTO Commodore USA, LLC
Opinions expressed are my own and not those of C= USA.
Commodore/AMIGA "Beautiful, High-Performance, Home Computers for Creativity and Entertainment."

@ChrisH

Or...

A audio/speaker control system with true seperation for not just
front.side,rear, but also top and bottom.

@ChrisH

Quote:

Here's a challenging idea for a Xorro plug-in card: A physics-calculation accelerator, the sort of thing that gfc-cards are now being used for on the PC. If you had a large array of XCore chips on one card, then it should be possible to make each of them work on a different bit of the physics calculation, all in parallel.

Interesting idea. There are companies like Ageia that do physics acceleration cards, though, and they were bought by Nvidia. I don't think an XCore would be likely to be able to compete with dedicated silicon for physics acceleration, since that's a workload that'd benefit a great deal from vectorized floating point math (i.e. calculate the new position of these 200 objects given these 200 current positions, mass and thrust)

Quote:

I suppose ideally someone would make a Xorro card with a lot of XCore chips on, and then we could use it for whatever we liked. Since single XCore chips are supposedly cheap, you could have 10 to 20 of them on a card that cost (say) £100. While the raw processing power of such a card wouldn't be stunning, it's ability to do vast numbers of operations in parallel could be...

The chips would run you from $313 to $203 for 10 of their largest quad core chips (depending on quantity - prices on the xmos website) or $209 to $136 for their smaller version (the difference is 256 vs. 88 external IO lines per chip). On top of that there's of course the card and extra wiring, build costs, overheads and profit margin etc. Perhaps $500-$800 somewhere might be doable.

10 of them would offer an aggregate of 16000 MIPS, which is well above what a dual core 2GHz PowerPC can be expected to deliver, but parallelizing many "normal" workloads is hard, though, so it's not really directly comparable.

_________________
Wiki for new/returning Amiga users - Projects: ACE basic compiler / FrexxEd / Git

@vidarh

Would it be possible to use the xcore,as some kind of decoder for mpeg 2,mpeg 4,h265,mp3, etc. ?

_________________