@eliyahu

Quote:

eliyahu wrote: @thread Another year, another show. We all had a terrific time this weekend, and I can only encourage those of you who are mostly 'online' Amigans these days to consider attending next year's AmiWest, Amiga38, or whatever get-together you can. It's always time well spent.

I'll see what I can do about attending next year, especially if my buddy @DiscreetFX will be there again.

_________________
All the way, with 68k

@All

So going from the news report at AmigaNews these look like the main highlights of interest:

* Camilla demonstrated how to program a GUI using CodeCraft on OS3.2.

* Sam Crow demonstrated programming with Hollywood.

* Steven Solie, the cheerful developer, apparently performed a live recreation of the Windows '95 release on stage when he demonstrated the ExecSG multi-core kernel crashing to the public.

I'll be sure to catch up on these when I watch the unlive stream. Also, the following showed up:

* Enhancer 54 will have AmiTCP as a TCP/IP stack. Not sure why this will be included. OS4 already had Roadshow which has a newer codebase than AmiTCP. And the AmiTCP AmigaOS did have with Genesis in OS3.9, which was the newest 20 years ago, did have an integrated GUI. Perhaps it's because it has a built in GUI that Roadshow lacks on OS4, and Roadshow disappears on OS4, since it isn't called Roadshow in OS4 and only generic as Internet. In any case Enhancer 54 still needs OS4 files so I will continue to misunderstand why all these replacements are put in place. All this effort for a glorified boot image creator?

* Novabridge will be a separate product on top of the already confusing Radeon driver situation. So, for modern 3d with old 3d applications and games, one will need to buy Warp3d-SI, Radeon driver or package with Radeon driver, a package with Warp3d-Nova. And finally install a library to actually run WarpOS software. Be good to have an integrated solution. Unless packages are combined into one at least two purchases will need to be made.

* CFE will have an update for multi core support. Which doesn't make sense because my old CFE boots Linux which has supported multi core in all that time. What does CFE need that it isn't already doing? Nothing appears to be said about what is more important right now, supporting UEFI firmwares, fixing the slow booting issue where it sits there for a minute before loading, and fixing USB so it can actually work.

I'm most interested in GUI creation right now. I've noticed I tend to complain about a GUI or lack there of while not providing many myself. A trend I need to amend.

@Hans

Quote:

That's more of a report on just A-EON's stuff.

Yeah, that's why my headline states exactly that

Quote:

I hope they'll cover all of the presentations and some of the stuff on the show floor too.

It's just a question of manpower. That single news item (and its translation) took nearly four hours of my time. Both my time and my motivation are 'somewhat' limited, so I need to concentrate on the part that most people are interested in. I'm not exactly a fan of AEONkits work - but their presentations are what most readers want to hear about IMHO, even if it's just because they want to make more jokes about this year's extremely funny LibreOffice status update.

Quote:

Looking at the schedule, there's a lot of interesting stuff. I'm skimming through the recordings right now.

Most other interesting stuff seems to have been cancelled? David Pleasance, Jens Schönfeld, the Italian IP thieves - none of that happened, AFAICT? I had your presentation running in the background while working on stuff, that seems to have been the only other interesting thing from a news service's perspective.

@all

Shut up.. manic depressive Babylon!

@Hypex

Quote:

Steven Solie, the cheerful developer, apparently performed a live recreation of the Windows '95 release on stage when he demonstrated the ExecSG multi-core kernel crashing to the public.

Please let there be video footage of this!!! 🤣🤣🤣🤣

@Hypex

Quote:

Hypex wrote: * Steven Solie, the cheerful developer, apparently performed a live recreation of the Windows '95 release on stage when he demonstrated the ExecSG multi-core kernel crashing to the public.

Which is something already very known: it already happened on AROS x64/SMP. YEARS ago.

And there's no magic wand that could be used to avoid those crashes: it's all about the intrinsic Amiga o.s. nature.

It's incredible how people are repeating the same mistakes without taking lessons of what was already done. History is ignored. And as well is ignored how the Amiga o.s. works.

But I can imagine that Steve dispensed a lot of laughs while taking about it...

@cdimauro

Quote:

And there's no magic wand ...

Sure, but with the right development team and funding, it can be made to work.

_________________
All the way, with 68k

@Hypex

Quote:

* Steven Solie, the cheerful developer, apparently performed a live recreation of the Windows '95 release on stage when he demonstrated the ExecSG multi-core kernel crashing to the public.

I don't think he demo'd it on-stage, but his machine was running the multi-core kernel during the show. I don't know if anyone recorded video footage of it.

Quote:

* Novabridge will be a separate product on top of the already confusing Radeon driver situation. So, for modern 3d with old 3d applications and games, one will need to buy Warp3d-SI, Radeon driver or package with Radeon driver, a package with Warp3d-Nova. And finally install a library to actually run WarpOS software. Be good to have an integrated solution. Unless packages are combined into one at least two purchases will need to be made.

Not quite. Novabridge will be part of the Enhancer Software Pack V54, which is taking a while to make. So, A-EON are going to let people buy it separately so you don't have to keep waiting. It's welcome news for me, because it was essentially done a year ago.

Novabridge makes the W3D_SI Warp3D driver obsolete, so you won't need that one any more.

Hans

_________________
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
https://keasigmadelta.com/ - More of my work.

@cgutjahr

Quote:

It's just a question of manpower. That single news item (and its translation) took nearly four hours of my time. Both my time and my motivation are 'somewhat' limited, so I need to concentrate on the part that most people are interested in. I'm not exactly a fan of AEONkits work - but their presentations are what most readers want to hear about IMHO, even if it's just because they want to make more jokes about this year's extremely funny LibreOffice status update.

Understandable. We appreciate your efforts.

Quote:

Most other interesting stuff seems to have been cancelled? David Pleasance, Jens Schönfeld, the Italian IP thieves - none of that happened, AFAICT?

Ah, so that's why old clips of Bill McEwen were in the stream instead of the scheduled talks. That's a shame.

Hans

_________________
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
https://keasigmadelta.com/ - More of my work.

@Hypex

Quote:

Steven Solie, the cheerful developer, apparently performed a live recreation of the Windows '95 release on stage when he demonstrated the ExecSG multi-core kernel crashing to the public.

That didn't happen. Good story, though.

Quote:

Enhancer 54 will have AmiTCP as a TCP/IP stack. Not sure why this will be included. OS4 already had Roadshow which has a newer codebase than AmiTCP. And the AmiTCP AmigaOS did have with Genesis in OS3.9, which was the newest 20 years ago, did have an integrated GUI. Perhaps it's because it has a built in GUI that Roadshow lacks on OS4, and Roadshow disappears on OS4, since it isn't called Roadshow in OS4 and only generic as Internet.

Yeah, a bunch of us were confused as to why this would be done. No one sees the 'value-add' for the customer, TBH; but that's true of many Enhancer components replacing the AOS4 ones. I'm sure there's a reason beyond having more ownership of system elements, and that will eventually be explained.

-- eliyahu

_________________
"Physical reality is consistent with universal laws. When the laws do not operate, there is no reality. All of this is unreal."

Hans wrote:
Quote:

Ah, so that's why old clips of Bill McEwen were in the stream instead of the scheduled talks.

First, the planned talks had various problems, i.e., the French developers forgot they were supposed to videochat during the DevCon, David Pleasance became ill, and Jens Schoenfeld never responded, even though urgent messages were sent to him saying that he was supposed to virtually appear. Second, as Bill Bosari said during the broadcast, he wanted to show some of the history of the Amiwest Show.

Truly,
Robert Bernardo
Fresno Commodore User Group - http://www.dickestel.com/fcug.htm
Southern California Commodore & Amiga Network - http://www.portcommodore.com/sccan

@thread

For folks interested A-EON has stealth-released NovaBridge on AmiStore, so if you wanted early access to this, you can purchase it now. I've been testing it on my X5000 with a SI card and my A1222 with a RX card, and it works very well indeed. It's a building-block for yet more ambitious graphics work coming, but just by itself, it's awesome to have both video-acceleration and 3D working on my little Tabor board.

A massive thanks to the very talented Hans and the underwriters over at A-EON for funding all of this work!

-- eliyahu

_________________
"Physical reality is consistent with universal laws. When the laws do not operate, there is no reality. All of this is unreal."

@eliyahu

Please can you detail the configuration on your Tabor? Both software et hardware(gfx) required for Novabridge I'd like to evaluate what I'm missing.

_________________
AOS 4.1 : I dream it, Hyperion did it !
Now dreaming AOS 4.2...
Thank you to all devs involved for this great job !

@abalaban

NovaBridge give you ability to have old Warp3D support (and so old MiniGL as well then) on any card supported by Warp3DNova : RadeonHD and RadeonRX ones.

While for RadeonHD there was W3D_SI driver before (which was exactly for that purposes - to have old warp3d support), it still didn't support everything as expected, and in some parts just slow.

Now, Hans made novabridge, mean that on both and RadeonHD, and on RadeonRX you can use old Warp3D/Minigl, without anything else. And as addon, all the stuff working over NovaBridge faster, better supported, and almost bug free (at least to the level we all can test it).

With NovaBridge and ReWarpd libraries (open source realisation of powerpc.library), you can run old titles such as Heretic2, WipeOUT, demos from MAWI (which will works even better than on original warp3d with voodoo cards) and so on.

See on my video where i cover it all some time ago:

https://youtu.be/WUIHDnJ947U

To summorize:

-- RadeonHD users have not _that big_ deal to have NovaBridge, but still, NovaBridge better and faster than W3D_SI driver. When installing on system with RadeonHD, be sure that you delete/rename W3D_SI one.
-- RadeonRX users have no needs to use hackish/crashing/bad minigl4gl4es, but instead NovaBridge


That all of course the same for all platforms, just if Warp3DNova works on, then Nova Bridge will too.

Last edited by kas1e on 25-Oct-2022 at 09:33 PM.
Last edited by kas1e on 25-Oct-2022 at 09:26 PM.
Last edited by kas1e on 25-Oct-2022 at 09:25 PM.
Last edited by kas1e on 25-Oct-2022 at 09:25 PM.

_________________
Join us to improve dopus5!
zerohero's mirror of os4/os3 crosscompiler suites

@RobertB

Quote:

First, the planned talks had various problems, i.e., the French developers forgot they were supposed to videochat during the DevCon, David Pleasance became ill, and Jens Schoenfeld never responded, even though urgent messages were sent to him saying that he was supposed to virtually appear. Second, as Bill Bosari said during the broadcast, he wanted to show some of the history of the Amiwest Show.

Thanks for giving us the background. I wasn't criticising. The AmiWest team have done a great job.

Hans

_________________
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
https://keasigmadelta.com/ - More of my work.

@agami

Quote:

agami wrote: @cdimauro Quote: And there's no magic wand ... Sure, but with the right development team and funding, it can be made to work.

I assume that you can also resuscitate dead people (without side-effects), since it looks like that you can do better than magic wands...

@cdimauro

Quote:

cdimauro wrote: @agami Quote: agami wrote: Sure, but with the right development team and funding, it can be made to work. I assume that you can also resuscitate dead people (without side-effects), since it looks like that you can do better than magic wands...

Now you're just taking the piss.

With technology, all things are possible. I'm not saying it would be easy, nor am I saying it would be pretty, but it can be made to work.

Now, should it be made to work? That is an entirely different discussion.

_________________
All the way, with 68k

@agami

Quote:

agami wrote: @cdimauro Quote: cdimauro wrote: @agami I assume that you can also resuscitate dead people (without side-effects), since it looks like that you can do better than magic wands... Now you're just taking the piss.

Friend agami, what else should I have done since I've clearly stated that not even a magic wand would be able to fix the mentioned issues, whereas you replied that money and a team could?

If the magic wand metaphor wasn't able to let you understand it, then I'm lost, because I've no other ways to explain the concept.
Quote:

With technology, all things are possible.

Here you continue on the same way and now you're forcing me to become a bit more brutal: your sentence doesn't apply on the real world, because things work very different from your hyperuranium.

It's clearly evident that you completely lack knowledge on Computability theory, otherwise you should have known that exist even problems which aren't calculable.

But even outside of Computer Science your sentence doesn't apply. Think about traveling back and forth from the past/future, traveling faster than the light, going inside a black hole and going back, resurrecting people, etc. etc. You got my point, I think.
Quote:

I'm not saying it would be easy, nor am I saying it would be pretty, but it can be made to work.

How? Do you know how the Amiga o.s. works and how its applications are running? What kind of "dirty" (but perfectly "legal", since Commodore allowed it) things they could do?
Quote:

Now, should it be made to work? That is an entirely different discussion.

First you've to see if this is possible and how.

As I've said, AROS x64/SMP has the same issues since years and they were never being fixed.

OS4 is facing exactly the same issues.

Do you think that it's a pure coincidence?

@kas1e

Thank you for your answer. I admit that I am lost with all these gfx related systems. So if I am not not misunderstood to have Novabridge running I need all the following:
* latest Enhancer pack to have all nova libraries,
* RadeonHD 5 driver (if I am using SI chipset)
* Novabridge (obviously)
Am I forgetting something?
Additionally if I will need Rewarp libraries but I was under the impression that it was the purpose of Novabridge to wrap Warp3D calls to Nova. Please can you detail?

_________________
AOS 4.1 : I dream it, Hyperion did it !
Now dreaming AOS 4.2...
Thank you to all devs involved for this great job !

@cdimauro

Quote:

Here you continue on the same way and now you're forcing me to become a bit more brutal: your sentence doesn't apply on the real world, because things work very different from your hyperuranium.

I don't accept the dismissive and defeatist attitude of the magic wand statements.
I am certainly talking about the real world, and I stand by what I said.

I have been on this merry-go-round before with people who think in terms of constrained engineering. We peel the proverbial onion one layer at a time, and in the end we reach the point where the engineer in question resorts to statements such as "yeah, but no one is going to do that" or "that would require millions of dollars".


Quote:

It's clearly evident that you completely lack knowledge on Computability theory, otherwise you should have known that exist even problems which aren't calculable.

That's not what we are facing here with the SMP challenge for running AmigaOS 4 apps, or SMP for running AROS apps.


Quote:

But even outside of Computer Science your sentence doesn't apply. Think about traveling back and forth from the past/future, traveling faster than the light, going inside a black hole and going back, resurrecting people, etc. etc. You got my point, I think.

I get the point where you are approaching these as an engineer of the early 21C.
Nothing you have presented are actual impossibilities. But lets leave the inner workings of our universe aside. SMP for AmigaOS/AROS is not challenging the limits of our collective knowledge of astrophysics and the nature of existence.


Quote:

How? Do you know how the Amiga o.s. works and how its applications are running? What kind of "dirty" (but perfectly "legal", since Commodore allowed it) things they could do?

I know enough about how Amiga OS works, and more than enough about how software works. How exactly would I make it work? That requires some workshopping with a team and a couple whiteboards. There's likely several ways to solve this problem, all of which require hard work, and no magic.


Quote:

First you've to see if this is possible and how. As I've said, AROS x64/SMP has the same issues since years and they were never being fixed. OS4 is facing exactly the same issues. Do you think that it's a pure coincidence?

As I've said, with technology all things are possible.
The fact that the small AmigaOS team, and the unknown number of people attempting to work it in AROS have not yet made this work is in no way commentary on the solvability of the problem.
It's not a coincidence because both efforts share in common key shortcomings:
- the lack of a larger pool of talented developers
- the lack of adequate funding to pay for the hard work required.

We can certainly agree that the problem is beyond the practical capabilities of the existing teams, especially given the low funding or free voluntary work. We can even agree that trying to make existing AmigaOS 4 or AROS perform SMP is not possible without a complete re-architecture and re-write of the OS. But that doesn't mean that it can't be made to work.

And by "it" I mean the running of modern AmigaOS 4/AROS software and old 68k software, running on a system that is able to distribute and balance the concurrent running of said software across the available compute resources of their given system, without clashing/crashing.

Last edited by agami on 27-Oct-2022 at 02:03 AM.
Last edited by agami on 27-Oct-2022 at 02:01 AM.

_________________
All the way, with 68k