@Britelite

That is normal. AmigaOS 2 wasn't 100% compatible with AmigaOS 1, AmigaOS 3 wasn't 100% compatible with AmigaOS 2 and AmigaOS 4 isn't 100% compatible with AmigaOS 3.

MorphOS is not compatible with AmigaOS because you can't run program compiled for MorphOS on AmigaOS.

_________________
www.exec.pl, www.amigaos.pl, eXec TV

@feanor

A poll on the various sites would be quick and easy for you to do
I suspect you will get a lot more responses that way.

_________________
If your nose runs and your feet smell, you're upside down.

@TiredofLife

Polls are anonymous, so not really an option. Well, what might work is I found a ratio, how many interested send vs how many interested do not send. That might be something I could use to extrapolate to the real numbers.

Last edited by feanor on 30-Sep-2009 at 10:23 AM.

@AlexC

Dude, all I did was cut'n'paste his questions and write the answers next to them. How lazy does one have to be anyway

@Comi

Troika were a bunch of poor *****. Except some hot air they had naothing to offer. I am deeply surprised they are still taken serous by some ppl (but there are also some ppl still taking A Inc serious..).

_________________
My programs: via.bckrs.de
MorphOS user since V0.4 (2001)

@feanor

I would still recommend to ask also in PPC-Linux communities for intrested people, you might find more potential customers there than you will find here.

EMAil send ...

Last edited by Arko on 30-Sep-2009 at 11:06 AM.

_________________
AmigaONE. Haha. Just because you can put label on it does not make it Amiga.

I borrowed this comments from here (#27 & #28):
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=38873&forum=2&start=20&order=0

@eXec.pl

Could you please move your babbling to another thread, this one is about new ppc hardware and not about what some poor **** say is teH ReaL AmigA.

_________________
My programs: via.bckrs.de
MorphOS user since V0.4 (2001)

@feanor

Quote:

feanor wrote: @AlexC Lol, at least you're honest. Ok, if that's the only reason, I'll think of preparing a form somewhere instead of a mail (you'd send an email btw, in less time that it took you to post this :)

I knew I had this one coming

_________________
AlexC's free OS4 software collection

AmigaOne XE/X1000/X5000/UAE-PPC OS4 laptop/X-10 Home Automation

@eXec.pl and anyone else taking this thread off-topic

I've made it very clear in this thread that we won't have court case debate and AOS vs. MOS OS debate. I've said to simply make a new thread for that it you are intent on discussing such. I've been quite patient on this. I will be no longer if anyone keeps it up.

Last edited by fairlanefastback on 30-Sep-2009 at 02:30 PM.
Last edited by fairlanefastback on 30-Sep-2009 at 02:29 PM.

_________________
Pegasos2 G3 running AOS 4.1 and MorphOS 2.0
Amikit user, tinkering with Icaros VM (AROS)
EFIKA owner
Amiga 1200

@feanor

'Altivec can make some things faster yes.'

As I suspected, and as the benchmarks show.

Now;

Would the Be/Haiku guys be interested in taking there gem into the AltiVec
realm ? Maybe not right away, but ........ as it matures.

And....

Would the PCIe slot be able to efficiently handle a (at least somewhat)
super-duper graphics card at;

1. 4x (P1022)
2. 8x (MPC8610)

And which, do you 'think / guess', would be better for longevity of the design ? Maybe this is one you should be asking yourself and not speculating about here.

Finally;

Could the overall use/design of libfreevec be more deeply rooted into
the various capable operating systems for further improvments / efficiency ?


P.S. email to follow soon.

@feanor

I'd be interested in a machine running Amigaos4.x. But haven't sent any mail or voted for a cpu.

_________________
I'm an antique. Don't light my fuse

@Mechanic

Quote:

Mechanic wrote: Would the Be/Haiku guys be interested in taking there gem into the AltiVec realm ? Maybe not right away, but ........ as it matures.

Well, you've just found my secret dream: port/integrate libfreevec into Haiku's libc which is MUCH MUCH cleaner than the mess that is glibc. :)

Quote:

And.... Would the PCIe slot be able to efficiently handle a (at least somewhat) super-duper graphics card at; 1. 4x (P1022) 2. 8x (MPC8610)

Ok, that's also a very nice question. I also was curious about this and did a little search and I found this link:

http://www.tomshardware.com/reviews/pci-express-scaling-analysis,1572-8.html

So, compared to the current situation a 4x is enough. But. The 8x is much better in the long run.
People would be able to at least plug in a nice ATI/Nvidia card and have graphics quality of only 1-2 generations before - and not 10 like it is now. The speed is not that important, because the CPU couldn't handle so much bandwidth anyway -ie it would make little sense to plug in an ATI 4890 on this system. But a 4670 or a Nvidia 9600 would be perfectly adequate. The only thing remaining would be the driver. From that aspect, I also though of including some basic gfx controller chip, but I'm not sure if that's such a good idea. Sure it would mean the platform is integrated and there is no need to worry about lots of extra drivers, but it's also limiting. I'd also love to hear what people think about that, because I admit I didn't put it in the poll.

Quote:

And which, do you 'think / guess', would be better for longevity of the design ? Maybe this is one you should be asking yourself and not speculating about here.

I think that a MPC86xx motherboard will surely be enough for the following years, but since the e600 core is becoming gradually EOLed, we should plan our migration to a newer platform. Unless we manage somehow to convince Freescale to continue the e600 core (I'd love that, personally, but I have to be a realist). Anyway, my personal feeling is that we'll go for a MPC8610 initially and then move to P1022 (or what's better then) in a few years.

Quote:

Could the overall use/design of libfreevec be more deeply rooted into the various capable operating systems for further improvments / efficiency ?

libfreevec is just a subset of a libc, all functions are POSIX so any OS that is POSIX or even has similar functions with similar names/functionality -which I think AmigaOS/MorphOS do, could make use of the libfreevec code. So far I haven't received any mail from either OS, though I have received mails from NetBSD users, but on the other hand, I guess MorphOS itself is quite AltiVec optimised so they already do this kind of work. Haiku does not use such code and I would like to contribute this code -ported to all 3 major SIMD engines, Altivec/SSE/NEON, so that the benefits will get integrated properly into the OS in a uniform way. I don't know if the Haiku devs will accept this, but I hope they do.

@feanor

Quote:

But a 4670 or a Nvidia 9600 would be perfectly adequate. The only thing remaining would be the driver.

I'm pretty sure there's someone on this forum who's got a mostly working Radeon HD driver for OS 4. There was a news item about it a while back. I don't have the link handy though.

BTW, emailed in support of either 8610 or P1022.

Edit: Here it is: http://hdrlab.org.nz/radeonhd-driver/

Last edited by Gleng on 30-Sep-2009 at 04:20 PM.

_________________

@feanor

I don't know, but if there's 5 silent persons to every vocal person on these Amiga related forums so if you'll get 100 emails there might be 500 in total who are interested.

_________________
Rock lobster bit me. My Workbench has always preferences. X1000 + AmigaOS4.1 FE
"Anyone can build a fast CPU. The trick is to build a fast system." -Seymour Cray

@feanor


Quote:

But. The 8x is much better in the long run. People would be able to at least plug in a nice ATI/Nvidia card and have graphics quality of only 1-2 generations before - and not 10 like it is now. The speed is not that important, because the CPU couldn't handle so much bandwidth anyway -ie it would make little sense to plug in an ATI 4890 on this system. But a 4670 or a Nvidia 9600 would be perfectly adequate.

I will certainly agree, especially with the 'long run'.

Quote:

The only thing remaining would be the driver. From that aspect, I also though of including some basic gfx controller chip, but I'm not sure if that's such a good idea. Sure it would mean the platform is integrated and there is no need to worry about lots of extra drivers, but it's also limiting.

Again, and I'm pretty sure most others, would agree. Perhaps some deal
could be worked out to ship a card with the board.



Quote:

Anyway, my personal feeling is that we'll go for a MPC8610 initially and then move to P1022 (or what's better then) in a few years.

Ahh, those magic words, 'in a few years'. In other words, not a moving target.
That is great. This would give the software people time to do more
than dink with constant upgrades/difference bugs, and the hardware guys time
to develop a next generation board while the present one puts some butter
on their (and your) bread.


Next question, and it must be asked. Will DMA be fully implemented ?

Finally.
As a (at this time) hobby computer, will technical specs/data be available so
those who like to play around and tinker not have to do so with blinders on ?

@Mechanic
Quote:

Ahh, those magic words, 'in a few years'. In other words, not a moving target. That is great. This would give the software people time to do more than dink with constant upgrades/difference bugs, and the hardware guys time to develop a next generation board while the present one puts some butter on their (and your) bread.

I'm not sure if this comment was supposed to carry any irony in it :)
In any case, well, if we manage to sell 10k units (I wish!) we could fund the R&D of the next system, which would include... er, I don't know, but let's cross that bridge when we get there :)

Quote:

Next question, and it must be asked. Will DMA be fully implemented ?

Implented where? In I/O operations like SATA? Well the controller we've chosen is fully supported in Linux -and iirc and Haiku as well. It would need an AmigaOS/MorphOS driver, but I don't know how easy it is to do DMA there.

Quote:

Finally. As a (at this time) hobby computer, will technical specs/data be available so those who like to play around and tinker not have to do so with blinders on ?

I'm a freak about openness (hate GPL btw, but love MIT/BSD licensees). While I don't know if I will opensource the full schematics, I will definitely send board info to anyone that wants it to do stuff. Opening specs about the board is one thing, opening specs about the board components (eg. SATA controllers, CPU, etc) is another and not in my power anyway. But I don't see why it should not happen.

Last edited by feanor on 30-Sep-2009 at 10:13 PM.
Last edited by feanor on 30-Sep-2009 at 09:48 PM.
Last edited by feanor on 30-Sep-2009 at 09:48 PM.

@feanor

. Sign me up.

@Tomppeli &Feanor


Quote:

I don't know, but if there's 5 silent persons to every vocal person on these Amiga related forums so if you'll get 100 emails there might be 500 in total who are interested.

Having official support from Os devs (AOS and MOS) could help too. What else are we doing of that board?

Petrol.

@Tomppeli

Quote:

I don't know, but if there's 5 silent persons to every vocal person on these Amiga related forums so if you'll get 100 emails there might be 500 in total who are interested.

Problem is, you get less then 5% of those will actually take money out of their pocket to buy the hardware. I ran a poll years ago before I formed Black Troll Technology for AROS. I had nearly 300 people state they were interested and I sure didn't get anything close to that so I'm going high end with 5%. If he wants to sell 500, he better look at a minimum of 10,000 responses.

_________________
Dammy

@damocles

LOL, thankfully I don't aim only at the Amiga/MorphOS market :D