> While I think the PA6T is one of the most interesting chips out there
> I think there a few things to consider: It is a dual core chip.

There's a recent statement by Trevor Dickinson which seems to indicate that the AmigaOne X1000 will initially use PA6T CPUs with only one functional core (and with the second one broken and thus cheaper to source?):

"the intention is for multi processor support to be added although the first AmigaOne X1000 will not support this."
http://www.commodorefree.com/magazine/vol4/issue43.html#ARTICLE3

Creepy that you posted that as I was just reading that interview last night.

I didn't draw that inference from that statement, but its a good guess as to the reason for not supporting both cores on the X1000.

Otherwise wouldn't "although the first AmigaOne X1000 will not support this" be "although at first the AmigaOne X1000 will not support this"?

> I didn't draw that inference from that statement

How did you understand it when you first read it?

> but its a good guess

Yes, I think so :-) But maybe it's meant to say something completely else and was just very badly worded.

> Otherwise wouldn't "although the first AmigaOne X1000
> will not support this" be "although at first the
> AmigaOne X1000 will not support this"?

Hmm, I can't really see how the last would make a difference to the first regarding my inference. Care to elaborate?

I assumed that he might be stating that dual core operation would not be supported at all on this AmigaOne system (only on future motherboards).

It had never occurred to me that the processors might have one dysfunctional core.
That's a brilliant theory.

I'd just assumed Trevor was either misquoted (and that they weren't supporting dual core operation yet) or that they just didn't want to attempt to support dual core operation on this version of the X1000 motherboard. A hardware issue hadn't crossed my mind.

An alternative explanation could be that the current MB doesn't correctly support both cores. But your pointing to hardware, no that hadn't occurred to me.

Good thinking. That was my point. Any other questions?

> I assumed that he might be stating that dual core operation
> would not be supported at all on this AmigaOne system (only
> on future motherboards).

This would mean that their "dual-core processor" marketing has been bogus all the time. That would be really bad.

> It had never occurred to me that the processors might have one
> dysfunctional core. That's a brilliant theory.

...unless Trevor didn't actually write what he meant to express ;-)

> I'd just assumed Trevor was either misquoted

I guess the interview was done in written form, which makes misquoting hard (if not done deliberately that is).

> or that they just didn't want to attempt to support dual core
> operation on this version of the X1000 motherboard. A hardware
> issue hadn't crossed my mind.

How could it be anything else than a hardware (chip level or board level) issue? Do you mean firmware level maybe?

Quote:

They only sold the ODW later. The first few batches of PEG-2's were quite a bit more. I paid over $1000 for mine, before the "OWB configuration existed". I think I got the 2nd or 3rd batch.

Also the PEG-2 was produced in at least one order of magnitude higher than X1000. X1000 is 100 to 125 units for beta test program.

PEG-2 production was >1000 units I believe, maybe far more. I don't know numbers have never been released.

But the same may hold true. You could find the end-user X1000 boards are a bit cheaper. Also as time goes on, companies need to reduce the price to spur sales, as the technology becomes outdated. (Not so much in the AmigaWorld though, since nothing really moves so fast in this market to make the older stuff obsolete...)

> You could find the end-user X1000 boards are a bit cheaper.

"we are making 100 Rev 2 motherboards available to Beta testers under a special discount program."
http://www.a-eon.com/news.html (under "27 May 2010")

Quote:

Yes. Final pricing hasn't been announced yet. But the final systems will be full computers, not just boards/cpu/ram.

We will see!

Yes, I assumed that they were unwilling or unable to overcome software incompatibilities or a firmware issue and therefore they were putting off addressing the issue for the current product.

And, yes, regardless of the reason, it seriously affects how they've marketed the system.

> But the final systems will be full computers, not just boards/cpu/ram.

I don't think that's what "special discount program" means here. Paying less for less doesn't classify as a discount in my book.

> I assumed that they were unwilling or unable to overcome
> software incompatibilities or a firmware issue

I still don't get what "software incompatibilities" would mean in this context, other than a firmware issue.

> If MorphOS moves to the X86 platform, I will stay with Windows.

"If the MorphOS developers decided to move to X86, I'd probably follow."
http://www.amiga.org/forums/showpost.php?p=577460

Seems you changed standpoint, eh? ;-)

Quote:

There are no hidden microphones in you house, Jim. There's no CIA satellite following you from the sky above. It's the mighty Andreas Wolf human surveilance system.

Amazing attention to detail, Andreas! Congratulations. I'd be honoured if followed the same way.

> I'd be honoured if followed the same way.

I didn't follow Jim per se but rather an amiga.org discussion thread he took part in. And I did follow you to imxcommunity.org, but I won't deepen in that site due to my general lack of interest in i.MX :-)

Quote:

As I would prefer to continue to use MOS, rather then switch to AROS or rely solely on OSX or Windows, in a word - yes.

Someone who's opinion I trust swayed me on this point.

However, I don't think we'll see the need to do this any time soon. First G4 Powermacs, then Powerbooks, then G5 Powermacs. After that, who knows. If we're lucky maybe there will be another PPC platform to move to.

If not, then X86 (damn) is a possibility. That isn't a decision I'm happy with (too much Deja Vu).

BTW - If you follow my comments on Amiga,org you're likely to be confused. Lately I've taken a position which POs Amiga purists. Since they're so offended by Barry Altman's Commodore USA, I've tried to point out that the Amiga name is nothing more than a marketable asset.

[ Edited by Jim on 2010/9/10 2:17 ]

Quote:

I'm not worried about Andreas' incredible recall. The only thing that concerns me is my own lessened focus on the topics discussed and the reduced quality inherent in my own postings recently.

I felt better about my input on this forum when I was more involved in researching/studying topics related to it.

>> Seems you changed standpoint, eh? ;-)

> in a word - yes.

Fine :-)

> There's a recent statement by Trevor Dickinson which seems to indicate that
> the AmigaOne X1000 will initially use PA6T CPUs with only one functional
> core (and with the second one broken and thus cheaper to source?):
>
> "the intention is for multi processor support to be added although the first
> AmigaOne X1000 will not support this."
> http://www.commodorefree.com/magazine/vol4/issue43.html#ARTICLE3

While that one by Trevor Dickinson is still an unsolved mystery, I came across another puzzling statement, this time by journalist impersonator Jeremy Reimer:

"The X1000 motherboard has a dual-core PowerPC family CPU (it is actually closer to IBM's POWER line than the PowerPC), PCI Express slots, and a new custom chip called the XMOS that is like a programmable vector processor."
http://www.amigafuture.de/kb.php?mode=article&k=3628

What could the part in brackets mean technically, considering that IBM POWER is actually an implementation of 64-bit PowerPC ISA?

(Commenting on nonsense like "custom chip" and "vector processor" isn't even worth the hassle I think.)

Having experimented with the XMOS line of processor, that comment about it being a programmable vector processor is laughable. I think your term 'Journalist Impersonator' is remarkably accurate. The man in question isn't apparently aware of what the processor on the X1000 mainboard is..

> I think your term 'Journalist Impersonator' is remarkably accurate.

Having read the whole "article" it seems like an entry to a "most factual errors per word count" contest. And I think it would have a good chance to win the trophy ;-)

The "article" was indeed "funny" in a way, probably even more innaccurate than what Jeremy Reimer already got us used to when he was writing for Ars Technica. The "Galleon3D" snippet was amusing, for instance. I won't comment about the bit about the XMOS chip being a "custom" chip: it's really not worth spending words...

Kind regards,
Andrea

Quote:

Ouch... I had Ars Technica high in my list, as I've red some amazingly detailed articles over there (don't remember author, though). What's your take on that site?

> The man in question isn't apparently aware of what the processor
> on the X1000 mainboard is.

I guess that even Reimer may get a clue now. At least Holwerda obviously has:

http://www.osnews.com/story/24151/A-EON_Reveals_AmigaOne_X1000_Processor_Through_Cake ;-)

There're some other people out there who I'm eager to see if they'll stick by their word:

"If they have a P.A Semi chip with Altivec and even if it's only a 1.6GHz version I will apologize in every forum about being skeptical, about calling the X1000 an over-hyped product and about calling the AOS4 fans naive." (*)
http://moobunny.dreamhosters.com/cgi/mbmessage.pl/amiga/169942.shtml

Anyway, I'm eager to hear the "quite a story" :-)


* Edit: So far he seems to have forgotten his promise he made 9 months ago:

http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=32910&forum=33#593514

(and he even keeps on with his "according to records the first CPUs in the A1X1K were used chips" story when in fact A-Eon said it was *sample* chips they used in the beginning)

[ Edited by Andreas_Wolf on 2010/12/21 15:11 ]

Now fully official:

http://www.a-eon.com/news.html

Still, no word on who the PA6T supplier, who A-Eon said they have their NDA which prevented them from revealing the identity of the CPU earlier with, is/was. So I guess there're still some information to be revealed to get the whole "quite a story".

Nevertheless this is good reading though! my question is without touring a.org/aw.net is how many units and how long of a production will this be?

The PA6T CPU paves the way for the introduction of SMP with AmigaOS4.x. Also check out the new multi-core CPUs being developed by Freescale, in particular the 64-bit dual core, 2.2 GHz P5020 and the 8-core, 1.5GHz P4080.

It would also appear they have a future plan together dispite what many have suggested, so with the socks/cake eating and all, what's ours?