Amiga One X 1000 and Xena/Xorro

[Dandy]

@Dandy

...
To see more done with Xena/Xorro, we need more people/developers to ask "what could I make this do?"

Exacly my point of view.

Of course, it would also help if we had more docs, tutorials, and a full dev-kit including compiler that ran on AmigaOS.

Hans

Yeah! (drool, slobber)
We urgently need more of these things to be able to get inspired!

[Dandy]

The whole Xena/Xorro thing was marketing BS and almost useless (as history has shown) unfortunately.

Hmmm - to me the "uselessness" rather seems to be a consequence of the lack of vision and inspiration among the community, which in turn partly is a consequence of the poor support for XENA/XORRO (documentation, tutorials, devkits, ...).

[Kronos]

(as cyrus+ does not have audio onboard.... perhaps a xorro audiocard could be used.

Or just buy a 10$ soundcard supported by allready existing drivers...

[KimmoK]

Or just buy a 10$ soundcard supported by allready existing drivers...

Sure. But 10$ sound card might not be able the tune the 3D sound environment according to head movements.

(btw. the cheapest sound card I found locally cost 30€, C-Media CMI8786 based, legacy PCI)

(also if xorro sound card would be made, it should have atari ports etc etc. as well for more easy access to xena I/O)

@thread

The lack of good xcore development environment on AOS4 is huge handicap.
But also, if "xena" would exist in some more affordable motherboard (with low latency CPU localbus&JTAG), it might get some users from http://www.xcore.com/ .

[Dandy]

Just curious why would a home hobbyist use a CNC rather than a 3D printer?

Maybe I missed something - but as far as I know there currently is no way of 3D-printing steel parts or other metal parts.
CNC machines are to my knowledge mainly used for metal processing.
Don't confuse 3D-printers with the Star Trek replicators...
 

[Boot_WB]

@Lyle

Thanks for the insights from your perspective on this, it's good to hear from the horse's mouth where development stands. :-)

Reply too long, continued here from previous reply.

And why haven't I bothered to finish my Logger? Friends, I have been busy with other coding projects. "Fun" stuff like Xena, and like my own pet project "Score", have gone completely untouched for well over a year now. Only two weeks ago I decided to stop all my other projects and work only on Score... for my own sanity if nothing else.

I sympathise immensely.
As someone who is renovating my kitchen, living room, garden and 'spare room' whilst simulataneously trying to run a business and set up a second one in parallel... I can seriously sympathise.
Last week I realised that in order to drive forward the improvements I want to make in some areas, I really need to get my home life back in order - so am taking the rest of July off to concentrate on finishing the WIP at home.
Disrupted kitchen = no cooking, no cooking = poor diet, poor diet = poor physical/mental condition, which undermines the work I'm trying to accomplish in other areas.

@Hans

Thanks for sharing your thoughts on this and I'm really not trying to be negative (I think Dandy and I had pretty much the same discussion on AW.net a while back).
The design and manufacture of a homebrew milling machine is well within my sphere of interest (although the control side of things make me think of Laplace transforms which makes me want to run screaming from the building) and I would certainly be interested in the work you do.
As you said, I think whenever I read ideas about Xena I'm applying the filter of "what are X-core's strengths?" and not seeing them being particularly harnessed.

Considering Xena as a hobbyist port is probably a more constructive way to approach it than trying to find _that problem_ for which Xena is the solution.

[KimmoK]

Maybe I missed something - but as far as I know there currently is no way of 3D-printing steel parts or other metal parts. ...

There exist 3D printers that can print out metallic parts. (very expensive devices, used by some prototype labs)
(I have held some printouts on my hand.)

(IIRC, there's also some experimental printers that print out concrete houses and someone made solar powered device that goes on sand at sahara and outputs some pots welded from sand and ....)

[Boot_WB]

There exist 3D printers that can print out metallic parts. (very expensive devices, used by some prototype labs)
(I have held some printouts on my hand.)

Shapes made of metal perhaps, but probably not useful metallic parts:

http://moobunny.dreamhosters.com/cgi/mbmessage.pl/general/35153.shtml

[TrevorDick]

Lyle, Thanks for your comprehensive reply. I was going to make a comment but summary was more than enough.

TrevorD

[wawrzon]

Hmmm - to me the "uselessness" rather seems to be a consequence of the lack of vision and inspiration among the community, which in turn partly is a consequence of the poor support for XENA/XORRO (documentation, tutorials, devkits, ...).

community is how it is, and how it became over time in course of events it had no influence on. We know that currently there is only very little technically educated, skilled people left with us, especially those with spare free time and motivation. so i consider it unfair that the consumer repeatedly gets blamed for every failure, in particular the lack of unconditional support or "vision", in the end. the x1k, as example, was designed and developed behind the closed doors without the community being involved or consulted and what concerns the inclusion of the xmos chip apparently both, without properly analysing the consumers expectations and without securing the software support in advance, that should usually be included with in the system. the current situation is nothing unpredictible, but a straightforward consequence of approach taken.

[kamelito]

@Lyle,

On the page below, they said that XMOS were(are?) working on an LLVM port of the XC toolchain and then it'll  be open sourced.
Not sure if they manage to finish this port.

http://www.openbusiness.cc/2010/01/15/xmos-limited/


Regards
Kamelito

[LyleHaze]

I just read my post from last night.. it does sound more like a rant than I would have liked. Maybe I should avoid posting after midnight.

When the topic was first brought up, even some of the XMOS regulars believed the toochain was open. parts of it are, but the "XC" specific extensions are not.
If you're curious, you can read more http://www.xcore.com/forum/viewtopic.php?f=26&t=2137

There are a few in the XMOS clan who prefer PowerPC processors, but the grand majority are there to play with the chips, and would prefer to just use the tools provided.

When I was bringing up the command line tools for AmigaOS, one of the snags I hit turned out to be missing clock configuration information in our compiled projects. The XMOS community (who are generally very capable with these things) had no idea what might be wrong. It turns out that all of the XMOS dev stuff is very thorough about autodetection and autoconfiguration without user intervention of any sort. Tools like this are convenient, but they do contribute to the end user knowing less about how things work. I finally worked it out with Seghers help, and the AmigaOS command line tools were modified to properly set the clock divider correctly. On the other end, I wrote up a config script specifically for Xena, and added it to the XDK, so when we go to start a project, "Amiga X1000_Xena" comes up as an option when selecting the target chip.
Of course, it's not shipped from XMOS that way, but it's (now) easy to add in.

More thoughts about Xena projects:
The ability to do VERY accurate time measurements would make it ideal for ultrasonic measurements.
It would also be VERY simple to make a IR Remote control recorder/playback programs, even better if you choose to support the Pronto format. This would allow loading/saving remotes from a large existing community..
Granted, neither of these is "the killer application", but both could be a lot of fun to play with. Honestly the Pronto project would be made SO simple with the programmable clocks in the Xena chip.

And about the compiler.. I have wondered a few times about "cloud processing" to help us out. I wonder how hard, or how expensive, it would be to set up a virtual machine hosted at some internet company.. and use it just for compiling Xena code for Amiga-folk. I have discussed similar ideas in the XMOS forums, and no objections were raised by the company. If it helps them expand the user base (or just shut me up) then it's good for everyone.
_somebody_ would need to pay some monthly fee for processor time used.. but I suspect the total CPU hours per month would be quite low.. it would certainly be worth looking in to, anyway.
This networking stuff is way over my skill set. I was proud to get ethernet working on my classic 4000, and have never looked deeper since then. I wonder if anyone in our community has the knowledge to discuss these options?

Back to work...

[KimmoK]

Shapes made of metal perhaps, but probably not useful metallic parts:...

Hey there, even paperweights are usefull at windy places!  

(but sure, printed metal parts are not yet as tough has normal steel as an example. But in some cases, like using it as heat spreader, it can be ok.)

@Lyle
" If it helps them expand the user base (or just shut me up) then it's good for everyone."

Nice!

[persia]

The question is not really whether XMOS is useful, clearly with the number of projects out there it is.  But all these applications are pretty much stand alone applications.  I can go out and buy a US$ 100 USB 2.0 program board and do that from a PC or Mac.

The real question is what advantage does connecting XMOS directly to the motherboard of the AmigaOne bring?

[Gulliver]

The real question is what advantage does connecting XMOS directly to the motherboard of the AmigaOne bring?

I believe it was just a marketing thing. Remember it was set to be perceived as an Amiga like "custom chip". So that it appeared more mind pleasing for those who had a real Amiga and big pockets, and were looking for a next generation amiga-ish computer.

[amigadave]

I believe it was just a marketing thing. Remember it was set to be perceived as an Amiga like "custom chip". So that it appeared more mind pleasing for those who had a real Amiga and big pockets, and were looking for a next generation amiga-ish computer.

I don't disagree with what you wrote above, but like Lyle, I get annoyed by some other statements that try to tell people what the Xena/Xorro is, or was supposed to be.  Lyle summed up everything in his long 2 part post earlier in this thread, so I will just quote part of what he wrote again for emphasis.

Finally, if you've been reading along this far.. I'd like to vent about  one particularly annoying rant.. Some people keep going on about what  A-EON _SHOULD_ be doing with Xena. Xena is there as a USER_DEFINABLE  expansion. It's not their unfinished project, it's YOURS.
If you want to do something with Xena. GO FOR IT!

I think that Lyle put it into perspective best with that statement.  Xena/Xorro can be what ever the user wants to make of it.  It was included to encourage users to experiment and be creative, which Amiga users have always been famous for in the past.  Call it a marketing gimmick if you want.  Call it useless and a waste to have it on the motherboard, instead of just including the USB version with every X1000 motherboard, if you want.  Trevor acted on the advice given to him at the time of the design for the X1000 motherboard by people who have much more technical knowledge than himself.  The additional cost for adding it to the motherboard was said to be negligible, so Trevor gave them the "Green Light" to proceed.  I don't see many actual X1000 owners complaining about the Xena/Xorro interface and processor being included, and I for one, think it is interesting.  I am glad that this interesting "user-definable" chip and interface is on my X1000 motherboard.  I also plan to "tinker" with it in the near future.  The one thing that should have been done before the release of the X1000, would be to have completed the porting of the XMOS toolchain to AmigaOS4.x, so owners could have begun playing around with the Xena/Xorro combination much sooner.  From what Lyle wrote, it appears that still more work needs to be done, before a complete set of tools is available to develop for Xena/Xorro, while using AmigaOS4.x, but the tools Lyle has made available to us are enough to begin many projects with.