@umisef

This thread is not about beeing practical or implementable (is the Amiga practical today?).

It's just about fantasy and ideas.

Then, after that step, we have to break down what ideas are good ones.

But thinking big means to allow all ideas - and I mean really all ideas.

No barriers...

_________________
Sometimes we give people a lot of credit just because they're writing nice sentences even if it isn't adding up to much.

If something is new and good and belongs to the software realm, sooner or later it's gonna be ported to other systems/architectures. You can't restrict the software to run on only one HW platform, it's (or it's smarter Google-ish copies) gonna spread 'world wide'. You want to make something extraordinary with X1000 itself, so you must concentrate on HW (betatesters? anyone?), but X1000 doesn't contain anything special, even XMOS card can be plugged to an ordinary PC.

EDIT: nothing special, apart form the fact XMOS is bundled with HW, so this is a way to establish a new hardware base but given the number of A1se/xe/micro/SAM-no-XMOS-included owners and a significantly smaller number of future X1000 owners, XMOS inclusion seems to be pointless

EDIT2: ...but I do want X1000

Last edited by nimrod7 on 10-Aug-2011 at 08:05 AM.
Last edited by nimrod7 on 10-Aug-2011 at 07:59 AM.

_________________
"Marxism must abhor nothing so much as the possibility that it becomes congealed in its current form. It is at its best when butting heads in self-criticism, and in historical thunder and lightning, it retains its strength" - Rosa Luxemburg.

@all
You want a market ?

Target children.

All the best do it.

Think about it for a long while.

_________________
Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway.

I serve King Jesus.
What/who do you serve?

@Birbo

Quote:

So if there is no clear direction, there will never be more coders than today.

You've pretty much summed-up the issue with this equation. What the Amiga needs, and the reason you're asking us to 'think big', is to hopefully divine some cool new innovative direction for the platform.

I've spoken about this before and have been shot down as a 'Negative Nancy'. A-Eon's strategy is "we'll build some expensive kit and we'll see where we go from there". Not much of a direction is it? Some read it as a first step to many other Xn machines, but like you said, as a developer I'm not going to spend what little time I have on something with no clear direction.

_________________
All the way, with 68k

@Birbo

Quote:

This thread is not about beeing practical or implementable (is the Amiga practical today?).

The Amiga today (whatever you consider "The Amiga today" to be) is obviously practical in the sense that it can be built. Mini space rockets are not.

Quote:

Then, after that step, we have to break down what ideas are good ones.

I think peace in the Middle East is a brilliant idea, just like a working fusion reactor.

But.. the Amiga Was a critical part of teh control of teh Space Shuttle, and Deltas and other launches.

NASA even said on quiet days they could put Multiple satelites through a Single A4000.

As for teh fusion reactor.. the rpesent low pressure plasma designs use arrays of small, High Speed magnetic field sensors and actuators to control the plasma fluctations.

The reason persent day reactor plasma lifespans are in teh seconds, instead of micro seconds, is purely due to the speed of response and processing power of the controlling computers attached to it.

At 8ns response time for the XMOS chip, theres nothing else at home within orders of magnitude the speed, and would hopefully be comparable with a lot of commercial stuff.

Now..how did that femptosecond switch work?.. oh that was it, sum of multiphase..Hey.. thats my old comms generator from the 90s.

Im getting old.. all teh weird and crazy ideas Ive had over the eyars are coming true. 8(

But I was hoping to give my stuffaway Open Source 8(

I think I need to add the one line neural net to my data analysis code. Real time raytracing or not, its taking a few seconds just to analyse Kick.rom. 8(

_________________
The older and more respected a scientist is, the longer it takes to prove him wrong.

Does there really need to be anything radically different ? Those have us that have stayed with the amiga and its derivatives have done so because there's something about the system that appeals to us. It has a few different paradigms that differ from other options, and for many of us that's why we're still here despite what others would define as "logic". Slowly but surely we're (os3.x/os4.x/mos/aros) makeing headway towards typical everyday use. Yes trends change as do standards, but there's a basic set of functionality that could be considered essential, and given time existing gaps should fill. Given how liquid trends are, I dont think it hurts too much not to have bleding edge stuff. Standards/trends typically need to time to not only mature, but for it to even be clear if its just a short lived fad or not. Within a few years most of the essential core functionality should be with us, by which time the newer standards that have stuck with us can be focused on.

The strength of amiga os lay in things like the fact the OS doesnt get in the way of the software, responsiveness, simplicity, being in complete control of your OS, arexx, datatypes and so on. This is of course in the eye of the beholder as to just how important such things are, but it is where it's individuality/personality is derived from. No gimmick or software will change that and to me it seems foolish to try to push something that can also be done elsewhere. If amiga is ever going to get even a modest increase in users it'll be through the same sorts of things that made us stick around in the 1st place. Shortcomings aside it really is a joy to use.

Last edited by fishy_fis on 10-Aug-2011 at 08:53 AM.

@RodTerl

Quote:

But.. the Amiga Was a critical part of teh control of teh Space Shuttle, and Deltas and other launches.

What people seem to forget is that if Amiga was great for something 20-15 years ago doesn't mean that it's still great for it. It might to come as surprise but technology has gone WAY forward on that time on other platforms.

@RodTerl

Quote:

But.. the Amiga Was a critical part of teh control of teh Space Shuttle, and Deltas and other launches.

Critical? Hardly, it ran some display board at KSC launch control room. It was as critical as the overhead lights working. I don't remember seeing anything about Amigas being used on the Cape Canaveral AFB side which launches the ELVs.

As far as Xmos, why not just use the USB version?

_________________
Dammy

@Rose

Quote:

What people seem to forget is that if Amiga was great for something 20-15 years ago doesn't mean that it's still great for it. It might to come as surprise but technology has gone WAY forward on that time on other platforms.

Didn't NASA sell of those old A2Ks like nearly five or ten years ago? NASA would be far better off, when they were still in the space business, to duct tape a smartphone with HDMI output to that monitor. Atleast it would have native networking.

_________________
Dammy

@damocles

Quote:

As far as Xmos, why not just use the USB version?

Oh wait, it was just pointed out to me that XMOS Dev board is USB 2.0, nevermind.

_________________
Dammy

@Birbo

I think the Amiga needs to be redesigned pretty much but building on what is its spirit. Such things are its spirit as responsiveness, non-limitating compact intelligent design (for its time), creativity freeing and enabling solutions. I would like to see the name Amiga on something that does things better than they are presently done by the other systems out there; and there are plenty of areas where things are done poorly in the computer world at the moment.

@gonegahgah

For example?

_________________
Sometimes we give people a lot of credit just because they're writing nice sentences even if it isn't adding up to much.

@Birbo

A wheelchair type vehicle for the very disabled with a heads-up display that
can be operated with minimal movement or sound commands that is also interfaced
to a harness for a companion dog trained to accept sonic or tactile commands.

It should also have robot arms and gadgets (one to give the dog treats) for
various needs and situations. Real time interfacing will be necessary in real
world activities.

ZOooooooommmm!

@AlexC

Quote:

What the Amiga needs the most is more coders.

+1

_________________
"Unix is supposed to fix that." -- Jay Miner

@Birbo

Quote:

New example: Using the Amiga as Control-Unit for Mini Space Shuttles. A new product, that allows everyone launching a Mini-Rocket in Space an controling them with the Amiga Control Station at home. (And watching the world from outer space...)

When highest volume website of platform starts to have this kind of posts guess what it does to credibility of platform....

Um, the Amigas were critical to NASA because they handled the downlink telemetry data, tens of thousands of channels, and the astronaughts life support data on the Shuttle.

They were used because they were fast, responsive, could reboot on fault and reaquire the data in seconds, instead of sitting around pondering for an hour as to what went wrong. Same reason the RAF gave for their weather telemetry downlink etc.

These days, if it was properly handled, the main aim of AOS and platform, shouod be security. Ok, so we use standard file formats, but AOS handleds it in a totally different manner.. what attack would just walk into a PC would have problems with the differing structures etc on AOS, If Properly Coded to handle such events In General. run the two machines side by side, PPC core and x86, and now you have to attack two cross checking architectures simultaneously with identical code, or its thrown out. Add an ARM security unit in there for a third manner, a FPGA for randomised hardwaer and encrypted code, and you have the crude beginnings of a very difficult to attack system indeed.

Especially if you then find out the computers lied to you all along, and youve been suckered into a honypot, carefully observed by yet another core that has no direct access from the honeypot at all.

Registers you can write to but cannot read from.. registers you can read from, but cannot write to, which return false values, like the 68000 CPU can use for keyboard access scan etc.. this is how you slowly make things secure.

_________________
The older and more respected a scientist is, the longer it takes to prove him wrong.

@Rose

Quote:

Rose wrote: Quote: New example: Using the Amiga as Control-Unit for Mini Space Shuttles. A new product, that allows everyone launching a Mini-Rocket in Space an controling them with the Amiga Control Station at home. (And watching the world from outer space...) When highest volume website of platform starts to have this kind of posts guess what it does to credibility of platform....

Boosts the FUN value!

@Birbo

I've been thinking along the same lines... which niche could the new X1000 fill that might sell more units and enable another rev of the hardware, so that the price point issue could be addressed. I'm still thinking... and some of my ideas are crap I know, but here goes:

1) low-cost factory/assembly-line automation

The xmos chip combined with some external interfacing solution (i.e. open source hardware solutions could play a part to keep costs down) could enable a company to build a low-cost solution that could compete against the super expensive assembly line automation equipment offered by Cooper tools, etc.

I really don't know how practical this is, but I know Cooper sells very expensive, custom systems that require experts to come in and program the master controller to understand your assembly line process. If the X1000 and XMOS could replace that master controller, there could be a much more cost effective means of tailoring the automation process to meet your needs.

Imagine an open-source assembly line, with the X1000 at the heart, controlling devices that advance the assembly line to the next work station, control the red/green lights, control the ejection process for bad items, turning on and off the powered bolt tighteners at the proper times etc.

Sure these things can be done with specialized solutions. But would the X1000/XMOS loosen the grip so that highly paid consultants aren't necessary to specialize the assembly line process for a particular installation? And maybe there are some other justifications for X1000/XMOS as master controller that I haven't considered?

I had considered this idea as a potential startup opportunity... but again, I'm somewhat out of my element, so I don't know if this even makes sense.

2) Exploit Debian Squeeze/XMOS to make this hardware more attractive for software development

X1000 runs Debian Squeeze, so if you could sell the board to that market, it is still a sale that furthers the Amiga cause. But why would someone buy X1000 for Debian Squeeze when there are so many low cost hardware options that will run that OS just as well?

I've given this question some thought, and so far I haven't come up with a great answer. But I do have a thought, that might trigger others to add to it...

Debian Squeeze is a great OS for build RubyOnRails (RoR), Python and Java-based web applications. What if the XMOS could be leveraged in some way to make it faster, easier to develop these solutions on an X1000 vs other hardware.

What I was thinking about was a kind of XMOS-based equivalent to Microsoft's Detours solution. Basically, the XMOS would be a container for your software that would instrument it, facilitate integration with and simulation of external systems, facilitate Unit or BDD testing and exception handling, and trap and halt runaway software that sometimes happens during the development process.

It is a young idea and not one I'm completely sold on. But hopefully it is food for thought about how XMOS might play a part..

3) XMOS as runaway process trapper

On Windows this would be more useful, and viruses, root kits and other nasties are hard to monitor from a compromised processor. But an external co-processor/XMOS that could recognize and shutdown a misbehaving application or rootkit would be a useful way to reclaim a system that is infected. Or even just to trap and halt a runaway application, that you can't otherwise figure out who is responsible.

On Linux, this idea is much less useful, however. I've had runaway processes on Debian that have consumed so much CPU it too me 15 minutes to type the "kill" command. An XMOS 'top', that automatically recognizes and stops processes that are out of control might still be a useful TSR type monitor.

Although I might be able to find this as some open-source tool, if I just looked.. Just thinking out loud here, so could be brain dead idea.

4) Kid-targeted Home automation toolkit

Kids are fascinated with home automation, and the XMOS chip, combined with some external systems integration and a home automation toolkit would be something kids would be thrilled to show off to their friends and parents.

This goes to an earlier thread I wrote some years ago. If a lower cost, X500 with XMOS home automation toolkit that kids could master and impress their parents and friends with could be built, I think there could be a market for something like that. That kids could master and learn on, instead of being another mind-numbed web surfer, gamer, and mp3 downloader like all their friends.

A follow-on idea to this would be a cell-phone games development kit, that that same kid could build Hollywood, python or ruby games on the X1000 and deploy those games to his friends so they could play them on their iPhone. Sure you could do that on other platforms... but if the X1000 or its lower cost successor could make computer programming fun and accessible for kids again (the way Amiga and Commodore did for so many of us back in the 1980s), these ideas could become bigger draws to bring kids to the platform once it becomes more cost effective.

But of course, this would require a second rev of the hardware to produce a lower-cost, kid targeted version of X1000/XMOS. Probably an X500. Still, the software could be developed meanwhile to enable a co-release.

Nothing else comes to mind at this time.. but maybe one of these 4 ideas could be a useful starting point for conversation?

Last edited by jingof on 10-Aug-2011 at 09:20 PM.
Last edited by jingof on 10-Aug-2011 at 09:14 PM.
Last edited by jingof on 10-Aug-2011 at 09:13 PM.
Last edited by jingof on 10-Aug-2011 at 09:09 PM.
Last edited by jingof on 10-Aug-2011 at 09:07 PM.

_________________
Vic-20, C-64, C-128
Amiga 1000, 3000
AmigaOne X1000

@jingof

Quote:

1) low-cost factory/assembly-line automation

Forget it. There are low cost automation solutions and companies wont buy products sold on obscure websites. It is really tough business and nobody is going to buy your stuff anyway unless you can guarantee your products are supported next 10 years at least.

PA6T alone (being EOL) makes it unsuitable for industrial use.

_________________
Amiga Developer
Amiga 500, Efika, Mac Mini and PowerBook