Here is where having talked to Paul Gentle (@ Varisys) before their connection to A-eon was announced, and since then having exchanged a few messages with Trevor helps.
AFAIK, the PA6T was never truly on the open market before the Apple takeover, at least not at the time of X1000, hence there was no real market functions (supply/demand) for setting the price, right? Apple bought and effectively closed PA-Semi in April 2008. The first "Nemo" motherboard prototype was built mid-2009. By August 2011 the first production run of revision 2.1 for "the AmigaOne X1000 beta test team" was made, more than three years after the CPU's fate was sealed.
"If someone would have told them in advance how volatile the pricing would be, I am certain that A-Eon would have picked a different processor"
:confused:IIRC Varisys had a rather limited stack of CPU's from the 2007 shipment, the price they asked for them was IMHO very high...
Wow! Who have YOU been talking to? Nope,
wrong!
PA Semi announced after the Apple buyout that it would only take orders from established customers. And that at a later date a final order date would be announced.
A-eon's 'purchase' was actually made by Varisys as they had worked with this processor before and were therefore...wait for it...
established customers.I won't delve into your little sermon on market economics, but the last PA6Ts were bought on the open market by Trevors firm because they had exhusted Varisys' supply. And they paid more for these than the original units (as no more were being manufacturer the price went up).
A cost, I might add, that Trevor bore on himself rather than pass it on to X1000 buyers.
I am actually kind of certain that the CPU choice was more of a "hype" thing (the new "Amiga" should have the most talked-about PPC CPU of the time), as well as the "Xorro"/"Xena" was a "hype" thing (Amiga Classic -> "Zorro", AmigaOne -> "Xorro"). It's just a feeling I have. Maybe because I simply can't identify any other logical reasons whatsoever.
Did you even read my post?
The PA6T was Hyperion's preference.
And, at the time, it was the best choice.
OK, if that's true I stand corrected about the engineering costs, but it's still all "wrong", I mean, you can't spend $200,000 on engineering alone for a motherboard that can only sell in a hundred or so units (there was always an upper limit on available CPU's, even if they would reach out beyond the OS4 community)! And *then* another $600-$,1000 for the CPU alone. And *then* everything else on top of that! The thought is absurd, and it's certainly nothing to applaud IMHO, someone should have hit the emergency breaks at the first price quote from Varisys, long before the development even started! And the complexity is hardly anything positive here either, why is a complex design with features that nobody asked for or even can figure out a purpose for better than a simpler but cheaper design like the 8610 concept? Especially when the latter will probably outperform or at least about break even with the former, and cost *a lot* less?
Um, just how much do you think it takes to keep a handful of engineers gainfully employed?
And, have you ever worked for a firm that built custom motherboards (because I did).
These figures are actually rather low.
But had bPlan developed the motherboard "in-house" for "themselves" (read: their own business), they would probably have "priced" the development costs differently, right? Then the *monetary costs* wouldn't have been $100.000 USD any more, it would be more a matter of unpaid work hours and materials. Like Jens Schönfeld probably also does, or Fab when developing Odyssey. Had Fab put a market price tag on all the work hours of his time (mostly evenings and weekends probably, making them even more "expensive") he spent on Odyssey based on the hourly salary of a senior western SW developer, it would hardly be open sourced at just a mere €7,500 EUR. And MorphOS would cost a lot more than the current €50-€111 EUR for the same reason. I have built my own house "myself". Granted, I paid some carpenters, electricians, plummer's, etc to do some (quite a lot) of the work, but I did a very much of the work myself. I did never put a price on my own "development", I invested my own time into the project. Had I paid craftsmen to do every single thing, it would have been a lot more expensive, and even if I could have afforded it myself, the price I would have to ask at a sale in order to not make a loss when/if I sell it would probably have been *way out* of consumer's reach in my area, hence the house would have ended up unsalable. But if I'd sell my house at the current going prices in the area, I would probably make a nice little "profit", since I didn't price my own work/time, or priced it very low, compared to a "commercial carpenter". Whatever I get between what I paid for the fixed costs of having craftsmen doing some of the work, and the final end-user price the new house owner pays me, will be my "profit", the cost of labor isn't there, it was never priced. Or rather, it would have become priced by that "profit", *that's* what my labor turned out to be worth. I actually think most small/independent HW/SW entrepreneurs reason a bit the same way, and are happy with that...?
Stated simply, you appear to have no concept how a business works, how costs are accounted for and handled, or the factors related to pricing a product.
And I assure you that you don't need the 16 credits of Economics I had to understand it (some people understand economics intuitively).
All costs have to be paid for.
You have very limited production.
AND a high cost for materials.
No conspiracy.
If you don't like it, don't buy it.
There is always Acube (on a value per dollar basis just as expensive, but a lower end cost).
Jim
