Im totally lost in this mess.
ExecNG, that is a vital part of AOS4 is done from people that seems dont know much about the company doing the OS, with people making firmwares that nobody needs...
A lot of people, companies and work but with no communication.

Sorry but all this is so weird from an outside view that yes...Im lost.

We're hitting confusion levels that shouldn't even be possible.

I prefer this kind of communication to no communication.
In the end it's all quite clear, and what emerge is once again a blackout in communication among parts, or better, no communication.

Anyway if there'll ever be a next time where Mark will offer himself to build a new version of x5k firmware, obviously explaining he about what would be addressed, IMHO could be better accept his help and work as a team, at least on this goal

Things seems going on on both AOS4 nd MOS sides, and this is another good reason to think positive for all amiga users.

[ Edited by Flash 29.10.2020 - 21:13 ]

These kind of problems makes mistrust in OS4 team even more, meanwhile MorphOS team works in silent without any kind of "awesome feature announces" that never arrive.

Quote:

To be fair, I don't think there is an OS4 team any more, just a bunch of volunteers completely unfamiliar with low level coding landed with a chunk of code that's probably not written to a standard where it can be easily maintained by anyone but its original coders.

I read this thread with my jaw open while shaking my head in disbelief. Wow.


Wow.

Quote:

It's not really my place to step into the middle of this, but I just want to address this one question of Trevor's, with the most likely answer, that everyone has read in Mark's initial post to this thread. Mark/Bigfoot posted this here because Solie and to an extent Trevor himself kept refusing to give Mark the answers to his questions in those emails. I think if it had been me, I probably would have resisted helping the ExecSG team even more than Mark did, and I would not have given them the source code, until they DID answer my questions fully. But that is just my opinion on how I would have reacted, and Mark is a much more helpful and "Nice Guy".

It seems completely reasonable to me for Mark to post this warning publicly, when he was kept "out of the loop" by Trevor and the ExecSG team, but learned what was going on and put the pieces together that seem to confirm his fears. I don't know if Mark made one last attempt to get the answers directly from Trevor or the ExecSG team, which I would have done before posting here, but it is done now, so why doesn't the ExecSG team and owner come out with an official statement to clear things up and ease the minds of the users. All of this "cloak and dagger" secrecy in our tiny community is silly. We are too small and way too far behind all other platforms to worry that anyone is going to steal the ideas of our development teams, and I can't see any benefit in keeping the users in the dark about what is going on. In fact, I think it hurts our community.

After all the help that Mark has provided to get MorphOS working on the X5000, I hope that it won't all be in vain because A-Eon decides to drop support for MorphOS, due to the current graphics card problem between the two OSes, and the apparent low number of users running MorphOS on an X5000.

The moment you realize that running abandoned HW with no chance of any firmware updates is a blessing.....

Quote:

:D

Regarding this "don't upgrade", is there something that prevents users from downgrading to previous version, if there are problems? (like firmware upgrades are done on modern games consoles, for example)

Also, an average user will probably think it's better to add as much features as possible to the firmware. But is there actually any benefit in doing that, instead of "doing as little as possible", and handing the control to the OS as soon as possible? (I assume the answer is "absolutely none") - would the OS actually run exactly the same, if firmware was completely erased once the system was booted up?

I think it's also good to point out that this firmware rom is mostly different from Amiga Kickstart ROMs. Both definitely do some initializations, but classic Amiga ROM (or ROM image in RAM) continues to play a significant role once the OS is up and running.

@Jupp3

Quote:
The issue at hand is that the possibility of downgrading your firmware version does not undo severe data loss that was caused by a prior upgrade, hence the need for a public warning before MorphOS users might be the victim of potentially irreversible damage.

> this firmware rom is mostly different from Amiga Kickstart ROMs.
> [...] classic Amiga ROM (or ROM image in RAM) continues to play
> a significant role once the OS is up and running.

This is because classic AmigaOS is actually two-part, disk-based and ROM-based, with the Amiga Kickstart ROM containing the ROM-based portion of the OS. The MorphOS equivalent of the Amiga Kickstart ROM is the boot.img, which is obviously not firmware, but OS. OS4 maintains a similar scheme.

Quote:

And yet, the official maintainer of the firmware for the X5000 systems is asked for help in compiling or setting up an environment that can compile the source code, and is ignored and kept in the dark, regarding projects that modify the firmware he is supposed to be maintaining.

WHY?

[ Edited by amigadave 01.11.2020 - 08:54 ]

Wow. Talk about mistreating a developer. This is so wrong on so many levels.

Ssolie:

”Board firmware is used to initialize the hardware to a known state so the OS can take over. Sometimes, stock firmware needs a tweak or two when the original authors didn't anticipate that it was going run some niche OS (e.g. AmigaOS).”

https://amigaworld.net/modules/news/article.php?storyid=8577&start=20#101195

I just watched ssolie's AmiWest 2022 talk on ExecSG, where he also mentions the need for a firmware update for the X5000 (and others) in order to make use of OS4 with multicore-enabled ExecSG currently under development. Summary as follows:

"Solie is also responsible for firmware updates. The U-Boot versions for X5000 and A1222 plus not only received various bug fixes, but will now also correctly initialize the additional cores of the multi-core CPUs used in these machines. Firmware updates will be distributed "soon" via A-EON. Once they're installed, the respective computers will then be "multi-core ready"."
http://www.amiga-news.de/en/news/AN-2022-10-00104-EN.html

Does the opening posting of this thread still apply, now that it's 2 years later?

Quote:

This is exactly what my initial post on this thread refers to, so yes, it very much still applies.

> yes, it very much still applies.

Thanks for the reply.

From what I saw of Amiwest, some backbone functions are being added to support multicore OS4, but there's still no idea how it's going to work in praxis. Because we know how discussions on having multiple concurrently-working CPUs on an OS with a task-shared memory address space usually end up.

Just the usual OS4 stuff, start designing the cushion covers for the house you're going to build, some day...

So probably best to avoid these updates if you plan running MorphOS, unless they do something useful like power down the unused cores.

[ Edited by KennyR 29.10.2022 - 17:05 ]

Quote:

Unfortunately not even that would be a good excuse. MorphOS already does power down the unused cores, and doing so doesn't actually require booting them and taking them over. You can do it entirely - and easily - from the boot core.

> MorphOS already does power down the unused cores

Very nice :-) Has this been the case since MorphOS 3.10 or been added later? I don't remember I already read this, at least looking through the release notes since 3.10 I cannot find anything with respect to this.
Btw, has this feature also been added for PowerMac11 (PPC970MP) meanwhile? (I know that generation isn't officially supported, and that MorphOS supports this on PowerMac7 (PPC970/PPC970FX) since MorphOS 3.10).

Quote:

It's been this way since MorphOS 3.10

Quote:

I don't know, sorry :)

Quote:

I can only hope the answer is 'no' as my QUAD consumes approximately 280W when idle.
If that would be on just one CPU core that would be insane.