Registering Interest for AmigaOne X1000

[vox]

Why this über-sensitiveness about topics all the sudden? Judging by your über-strict views on topic-discipline, only post #2 was on topic (since it was Darrin "Registering Interest for AmigaOne X1000",

All true, incl. benchmarks, but as we all know X1000 performance might really improve by new OS version, better Nemo board drivers and using second core. That is quite potential. And I am sure most of buyers are making informed decision knowing what they are getting now.

It would be way better if we had oportunity to use MorphOS 3.x on SAMs and X1000 and figure the difference on new hardware already purchased. That is way better then fighting it just on machine vs machine performance ratios.
As well as it would be great to have AmigaOS for PPC Macs which are not junk at all.

To my knowledge A1X1k is the only desktop computer to use PA Semi, which is unique, no matter how old the CPU is. Correct me if I am wrong.

[vox]

Buying ONE A1X1K for $3,000, or buying THIRTY computers of roughly A1X1K performance for the same amount of money, or buying one computer .

No one is buying 30 computers, but since you are trying to equalize the worth please include 200 eur software licences to each Mac (OS 4.1.5 and 4.2 licence) or at least MorphOS licence per computer. Then you will get to more realistic 10-20 machines. But even that is not the point, just like with past AmigaOne and Pegasos boards, new PPC boards are expensive and come rarely. Enjoy the moment.

If you really think of comparing it with PPC MacMinis, take into account X1000 has dual core CPU that is yet to be used, much faster RAM, R700 video chip with a ton of VRAM, much more and is quite expandable board, especially when onboard drivers will be done. Board also supports 8GB RAM I believe, so might drive OS4 to brake its own limits in many, many ways (from RAM barreir, SMP, XMOS support, RadeonHD drivers ...). Hell, even supporting AMD bridge chip and RadeonHD is step forward some future AMD Fusion port :-)

[vox]

Yeah, it's always this "is yet to be used", "might", "future", yada, yada. Anyone knowing Hyperion, should really know by now that they love to sell unhatched chickens with great promises, long before the chickens are hatched. Sometimes it turns out in the end that the eggs never was laid, and never will be. Was the Micro A1 GX ever sorted out/fully supported?
Will you *never* learn I ask myself?

original A1 board was supported including Altivec, only some problems with really defective USB I believe, but PCI cards with USB got full support. SAM 440 is fully supported by now (incl. FPGA) and only sadly SAM 460 support is lacking now. Yes, USB 2.0 took ages and didn`t give real fast results, it even took time to reach stability with update 4.
There are lot of things to be angry for (e.g. SAM 460 still having beta SATA, sound and RadeonHD with no 3D) but its also clear to see AmigaOS got more development, team leader, beta testing, website, forum etc. in last two years then ever before.

If there would be no SMP and 3D as well as other optimizations and drivers expected out of AmigaOS 4.2, there would be no need for X1000 at all. Single core G4 or AMCC 460 would be enough. I understand the frustration of releasing hardware first and then finalizing the drivers later, agreed I like the MorphOS approach of testing hardware and developing drivers first and then making public appearance. But there are no doubts X1000 (and SAM 460) drivers will be developed in time, assumingly with some driver release for OS 4.1 users and with initial OS 4.2

If there were no PCI-E PPC Amigas like SAM 460 and X1000 RadeonHD driver would not make much sense. MOS team builds only MorphOS and yet is late with MorphOS 3, while Hyperion does support and bugfixing of current AmigaOS 4.1, development of new features of OS 4.2 and Timberwolf simultaneously. Would love to see them throwing

All in one, nevermind the price, X1000 is attractive hardware for PPC Linux and AROS and should really drive AmigaOS 4 forward. Should be hardware viable for some MorphOS 3.x update - who has money to buy X1000 might have interest in paying 100 euros for "AmigaOS done well" IF it would fully support X1000 hardware. There is a place for MOS team to compete, and I am sure Trevor would donate a board for developers.

Any set dates for MorphOS 3? (none for OS 4.2)

So instead of attacking its price / performance side, use it to your best interest.

Surely there will be some solution for SMP in both MorphOS and AmigaOS 4.x. It might make 68k compatibility go away, but since such hardware as SAM 460, MacMini or X1000 should run 68k software well with JIT - that could not be a shame, but way forward. Before X1000 there was no need to consider it at all.

[vox]

just FYI, at least with the 'first contact' systems, you had to purchase the OS license additionally. the free copy of OS4 is the copy of OS4.2 when it ships.

-- eliyahu

True and not true. Base price for X1000 doesn`t include the OS 4.1 licence to make promised 1700GBP price tag which is what Trevor gets but excludes everything else ... (excludes VAT, shipping and the OS, as well as more RAM as 2GB was first promise). But in reality for price of one OS licence you get both OS 4.1.5 now and OS 4.2 in the future.

This is just made to make system base price look cheaper and might change with final system just because OS 4.2 should be done by then as well as drivers for motherboard, rendering sound and network PCI cards not needed. This could open the space to make system a bit cheaper, or at least with OS included. That, or higher order rate for batch. But not too much anyway.

But one thing is right: $3000 price calculated for X1000 includes VAT, AmigaOS and shipping while cheap ebay prices of MacMinis exclude shipping and MorphOS licence making not such a huge difference, rendering ratio almost to half.

Also, the trick is in comparing new and used hardware. Surely, you could get 4-5 SAM 440 766-800Mhz systems used for price of X1000 running AmigaOS 4 but they would not perform not even close to it. But maybe 3-4 A1 or Peg2 systems that would perform 20-30% slower.

But what if people want expand ability, and ability to use dual core in the future and only X1000 offers it?

[vox]

If you go that way, MorphOS team also bugfixed 2.x and developped 3.x at the same time. Many of the developers of MorphOS team (including myself) have "pet" projects/applications beyond the core of MorphOS (that sometimes get integrated to MorphOS, course). And actually, Odyssey is part of MorphOS distribution.  

Haven`t heard of anything new after 2.7, and 3 was announced afterwards. OK, having similar situation should help the understanding why some things are not yet done. MorphOS is done better, surely, but has longer career of development and no frozen time of court case. We are all looking forward to see how OS 4.2 (and MorphOS 3) will improve and yes, will difference increase or decrease at least on Peg2 (I don`t expect OS 4 to become superior soon).

It is a said fact that beyond Blizzard cards (MOS 1.x, AOS 4.1) and Peg 2 (MOS 2.x, AOS 4.1) there are no more PPC systems supported by both.
Or even better OS4 Emu and kind of MOS Emu wrappers or some code trade-offs (e.g. Ambient for OS4). Having OS4 reached MUI4 compatibility was a great thing and brought MUI OWB (thanks for OWB development!)

If 68k compatibility has to be dropped, then going x86 is a much better solution, don't you think?  

Woud like both to evolve to x86 board with RadeonHD cards and some newer Sound Blaster. At least MorphOS team and Hyperion would prove to be software only companies looking to offer OS to wider hardware base. But meanwhile, buying what is available is only option.

[QUOTEAs for SMP, assuming it would be possible without breaking everything and/or having bad performance, MorphOS would also have another potential
target with the dual G4 boards for PowerMacs, or the dual/quad G5s.[/QUOTE]

Would love to see MorphOS (and OS4, PPC Linux, AROS) using Mac G5.
Cheap and mighty!

[vox]

Isn't a platform migration the natural thing to do at that point? There is no "Power" in PowerPC, not in the year 2012 and beyond! ARM and/or x86, but not PPC! Since those features would require a break from the past, why continue with PPC at all? Why not bring the new platform onto a modern architecture? For example, if MorphOS developers would want to start exploring 64-bit support for the future (as suggested by Fab in his presentation of MorphOS future), why would they do that on a *dead* G5 platform (or PA6T for that matter)? I mean, if the legacy is to be broken anyway, why not migrate to greener pastures while you are at it?

There is still enough power in POWER in 2012, wouldn`t dismiss something living and developing. Is it a top leader? NO. Is it developed and usable. YES. Somehow just like AOS and MOS :-)

Forgot to mention X1000 is first platform to offer 64-bit transition path to AmigaOS. In my opinion, its more likely both advanced features might first be used on existing PPC systems being X1000 and likely G5 for MorphOS then we know of any real x86 transition.

Surely, would be happy at least to see some PPC emulation box for masculine and overpowered x86 multicore systems before the transition could be over. So all those arguing FOR the transition should think of bounty to get Apple Rosetta PPC emulator or to fund QEMMU / PearPC or some "Amithlon" kind of emulation box as emulation solution before the full port OR a realistic estimation would be what time is needed for other CPU port.

OS should develop meanwhile, and Hyperion decided to team up with new hardware. Hope they will do the net top as portable Miggy and low cost one, so the range can be from reasonable to heavy priced X1000.

If development should be continued on PPC and SMP introduced, next cheap and mighty thing are Playstations 3: soon will be replaced by PS4 and have enough horsepower to be a good computer.