> go to bPlan, have them desing a REALISTIC computer for you, and insist in
> making them solder down one of those rare Transputer things, to have that
> "unique" spec no one else has.

I see no way how bplan could have developed a PA6T and XCore based board significantly cheaper to build than Varisys did. Likewise I don't see why Varisys couldn't or wouldn't have developed an MPC8610 (or similarly priced processor) based board (with or without XCore) if someone had commissioned them to do so.

"The following core designs are on the shelf to expedite the custom design cycle: PA Semi PA6T-1682M Freescale PowerPC MPC8641, MPC74xx, MPC86xx [...]".
http://www.varisys.co.uk/services.html

Bottom line: I don't see anything bad per se in choosing Varisys over bplan.

> the Efika came. It's true it took a very long time to port MorphOS, but I think there were
> also hardware problems that delayed it (see that "B" letter in the CPU name).

Didn't the MorphOS port to the Efika 5200B commence not before Genesi paid for it in January 2007, i.e. when the transition from MPC5200 to MPC5200B was already done?

> it was done ten years ago! It was called Thendic

Ten years ago there was no Thendic in a MorphOS context. They came into play a bit later.

Andreas has a very good point here. Even before A-eon's announcement about their relationship with Varisys, if you'd given me the task of finding a company to design a MorphOS motherboard Varisys would have been one of the first companies that would have come to mind.
Their website mentions custom designs available and in addition to the PA6T the MPC86XX is mentioned.
It wasn't posted here, but a long time ago I'd been discussing with Andreas the potential of a 8640/8641 based motherboard using an AMD SB600 Southbridge (the Southbridge that is on the X1000 MB).
The problem is we're talking about a LOT of money (needed to finance such an undertaking).
Bplan and Varisys won't do this kind of thing out of the goodness of their heart.
And I'm confused as to everyone continues to focus on Bplan (perhaps their past connection with our market) when there plenty of talented alternatives that could build a new PPC board for our market.

I just sent this to Altman (to see what happens):

Altman,
Why, as an Amiga fan should I buy your product? What does it have to offer over an Amithlon equipped PC I can get for $10? Furthermore, I can equip said PC w/ AROS, yes, the very same OS your PC does, for free! Also, how do you discredit the sources saying Amiga Inc. (and by extension, Bill McEwen) have no rights to the Amiga Name.

Yours Truly,

Redfoxva

P.S. My balls are not for your enjoyment thanks very much

Quote:

I think they won't even make ends meet on this. And obviously they don't thinks so themselves, hence the pre-payment scheme, even for *beta pre-release hardware*. But *I do* think it's obvious they have calculated on hefty profit margins per unit. Multiplied by dreams. And dreams are infinite...

Quote:

Thanks, I hadn't had a good laugh today yet!

BTW - I haven't posted his phone number because he only asked that I relay it to one other person, but don't be surprised if he offers to talk to you directly.

Quote:

Aye!

Quote:

Personally, I think the Natami is very interesting. For some "nerdy" POV only. I will probably buy one, when (if) it gets released.

However, I realize that it offers:
1) more than the retro crowd would require (hence it's not suitable for this crowd), and
2) less than the "NG" crowd requires (hence it's not suitable for this crowd either)

"Stuck in the middle"...

Quote:

At least some key MorphOS developers acknowledge that PPC wouldn't be the choice today if they had known back then what they know now...

But under the current circumstances, the optimal decision would be to support the best of the most widely spread PPC hardware out there.

Which the MorphOS team obviously aims to do.

I couldn't agree more!

Quote:

Well, that's the question, isn't it?

The OS created by some of the people responsible for taking Amiga beyond "a 68k based games console with a keyboard" (RTG, PPC, USB, MUI, RTA, and so on), the fastest, most Amiga compatible OS with the best features, *or* the OS created by the twin brothers taking a crash course in OS design on their local university, utilizing *the left-over* Amiga standards, incorporating Linux bloat as soon as they can, etc. Hmm, difficult decision...? I think not...

Quote:

What we are talking about (in the recent threads about "Commodore Amiga"), is really *the regular AROS system*, right? "PC", yeah? What's new?

The new thing is that the regular AROS system (if AROS will be made to support this particular HW) could now be called Commodore Amiga!

Personally, I have absolutely *ZERO* problems with that! I emotionally abandoned those brands a decade ago, but I would think it to be cool if some new Amiga product would materialize using these brands!

Nothing that would affect my views and devotions to MorphOS though!

Quote:

No?

Quote:



Quote:





Quote:

Bring on the popcorn!

Quote:

What you said doesn't make sense. You say they don't think themselves they would make ends meet, then you say they priced in a hefty profit.

So which is it?

Sure, there may be some profit priced in (that is what business is about - making money), but I doubt it's hefty. It's probably enough to make a small return, and on the end user machines make enough so they can make some money and the dealers, too.

At least they most likely are not planning to lose money. But they have every right to make a profit.

I would have to disagree that they are planning on making a profit. Hoping to make back their investment of time and money is more like it and trying to minimize the amount of their loss is most likely.

The investment to create this custom computer has been large and the number of sales might be very small. So small that A-Eon may not even come close to recouping their investment capital. We will just have to wait and see how many pay to beta test it and how well it sells to the Amiga public after the beta test program is over.

I hate to be a pessimist but given the facts, it doesn't look good for making any profit.

> the twin brothers taking a crash course in OS design on their local university

Maybe you mean it and are only expressing yourself in a strange way: The Friedens studied computer science at the University of Trier and attended an OS design lecture as part of their schedule. "Crash course" is an improper term to describe a regular university lecture, I think.

> I would have to disagree that they are planning on making a profit. Hoping to make
> back their investment of time and money is more like it

To just make back their investment they'll have to make a profit on each machine sold.

Your perception that your definitions are always the only ones recognized, is incorrect.

>> To just make back their investment they'll have to make
>> a profit on each machine sold.

> Your perception that your definitions are always the only ones recognized,
> is incorrect.

Which "definitions" are you talking about specifically?

Why do you guys bother to mail or phone this Altman guy? If you don't like the thing, don't buy it. It is their business, let them get their bloody nose themselves or eventually get rich with their plan. I couldn't care much lesser. Heck, I am even not much entertained by the aw.net threads regarding this topic...

At least they are giving the guy who wrote "Over the Edge" some new stuff to write a sequal.

Quote:

Title: "Rotting in the canyon"

Quote:

Actually it is .....

If their whole study had been about OS-design they would (could) have come out as experts. With single (half year ?) lecture they only learned what to they need to learn to design an OS.

Comments made on ann.lu and early design concepts of OS4 clearly showed that they lacked the experience to see the difference between "cool concept" and "will work in real life".

>> "Crash course" is an improper term to describe a regular
>> university lecture, I think.

> Actually it is .....

Yes, improper that is.

> If their whole study had been about OS-design they would (could) have come
> out as experts. With single (half year ?) lecture they only learned what to they
> need to learn to design an OS.

I never said that their attendance of one (or more) OS design lecture(s) made them experts on that topic. I merely objected to the usage of the term "crash course" as that clearly doesn't fit into the concept of university lectures at all. On the one hand you have one (or more) basic lecture(s) on certain topics and on the other hand advanced lectures which build upon these basic lectures and go deeper into different sub-topics. I have still no idea how an alleged "crash course" at a university would relate to that.

Quote:
I'm curious - did you go to university? I took a couple of courses on operating systems back in the days, and I think "crash course" covers them quite well.

[ Edited by kolla on 2010/9/6 4:05 ]

> did you go to university?

Yes, that's where my experience regarding university lectures comes from. Although I didn't graduate in computer science but in another area my schedule contained some compulsory and elective computer science lectures as well (no "OS design" though ;-)

> I took a couple of courses on operating systems back in the days,
> and I think "crash course" covers them quite well.

It seems the quality of lectures can vary between universities ;-) Just to present one random definition of the term "crash course" (note the text in brackets):

"a rapid and intense course of training or research (usually undertaken in an emergency)"
http://www.thefreedictionary.com/crash+course

Quote:

Sure, but I used the generic term "computer" with intention. Why targetting that specific CPU? Who cares about CPUs nowadays? Or it's a firm sales guarantee?

Also, I mention bPlan because, what a coincidence, Gerald Carda and Thomas Knabel designed glorious hardware for the Amiga back in the day. I thing that would indeed be an incentive for die hard amigans, if used in marketing. In fact, it would mean the only connection to that glory days every one wants to take a buck out of.

Quote:

Indeed. Is the PA Semi part the thing that's making the X1000 so bloody expensive? Then, it's not an advantage, it's the biggest mistake of the whole project. Is it a very fast CPU? Heck, who needs that in an Amiga? To brag about againts PC friends? That'll be so childish it hurts even thinking about it.

Quote:

Another mistake. Who really believes this weirdo coprocessor is going yo mean anything useful? You are going to plug in a GPU a dozen times more powerful anyway. Oh, but you can't code it to its limits, pity.

Quote:

I wonder about how much time they took to decide the hardware specs. Perhaps it was in a ten minute rush, on one of those nostalgia afternoons with other amigan pals. It usually ends up in something like "grab the biggest there is, whatever it costs".

Quote:

Nor anything good. No amiga market experience, for a start. But yes, of course, there are many computer design firms. But bPlan's experience might have been the only sane part of the project.

Quote:

I don't know, but I seem to recall comments from Matt "Neko" Sealey on "powerdeveloper.org", saying they indeed had problems with the first version of the MPC5200. It's possible I'm putting those into the MorphOS port, and they weren't.

Quote:

Yes, MorphOS started as a pure software project for phase 5's PowerUP accelerators (made by the same people that formed bPlan afterwards). Then, bPlan made the Pegasos computer, and for a while, this Pegasos/MorphOS combination existed withoout the Thendic term existing. Soon, Bill Buck and Raquel Velasco (BBRV), known from the VisCorp affair, came into scene. But could it be that bPlan made the Pegasos out of Thendic's wallet, to some extent?

Quote:

Indeed. It looks as if we just wanted to be cruel to some guys. That attitude come out of frustration many times, and I don't see it much in this MorphOS land. In that regard, we're no amigans.

@ Andreas_Wolf

A lecture does not have to be officially labelled "crash course" to become one.

This depends entirely on the intention of the attendees to use the gained knowledge in real-world situations despite the fact that most people would consider the completed lecturing to be insufficient (to complete the tasks at hand in an adequate manner).

If somebody attends a cooking course for beginners and then decides to open a restaurant and do all cooking, it is fair to call the cooking course a "crash course" as long as you specifically refer to the education of this particular individual.

> I used the generic term "computer" with intention. Why targetting that specific CPU?

You don't have to, albeit it's fast in terms of Power Architecture (but expensive also). For other Power Architecture CPUs see the sentence I wrote next, starting with "likewise...".

> Who cares about CPUs nowadays?

The OS does, as well as its transparent m68k emulation. That's why it must be a Power Architecture CPU, for now at least. And additionally, the CPU shouldn't be too underperforming. I consider the performance of my 1.5 GHz G4 the absolute minimum for my needs in terms of desktop computing (while a netbook or similar could go as low as an 800 MHz e300 to satisfy me).

> I mention bPlan because, what a coincidence, Gerald Carda and Thomas Knabel
> designed glorious hardware for the Amiga back in the day. I thing that would indeed
> be an incentive for die hard amigans, if used in marketing.

Remember, these "die hard amigans" chose relabelled Teron boards over bplan's Pegasos. I doubt it would make any difference to them if a new "AmigaOne" board was developed by bplan instead of Varisys as long as it (or the full system containing that board) is labelled "Amiga something". And honestly, it wouldn't make a difference to me today either (except for the name part of course, which has never mattered to me). A commissional work is really just that, no need to get emotional. If Varisys is as good as bplan at designing a board to the predefined specs, then I see no advantage in not choosing Varisys. And as I can see, Varisys has past experience with PA6T and MPC86xx, which bplan may not have.

> Is the PA Semi part the thing that's making the X1000 so bloody expensive?

In the interview you've linked to it's said that "just the CPU is 400-450 euro". So it definitely has it's part in the high price of the X1000. But I must admit that I really don't know why the beta Nemo board alone (including RAM) is as much as 1400 EUR, which is a discounted figure even (according to A-Eon). Can we assume that the insane price is mainly due to A-Eon's high profit margin (in addition to the small batch size)? I don't know. In any case, I doubt that the price is due to Varisys (and not bplan or else) being the board developers.
Let's imagine for a moment that they'd get the PA6T part for free. Then the board would still be barely below the 1000 EUR mark. Still no sane figure for a mainboard with RAM.

> Is it a very fast CPU? Heck, who needs that in an Amiga?

If you intend to use this "Amiga" as your everyday computer (like I do with my MorphOS Mac mini, which is easy because I can temporarily switch to MacOSX or Linux for applications MorphOS is lacking) then a CPU as fast as possible can't hurt for sure. See my arguments there:

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=7186&forum=11&post_id=74398#74398

>> with or without XCore

> Another mistake.

Yes, probably. I only mentioned the XCore because you wrote: "insist in making them solder down one of those rare Transputer things, to have that "unique" spec no one else has".

> I wonder about how much time they took to decide the hardware specs.

According to Trevor Dickinson at the VCF in June, Hermans got the very rough idea for an own hardware at (or shortly before) the time Hyperion was sued by Amiga Inc., i.e. around or before April 2007. About one year later they decided for the CPU and the (still unknown to us) CPU supplier, who suggested Varisys as a design company with experience regarding that CPU. In early 2009, Varisys was commissioned to develop the board according to the specs and the real development began.
Just for comparison: Sampling of the PA6T started in February 2007, volume production in early 2008 (at TI's fabs, I believe), but only for a very short time, because in April 2008 P.A.Semi was purchased by Apple. In July 2008 it became known that Apple would supply the PA6T for at least 3 years (and upto 5 years) and explore selling the design to a third party after that time. In September 2008, Apple started volume production of the chip again.

>> I don't see anything bad per se in choosing Varisys over bplan.

> Nor anything good.

But there is: past experience with the PA6T.

> No amiga market experience, for a start.

That's not needed for the company designing the board as a commissional work according to predefined specs. The "amiga market" is to be entered by A-Eon and AmigaKit.

> bPlan's experience might have been the only sane part of the project.

Why do you think so? AFAIK, bplan don't have any past experience with the PA6T.

> I seem to recall comments from Matt "Neko" Sealey on "powerdeveloper.org",
> saying they indeed had problems with the first version of the MPC5200. It's possible
> I'm putting those into the MorphOS port, and they weren't.

As you probably remember, in early 2005 there was an internal MorphOS port to Freescale's Media5200 system with Lite5200 board, which at that time had the MPC5200 (without "B"). The switch from MPC5200 to MPC5200B took place in the second half of 2005, and the Lite5200 became the Lite5200B. There's a chance that MorphOS had the problems you remember on the Media5200/Lite5200, and not on the Efika 5200.

> for a while, this Pegasos/MorphOS combination existed withoout the Thendic term existing.

But only as an announcement, not as a product. There was no Pegasos/MorphOS combination as a product before Thendic entered the game. And even after that, it took a while for the Pegasos/MorphOS combination to become a product.

> could it be that bPlan made the Pegasos out of Thendic's wallet, to some extent?

"Made" as in "manufactured in more than single prototype batches"? Sure. "Made" as in "developed"? Not quite. Thendic began financing bplan, who had announced the Pegasos one year before, in November 2001, presumably after them visiting the AMIGA 2001 show in Cologne where bplan were publically showing their non-prototype (i.e. microATX) Pegasos board for the first time (and running MorphOS). I'm sure though that without Thendic's (and later Genesi's) money bplan wouldn't have survived the Articia/MAI/April mess which began in September 2002. They would have been economically forced to either close down without releasing the Pegasos at all or release the Pegasos without the expensive but required April fix (and close down subsequently due to warranty claims). Thendic/Genesi were the ones making bplan and the Pegasos survive this mess, no doubt about that.

Quote:

Understandable. But also, one can't choose the amount of bloat in the information received everyday. For example, many times I'd settle with a slower computer, but some bloated document gets in, or some inefficient process has to be done, and I need more power than neccesary, if things were better done.
Think about that guy stuck on an error message, and instead of copying and pasting the text, sends you an e-mail that contains a two megabytes file that contains a Word document that contains an uncompressed screen capture that contains the error.

That's why the phrase "I need a faster computer" has always made me smile. With nostalgia.

Back on topic...

Quote:

Sure. In fact, I was trying to point out that the chosen CPU is a mistake. Powerful (for what?), but very expensive, hard to get, and already at its end of life: I forgot something that, thankfully, you mentioned:

Quote:

Where is this CPU used in volume? I thought it was kept because of some contracts with US army, that have systems that use it, and thus the manufacturer is forced to keep the part in stock.

Quote:

Sure? Would love sources about this.

Quote:

Indeed, there's something very wrong about this. It's almost impossible not qualifying the X-1000 as a cash-in trap.

Quote:

WHAT? That would be a computer designed by a lawyer!? Just joking, I've seen that "very rough idea" words. But couldn't resist... Now, really, that would explain why the specs can't be the ones chosen by a sensible engineer.

Quote:

That's my way to insist in taking that fantaaaaastic CPU out of the picture.

Quote:

That's why I mentioned Thendic/Genesi in my previous post. That was a way to do things:

1.- Realistic specs.
2.- Proven Amiga experience.
3.- Both hardware and software running.

Not to mention something still more important: It was TEN years ago.

Quote:
PA6Ts are used by many mil products (more than 10 different ones). E.g. by Lockheed Martin or Raytheon. There is demand for several 10s of thousand of these processors. No big volume production, but still volume production. The different wars and fear of terrorism/phantom menace keeps demand steady.