@Caveman

Don't know... Possibly - I did some searches earlier and saw someone was working on porting an ARM based mp3 decoder at least.

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

@ A-Eon

Quote:

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?

Here's another thread
http://www.xcore.com/forum/viewtopic.php?f=5&t=114

I guess there are/will be more.

people outside the community starts to take notice. Someone at A-Eon should try to sustain there attention or they might forget about it.

_________________
"Great spirits have always encountered violent opposition from mediocre minds" - Albert Einstein

@ChrisH

Quote:

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.

And an Amiga joystick interface too!

_________________
Amiga user since 1985.

@Lynx

Quote:

Lynx wrote: @DAX Quote: We are indeed asking how it communicates with the rest of the system You're right. Actually, this is THE crucial point which will decide whether Xena is a genius idea or a bad joke, and it's not really discussed on the A-Eon website. For what we know, Xena's I/O lines are routed to the Xorro port. I hope there is more than that! We're all (including A-Eon) talking about the Classic Amiga's custom chips, but Jay Miner and his team didn't think : "Ooooh, crap, we'll just put those chips somewhere on the board, give them some kind of vague connectivity, and programmers will do what they want with them. I, for one, don't have a clue about what they might be useful for." So the question is (Rogue, TrevorD, if you're reading this...) Does Xena have priviledged access to memory? DMA? Can it directly access the peripherals using the southbridge? Can it send interrupts to the CPU? In a single, somewhat provocative question : What can Xena do that a XMOS chip embedded on a PCI-e card couldn't?

I quote this lynx's questions, we need answers to these before arguing what will it possible to do with this new Amiga.

_________________
God created Paula so that Allister Brimble and Dave Whittaker could do music

Check my Amiga gameplays (ITA)!

@ErikBauer Quote:

What can Xena do that a XMOS chip embedded on a PCI-e card couldn't?

Thoughts:

1. Guarantee that every X1000 owner has an XCore chip, and that there is therefore a market for software/hardware designed to use it. If it were an add-on card, then no-one would buy it - see the failed PhysX accelerator cards in the PC market. It avoids this chicken & egg problem.

2. By coming with the system, it is there to tempt people to use it, rather than them having the barrier of having to buy an add-on card.

In theory the Classic Amiga chipset (and perhaps processor) could have been an add-on card for PCs, rather than being a whole computer. But I'm skeptical it would have been as successful. (For sake of argument, I'm assuming here that there would be no technical downsides to doing this.)

Last edited by ChrisH on 07-Jan-2010 at 12:00 AM.

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

@ChrisH

As Hans said in post #21 of this thread:
Quote:

For low latency stuff on a traditional CPU, you stick that code straight into the interrupt service routine. It looks like the advantage is in the fact that there is hardware support for eight threads, each of which can be triggered by events (interrupts). Other than that, it still looks like interrupts.

I suspect it's used for some of the bus-control pins. Particularly since AmigaOS has to have drivers custom written for every device we get. Why not have some of the interrupt service routines offloaded into the local-store memory of the coprocessor as a sidekick processor chip?

@ChrisH

Quote:

ChrisH wrote: @ErikBauer Quote: What can Xena do that a XMOS chip embedded on a PCI-e card couldn't? Thoughts: 1. Guarantee that every X1000 owner has an XCore chip, and that there is therefore a market for software/hardware designed to use it. If it were an add-on card, then no-one would buy it - see the failed PhysX accelerator cards in the PC market. It avoids this chicken & egg problem.

PhysX accelerator was killed by NVIDIA CUDA based PhysX accelerator .

_________________
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)

@Hammer
No, NVidia only put the final nail in the coffin, by buying-out Ageia. PhysX (from Ageia) never took off, and the only way they could avoid loosing masses of money was to sell it to NVidia.

Last edited by ChrisH on 07-Jan-2010 at 01:42 AM.

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

@ChrisH

Quote:

ChrisH wrote: @Hammer No, NVidia only put the final nail in the coffin, by buying-out Ageia. PhysX (from Ageia) never took off, and the only way they could avoid loosing masses of money was to sell it to NVidia.

There's very little need for PPU with CUDA PhysX.

If you read [url= http://http.developer.nvidia.com/GPUGems3/gpugems3_ch16.html]CryEngine2[/url]

"vegetation react to global and local wind sources, and we bend not only the vegetation but also the leaves, in detail, with all computations procedurally and efficiently done on the GPU"

With CryEngine2 (Crysis), physics being done on GPU's shaders/stream processors.

DX9 vs DX10 on physics

There's very little need for PPU.

Last edited by Hammer on 08-Jan-2010 at 09:40 AM.
Last edited by Hammer on 07-Jan-2010 at 10:45 AM.
Last edited by Hammer on 07-Jan-2010 at 10:43 AM.
Last edited by Hammer on 07-Jan-2010 at 10:42 AM.

_________________
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)

@ChrisH

Quote:

ChrisH wrote: @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.

Link

"CUDA-enabled GPUs are ideal for real-time signal processing".

_________________
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)

@Hammer
The presence of Xcore does not exclude GP-GPU on the X1000 (Hans said he is very interested in OpenCL for instance), we might even have applications that handle different tasks/threads using one or the other simultaneously.

_________________
SamFlex Complete 800Mhz System + AmigaOS 4.1 Update 4
Amiga 2000 DKB 2MB ChipRam GVP G-Force040 Picasso 2 OS3.9 BB2
AmigaCD 32

@vidarh

Welcome to AW.

Just let say that we are all fallen in love with Xmos. :)

Some question: it is posssible to create the Xena on PCI. Could the Xena be used in connection with FPGA?

Again wellcome on board.

_________________
retired

@AmigaBlitter

Thanks...

Quote:

Some question: it is posssible to create the Xena on PCI. Could the Xena be used in connection with FPGA?

I don't see why not, with a couple of caveats. One is that we don't know how tightly they're integrating the on-board one. It may not be possible to achieve all the same things from one that's put on a PCI card.

If you put it on the Xorro card it is *likely* based on what they've said that you'd be able to hook the XCORE's together using "Xlink" (but again, I don't know yet for sure based on the info that's been released) which is XMOS' way of connecting the chips together using a fast, low latency protocol, in which case the Xena on the motherboard could be used as a "gatekeeper" for any resources on the motherboard it's integrated to.

As for using it in connection with an FPGA, sure - an FPGA looks like any other chip. But it'd mean a custom card, of course. I'm not sure what applications it'd be suitable for, though. Got any exciting ideas? :)

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

Quote:

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

Well it certainly can be used for MP3 decoding. I haven't found anything concerning video yet. I suppose it's possible if you chain enough XCore processors together.

_________________
Unless otherwise explicitly stated, this message is not meant to affirm nor deny, defend nor offend any faction within the 'Amiga' Community.

@JonathanMay

Quote:

JonathanMay wrote: Hey - I'm now allowed to post freely and will try to answer questions: I also recommend you head over to our official community if you have questions about the XMOS processor. http://xcore.com Just to say up-front, I don't know how the XMOS chip is coupled to the main host CPU, so can't help you out with info there. Looking forward to seeing how this develops.

Welcome to AW.net. It's hard to know exactly what questions to be asking. I guess that there's my what is "event driven architecture" question. What do you think the Xena (L1 XCore chip) and Xorro (the associated expansion port) would be good for? Oh, and do you have any Amiga background?

Hans

_________________
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
https://keasigmadelta.com/ - More of my work.

Here is the SID emulation video.

XMOS playing Commando.sid


I think I'll hook up some leds around my room. Really nice looking :)

Last edited by Ancalimon on 08-Jan-2010 at 01:08 AM.
Last edited by Ancalimon on 08-Jan-2010 at 01:07 AM.

@ShadesOfGrey

Quote:

I haven't found anything concerning video yet. I suppose it's possible if you chain enough XCore processors together.

The best you could possibly hope for would be a streaming MJPEG decoder that outputs the video as it decodes it.

MPEG, H264 and all similar formats employ motion compensation. That requires having at least one (and these days typically two) reference frames in the decoder's memory. And 64kB just ain't gonna hold those frames for any non-trivial video resolution.

@umisef

Thanks :)

_________________

@vidarh

Quote:

I'm not sure what applications it'd be suitable for, though. Got any exciting ideas? :)

I'm thinking about to use a Xena card on PCI on the Sam440 Board. The Sam440 have a Lattice XP FPGA programmable on the fly with jtag connectors. I'm not an FPGA expert, but i think that this FPGA could be used with the Xena. Just wondering if a Xena card could be fitted in a PCI 33 slot. Btw, the LatticeXP could become a PCI endpoint too.

_________________
retired

@Hammer

"CUDA-enabled GPUs are ideal for real-time signal processing"

What kind of I/O do they have for that? (RF in x1?)
What OS they run?
(I bet they mean video signal processing, not general *.)

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