Quote:

agami wrote: No one is saying that AmigaOS on x86 is the solution to all the problems. And your eagerness to bring up AROS as an example shows your lack of argumentative maturity, a.k.a arguing from ignorance. ..... "A house divided against itself cannot stand" - Abraham Lincoln So the core benefit of AmigaOS, MorphOS, and yes AROS as well being on x86 is that the net pool of users is much larger. More users means larger market for Amiga software.

excuse me but this thread is about wanting to get a bounty to fund a 'cheaper' PPC motherboard than those offered from A-Eon/ACube and I am not 'arguing' with anyone as I clearly stated I am outlining my own personal experience with a control group of friends.
My 'eagerness' lol as you say to bring up AROS is because I use AROS on my notebook and enjoy using it and it just happens to run on cheap x86 hardware as well as PPC and other platforms and where exactly did I say AmigaOS on X86 is the answer

You should learn to comprehend what you read before you make such immature comments.

Nice you think that all camps should get together though,& many of us surely agree, but thats been debated to death and for another thread.

Last edited by klx300r on 05-Nov-2015 at 02:55 AM.

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

@Fransexy
But OS4 on a dead and OVERPRICED platform is going to??
I never said it would ever become a mainstream OS, but having Amiga OS on a affordable platform with high end boards MIGHT have the chance in at least keeping it alive as a niche platform.
The fact is that Amiga OS has no chance on the dying PPC platform even as a niche platform.

@BigD

Quote:

BigD wrote: It seems that way. I can't see it changing to be honest. Hyperion, A-Eon and ACube have fought for this situation and porting to x64 is a huge gamble, even bigger than the X1000 & X5000 projects. Would they recoup the expenditure? Would it kill the AmigaOne hardware projects?

AmigaOne is (was?) an open platform, not bound to PowerPCs. Ask pavlor for a confirmation.

The trademark is owned (exclusive license) by Hyperion, but they said that they aren't bound to PowerPCs for the Amiga o.s..
Quote:

Would it further fragment the market?

Why? Once you move to a more powerful and cheap ISA (don't say x64, otherwise olegil becomes nervous and starts his nazi harassment), doesn't make sense to continue supporting the old one.
Quote:

Would all backwards compatibility and hence all unique selling points be lost?

You can use sandboxing for running old, legacy applications, with a certain amount of shared resources between the two worlds.
Quote:

These are the considerations and since PPC is not quite finished

Well, it is, and from very long time. It simply should be accepted... or post an official roadmap which shows other (micro) architectures which are coming after the e6500...
Quote:

the decision does not have to be made right now just viewed as a contingency. Also, all the new money invested into the Amiga scene is coming via A-Eon a hardware company. Hyperion is basically working for A-Eon and ACube and by porting to x64 they would be destroying their best customers by giving people the option to just buy from PC World rather than Amiga companies. A-Eons software projects are not mature enough to keep the company going without the hardware sales!

What the history after the Commodore bankrupt should have been taught, is that the only thing left and for which makes some sense to invest, is software. The hardware should be a "commodity": you don't have to be strictly bound on it.

@klx300r

Quote:

...and where exactly did I say AmigaOS on X86 is the answer...

You didn't, but yours and Fransexy's response read as those to someone who claimed that, though no one did.

Someone mentions how @Raffaele would be better off spending his time and energy on getting Hyperion to port the existing OS to existing value-for-money x86 hardware, and almost like clockwork people bring up AROS "Well if moving AmigaOS to x86 is the solution how come it's not solving AROS' problems. How come AROS isn't a runaway success? I showed AROS to friends and they were like cute but meh." Implying that since x86 isn't doing anything for AROS it certainly won't do anything for AmigaOS 4.

Whilst I agree the overall effort, time, and money would be better placed and more worthy if it were aimed at an x86 or ARM port, the reality is that neither of these options are realistic, the third board or the port.

So sorry if you feel that you have been unjustly caught up in my rant-ish response, but you did cite how your AROS presentations are usually met with lacklustre interest.

_________________
All the way, with 68k

@BigD

Quote:

Also, all the new money invested into the Amiga scene is coming via A-Eon a hardware company... A-Eons software projects are not mature enough to keep the company going without the hardware sales!

I'll take a contrarian stance and argue that new hardware investment is the absolute worst thing that could happen to this community. I have great respect for the initiative that Trevor and company have taken but things continue to move sideways.

There are innumerable software and hardware solutions for people who want to continue to embrace the status quo and stick with the past. There is a demand. We all respect that.

But the real problem is the *s-o-f-t-w-a-r-e*.

It simply isn't feasible to bring the existing OS up-to-speed.

It would be an order of magnitude less involved to reinvent the OS by grafting the "what makes an Amiga an Amiga" elements onto an existing core.

We already have that core. It's spearheaded by someone who has been involved in this community throughout its evolution and is eminently qualified to move things forward. That technology evolved with sweat equity. $0. And it's current. Like 2015 current.

Count me among those who believe a NG OS is critical and should be a priority.

@Raffaele

>by a China Manufacturer

Some thoughts...

Not sure if it would be possible to base HW production on a Chinese manufacturer.

From my experience you need personnel to the china plant to monitor the production, otherwise you end up getting a lot of sh*tty boards. So, in that way it would seem impossible option for niche product.

It is different for simple boards that are ordered just once from some manufacturer.
(perhaps ok for some development versions of a NG board)

So. Most likely, some western manufacturer is more realistic option or a manufacturer that has production both in western world and in China.
For example I know an ex-amigan who owns (part of a) company that does prototype HW production in Europe + has good relations with Chinese company. They do below 10000 unit productions in Europe and ~100000 units runs in china.
(+the size and complexity of the boards also determine the manufacturing sanity)

Last edited by KimmoK on 05-Nov-2015 at 07:30 AM.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@billt
I don´t understand why you will use a southbridge. Sure it´s nice to have but not necessary.

If you want to keep the board low cost. It is much cheaper to use 2 Sata and 2 PCIe x1 connectors. The user can buy what ever he wants and use the PCIe x1 connector.

If you use a Southbridge it must be designed in, tested and driver must be written. That will add costs to your design.

Most people only need Audio from the hdmi port of the graphics card.

@agami

Yes AROS is not AmigaOS and will never be

But what is a OS? It is a desktop (or today mobile GUI) and a API

The API and structure of AROS is very similar to Amiga OS 3.1 (with many components that were patches or third party in that time). So you can configure it to behave very similar to 3.X when having a configurable desktop or make it much better or different however you call it. AROS is also not only X86 (even if most people think of X86 when talking about AROS). When using the 68k branch of AROS I do not see differences or advantages for AmigaOS. But that is my personal view.

Regarding AROS... AROS is source compatible not binary compatible (except Aros 68k) so both AmigaOS and MorphOS offer more "Amiga feeling" than AROS could. On AROS you only can use what is compiled for the target hardware. You cannot mix 68k and X86 f.e.. That is a disadvantage and even more was more than 10 years ago when both AmigaOS and MorphOS became reality. Additional AROS was not far enough developed at that time. Most developers and users joined either AmigaOS or MorphOS or left. Unfortunately even in the coming years most users and developers left instead of joining another platform. So it lagged for a long time. That was caused by not enough manpower and not by the hardware (except of course that you have nothing like Petunia but even that could be solved but here again missing manpower)

Just a offtopic comment...

why I reacted at all when "X86 is no solution" came up again because I disagree there. I read the same in comments from both Ben H. and the Frieden brothers long time ago. The thesis was BeOS failed on X86 because it competed with Windows there. Thus AmigaOS would fail too on X86 because of that so it is better to be on another hardware platform. I am aware of the negativity of Amigans regarding standard hardware in the past (I was one of those too) but time have changed and most people have moved on. "Intel outside" is out too today, most people use standard hardware at home. My thesis is BeOS did not fail because it run on X86 but it had no competitive software base. And I have also the thesis that it is wrong to assume AmigaOS does not compete with Windows or Linux just because it runs on exotic hardware. People always compare and decide. The software base is always smaller on a exotic OS and worse it runs on weak expensive hardware. But most software today is developed to run on fast hardware, nobody is optimizing it for embedded hardware or doing necessary changes that it can be ported to exotic hardware almost noone uses today. My view is PPC was 2003 the wrong choice and it is even more wrong 2015 or 2016. If people want a future for AmigaOS (not only in UAE) then they should push for a ISA change.

Last edited by OlafS25 on 05-Nov-2015 at 10:26 AM.

@OlafS25

I agree with pretty much everything you've written.
Other platforms' negative experience of moving to or being on x86 is not a rule, and it shouldn't be used as an argument against another platform moving to it.
If anything, they should be used as valuable lessons. Actually analyse what didn't work. Like you said, BeOS' strategy to move to x86 is not what undid them.

I like AROS, I'm impressed by the work that has gone into it over the better part of two decades, and if there were some apps for it I would be using it right now. And when I say apps I mean the apps I use to do the work I need to do on a daily basis. One day perhaps.

_________________
All the way, with 68k

@agami

it is impossible to compete with the software base on windows, mac or even Linux. At best there should be certain segments covered with special applications (or modern: Apps). And here I mean unique software, not ports or slightly updated amiga software. That is enough to keep existing users at best but will not bring new people from outside. But of course that is easier to ask than to do

I do not want to spoil people the fun... if there are really enough people that would invest in a new PPC hardware then Raffaele should try.

Last edited by OlafS25 on 05-Nov-2015 at 10:58 AM.

@OlafS25

Quote:

I reacted at all when "X86 is no solution" came up again because I disagree there.

That is a point I agree on.
I initially supported a move to ARM, but why not use the most successful ISA?

@fjudde

Quote:

@Olegil Your missing out on the fun! iggy and billt have already started discussions about the design!!!

I've had this discussion with billt so many times it's not even funny. His insistence on using THE BEST at all times is driving me mad.

And iggy has been corrected on the "number of PCIe lanes available on a T10x2 design with a SB" issue several times so far without learning from it.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

Sorry for the slight out of topic, (and for my unexperience on this matter) but rather than design a new board, would'n be easier to design an adaptor board with the ppc pin compatible to the amd/intel cpu in order to use the commerchial mainboards?

@ilbarbax

Not really

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@Rudi

x1000 uses an AMD southbridge, ergo some drivers already exist. And it's known to physically fit. And it provides the necessary interfaces to overcome a bit of a shortcoming in the T10x2 design.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@olegil
I have nothig against of using a southbridge.
But I want to know what advantage do I have using it and what will it cost.

@billt

Black Mamba, 8640d or 8610d?

These are just examples inviting you all to post the specs you desire, in order to choose the best ones for the community...

_________________
"When the Amiga came out, everyone [at Apple] was scared as hell." (J.L. Gassée, former CEO of Apple France and chief of devs of Mac II-fx, interviewed by Amazing Computing, Nov 1996).

@pavlor

Quote:

pavlor wrote: @cdimauro Quote: First, you didn't ever took a look at the processors' data sheet to see if they are compatible with the motherboards specs that you proposed. Well said! @Raffaele I know you mean well, but your proposal is not based on reality. Recent Freescale SoCs (10XX) would be much better suited for such low-cost board. Note: Black Mamba is 32 bit. Both your SoCs of choice don´t support features you want.

>Example 1<

>Example 2<

Ahhh English... This always misinterpreted unlucky language...

_________________
"When the Amiga came out, everyone [at Apple] was scared as hell." (J.L. Gassée, former CEO of Apple France and chief of devs of Mac II-fx, interviewed by Amazing Computing, Nov 1996).

@jaokim

Quote:

jaokim wrote: Why not instead of starting a new project, join this: http://www.powerpc-notebook.org/en/ Quote: It's the right time to make new choices, a new PowerPC Notebook

I joined... No news from last May 2015...

_________________
"When the Amiga came out, everyone [at Apple] was scared as hell." (J.L. Gassée, former CEO of Apple France and chief of devs of Mac II-fx, interviewed by Amazing Computing, Nov 1996).

@Rudi

>If you want to keep the board low cost. It is much cheaper to use 2 Sata and 2 PCIe x1 connectors. The user can buy what ever he wants and use the PCIe x1 connector.

One must also note that it will take time before SATA drivers are done for the SOC.
One alternative is to use a supported SATA chip of a south bridge or on PCIex1.

>If you use a Southbridge it must be designed in, tested and driver must be written. That will add costs to your design.

Yes. Drivers are needed any way. Some ways it might be simpler than via some other.

>Most people only need Audio from the hdmi port of the graphics card.

yes. So far we do not have drivers for Radeon Audio.
Alternatives are:
-in P10xx case use the codec on i2S bus, T10xx does not have i2s
-in T10xx you can use audio card on PCIex1, USB etc. BUS.
-One way to proceed is to use southbridge on a PCIe card.

I have visioned two main alternatives.
Nano or mini-itx size board with 2...4 PCIe slots.
A) Board with 2 PCIe slots could use one PCIex4 slot to connect to southbridge & busboard.
A) Alternatively the slot could house a FPGA to do our own southbridge magic (initially this FPGA surely would not be available, just an option to develop later)
A) Second slot would be used for 3D GPU or for FPGAGA GFX+Audio
A) nano would have slots in non-standard locations
A) ITX would have one slot1 in standard and the other in non-standard location
A) MXM might be an option for wiring the slot2
B) ITX Board with more slots would have PCIex4 (physical x16) in standard position for GPU
B) one slot would be normal x1 slot (physical x16) in non-snadard location
B) one slot would be mini pcie x1, possible to be wired as ePCIe to busboard with southbridge (or busboard with PCIe to PCI bridge + SIL SATA + audio +++)
B) another miniPCIe slot would be configurable to be used as SATA or PCIe etc....
B) would have one SATA + one shared SATA/miniPCIe
Both A and B designs would have USB hub on board and geeky port with GPIO I2C and and etc... 8-P

This "vision" would enable building of tiny computers (PPC-PI/CD32/"CD64"/CDTV/"SDTV"), keyboard computers (like A600), larger expandable computers like A2000 & A1500 & A1200T.

(just need EUR1M to do it fast, EUR2M to release it with AOS4.3/AOS5 in sane time frame)
(with EUR50k I would have prototype in half a year time... perhaps, depending on time budget of some HW gurus I know)

UPDATE:
mini-DTX is remote option C). There one has two standard card slot positions, but mini-DTX cases rarely have enough room to sanely use both slots. Already mini-ITX board with latrge GPU card needs a mini-DTX case to fit in.

Last edited by KimmoK on 05-Nov-2015 at 02:02 PM.
Last edited by KimmoK on 05-Nov-2015 at 01:56 PM.
Last edited by KimmoK on 05-Nov-2015 at 01:51 PM.
Last edited by KimmoK on 05-Nov-2015 at 01:50 PM.
Last edited by KimmoK on 05-Nov-2015 at 01:49 PM.
Last edited by KimmoK on 05-Nov-2015 at 01:46 PM.
Last edited by KimmoK on 05-Nov-2015 at 01:43 PM.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?