@Signal

You mean the OS that was designed for "normal" PPC ?

Or maybe and I really mean maybe one should check whether a CPU is really suitable for a job before creating HW.


Haven't changed my signature here in years, and still it turns out spot on every other day

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@Kronos

Quote:

maybe one should check whether a CPU is really suitable for a job before creating HW

Or ask everyone to alter the job to suit the new hardware!

Last edited by iggy on 12-Oct-2015 at 10:00 PM.

@iggy

Or maybe, not definitive hardware. :)

@Antique

Why anyone would believe that OS4 wasn't available to see and try at Nuess, when it had previously been presented at Amsterdam and Peterborough, is beyond me.

I think that this is Raspberry Pi 2 for PowerPC or as close as you
can get to it with current Freescale policy.

There is no particular reason why you couldn't make 200 Eur low
end PowerPC motherboard for end user and make profit if you don't
outsource hardware development to third party (Varisys) and if you
can buy cheap PPC CPUs from Freescale (which somehow is always
problematic).That way you could could make larger investment (like
in Pi's case) and produce large production runs (100 000 for example)
of cheap PPC entry level hardware. Things are not going that way though
and Power missed it's chance to revive in favour of Arm.

Also , primary focus would have to shift to Linux (same for Pi , they didn't
market it as new RiscOS machine) with side effect of it being able to run
AmigaOS.

@blizz1220

Quote:

There is no particular reason why you couldn't make 200 Eur low end PowerPC motherboard for end user and make profit if you don't outsource hardware development to third party (Varisys)

Indeed, we should only buy from A-eon if Trevor himself designed the thing... He knows how to do that, right?

_________________
All glory to the Hypnotoad!

@billt

No , but maybe Trevor could make some money that way and
maybe more users would come.I'm not saying I see solution
for companies making last PPC hardware for use in desktop
just stating how I see it.

These things are usually done in large scale and with larger
manufacturer support.Varisys didn't even bother to make
drivers for AmigaOS , I wonder how much would they ask
to deliver fully AOS4 supported hardware.If this was done by
larger company that would order magnitude of CPUs (which
are still way to overpriced for what they offer and sometimes
prone to changing prices/availability) and if that company did
the design and drivers it would be single entity.

I wonder who made Raspberry Pis and why not try to do the
same thing with Raspberry PPC and Kickstarter where AEon
would maybe fund 1/3 or 1/5.People still seem interested in
PPC as they were once interested in 68k but that won't last
forever.

Not that I think that this ship can change course now and it
was heading toward rocky bottom from the start.

Quote:

blizz1220 wrote: ... There is no particular reason why you couldn't make 200 Eur low end PowerPC motherboard for end user and make profit if you don't outsource hardware development to third party (Varisys) and if you can buy cheap PPC CPUs from Freescale (which somehow is always problematic)..

sounds like you really know what you're talking about and believe it can be done SO there is no particular reason why you can't give us more info when your 200 euro low end PPC MB is going to be available for purchase

Last edited by klx300r on 13-Oct-2015 at 02:00 AM.

_________________
____________________________
c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII
! My Master Miggies- Amiga 1000 & AmigaOne X1000 !
mancave-ramblings
X1000 I BELIEVE

@klx300r

Er I wouldn't say I'm an expert and I think I said I see
no way of doing it NOW.

There was a last last chance with AmigaONE x1000
in my opinion but that was that.

@pavlor

Quote:

pavlor wrote: @TRIPOS Quote: Yeah on Windows, but this is a slow, extremely low-end CPU compared to the fast Core i5's etc people use today (dimensions ahead of these kinds of PPC CPU's). Do you get PowerPC CPU is emulated in WinUAE and integer performance is slower than SAM - comparable to G3 500 MHz on Core i5-2500K 3.3 GHz?

It depends on the test. On the other thread which reports some benchmarks, it's more or less close to the SAM460.


@pavlor

Quote:

pavlor wrote: @K-L Quote: So, 8 GB Ram ? What board you quote? This is not "Tabor". Here is P1022 Fact Sheet It clearly says: 36-bit physical addressing, One x4 PCI Express interface (converted to x16 like in SAM460).

36-bit is the internal limit of the architecture. It has to be seen how much memory supports the memory controller.


@xeno74

Quote:

xeno74 wrote: Hi All, The Tabor board doesn't have a classic FPU with dedicated FPRs found on most other PowerPC systems. It is replaced with a set of SPE instructions which perform floating-point operations on the integer registers. Important: The SPE unit can execute floating-point operations very fast. We have tested a lot of programs with FPU operations this year.

I suppose that it happens only with applications recompiled to directly use the SPE unit, because trapping is a slow process itself (and it happens also for the PowerPC G5 and the PA6T, for some instructions which are missing and that are used by some software, mostly emulators; it might be also the case for this processor too).

Since the SPE unit uses the integer registers, it probably means (I don't know how the SPE unit work: just guessing here) that, when running in 32-bit mode (which is the usual way on post-Amiga o.ses) you cannot use them for emulating double precision FPU operations, because there's no guarantee that the upper 32-bit of each registers is preserved, unless you decide to disable and enable interrupts et similia on each critical section, which slows down the execution.

Anyway, you need to fork software here: provide ad-hoc compiled binaries specifically for this board, if you want to get decent FPU performance.

Quote:

I think that this is Raspberry Pi 2 for PowerPC or as close as you can get to it with current Freescale policy.

Except that the RaspBerry costs 35$, only requires an SD card and USB/HDMI cable, and comes for free with Linux & RiscOS (and a bunch of other OS of course).

You can spend 35$ just for the fun of it. You won't spend 1000€ just for the fun...

I really wonder what market they target with such a board (do they even have one ?).

_________________
http://www.warpdesign.fr/

@blizz1220

Quote:

blizz1220 wrote: I think that this is Raspberry Pi 2 for PowerPC or as close as you can get to it with current Freescale policy.

The "Rasperry Pi 2 for PowerPC" existed years before the "Rasperry Pi 1" was eventually released. Its name was "EFIKA", it came bundled with Linux and later had MorphOS ported to it.

Quote:

There is no particular reason why you couldn't make 200 Eur low end PowerPC motherboard for end user

See here: 189 EUR at Vesalia

Please note that the listed price included 19% German VAT...

Quote:

and make profit if you don't outsource hardware development to third party (Varisys) and if you can buy cheap PPC CPUs from Freescale (which somehow is always problematic).That way you could could make larger investment (like in Pi's case) and produce large production runs (100 000 for example) of cheap PPC entry level hardware. Things are not going that way though and Power missed it's chance to revive in favour of Arm.

There is simply no market for 100,000 general purpose 'desktop' PowerPC mainboards. Considering how frequently large companies are being shamed for outsourcing work to save costs, I think it is safe to say that "outsourcing" is not a cost-increasing measure per se either.

Quote:

I wonder who made Raspberry Pis and why not try to do the same thing with Raspberry PPC and Kickstarter where AEon would maybe fund 1/3 or 1/5.People still seem interested in PPC as they were once interested in 68k but that won't last forever.

Crowd funding has already been tried to fund the development of a more expensive (yet still affordable compared to the alternatives) MPC8610-based mainboard. The hardware design was billed to cost around 65.000 EUR, was to be made open source and would have been handled by the developers of the EFIKA. There was insufficient interest to get anywhere close to funding the project...

Last edited by ASiegel on 13-Oct-2015 at 08:21 AM.

@blizz1220

Yes of course producing 100.000 units is much cheaper per unit than producing and selling only 100 per batch but... who will sign the bills to produce the 100.000 and take the risks? Fortunately for the AmigaOS camp Trevor D. started to prefinance expensive hardware projects otherwise AmigaOS would only run on Acube hardware. I personal propably if I had the money would have invested it in ISA change (f.e. to ARM) instead of producing custom PPC hardware but it is of course his decision and money. He is (as I understand it) prefinancing projects, prefinancing means he tries to reget his money and cover the expenses, if it runs good he makes a small profit again investing it in related projects. I do not assume he would even think about producing huge volumes of hardware with the risk to make huge losses. What would be the advantage of cheaper PPC based hardware compared to ARM or X86/X64 based except running AmigaOS (what is no advantage for most people out there)? If you can name none that is obvious and justify buying there is no chance.

Last edited by OlafS25 on 13-Oct-2015 at 09:20 AM.

@TRIPOS

Quote:

TRIPOS wrote: ... No "Xorro"/"Xena" on either of them (double-plus good)! ...

Bummer!

_________________
Ciao

Dandy
__________________________________________
If someone enjoys marching to military music, then I already despise him.
He got his brain accidently - the bone marrow in his back would have been sufficient for him!
(Albert Einstein)

@pavlor

Quote:

pavlor wrote: @TRIPOS I will arther wait for actual benchmarks.

Care to explain what "arther" means?

_________________
Ciao

Dandy
__________________________________________
If someone enjoys marching to military music, then I already despise him.
He got his brain accidently - the bone marrow in his back would have been sufficient for him!
(Albert Einstein)

@OlafS25

I would rather he produced 10 000 motherboards that have
200 Eur price tag on them than 200 motherboards that have
10 000 Eur price tag.Somehow these things sell (It's Amiga!)

Efika is good example and prices of components and
manufacturing went down from the times those were last
made so I guess it would have been possible.Outsourcing
is best done when you do it in China and Taiwan to cut the
costs down and large companies there (that make Varisys
look like ant) have coders , designers and all that is needed.

But off course porting and upgrading the OS was the most sane idea.

@itix

Quote:

itix wrote: @Everblue ... People applaud for more cores even when they are not able to use it.

Wasn't it said that Linux already runs on it and can use both cores?

_________________
Ciao

Dandy
__________________________________________
If someone enjoys marching to military music, then I already despise him.
He got his brain accidently - the bone marrow in his back would have been sufficient for him!
(Albert Einstein)

@Chain-Q

Quote:

Chain-Q wrote: @iggy ... I have a Linux Pegasos II/G4, running 24/7 since mid-2007, and I'm slightly worried that box might give in any day... Simply because of the age. ...

Hmmm - you worry about a 7 year old System?

I still have an Amiga 4000 PPC running - it's motherboard is already 22 years old - so more than three times as old as your system...

So keep cool...

_________________
Ciao

Dandy
__________________________________________
If someone enjoys marching to military music, then I already despise him.
He got his brain accidently - the bone marrow in his back would have been sufficient for him!
(Albert Einstein)

@blizz1220

I do not know if you could sell custom hardware for 200 EUR when producing 10.000. Assuming yes, the whole community (including all people using emulation) is somewhere between 10.000-15.000 and I think many of them would not buy it. So you would need to attract a lot of people from outside. It is very risky. Would you do it? You could make debts and invest it in such a venture if you are really convinced. It is always easier to give advice when it is not your money.

Time for PPC is long over, today nobody needs it and it will slowly die. Perhaps it will survive in some niches for some time but on a larger scale nobody needs it.

Last edited by OlafS25 on 13-Oct-2015 at 11:01 AM.

@OlafS25

Additionally; you need space for storage. Storagespace means often rent etc, so there are several things to consider.