Amiga One X 1000 and Xena/Xorro

[OlafS3]

Thanks for that AmigaDave.

We have produced Xorro development boards (which are available) and are working on a couple of Xorro specific projects. We are also including an improved Xena/Xorro combination in the AMIGAone X5000, thanks to the specification and hard work of Lyle.

As for complaints, most of the comments appear to come from none AMIGAone X1000 owners?

Trevor

You must admit that up to now there are no many real applications available or known for it. It was marketed as "custom chipset" (if I remember right) and led to wild speculations. It is in any case making the hardware more expensive. It is a industrial component, X1000 (and X5000) are bought by consumers, do you really have possible applications that make it worth to integrate it again?

[OlafS3]

IMO: That's childish thing to say.

We know the xcore idea came from Varisys. Not from any Amiga custom chip lover or marketing man.

I would love to hear what was the idea by Varisys people when they suggested that core should have direct connection to CPU.  Lower latency (than over PCI or USB) between CPU & xcore? Being able to have 100% control of xcore without extra tools? Being able to reprogram xcore on the fly per desktop application request? All of those? More?

If the idea is from Varisys and they should have known that you need certain software and drivers for it that are closed source they should have had a plan how to support it in future. If not why integrating it? And why integrating it in X5000 yet again?

[OlafS3]

Curently, people seem to be having a hard time finding uses for the XENA that require the chip to be inside the Amiga(one) case, putting XENA plus a prototyping area on a ZORRO card is intended to encourage "inside the box" thinking.
It may be that classic Amigas need XENA more than the Amigaone does.

classic Amigas need XENA? :confused:

XENA without official support by XMOS including the development tools is useless, on whatever form

[OlafS3]

Constructive criticism.  Thank you.  Not everyone will agree with your opinion that it is "far too expensive", but that does not change the fact that your criticism is constructive in nature and suggests a possible helpful alternative to increase interest in Amiga + XMOS projects, as not all Classic Amiga owners have added USB to their Amigas, so the option of using a standard XMOS USB development board is not available to them.  But remember, it is a waste of time to use XMOS chips with your Amiga, as you can do it faster and better connected to your PC.

Edit:  Just to remind some of you who don't know me, I am the guy who promotes MorphOS at 5 out of the last 6 AmiWest shows, not an AmigaOS4.x "fanboy", or someone who defends everything that Hyperion does.  I just happen to also own an X1000, but my current favorite Amiga inspired platform is still MorphOS3.6 on my 17" 1.67GHz G4 PowerBook, over-clocked 1.5GHz G4 PowerMac, and Dual 2.7GHz G5 PowerMac (400MHz Efika too, but it has not been turned on in a couple of years).  I choose to support all flavors of the Amiga experience, and think it is a waste of time to tear any of them down.

Nobody doubts that there are use cases for the chips, otherwise they would neither be developed nor sold but I am wondering for what consumers can use them (and both X1000 and X5000 are consumer boards). If it would be sold to people using it professional then it might make more sense (if it would be full supported) but companies will certainly not buy it because it is too expensive, missing drivers and development environment for it, and no support nearby and thus problems if f.e. the system must be repaired. The last point alone would be a no-go for a company. So X series is for amigans only. Trevor integrates Xena again in the new X5000 despite the experiences with X1000 so he has either ideas people here not have or it is just a small gimmick to make the board special without any real use. But every gimmick makes it more expensive so I personal would have left it away if I realize that something is without real use but it is Trevors decision and the money of the people that buy the boards...

[OlafS3]

It's possible, it would be interesting to see what uses people could find for it in a classic Amiga.
Audio applications have already been suggested, XENA could give us a 16/24 bit sound card, speech recognition perhaps, can XENA implement an Ethernet port?, how about an SD card interface, USB port, it could drive a pair of ultrasonic transducers to give us similar functionality to that MS device whose name escapes me at the moment. EDIT: Kinnect.
XENA could give us a multifunction card that could evolve as people developed the software to implement new interfaces and ideas.
Software support would evolve faster if more people were trying to use XENA.

a soundcard? As I understand it XENA is a special chip to manipulate streams independent of the processor so you basically develop a program and upload it in the chip where it is executed independently and reacts on events but no interaction with the host system (if I am wrong then someone can explain it). So how do you want to program that soundcard then? In a FPGA it is different because everything is included and interacts with each other but XENA works different. You can automatically manipulate sound streams (perhaps convert them?) but it would be not a real soundcard.

[OlafS3]

One core use case for xcore is to use it as a bridge chip. In that way it could enable limited classics to get interfaces to modern technology.

Another way would/could be a FPGA card for zorro & clockport.

(neither make sense businesswise, but that's no news on our niche)


For cheaper access to xena... it would be cool if every NG HW had two or more (shielded/protected) D9 ports that have xena I/O.
Then we could use legacy joysticks etc very easily + use those pins for instrumentation & home automation etc. without too much extra HW.

I think the new accellerators by apollo team with LAN and modern output are already available in near future, I also read from a new extension for classics that make it possible to use USB mouse and joysticks at classics (the card converts signals automatically). Additionally the Apollo team is also planning for a standalone system in future and there is also FPGA Arcade. So I think the 68k community will be quiet well supported :-), even without XORRO

videos about the adapter(s):
http://www.youtube.com/watch?v=BSdt4vLx9CI&feature=youtu.be
http://www.youtube.com/watch?v=B_yfLJftOGM&feature=youtu.be

looks cool :-)

[OlafS3]

Digital audio is just a stream of 1's and 0's, XENA has one third of its pins dedicated to a local bus for communicating with the system, downloading the XENA program and then transferring data and instructions as required, it could be paula's big sister., with a Retargetable audio driver, data compression/decompression optional.

I think it is more propable that this will be done in a FPGA where everything is included. I see the problem how to change it on the fly like it is done in amiga with its registers. It could be a kind of extension but you would need someone to do the card. And then you have still the problem that our platforms are not supported by XMOS

[OlafS3]

I have 2 hooked up together permanently via serial cables which also allows me to run different revisions of the AmigaOS and have different types of graphics cards installed, for example one currently has the 7970 and the other a R9 280X; all for beta testing.

The 3rd is a spare, it's a long way to the UK if something goes wrong

three X1000? Wow

[OlafS3]

I started this thread to get more insight in the capabilities of the XENA/XORRO thingie and possibly to develop some neat ideas on how to take advantage of it.

This requires some sort of inspiration and vision, aside from knowledge.

Those lacking the required inspiration and vision are expressively NOT invited to take part in this discussion, not to mention those that combine their lack of vision and inspiration with the need to polish their ill egos by trying to bring things others are interested in into discredit.

I would really appreciate if such charcters would refrain from participating in this discussion.
No-one forces them to buy an A1-X1k and to use XENA/XORRO.
If they don't like it - fine. They just have to stay away...

In this case this thread isn't for them - they're better off with classic Amiga stuff or WIntel stuff.



And the best you and similarily unimaginative minds can do is to stay away from this thread instead of spamming it with repeated trolling attempts.

Peace please, now you are starting to bash others... expecially with using phrases like "characters" this could be seen as insulting

and commenting "they're better off with classic Amiga stuff". New FPGAs are used all over the industry so this "we are the best, the modern, the NG owners" and there are the "retros 68k old fashioned gaming people" is not true (and if anyone wants to win more customers it is stupid)

[OlafS3]

My hope is to gather as much XENA-information as possible with this thread.
Who knows - perhaps it will evolve into some sort of "XENA/XORRO support"-thread one day.

If there is no central point where we get get information from, then let's create one here!
Let's collect any info we can get about XENA/XORRO here!

As AEon recently acquired AOrg, I thought this might be the right place to get answers from those who have the most experience with it/knowledge about it.

A better support for Amiga-XMOS (XENA) - including the dev tools - is definitely desirable.

Perhaps this thread can help to increase the support or at least to serve as a central knowledge reference regarding XENA.

From my point of view it can only help sparking the interest, if more and more information about XENA, XORRO and the JTAG programmer becomes available.

Perhaps someone can put informations here. The only problem it is very specific so someone has to know both Amiga (AmigaOS and its hardware) and the XMOS chips well. Developers that normally use XMOS chips are certainly using one of the officially supported platforms. I have programmed a simple controler some time ago and it is not fun without appropriate documentation. For XMOS the AmigaOS community is too small to support propably and everything (including the software) is closed. There I see the problems.

I have read first time about JTAG. If I understand it right it is only to test everything.

http://de.wikipedia.org/wiki/Joint_Test_Action_Group (German)

here is a interesting discussion with one known person involved :-)

http://www.xcore.com/forum/viewtopic.php?f=44&t=2477

[OlafS3]

I'm just trying to get rid of those in this discussion who are permanently bashing the implementation of XENA/XORRO and the people behind this idea.

Not sure what should be insulting with the use of the word 'characters'...



If someone isn't interested in the NG stuff, but is permanently around in NG threads and gets on the bad side of NG interested people by bashing them for their interest, then he really would be better off if he cared more for the classic stuff, wouldn't you agree?



I don't mind...



I never claimed that - after all I'm a classic user myself (Amiga4000 PPC w. OS 3.9 & WarpOS 16.1 /experimental OS 4.0 classic installation).

hmm ok

the thread I mentioned but hard stuff:
http://www.xcore.com/forum/viewtopic.php?f=44&t=2477

shows how specific it is

[OlafS3]

The original purpose of JTAG was to allow board manufacturers the ability to test assembled boards with a minimal hardware interface. By chainging them serially and setting a common interface spec even dissimilar chips can be polled and verified in a single JTAG loop.

As used with XMOS chips like Xena, which usually keep the executable code in RAM, it allows that ram to be programmed, and also allows setting of various configuration options such as links between cores(tiles), PLL clock divider settings, and other various details required to get a XS1-L16A chip up and running.

Within the realm of hobbyists, the XMOS chips are almost exclusively run this way, since the OTP ROM is, well.. One Time programmable. It'as not flash, and once it has been progframmed, that chip is committed to that code forever.

Obviously Flash would be more hobbyist-friendly, but it would also create security concerns for any commercial products made with XMOS chips.

Getting back towards the topic(s)..
The non-Amiga XMOS hobbyists are largely oblivious to all these JTAG details. The XDK suite provided by XMOS will seek out XMOS compatible JTAG programmers on startup, then access those to get ChipID's, and often board IDs from the JTAG device. Net result, it's all super-automatic, hands-off, and self managed. Just plug your stuff in and everything is configured automatically. All of this is wonderfully convenient, but it leaves those with non-XMOS kit out in the cold, so to speak.

The discussion you referenced is about one of the "new" products from XMOS. They are now building the USBtoJTAG directly into their newer kit, and doing it in a way that can NOT be easily reached from an external JTAG programmer. This means that we will not be able to use these new "StartKits" from an X-Series AmigaOne, because there's no place to connect JTAG to it.
They are meant to be used only on XMOS supported platforms with XMOS authored software.
Eventually a few hackers found a way to (somewhat) get around this, but not without some loss of function.

Veering back towards a topic once more.. JTAG on Xena is required to load any Xena program you wish to run, The JTAG loop is extended to the XORRO board, so that additional XMOS chips can be added at will. The XMOS tools that we have can load code into any number of XMOS chips through this loop, as well as freeze execution and read the current state of all the chips in the JTAG loop. This can be useful for debugging. We can dump RAM, registers, switch states, ROM, lots of stuff I have not yet documented. There is also a XReset command that will drop all connected (or any one connected) chip back into it's inert power-on state. While these are technically command-line tools, they work with tooltypes as well. Any XMOS program that sets XRunXE as it's default tool with load up when double-clicked.

About adding DB connectors to the Xorro board, that is possible, but I would discourage it. Perhaps I am overly cautious, but anything I connect that comes outside of the case will be optically isolated. While this may be overkill, I think it a prudent choice. I have an SD card there now, no need for isolation. I have a MAX3232 chip, which is connected to the serial port inside the case. again, no need for isolation there. But anything that comes outside the case, for joysticks or whatever else, will be isolated on my machine, and I would suggest that others consider it as well.

Intersting thought.. the S/PDIF in and out connections are optical by design, so adding those should be as safe as anyone could hope.. a very easy way to get high bandwidth audio in and out of the Xena chip.

I am being called away..

LyleHaze

Interesting stuff and many thanks for your work on it

So you can load code in it? Can you use the official tools (f.e. on windows) to create it? Then you would still need a kind of chain to transfer it from PC to X1000/X5000 to test it there. Or propably more convenient develop and test the code on PC and then transfer it to Amiga and test it there. What XMOS chip is used? Perhaps better to use the same chip and configuration on PC for development and testing and then transfer it and test it on Amiga?

[OlafS3]

Hmmm - here at my end it tells me:
» Amiga.org » Amiga computer related discussion » Amiga Hardware Issues and discussion » Amiga One X 1000 and Xena/Xorro

No mention of "classic" - sorry to disappoint you...

Please stop!

[OlafS3]

From the thread so far: you can load code to xena from AOS4 command line.
But "compiler" runs only on windows.
There's a "study" of putting up a server to handle that compilation for all x1000 etc. users.
(another option might be emulator to run windows tools on AOS4, but so far no luck)

For the second you would need either something like VMWare or WINE (and I am not sure if it would work with WINE at all). More realistic (personal guess) is to either create a kind of chain, develop and compile on Windows and then automatically transfer it to amiga or develop it on Windows and just finally test it on Amiga.