> If you like to quote from over 2 year old threads

I do :-)

> you probably should add the original datestamp to your quotes

I expect anyone who leans towards drawing conclusions from these quotes to click the links beforehand. And be it just to know who exactly wrote those lines :-)

> at least use past tense ;)

While I'm not a native English speaker I think that past tense in English language implies that the action is completed, i.e. that CISC now doesn't feel this way anymore. But I don't know if CISC's stance regarding a Wii port changed over the past 2 years as I haven't read anything to that effect nor to the contrary. Do you know if it changed?

Quote:

I can assure you that CISC didn't hold his breath for 2 years waiting for a sponsored Wii to get MorphOS working on it.
Unfortunately, CISC's time to work on MorphOS is very limited nowadays, so you can definitely put his formerly planned actions into past tense ....

> I can assure you that CISC didn't hold his breath for 2 years waiting
> for a sponsored Wii to get MorphOS working on it.

Hehe, CISC waiting for getting a sponsored Wii to happen is not even what my use of present tense implies anyway :-)

> you can definitely put his formerly planned actions into past tense

I surely didn't even go as far as reading his past comments as "planned actions". In fact, I only wrote about his general opinion on the Wii's suitability for MorphOS. But interesting to know that he had actually planned to do a MorphOS port to the Wii. Thanks for giving this information.
Back to what I really wrote: Him not being able to work on a Wii port anymore (as you say) wouldn't necessarily mean that he doesn't generally regard the Wii as a suitable target platform for MorphOS anymore, would it?

Quote:

Not so much a plan as a wish. ;)

Quote:

Well, perhaps not a very suitable platform, but a fun exercise, and at the time indeed the only viable one. Now there is the possibility of PS3 due to recent efforts, but given Sony's stance on the matter (even going so far as suing end-users) it's still as unlikely as ever.


- CISC

> Let's not look what *might* be possible, but rather use what is
> available now. And currently Apple G4s are the better target
> (port done, cheap, powerful enough) than a PS3.

If "port done" was a valid aspect for evaluating potential target platforms for MorphOS then we'd still be stuck with MorphOS for PowerUP hardware at best ;-)
Regarding "cheap": Currently, the PS3 starts new at 240 EUR (with games) or 295 EUR (without games) in the EU.
Regarding "powerful enough": Yes, a new 1.6 GHz G4 processor upgrade for the PowerMac G4 starts at (converted) 113 EUR currently, which leaves 127 EUR for a used PowerMac G4. That's assuming the 3.2 GHz Cell PPE is about as powerful as a 1.6 GHz G4, which is just an estimation from my side.

> Not so much a plan as a wish. ;)

That's what I thought prior to pega-1 telling me otherwise. Thanks for clarification.

> perhaps not a very suitable platform

So am I right in concluding that your stance regarding the Wii's general suitability for a MorphOS port in fact did *not* change? (Back then you told to the effect of "suitable, but with limitations", i.e. not "very" suitable, but still).

someone explain how firmware (or upgrades) affects ps3 port? This is just for clarification. I assume you do a fw-boot as in efika, and install Mos. The issue would then come under sony upgrades then, if you had sony'os boot ps3? what if we stripped that as in one os (mos?). as pega-1 said dead issue just want reasons in clarity ,as I don't own a ps3, but would consider it if morphos could be on it.

@xyphoid

Check the link in the first post of this thread.

Quote:

Sure, what I said back then is still valid; among current-gen consoles the Wii is still the most likely candidate for MorphOS, if MorphOS were to be ported to a console, ie there are no plans to do so, and even if there were it would be for specially interested people only.

It's worth noting some of the limitations I listed back then have changed though:
Quote:

Indeed USB 2.0 and SDHC was simply a matter of tweaking the right registers (and there are even official Nintendo IOSes now that support it), which makes it easy to get all the storage you need onto the Wii. The remaining limitations (for which there are likely to be no solution) are the 88MBs of RAM and the 480p/576i resolution...


- CISC

> what I said back then is still valid

Thanks for confirming that my use of present tense when quoting you was justified, albeit needlessly criticized by some.

I can understand Pega-1's stance on the PS3, but even with the problems presented by Sony and the difficulty in porting to a hacked console I'd buy one in a second if MOS was ported to it.
From a hardware stance its much more powerful then the Wii.
And a 3.2 MMhz Cell processor might be the most powerful CPU we could run MOS on.

> a 3.2 MMhz Cell processor might be the most powerful CPU we could run MOS on.

Ignoring the Cell's SPEs, I think that a 2.0 GHz G4 is faster in computations where the G4's low memory bandwidth is not a bottleneck.

I'm looking for something compact as the mini, but more powerful, which can also be used as my entertainment system. ps3 fits the bill, unless a port to some micro tower 2+g4.

@ Andreas Wolf:

I didn't say "port done" would imply to think about (and actually *do*) future ports. but the workbook for the foreseeable future seems to be well filled: Powerbook and G5. I think there's no room for more hardwrae support for the next couple of month (now to 48 months).
Maybe I worded teh sentence a bit unsharp (I guess most ppl are able to interpret these unsharpness corectly), but the "port one" part of my sentence was primarily focused to those ppl who don't take what's available now, but project everything into a coming future (not you, you enjoy the mini probably as much as I do). But this attitude "I don't buy now because in two more weeks (tm) comes something much more potent " is what this micro market hurts quite a bit.
And I think in this regard Pega-1's statement was very clear. PS3 port will not happen - at lest not in a foreseeable future. So if you're interested in MorphOS you don't need to wait, but can use it today.

[ Editiert durch Zylesea an 2010/10/23 21:38 ]

> I didn't say "port done" would imply to think about (and actually *do*)
> future ports. [...] the "port one" part of my sentence was primarily focused
> to those ppl who don't take what's available now, but project everything
> into a coming future

Then it seems that you didn't say what you meant. You wrote:

"currently Apple G4s are the better target (port done, cheap, powerful enough) than a PS3."

You clearly used "port done" as a point for G4 Macs and against PS3. But as I said that's a non-argument really in a discussion about whether a port to PS3 makes sense or not. And that *currently* G4 Mac is better than PS3 for running MorphOS is a truism because *currently* MorphOS doesn't run on PS3 at all (which in turn is the sole reason this very thread exists).

> the workbook for the foreseeable future seems to be well filled:
> Powerbook and G5. I think there's no room for more hardwrae
> support for the next couple of month (now to 48 months).

In the context of what's needed to port an OS a fully hacked PS3 would be no more different from G4 Mac than G5 Mac is. I guess the people who want to see MorphOS on PS3 would have no problem with seeing it instead of (rather than additionally to) MorphOS on G5 Mac.
Btw, you think that MorphOS for G5 Macs will only come in 4 years time? I really hope it will be there sooner.

> you enjoy the mini probably as much as I do

You betcha :-)

> this attitude "I don't buy now because in two more weeks (tm) comes
> something much more potent " is what this micro market hurts quite a bit.

That's surely not the case here because even before pega-1's statement there was no indication that a port to PS3 is even considered in the slightest way by the MorphOS Team, quite to the contrary. Else this very thread wouldn't exist. But I can easily imagine there're people delaying the purchase of a MorphOS license waiting for the G5 Mac port to appear ;-)

Quote:

It would prove an interesting comparison, the Cell does have a speed advantage but as an in order processor it might lose to a slower G4.
However, the SPEs could be used to offload some of the processor intensive tasks (like video encoding and decoding). Everyone keeps thinking of the units as additional processor cores when its probab ly more accurate to think of them as specialized coprocessors.

And unlike previous posts I've seen here, the SPEs do not look that difficult to program.

The PS3 would give us some capabilities that would be difficult to match on conventional PPC based systems.

> It would prove an interesting comparison, the Cell does have a speed
> advantage but as an in order processor it might lose to a slower G4.

I found some DMIPS figures comparing G4 and Cell PPE:

------------------------------
PS3 Cell 3.2GHz: 1879.630
PowerPC G4 1.25GHz: 2202.600
------------------------------
http://forum.beyond3d.com/showpost.php?p=880950

So in terms of DMIPS the Cell PPE would equal a 1.067 GHz G4.

> Everyone keeps thinking of the units as additional processor cores when
> its probab ly more accurate to think of them as specialized coprocessors.

The only people I've seen calling the SPEs as complete processor cores have been the ones who are either uneducated on this matter or have been uttering marketing verbiage (or both). But that's surely far from "everyone". I guess this is the same kind of people calling the MPC5121e a triple-core processor ;-)

Well I'm glad at least you get that point. It confuses me when people bring up the SPEs as if they're some sort of disadvantage.
I wonder how many PS3 game programmers ignore them completely? They don't have to be used and with an OS like MorphOS that only runs on one core these little sub units could be a neat way to off load tasks from the core processor.

Its a pity that this idea keeps getting nixed by the developers. The prospect just fires up the hacker lurking in the depths of my soul.

> I wonder how many PS3 game programmers ignore them completely?

I guess very few. I don't know much when it comes to current console games but I've been under the impression that games for the PS3 are about as demanding as games for the XBox360. Xenon's core is similar to Cell's*, but has three of them as well as improved AltiVec (VMX128). To make up for that I doubt you could leave the Cell's SPEs untouched.

* http://www.gamasutra.com/view/feature/3904/processing_the_truth_an_interview_.php

Probably true. Comparing Xenon to Cell B.E. (and then that processor's recently cancelled sucessors) it makes you wonder what IBM could create given the die sizes used by Intel or AMD.

> Comparing Xenon to Cell B.E. (and then that processor's recently cancelled sucessors)

While PowerXCell 8i's successors PowerXCell 32ii (2 PPEs + 32 SPEs) and PowerXCell 32iv (4 PPEs + 32 SPEs) were indeed officially cancelled a year ago it's apparently not over yet:

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=6768&forum=3&start=35

> it makes you wonder what IBM could create given the die sizes
> used by Intel or AMD.

Things like that:

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=7001&forum=3&start=431

...and probably also even better things :-)

I reread my earlier post again. With the "port done" pharse I also wanted to point to the fact, that other Apple hardware ports may benefit more from the ports done yet than a PS3 port. But I agree that was at best half hearted worded. Don't mind that post too much anyway.

And: The "48 month" were a typo/miscalculation. I meant 24 months (which is not meant literally, but more like a "till about end of 2012 the workbok is filled with Apple ppc hardware support". I think we cannot predict what the market ("our niche", ppc and - to a lesser account - computing in general) is in two years tme and what approach will be worthwhile to follow by then.

And as Pega-1 a few times indirectly indicated already: continued work on MorphOS also depends on success of the current work. I.e. if no new users arrive, the motivation to continue development will probably not become stronger.

> With the "port done" pharse I also wanted to point to the fact, that other Apple
> hardware ports may benefit more from the ports done yet than a PS3 port.

That's true for PowerBook G4 for sure. But as I said, I doubt that the existing port to G4 Mac benefits the porting to G5 Mac any more than it would benefit porting to PS3.

> The "48 month" were a typo/miscalculation. I meant 24 months

Okay, 2 years is an estimation that's not too pessimistic I think.

Quote:

I think in a market as tiny as the MorphOS market a standard hardware is of particulr benefit. Currently we have ppc w/ and w/o Altivec. Better if all MorphOS maschines had Altivec. It helps to reduce the development overhead. I mean support for Cell's SPEs is not done by a simple Compiler flag. And while maybe relatively easy to program,it adds development overhead. As coder you must maintain code for a vanilla ppc, ppc + Altivec, ppc+SPE. I don't think taht is helpful.. Unless the market doesn't grow significantly I think it is better to focus on one standard as much as possible.
It is more or less a similar arguement with the XMOS X1000 stuff. The hardware basis gets too heterogene. Applications need to take care what actual hardware is present, users have to keep track what software supports their hardware, developers must decide what hardware they demand for their application and so on. The butterfly country is too little for that. If one day there are significantly more users, there may be more resources to spend to maintain variability. But not now.

Ah, but the possibilities Zylesea!
I swear there must be a German somewhere back in my Irish ancestry because the thought of using such technology practically gives me an erection.

BTW - Thanks for the references Andreas. I've spent so much time studying this stuff that its heartening to see its not going to be abandoned.

Oh, and Zylesea, the Cell (if you ignored the SPE's) should run the same PPC code. The Nvidia GPU, that's a problem considering the developers aversion the Nvidia hardware. And we may have some variants in our future anyway if the developers ever decide to tackle SMP.

One last re-edit: Cell and Power8 technology? Damn Andreas, that sounds far too optimistic.(although the though gave me a cold chill). Then I think back on your track record of predictions so far...
Damn, let's find someone with deep pockets build something with this and pay the development team for MorphOS4 (Qbox and all) and take back PC from those nimrods at Intel.

[ Edited by Jim on 2010/10/25 4:49 ]