Pros and Cons of Amithlon type systems

[Heiroglyph]

I skipped over Amithons heyday and got into it later, but I'm very interested in it.

Would you still use something similar?

What would a modern version need to change?

[Heiroglyph]

Basically it has a minimal Linux underneath that exposed the PCI bus and many devices to the AmigaOS as real hardware.

You had as direct as possible access to the hardware so it acted much like a Draco in that you could only run OS legal applications with Picasso96, no hardware banging, although at least the CIA's were there.

I'm not sure that the chipset emulation would slow it down by much on todays PC's, but that was some of the justification.

You could also mix and match x86 libraries and executables.  For example you could use an x86 native datatype and 68k apps would benefit from them.

Part of the idea was to slowly move more and more of the OS to x86 without having to do it all at once.  It also made more hardware available, giving a huge speed boost, more RAM, RTG, AHI and network card at very little cost.

At the time it seemed to be a great option, but the licensing wasn't done properly so it was killed in court.

[Heiroglyph]

@XDelusion:
For system friendly/RTG type software it seems better than my tricked out 4000T.

Considering you can run UAE on it at full speed, the lack of chipset emulation doesn't really bother me.

There are updates that support some newer motherboards and video cards, but it's still very hit and miss.  Very hard to get set up with the updates too.

[Heiroglyph]

So back to my original question, is there still interest?

Is it just the legality keeping it from use?

Is it the fact that it's somewhere between hard and impossible to set up on current hardware?

Maybe it sounds good on paper but you didn't use it?

Has Aros negated the need?

[Heiroglyph]

@J-Golden

Thanks for the answers.

In my case, the lack of hardware was a big stumbling block.  I just don't have access to that much old PC hardware and it's super picky about what works.  My one old compatible PC isn't holding up as well as my long lived Amigas, it randomly doesn't post anymore.

I like to think that AROS will get there eventually, I'm just finding along the way that I'm not enjoying AROS as much as I thought I would. It's largely still just a fast subset of OS3.1 without much software.

Amithlon seems to give me all my old software plus a super RTG Amiga and right now that's what I really want to use.

Long story short, that's why I'm looking at reviving Amithlon without the IP problems but I was curious as to what people liked/disliked about the original.

[Heiroglyph]

Bernie got so screwed in that situation.

It's a shame, he's an awesome developer and has great ideas.  Plus he's just too nice to have that happen to him.  Where is karma when you need it?

[Heiroglyph]

From what I remember, in this case it was either greed making them think they could avoid licensing costs for the OS they were shipping or believing that they had the right to do so.  I won't try to read their intentions, but regardless they were proven wrong.

Bernie got caught in the middle and bailed out to cover himself from the lawsuits.

[Heiroglyph]

I can do it, I write more complicated things for a living, I'm just seeing if it's really worth the time investment.

The main missing bits are the glue between the emulation and AmigaOS that give you native x86 code and direct hardware access.  Luckily those are pretty straightforward after seeing the headers you compile against and the patches to UAE and the Kernel.

The things I'm still weighing out are how much to emulate vs. pass through to the Amiga side (all PCI? SCSI? USB?) and which OS to use as a base.

Bernie was already rethinking how devices were handled for v2 and seemed to be going more emulated to increase compatibility. (No special Amiga side drivers to write)

Linux is a proven host but isn't interesting IMHO.  The big benefit there is that Bernie already patched the kernel so that you could get a lot of the memory mapped to the correct locations for a speed increase, but is that even needed with multi-core computers with memory faster than the CPU's Amithlon was designed to run on?

There are just a lot of variables to take into account before I decide to commit to it so I thought I'd gauge interest.

[Heiroglyph]

@B00tDisk:
If I remember correctly, there were native versions of ImageFX, Aladdin 4D and a few other Amiga apps, but not a lot of new ports.

@XDelusion:
I agree that Aros has been going pretty slow, but you have to give the guys a LOT of credit for starting from nothing.

Toni and Jason get a lot of credit recently for giving the Amiga version a huge boost, but they all deserve a lot of recognition.

[Heiroglyph]

I'm on the AROS mailing list and have source code access, but this is my fun project for after work and for some reason working on AROS just isn't interesting to me.

I've even made donations because I believe in it, but it just doesn't pique my interest.

[Heiroglyph]

Personally, I think compatibility is one of the things that is holding back  Amiga derived systems. IMHO, AROS is stuck without MP, SMP and useful drivers because it aims at 3.1 compatibility. I feel OS 4.1 and MorphOS are similarly stuck because of PPC compatibility.

I didn't make it clear, but I meant better hardware compatibility, not software compatibility.  Beyond OS legal apps isn't something I'm interested in, that's what full system emulators like UAE are for and they are beyond reasonably complete.

I agree with both you and B00tDisk, 3.1 was great 20 years ago, but there needs to be forward momentum. I think the lack of forward progress is one of the things I disliked when I was working on AROS, I felt like I was reinventing a wheel rather than adding value to something.

[Heiroglyph]

Removing the original OS would be a bad idea at first.

1. Then why not just run AROS?
2. Compatibility with old apps wouldn't be good
3. You're just going the AROS route again with 3.1 on x86.

I was looking at something more like:
Take a more advanced OS that matches well, say Haiku.
Get Amithlon working on this rather than Linux.
Make use of ClassicWB or other distribution and the users own ROM.

You could start releasing right here.

Amithlon would become more of a UI than the OS, like early Windows or XWindows.
Instead of peeking through a keyhole with special x86 extensions, expose the underlying OS to AmigaOS.
Make use of AROS code to slowly integrate them better making Haiku more Amiga like and AOS more native, eventually negating the need for ROMS, etc that can't be distributed.

Part of my reasoning is that Haiku is more of a spiritual successor than most other OS's, but is far more advanced.
It has SMP, memory protection, virtual memory, datatypes, responsive UI, it's multi-media centric and has a good selection of drivers.  It also boots to a usable desktop in 8 seconds. (not counting BIOS crap that can't be helped)

Anyway, that's what has been kicking around in my head.

[Heiroglyph]

PS: I see the usual idiots who lurk around here but are too afraid to say anything have started a thread on PooBunny dissing your idea, gawd what a sad bun of tards some of these tubes are... :rolleyes:

This may be the stupidest idea ever, who knows.  Wouldn't be my first, won't be my last.

It sounds fun though, that's about all that matters anymore.

It's not like I'm trying to make money from it, it's just an interesting challenge with open source code to get a head start from.

[Heiroglyph]

I don't think 3.9 was that much slower, but I think it feels slower.

Feel is a lot of what makes AmigaOS AmigaOS IMHO.  OS3.9 has this intangible slowness that certain people are more sensitive to.

The multiple reboots are just inexcusable.  I have no idea how that made it out the door.

[Heiroglyph]

Doesn't Haiku have the same limitation that AROS has in regards to drivers? I assume you'd want to support a limited set of architecture first, but wouldn't the ideal be to ultimately support as broad a range of hardware as possible? Am I missing something here?

I have nothing against Haiku and I'm definitely not a Linux fan, but wouldn't you want to start with something that had a fairly broad foundation?

Just curious.

It's not exactly a logical project to work on as much as an emotional one to begin with, but Haiku has surprisingly good hardware support and I like the actual OS.

Haiku is just far more like what I want an updated AOS to be than Linux is.  I'd spend all my time covering up Linux-isms instead of weaving the best of them together.

With Haiku, if it shows through the emulation it's still basically a feature.  Plus I'd rather give my changes back to Haiku and AROS than Linux.

(BTW, I use Linux a lot, I just don't have an emotional attachment there)