Given x1000 news is everywhere.....but nonchalance prevails

[DAX]

@The BilgeRat
But you can drive similar comparisons of that same HP workstation with iMacs or $5000 Alienware machines, where if you buy the same pieces yourself, you can get the same specs at half the price (I could build up another godzillion comparisons) still they don't run AmigaOS do they?
This is a machine for AmigaOS those interested will buy it, those not interested will not (no need for apple to orange comparisons).
Talking about the price, it is not known yet, I saw much fuss done for a post made by user Tommo1975 but nobody cared for user djrikki who in the very same thread says he heard "less than €1500" and with all THAT CONFUSION where everybody is saying they couldn't clearly hear a thing, it's quite understandable.
Anyway, the X1000 is aimed at developers and power-users according to A-Eon, meaning they don't plan any mass market sales for now .

[DAX]

Double

[DAX]

about Aros/WinUAE, the first is not very usable at the moment (I have icaros desktop worry not), and the second does not empower its users at doing modern stuff on Amiga.

I won't start talking about modern  browsing, FLASH, what  kind of 3D is possible today etc. etc. (and don't start with "i  do that in windows" silly argument, the point is doing that ON AMIGA) I  will instead point your attention to the fact that after porting OS4.1  to the X1000, they will start modernizing the OS in order to take  advantage of it.

Yes we all have our estimates, but they said they will look at utilizing  what's on offer when the OS port is done so that's all there is to it  for now.
I believe that there is people here that is trying to push into everyone  else throat that this was going to be an A500 MAINSTREAM cheap revolution, and that they  failed to deliver, however this was never announced as such (machine for devs and Power users they said).
They are not gonna fail because there are enough AmigaOS users  interested and that's the only crowd this machine is aimed at at the  moment.

When the OS will be more mature and if the sales of the X1000 will go as  planned with the above indicated crowd, they might decide to do  something a little more "mainstream" ("Amiga mainstream" not "mainstream"  in general) probably a couple of years from now.

[DAX]

It is really the end of the road for some people when a product like  this disappoints.  

Not exactly, the "product" is not finished yet, it was brought there to show its tangible existence.
At any rate it only managed to disappoint other "factions" followers, which was to be expected.

[DAX]

I wouldn't be making too many comparisons in regards to usability of OS4 running on a A1X1K right now vs Icaros Distro. ;-)  Second does empower what the vast majority of "amiga" users want, to play their old games.




Since it took, what, six years to port AOS to PPC, I wouldn't be holding my breath for a 64 bit, SMP, full MP OS4 in the near future.  Does OS4 currently have 2D/3D drivers?

Usabilty will be up and running soon worry not, and about the time it takes to develop 3D drivers, Hans De Ruiter  responded to Fab asking the same question a few months back like this: the good thing about writing them for a family of similar cards (ATIR500/600/700) is that you don't have to do them over and over again from scratch. He will finish his R700 drivers, soon. Time doesn't go back, much work has been done, they will build upon it.

[DAX]

@piru
OS4 developers confirmed Hans is working with them on the X1000 R700 drivers (aside from his old personal project).

[DAX]

@dammy
If we look at how the HW world changed from 1980, I believe the X1000 is as "Amiga" as it gets from a HW standpoint.

Fabled chipsets are no more (even Commodore engineers and J.Miner himself suggested expandability without any chip sets) , replaced by what engineers do on GPUs today, moreover we are entering the era of heterogeneous computing and the X1000 lends itself well to this concept by allowing integer crounching power expansions via XMP and Floating Point expansions via GP-GPU (yes both not avaialble at this stage, but considering the actual software doesn't need much power it is only an option for the future).

Nemo is no off-the shelf mobo, it took a lot of effort to develop, you can read about it here (from page two).

As for the rest, yes, you get off the shelf components with some Amiga customization on them, which is a good move at this early stage.

[DAX]

With an embedded (SoC) CPU?  It's the anti-Amiga 1000 then.



But there is nothing on the A1X1K's mobo that someone else couldn't design and sell.  It's a collection of standard mobo parts for a SoC CPU.  So what hardware on the A1X1K has "amiga customization" on it, the ATI gfx card has some extra Amiga logic circuits on it?  how about the RAM, extra Amiga logic circuits or chaches?  How about the disk drives, special Amiga caches on them?  Adding decals or silk screening does not make them customized in my book, just rebadged.

Soc is an intelligent way of integrating memory controllers and other circuits inside the CPU, and quite frankly, it's the way of the future, Intel is going the same route adding stuff "inside" the CPU all the time and Soc is their future as well (if you care to disagree call their engineers and explain them why everything should stay separate instead).
In the X1000, the CPU should probably be this one check the whole paper and you will see it is quite similar to other very modern designs: LINK

As for "there is nothing" on the mobo that other could not manifacture and sell, I disagree 2 folds, the first is that the way Xena/XMP and system communicate is a proprietary design Verisys will patent for sure, moreover there will be no home computer as "different" as this one from any manufacturer, which also includes the fact of using a PPC CPU in a newly born Home Computer (if you know of a project feel free to fill me in).

Basically it is as different as it gets for modern standards, the reasons of why this is the case are further explained below).

The second point is that my A2000 uses a ton of off the shelf parts too (the case was even recycled from another C= project) there is not a special part inside that could not be manufactured by other vendors if we exclude the obsolete OCS. The curious thing is that J Miner himself replyed that for pro apps on his 2 A2000s he needed RTG cards and couldn't understand why they didn't do anything like it in the A4000 (in a famous old interview at CES) OCS/AGA were already obsolete in 1992 and AAA was admitted to be as being too little too late.
Commodore engineers wrote a long paper as to how Amiga would develop after AAA and if you read it you will see the X1000 is what they describe (I kid you not).

[DAX]

For the Atom and to a lesser extent their i3/i5 chips yes, more things are being shifted over into the cpu. However, these are primarily aimed at the desktop/nettop/netbook markets. The PA6T was designed from the start as an embedded microcontroller along the same lines as the ARM and MIPS families.

Not at all, they already have designs with additional parts being integrated in the CPU for TOP models as well (and where do you think this continuous process will lead in the long run? at having everything separated again? get real please).

I'm sorry what? A bit of glue logic that anyone with access to an fpga prototyping kit could knock up in a few weeks? That's your argument? Dear gods sir you need to lay off of whatever the hell it is you've been huffing. It shares the vast majority of its arch with a bog standard PC. It has a circa 2002 desktop performance SoC sat on top running the show at heavy duty workstation/gaming rig prices.

A 2Ghz CPU has the same benchs as a similarly clocked Core2Duo (as posted by Karlos) if you had that power in 2002 lucky you.
And the Nemo board is still the most "different" motherboard you will ever see in a personal computer from today to eternity, might as well like that a little bit?
 

Only if you're blind. I've read the paper, it includes a move to double up the AAA chipset, adding PCI in place of Zorro and using a PA-RISC cpu.

In short a road map to use as much off the shelf gear as possible as a means to keep down costs associated with development. Note too that AmigaOS was EOL. Commodore were going to be running WindowsNT on those PA-RISC boards.

At no point was PPC a feature on C='s roadmap.

I don't remember anything about PC-Risc and doubling of AAA, which means you are referring to something earlier.

If they had survived long enough to follow that road map and were still around today, they would have ended up being just another beige box x86 supplier as HP dropped the PA-RISC some time back in response to Intels request for the adoption of the abortive Itanium cpu.

probably x86 and like Fujitsu and Sharp in Japan, they would have adopted windows as well, that's why the X1000 it's not bad considering what happened.

[DAX]

Amiga_Nut
Your post is extremely insightful and I agree with it 100%.

The problem is that "WHEN" those decisions were to be made, they decided otherwise (as you so clearly explained) and the time arrived where everything was in jeopardy.

And as you said, today, if Microsoft can't...no brainer.

Basically the point is, in 2010 Nvidia+Ati rule, no need to cry for fabled chipsets (today I mean, not in the past, where the right decisions might have led somewhere nice) as they (fabled chipset wannabe makers) would never compare with all the R&D these companies throw at it every year.

[DAX]

Certain things are being integrated for sure, on the low end market, especially the nettop/netbook market this makes a huge amount of sense both in terms of cost and power usage. However at the top end there are different stresses - the integration of the memory controller for instance is done to reduce latency. It's unlikely you'll see the gpu moving onto these parts any time soon simply because there isn't the need either on the generic server markets or workstations or high end gaming rigs.

You repeated your reply but not answer my question: where this integration process will eventually and inevitably lead in the coming years considering that Hi-end gaming PCs are selling less and less every year ?

As Karlos stated, he was unsure if the Core2Duo benchmark was based on a single or both cores. Regardless, C2D based systems can be had for a tenth of the cost of an X1000. A tenth.

You seem to forget that this is a premium PPC machine for AmigaOS developers and users with deep pockets (a lot of them around as they got 150 people that want to beta test) in order for them to expand the platform. Those that think in your terms are not the target audience, the latter instead DOESN'T think in your terms. This is just a first step, if all goes according to plan newer machines will be done later.

Different!=Useful.

There was a post on another thread that I think perfectly summed it up, it said something along the lines of "£1500 to run an Alpha of Firefox?"

see above.

Or that you've got no idea what you're talking about. I suspect the latter. Here: Commordores last gasp - Hombre. There was nothing later than this as C= down long before it ever got off of the drawing board.

Instead it is as I said has the paper i mentioned as nothing to do with Hombre. it was posted many times at AW if you want I can go there and ask for a copy/link.

You do if that embedded microcontroller is a key part of a telecommunications node. That is where PPC is aimed at these days. Not desktop, not generic servers but telecoms. PPC for desktop (and lets face it, generic servers) is dead, it exists only in niche products in niche industries.

You often find PPCs tied with FPGAs these days (thankyou the fella that pointed me to that a few months back) within this market.

And who says? PPC cores are used in the Xbox360 (and more) and now in the AmigaOne X1000, if the CPU does what you need it to do, than it's all good.

[DAX]

I'm happy to ascertain that we agree on the fact that we cannot be sure about many things until we actually wait and see.
So i have my hypothesis you have yours.

That said i understand you were talking about Commodore "gamish" HW named Hombre (an home console that was never produced basically
while I was talking about their plans about future Amigas (nothing to do with Hombre) :

1)Hombre: an uncompleted project that had nothing to do with Amiga because (Dave Haynie speaking below):

Strictly speaking, Hombre is not an Amiga chip set.  While it supports some
of the Amiga ideas, it's no more Amiga compatible than an SVGA chip (less,
actually, since all SVGA chips support planar as well as chunky displays,
at least up to 4 bits/pixel).  

and...(Haynie again)

The Amiga OS was not to have
run on this system in any form.  

On the other hand I was talking about the devCon paper done in 1993 more or less simultaneously to Hombre's development starting point which described a similar to X1000 machine:

The  primary goal of an advanced Amiga system can be summed up in one word:  modularity. Such a new system, both logically and physically, is  composed of several interchangable subsystems. No one piece has any  unnatural dependence on any other; interconnections between the system  components have to be based on intentional system standards, not chance  implementation details.

The  motherboard for such a system contains just the basics that will be  needed by every system. This will certainly include a number of basic  I/O chips for the standard ports on that machine.
Not only does this make  motherboard upgrade much easier, but it allows several different  motherboards to be designed using the same plug-in modules, and it  allows Commodore to easily support more options in system and processor  makeup.

If you add to that that not even MS can afford to challenge Ati and Nvidia R&D departments, you can clearly see where this all would have led...

[DAX]

@The_Leander
I'm not saying it's the prodigal son, only responding to "classic" users that there would've not have been any fabled chipset Amiga machine. And by the way, hombre was WAY behind in schedule, Haynie said:

The design called for two chips...Neither design was finished;

which basically mean all you got there is the AAA situation all over again, too little too late.
The whole chipset idea was to be scrapped as there were graphic processing units manufacturers pumping out new designs every 3 months back then (where there were around 10 different makers).

At the end of the day the X1000 is still more Amiga than anything else, the barebone mobo allow to upgrade several parts in case you upgrade the CPU, (instead of having just a CPU change and keep the old controllers)  Xena could lead to nice ideas, it can still produce some nice Int expansions (XMP), and is something more (not something less) that adds in to the fun.

No matter how you put it, old chipsets are (and were) no more and this is as good as it gets for now.

I underlined that, because both Varisys and A-Eon hinted at this being a beginning not an an end, and since we are just a few months after the settlement it's a nice start, way better than anything we would have seen from C= which as you said, it would have been absolutely nothing, ( just some Wintel clones running NT, yes no Hombre, as it was already obsolete while in highly unfinished state, and would've been scrapped for some cheaper and more performing GPU)

[DAX]

@the_leander
Look I leaved the Amiga scene around 96/97 (ah, bought my A2000 in 1988 )  but my business were GPUs (I dealt with Elsa and 3Dlabs) so let me tell you, HOMBRE might have been on par with certain options day one if it made day one, even less options after 3 months, and dead obsolete after a few more months and that's ONLY IF it existed (read: it NEVER existed, wasn't even close to completion).
What part of "it wasn't completed and thus it would have been surpassed several folds if they continued to drag on it" you do not understand? ()

Look I'm tired, I will repeat again the x1000 is a premium machine aimed at developers (and Power users with no budget constraints).
it is not uderpowered at all for an Amiga machine, it actually sports a better than G5 CPU, and when you buy it you are not searching a bargain,+ mainstream is not a target.

Let me just say (and I'm finished) that those that will buy it will not do so with your money, so you may relax a little bit

[DAX]

Which part of your "gpu business" dealt with the console market, exactly?
Playstation 1, released 1994 was still being actively developed for some  10 years after its initial launch and is one of the most popular games  consoles ever to have been sold.

Do you think for one second that the PS1 had state of the art graphics  even a week after its release? No, it didn't. Did it make a blind bit of  difference to it's success? Not in the slightest.

You've tried repeatedly to conflate AAAs target market (mainstream  computing) with Hombres (a games console concept), you can stop now.

1st) I was taking in consideration your fabled HP-Hombre WinNT computer (which might have dealt with GPUs very much),
2nd)You might be unaware that GPUs at the time derived from SGI technology papers and that Nintendo64 (same generation-1996) uses one of those GPUs for openGL acceleration.
In 1998 the Dreamcast came out using a PowerVR GPU. (but again I had in mind your "phantomatic" Hombre computer there, read more about consoles below).

In the computer market, sure, which is why AAA and your "future amiga" got canned. But as a games console the system would have been a fine competitor. I'm well aware of the fact that Commodore folded, but I'm not the one claiming that the Hyperion (with all the crap they've pulled over the years) and an as yet untested and unproven AEON are better for the Amiga market than Commodore.

Sure, and I said that it was so late they would've dragged it until it was probably 1 year earlier than DC if C= didn't fold (which would have kicked its a*rse big time not to mention PCs were getting Nivida love by that time). If it was released in time sure, but I never argued about it, go re-read what i wrote, all I said was, they were "late" which is a fact regardless of C= status, and by time it would have been ready for "REAL" it would have been obsolete.

Power users? Where? Seriously where are these mystical power users with more money than sense all going to mysteriously appear from? And even if they do, as gazgod points out, wtf are they going to run on it? An Alpha of Firefox and Blender simply do not cut it!

I still get the impression you are somewhat exited...
Anyway (@Gazgod too) they have already 150 people booked for 100 beta systems so they will have to discard 50 of them, but try to understand once and for all a simple fact: all they plan, is to sell a limited number of machines, get their money back+some and re-invest.
That's it.
You should visit other sites every once in a while you will meet many of this guys...(not to mention all those that solely post in their own country due to language barrier).

If your offering to these mythical power users is an early firefox port and blender, the question then is, why bother with OS4 when you can do that faster for a tenth of the price on a PC?

Enthusiast with enough money or will do it just for fun (early adopters type), and they will be the first to enjoy what the developers will do with the machine.