After a little chat on the MorphOS IRC channel it seems that Bigfoot has no more access to a G5 machine. This has due to logistic reasons (it seems he is moving far away from his home). Obviously without a G5 machine, the hopes to having him working on a G5 port fall down. .

Now I know that at the moment priorities are others (one for all, MOS 3.0 is still not out and must be finished), but please can we seriously speak about the possibility to have a G5 port after 3.0 and before future moving plans to port MorphOS on other architectures like x86 or Arm? If so, in the meanwhile, we (users) could start to organize a bounty to collect money to let Bigfoot and/or other Mos developers have a G5 machine.


I know that G5 is just another old technology, but we all know that porting MOS to it will surely take much less time and it will be surely less traumatic for compatibility than porting MorphOS to a totally new architecture like x86 and ARM. So please, seriously consider this MorphOS Team. Having MOS running on a very fast and cheap machine like PowerMacs G5s will ensure users to have a long term usable machine while the long PPC->x86/ARM transition tunnel will be travelled by MorphOS Team, and we now this tunnel will take a long long long time..

[ Edited by Miky060 10.02.2012 - 12:35 ]

a) there are other team-members who have access to G5-HW and the capabilities to adavance such a port
b) no firm commitment to G5 has been given, so it's all a bit premature
c) AFAIR there were some statements that the money collected on MorphOS-licences was to be used to cover (amnong other things) the costs of HW the team needs for developing/testing

Kronos,
Quote:

Nice to hear this. But this does not prohibit community to give a contribute with a bounty if this can help to improve the time needed to reach this goal. Other bounties has already done to donate hardware to external developers to let them develop for morphos, so we could do the same for this. More people inside the MorphOS Team will have a G5 machine sooner will have MorphOS ported to them.


Quote:

I know nothing has still be decided, I am indeed saying that we should start to seriously "speak" about this, so that when current priorities will be finally reached, they can start to work on this other important goal.





[ Edited by Miky060 10.02.2012 - 13:58 ]

Kronos,
Quote:

Well, we can always hope.
And I'd support an additional bounty.

Kronos, Quote:

I'd still be willing to offer what financial (and hardware) assistance I can.

I've promised R400 cards to Frank, but I ran into a hitch when I found out that some FireGL X3s don't re-flash correctly.
Amigadave and I (with the assistance of a third party) are working that out right now.

I'm not sure that we can be of that much assistance, but I think Miky is right, we (the community) would like to help.


[ Edited by Jim 10.02.2012 - 13:28 ]

Not to be the party pooper but.... how much do you expact to collect for the bounty in a community that has to choose between so many bounties ? And are people that have supported this bounty going to be willing to pay the full price for a license ?

I was ready to start a similar topic -MorphOS G5 support-, count me in for a bounty.

>are people that have supported this bounty going to be willing to pay the full price for a license ?

At only 111 euros I don't see why not.

[ Edited by Jim 10.02.2012 - 17:41 ]

I'd rather see them work on a next generation MorphOS that could run on any architecture than see another PPC port. Of course a PPC port would take less time, but the more time is spent on another PPC port, the later the new MorphOS will appear.

So: better start now. I wouldn't be surprised if such a port would take 6 months-1 year. That's way too long.

G5, ARM, Intel... Always the same topic, just the CPU name changes

The CPU is not the problem in general: most of issues delaying MorphOS releases on new hardware are related to peripherals (and so drivers) !!

A board is not limited to its CPU.

>I wouldn't be surprised if such a port would take 6 months-1 year. That's way too long.

I can wait that long. I can continue to use what I have now for awhile.
What I'm not sure I can wait for is support for a new ISA.

And I'm building an AROS system, so MorphOS X86 has limited appeal to me.
ARM would be cool, but it changes so fast I'd be afraid of investing in a rapidly obsolescent platform.
A 2.0-2.7 GHZ G5 would remain relatively powerful (for the couple of years or more its going to take for an ISA jump).

If the MorphOS Team needs to buy SW, HW, licenses or whatever, they can do so, they have money (or should have at least). It's not that it's a lot of money either. It's hardly a development machine that stands in the way of a G5 port of MorphOS.

I agree that it might be premature to start a bounty for this, but as Miky060 stated, it is not too early to discuss it.

I am not convinced that a bounty is needed, as supporting the G5's is in my mind the most logical next step for the Team to take, that will give added performance and possibly HD video playback, with current, or future video card drivers. Supporting the G5's will also mean new SATA controller support and other drivers required to fully support the G5 hardware.

As Yomgui stated, bounties for other drivers might be more productive than a G5 bounty at this time.

As an owner of a dual 2.7GHz G5 PowerMac, I will support any initiative to get a port of MorphOS that works on it, but we should not ask for it until after the MorphOS3.x version is released, that also includes the new 3D driver support and the other new features that the Team has planned, but won't be finished in time for the initial release of MorphOS3.0. Who knows when that version of 3.x will be finished and work can start on supporting any G5 models.

After that, I don't have any problem with the port taking 6 to 12 months, and don't doubt that the port will take approximately that amount of time to complete.

There are probably many MorphOS users and Team members, who think that MorphOS3.x's support of the 1.67GHz G4 PowerBook's is sufficient while we are waiting what ever amount of time it takes for the transition from PPC to a different architecture. There might be some other users and maybe even a few Team members, who think that until there is more advanced software for MorphOS to take advantage of, there is no need to add support for any G5 models. My answer to them is "what about HD video decompression & playback?", unless Fab, or someone else wants to step forward and tell me that supporting the G5 PowerMac will NOT give us 1080p video decompression & playback capability. (I might be using the wrong terms, but I think you know what I am writing about)

What software are we going to be able to run when the initial port of MorphOS to ARM or x86 is completed, unless there is some magic that allows it to run Linux software, or has a built-in "Wine" capability, other than running old 68k Amiga software through UAE?

Many users are unsatisfied with the amount, or quality of software we have right now, or would at least want more software. How is that situation not going to be worse when MorphOS is first ported to ARM, or x86?

With the current speed and efficiency of MorphOS and how it runs great on old and slow G4 computers, plus the possibility of porting it to several G5 Mac models in a relatively short period of time to gain even more speed, coupled with the fact that FreeScale and IBM are still working on designing and manufacturing new PPC (or Power Architecture if you prefer that term) CPU's, maybe MorphOS will never NEED to be moved from PPC to ARM, or x86? Of course that depends on a lot of variables and would need those new QorIQ CPU's, or SOC's to be available to the general public at reasonable prices, and preferably, a desktop motherboard designed, mass produced and available for purchase by MorphOS users, at a reasonable price, so we don't have to pay several thousands of dollars or euros to buy one because it has been made by a company like A-Eon, or ACube in such low quantities.

My perspective is a selfish one, as I have my fast G4 laptop and very fast G5 PowerMac waiting for MorphOS3.0 and MorphOS3.x w/G5 support, plus I have my X1000 which should be delivered on Feb. 15th, so I am not thinking of buying any more new (or used) computers for my AmigaOS4.x & MorphOS3.x hobby for a very long time.

The concern regarding what software will be available (without counting software run through UAE) for a port of MorphOS3.x on ARM, or x86, and how new software for either of those ports will be able to run on our existing PPC MorphOS computers, is a real concern of mine, as either of those ports can be seen as a further fracturing of our already small community. My thought is that we already have AROS for x86 & ARM, and those that want MorphOS to be ported to those architectures, might be better off supporting AROS and bringing it up to the same standards of quality that MorphOS currently enjoys.

Sorry for the long post.

Edit: I hope the situation of "Big Foot" being without a G5 to use and develop on when he moves, is a temporary problem that will be resolved before all work on supporting G4 PowerBook's and new 3D (plus any other announced features for MorphOS3.x) are completed. Hopefully by the time the "Team" is ready to move on to supporting any G5 Mac models (if they ever decide to support any of the G5's), "Big Foot" will have a G5 Mac model again to use.

[ Edited by amigadave 10.02.2012 - 15:52 ]

Quote:
Every app that is already ported to x86 (which means most apps that are currently available: including MPlayer, OWB,...). And guess what ? You'll be able to play 1080p ;)

> unless Fab, or someone else wants to step forward and tell me
> that supporting the G5 PowerMac will NOT give us 1080p video
> decompression & playback capability

Fab has already made a statement regarding 1080p video handling on the 2.7 GHz PowerMac G5:

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=7372&forum=11&start=19

takemehomegrandma,
Quote:

No, it's definitely not a development machine that stands in the way. We stated before that some work on G5 was done as some proof of concept work.
The problem with more hardware support is that it also adds more complexity to the software and its support. Does a Mac G5 tower really have that much appeal to be worth supported ? I'm definitely not sure if many people want to place some noisy and bulky tower with a weight of roughly 20kg under their desk just to play 1080p videos (unless you already have such machine waiting for support).
Last but not least our resources to write drivers for all kinds of hardware is rather limited. Getting PowerMac G5 supported requires at least a new set of SATA drivers and all the bells and whistles required for thermal management. I currently can't locate anyone in MorphOS team who is willing to take that challenge.
Money is definitely not the driving force to get things forward in that regard for the majority of its members. So I don't see how a bounty would help here.

But Sata drivers will need to come sometime, I dunno if the G5 Mac's cooling system is unique, but every active cooling system will need some form thermal mangment sofware support, irrescpective of the CPU family?

Perhaps they have enough on the 'to-do' list in the immediate term re: 'complete enough' support for laptop(s). This should bring some framework for active cooling management anyway, and hopefully frequency stepping at some point.

More importantly, has anyone designed a 'Wilkommen' animation and CD cover for MorphOS 3.0 yet?

I am sure that there are more than the two issues that pega-1 mentioned that need work to make a port to any G5 possible, but instead of starting a bounty for G5 support, maybe we need to find 2 to 5 new developers that would be willing to take on the task of writing a G5 PowerMac compatible SATA driver and the required thermal management software?

Not an easy task to find qualified programmers who are interested in doing such work, even if we also raise some money to pay them for their work. Then the MorphOS Dev. Team would have to approve of these developers, as they would probably need some access to parts, or all of the MorphOS code in order to integrate their work into the OS.

Sounds difficult, but not impossible to achieve, if we can find the right programmers, who have sufficient skills and free time. They could even get started working on the two items pega-1 mentioned for MorphOS without negatively affecting the current work to complete MorphOS3.0 for the G4 PowerBook's. Of course they would need permission and acceptance from the rest of the MorphOS Dev. Team first.

I know, just wishful thinking.

>I currently can't locate anyone in MorphOS team who is willing to take that challenge.

Pretty straight forward. So Miky's idea isn't likely to change this. OK, for the time being, no G5 support.

> Sata drivers will need to come sometime

SATA drivers have been in MorphOS since version 2.5.

> I dunno if the G5 Mac's cooling system is unique, but every
> active cooling system will need some form thermal mangment
> sofware support, irrescpective of the CPU family?

I doubt that the active cooling system of my Mac mini G4 is controlled by any software component of MorphOS.

I am not sure how the "Proof of Concept" was done to boot MorphOS on the dual 2.7GHz G5 PowerMac, without some kind of rudimentary SATA support for that machine, but it appears that more work is needed to perfect it, or pega-1 would not have mentioned it as one of the two things that needs a programmer to work on to provide support for G5 Mac's. Unless he is referring to the SATA controller in a different G5, like some of the iMac's?

As for the thermal management software, if is obviously more important in a G5 system that is capable of producing considerable more heat than probably all G4 systems.

If I had to guess, I would say that thermal management software might be one of the reasons for the delay in the release of MorphOS3.0 for the PowerBook, where thermal management is also an important concern, if you want to retain long system life without having the fans running full speed all the time. Maybe not, it might be done already and they are working on something else, which is causing the delay.

Edit: As for pega-1's comment about the G5 being noisy, I find my dual 2.7GHz G5 is much quieter than my dual 1.25GHz G4 MDD was before I replaced the main fan, and is probably still quieter most of the time, until you use some software that ramps up the CPU usage. Then the fans speed up and are noticeable. The rest of the time the G5 noise level is negligible. It is large and heavy, but I don't mind that for a desktop system under the desk. It makes for a great foot warmer during the Winter months.

[ Edited by amigadave 11.02.2012 - 16:32 ]

Andreas_Wolf,
Quote:

Silicon Image SATA drivers, not those for the G5.

Andreas_Wolf,
Quote:

If I remember correctly, the initial G5 port left the cooling system running at full speed. On a G5, this could be quite noisy.

Frank has some legitimate points.
AND, if none of the developers want to address them, well...for the time being we're done.

when I could be bothered with Linux I tried it on G5 Powermac, during the install the fans all came on full, yowsers, No one with ears would want that near them...

And, looks like Jim's last point is correct

@AW

When bigfoot did his initial MacMini-demo in Bad Bramstedt all those years ago he had bolted some big fan onto it (after removing the topcase), so there surely is some SW in MorphOS to keep the Minis from burning out

AFAIK this is done by throttling the CPU under small loads. Even if there is some direct control of the fans, it might well be that the components used by Apple for this are very different between Mini, PB or the various G5s.

@amigadave

All G5-demos I witnessed were done booting from USB-sticks.

Thans Frank for clarification. So guys, can we focus on software side now? :)