Quote:

I think Zylesea has a strong and valid point there...

@ Zylesea

Quote:

Others might feel and fear a lot of things which may or may not be the result of fact-based rational thinking. In the end, what really matters is that you offer software that people like so much that they are willing to buy / use it. It is really as simple as that.

When people do not actually like the current product offering enough to justify the expense and need a lot of promises regarding the future to lure them into a purchase, they are most likely best advised to just stay away for the time being and perhaps reevaluate their decision in the future in case they are still interested.

It would be nice to have MorphOS running on all kinds of different hardware, including the newest Efika, brand new x86 PC's and Macs, NetBooks and laptops, but the reality is that the size of the MorphOS development team is, what, 20 people, 15, 10, or is it only 5 that can put up serious hours of work on it with the rest just being able to contribute a few hours here and there when they feel like it? They certainly aren't doing it for the money because there isn't any to be made from such a tiny user base. Just some pizza & beer money when you look at the amount that has been made and then spread it over the number of programmers and the amount of time it has taken to make MorphOS and also complete the free updates.

MorphOS started out on PPC for various reasons that made some kind of reasonable sense at that time and the path of least resistance has kept it on PPC, probably at least partly because of the tiny size of the team that is doing all the work on it and the reluctance to take on the increased workload that would be required to port MorphOS to another architecture. That is just a fact of life that there are limited resources (men and hours available to work on this hobby OS) and that it is more work to port the OS to another CPU and architecture than it is to continue porting it to any PPC computers that are available. It is easy for others who only sit on the sidelines watching MorphOS progress slowly to second guess the team and spout their views of what should have been done and which platforms should be ported to next, but none of them are actually doing the work for next to no pay, so they have no idea of exactly just how much trouble and how many hours are involved in doing what has already been completed, or how much more it would take to port to a different architecture and CPU.

Yes, it took maybe 18 months to get a stable port of MorphOS for the Efika and only the development team would know how much longer (or shorter, but I doubt it) it would take them to do a port to ARM, or any x86 hardware. I would think that everyone should take all of these things into their considerations before they criticize the MorphOS development team for not switching to some other platform. I am sure the team would like to have MorphOS on more modern and better performing hardware, but they must weigh the amount of time they think it will take them to complete such a switch against the amount of manpower and time the team has available to perform such a port to a different architecture. I am confident that they will make a switch some day if they can keep the team together, or perhaps get more help to make such a big step possible in a reasonable time frame. Only the development team can know what their resources really are and when the time will be right FOR THEM to make any change.

end rant

Amigadave: amen

[ Edited by pampers on 2009/8/28 17:09 ]

Quote:

That is not what I read from my long argument with Matt months ago? Yes, there might have to be other hardware changes besides just adding, or replacing the original RAM, or both to have an Efika with more RAM, but the pcb itself would not be different and the changes would be POSSIBLE, just not cost effective to make any change probable. The Efika would not have different pcb's for each different RAM configuration and everything else on the board COULD be replaced, or altered, but it is just not practical or cost effective. The design decision to only produce the Efika with 128mb of RAM in the first place was a mistake in my opinion, as I wonder how much difference in price it would have been in the beginning to fit it with the maximum 512mb of RAM? At least it was a mistake for us MorphOS users, but then we were not the target market for that board, the embedded auto makers were the market and they most likely did not need the additional RAM, so I understand the decision. Saving what ever amount of money on each board was additional profit in the pockets of Genesi and less capital outlay of funds to manufacture the Efika. It is just too bad that it did not make economic sense to make a smaller production run of the Efika with the maximum 512mb of RAM installed so the MorphOS users would have had a choice between the 128mb and 512mb boards.

It is my understanding that if someone, or some group made the firmware changes and the correct RAM (and possibly some other components that might be needed) could be found for sale, existing Efika's could be altered to use more RAM. This would be in violation of the firmware agreement that prohibits alteration and it is not likely that anyone or any group would do it because the benefit would not justify the cost in time, effort and parts, unless that someone, or some group got permission to do so and was willing to do it for free and the parts could be found.

I know that this is not going to happen, but look at the guy who spent thousands of Euros of his own money to redesign a new 030 motherboard for the A1000. I think stranger things have happened in the world of Amiga over the past 20 years, so anything is possible.

Quote:

where did that get from?

Quote:

I'm sorry but this is not the way it happens. I began to look into hardware myself lately, so let's give an example: let's say you have 150k (?,$, whatever it doesnt' matter) initial budget and you want to design the motherboard of your dreams. Let's say for the sake of argument that the PCB design phase with prototypes will cost you arount 80k? (it can be lower or higher, depending on the design of course). The factory tells you that production runs happen in batches of 500 -or similar quantity, for technical reasons but also for parts price reasons- , so you have to see that the remaining money (70k?) will last for the production AND the parts. For 500 units, this means you have to make sure that the cost/unit will cost at most 140?. But you're short of some cash, so you have to change the design and lower the specs somewhat. Now this varies greatly depending on the PCB, but let's say that RAM is easy to dismiss, because RAM chips are expensive and more expendable, the RAM capacity is variable. In an embedded board where most peripherals is on a SoC, you can't dismiss ethernet, you can't dismiss audio, you can't dismiss the CPU(!), so you have to take down on RAM.

Well, theoretically, you could lower your profits and use that amount to increase RAM, but that's really bad, bad, bad business practice. A product that doesn't make a profit is doomed to fail.

Disclaimer: I don't really know what happened with the Efika, and this is NOT an explanation of why it has only 128MB or RAM (which imho are just fine, the Efika is not a really desktop system). But it's a problem I've come unto myself and I think pretty much everyone that produces soemthing (and software of course, specs minimizing are not only hw-related).

Quote:

Indeed. But what if it happens? What if someone releases tomorrow or in 6/12 months a modern PPC motherboard that fulfills your wishes? Would everyone here rush out to buy it? I don't think so.

Quote:

Do some research about where most of the Efika boards sold to and read the comments about the 2% to 5% of the Efika boards sold having MorphOS licenses. Only makes sense that the remaining 95% to 98% of the Efika boards produced sold to the embedded and auto maker market, unless they are still sitting on supplier's shelves, but Genesi is sold out of their stock, so they must have gone somewhere.

[ Edited by amigadave on 2009/8/28 10:30 ]

Quote:

lol, I have 5 efikas here and I'm in no way related to the auto market - I do however work on embedded projects occasionally. If you were referring to embedded AND/OR auto markets, then ok, I agree 100% and there is no argument, my objection stands only if you were referring to the auto market alone.

Quote:
Amen to that! People can try MorphOS on the machines it is available, and if they like it as it is now, they should register, otherwise it would mean MorphOS is not the solution for them right now. Communications right now about any possible log term plan are out of the loop and do not change the way I can use my MorphOS machines right now.
On the other hand, however, some more inputs about the short term plans would not do much harm, IMHO. Having a better idea about the G4 Mac support for sure is not going to harm sales of Pegasoses or Efikas, nor registrations: I am pretty sure that most of us that were interested in that, had more than enough time to find, set up and register their machines.

And just to be clear: I'd like the team to know that I like and enjoy MorphOS as it is now (2.3), and appreciate their past and present work and efforts. In fact I did not register 2 copies of it for my two machines just to generically "support" them, but because I use said machines, and like doing so.

Kind regards,
Andrea

> If you're designing a car from scratch and not just plagiarising
> chassis from old models and components from old models, any i.MX
> chip would do the job.

I'm really no expert regarding automotive semiconductors but the following Freescale automotive documents:

http://www.freescale.com/files/automotive/doc/brochure/BR32BITAUTOMP.pdf
(no mention of i.MX at all)

http://www.freescale.com/files/microcontrollers/doc/roadmap/BRAUTOPRDCTMAP.pdf
(don't miss the smallish table near bottom on the right of page 2)

http://www.freescale.com/files/microcontrollers/doc/selector_guide/SG187.pdf
(pages 20 to 24)

...don't make the impression to me as if i.MX could currently replace each and every MPC55xx/56xx for each and every use case, even if the car is designed from scratch.

> Power familiy (from mobile phones to Super Computers)

Mobile phone with Power Architecture CPU? Where?

Quote:


Agree

Quote:

Strangely enough, Freescale lists all their m68k chips as active/available, allthough sometimes with insane prices - but the 7447 used in pegasos2 is listed as .. dead

The rest of your post, regarding morphos and legacy... sounds like me in 2002

Quote:

Well, if it was so, it would just mean that freescale has done it job right.

Sorry for the cheap shot, just couldn't resist.

PowerPC deserved better. Who knows, perhaps the "Amiga curse" will only end when its endless stream of bastards descendants forget about their Motorola heritage.

> if it was so, it would just mean that freescale has done it job
> right. [...] PowerPC deserved better.

Sorry, I don't quite understand. Do you think ARM (i.MX) should replace Power Architecture completely within Freescale? Or do you think Power Architecture should remain as a Freescale supported architecture?

Quote:

Sorry, cheap joke, and badly worded from my part. What I pretended to mean is that, if freescale had ARM substitutes for every PowerPC part, it would mean they had done their job right. In my opinion, freescale isnow a good provider, and I blame them for many failures in Amiga related history. Which qualifies me as some kind of ungrateful bastard, but I can't help thinking that way.
Back to my "joke" I realize it's not funny at all, even if I got the meaning right. Andreas, do you have a category for stupid posts in your knowledge database? And most of all, why you never answer, to the several times I've asked you about how you can organize all those URLs you have?
What are we going to do the day you are not here?

Quote:

PowerPC mobile phone

> PowerPC mobile phone 8-)