Wither Natami?

[bloodline]

JimS wrote:

bloodline wrote:
If the reimplementations in the FPGA were gate level accurate, to the Original Chips, I would agree.

But they are not.

The accuracy of the reimplementation in the FPGA doesn't have anything to do with whether the minimig & friends should be called "emulators" or "clones", "reimplementations" or what have you. It's the technology being used that I'm quibbling about.

Ok fine. We can call it Brian if you like :-) But that doesn't make Hardware clone of an Amiga more or less accurate than Software clone of a Amiga. Some people get confused by that simple point.

[bloodline]

freqmax wrote:
@bloodline:
UAE is an software emulation. And software just won't get it cycle exact to realtime. And for some applications software emulation won't cut it.

It can't? I would appreciate a list of these applications. Cheers.

[bloodline]

The level of ignorance in this has become too great not to address.

Here's a few words of advice for those of you who lack technical knowledge.

Think before you post. Formulate a response, try and establish the point you want to make. You need to make sure that know why you are posting.

Now head over to Google and try and learn a little about the issues involved. Don't for get Wikipedia, which should be your first port of call after reading any post by a member who clearly has more experience with any given topic.

Now, just go back, edit your post appologize for wasting people's time and enjoy the learning experience.

Back on topic, since a number of posters seem to have missed my post on the subject... A Hardware clone of the Amiga is inherently no better or worse than a Software clone. BUT we are in a situation where the Software clone (UAE) is MUCH more mature, and thus compatible than even the most mature of the Hardware clones, MiniMig.

[bloodline]

Jakodemus wrote:
@bloodline

Yes, UAE is fast and very compatible, but you have to run it on top of windows/linux/morphos etc. With hardware you can boot straight to the system we all love.

With a minimal Linux set up or Amithlon... you would never know the difference.

If you are happy to use UAE as your primary Amiga, it's fine for me and I suppose that it's fine for everyone else too, but some of us want to have new hardware dedicated only for Amiga and you shouldn't have any objections against that.

Find one objection from me regarding the Hardware clones. You'll find none, in fact I would like to get one, I just wish people who don't understand anything would stop banging on that Hardware Amiga clones are inherently better than software! Also if they could stop posting tenchnical nonsense as fact that would cheer me up too.

[bloodline]

WotTheFook wrote:
If it runs Amiga software, it's an Amiga, if it's hardware or not, the rest is just semantics.

So a PC running ReactOS is Windows? A Linux box runing Wine is also Windows?

These meet your requirements... but they are Not Windows, they are clones.

Every AMD or intel box running MSDOS or FreeDOS, is an IBM PC?

Again... they are not IBM PCs they are clones.


The key difference is that the behaviour of the original is known and well documented. The clone is a new system, which is compatible with the original.

The hardware purists need to realise that is isn't 1988 any more, you can't get the original hardware, so if you replicate it in another chip, what's the problem?

There are subtle differences with hardware... have you forgotten the long thread where myself and Karlos tried to work out the actual response of the Paula DACs... which is nonlinear, due to the now long obselete chip Fabrication process used by Commodore! No modern chip suffers from this nonlinear nature and as such sounds different.

The nature of the MiniMig, for example, not being a gate level copy of the Amiga, means that it is still full of bugs. While the chip is compatible with the Amiga chipset, the subtle interations of the various circuits is different, which upsets certain software.

I'm not a Hardware purist, by any stretch, as a man who sold his hardware synths for softsynths... but I do get annoyed when people make false claims.

The software purists, ditto to the above. If you couldn't tell by looking at the screen and using the mouse etc, it's an Amiga, isn't it?

No, it's not... for all the same reasons as I stated above... it's a clone. A workalike. My pefered Amiga option right now, due to advanatages too numerous to mention.

Also, if you replicate the Amiga chipset and add "AGA with knobs on" like Natami are planning, how can you claim it's an Amiga, since you have modified it? If I mod my A1200 by adding an IDEFix and a laptop DVD/CD drive, is it still an Amiga?

Because one is an Amiga with bits bolted on... the other is a new system, that tries to be compatible with the Amiga Chipset.

If you made a Gate level copy of the Amiga chipset, with accurate timings, and tried to match the electical properties as best as posible WRT the DAC/ADCs etc... then yeah, I'd say you had yourself a real Amiga.

If I were to take a Lotus and then put a shell on it to make it look like a Delorian, and fitted out the interior to look like a Delorian... that's not a a Delorian.

If I were to take the Delorian plans and build one using new parts, to the original specs... then yeah... that's a Delorian...

[bloodline]

WotTheFook wrote:
@ bloodline

I am not going to copy and paste comments here for the sake of an epic post. Suffice to say that you all appear to be arguing over what appears to be a Turing-type test, where the user tries to work out if the machine at the other end of the mouse and keys is an Amiga.

If all you had was a screen, with a mouse and keyboard that looked like the Amiga ones, could you really tell without seeing the hardware... This is the test they applied for artificial intelligence, it should surely be good enough for the Amiga....

... except that such a test doesn't exist. Oh dear....

All this hardware-versus-software debate is devisive, we all love the Amiga, isn't that enough?

Ok, this is getting annoying because I agree with you that it doesn't matter if you are running a real Amiga, an Emulation or an FPGA... since that was my original point!!!

Then someone stated that Emualtion was not as good as the real thing, so they wanted NatAmi or MiniMig... This is why I am now having to argue that the FPGA clones are not the real thing. I am not saying they are better or worse than Emulation, or better or worse than a real Amiga. But they are different.

[bloodline]

TheDaddy wrote:
@bloodline

Come on, we are splitting the hair now.

Millions of cars share millions of parts but they are still Fords, Fiats, Opel, BMWs, Audis and so on.

Millions of Computers are built using the same capacitors, resistors, transistors, gates, 74xx ICs, etc... What matters is how they are put together! :-)

I would like to say that if the Minimig/Natami can, to a certain extent, be 90% compatible with old Amiga software then they are Amigas.

But If I need them for a specific Amiga feature... lets say the nonlinear audio reproduction... then with the MM or NA I'm stuffed... I wouldn't call it a real Amiga then.

I was also thinking would you call an Amiga a real Amiga if it couldn't use hardware add-ons designed for a specific Amiga model? I would still call it an Amiga.

Irrelavant... the clone may or may not have compatible interfaces, these are often the easiest part of a system to clone since the spec must be very detailed and tight!

If someone said to me: "From tomorrow we have three new Amiga models but they are not 100% compatible with old Amigas/Amiga software", I would be enourmosly happy and thankful. It wouldn't even cross my mind to check if and why they are not 100% compatible.

For me then it would be pointless as an Amiga... One of my favourite projects is posibly the least Amiga compatible Amiga Clone project... I don't think of it as a real Amiga though... I see it as a learning tool and a hobby!

[bloodline]

WotTheFook wrote:
Amen, Persia, here endeth the first lesson.

 :bow:  :bow:  :bow:

 :cheers:

While Persia is totally correct, that actually has nothing to do with my point :-)

Anyway, a computer platform is defined by it's software. The problem with the Amiga is that all serious software development ended once software houses realised the Amiga platform was in trouble... perhaps around 1995 at the very latest.

The Mac and PC, bear no resemblance to their historic ancestors, but that doesn't matter, because the software and applications grew with the platforms. New verisons of old software took advantage of the direction the platfomrs went in, so developers and users could grow with the platform.

The Amiga is now frozen in an early/mid 90s snapshot... the applications require technology that is now several generations behind anything we have now... and as such a modern Amiga would have to support a massive Gerneration bridge in order to make any sense... such a bridge would be pointless.

[bloodline]

Jakodemus wrote:

bloodline wrote:

The Mac and PC, bear no resemblance to their historic ancestors, but that doesn't matter, because the software and applications grew with the platforms.

PC is still compatible with IBM PC. You still can boot to MS-dos 2.0. Your ultra-fast graphic card, with the latest pixelshaders and high speed GDDR5 memory chips, can still display CGA-graphics. Hardware has maintained its backward compatibility. Only thing that has changed is the OS. Microsoft has dropped support for 16bit programs, so nowdays you have to use dosbox.

Yes you can run your old MS DOS 2 apps on a modern PC... but if youy read my post carefully... you will notice I am talking about the environment that has evolved.

If Commodore had kept going, every year the Amiga platform would have evolved to keep it in step with with latest generation of software and hardware technologies. But it didn't so now our next gen Amiga's are still stuck in the 1994 time warp... where really the Modern Amiga should be as different from AmigaOS 3.9 as MacOSX is from System 7... but we've missed those 15 years of dev time. In that regard "It's all over!" - MegaloMania 1991

[bloodline]

Hans_ wrote:

persia wrote:
For a decade and a half the Amiga has been leaderless, slowly losing it's user base.  WHy do we need to tear at each other?  We're all that's left, maybe a few thousand scattered across the globe.  We have different visions, different ideas  We disagree on what we want, but we don't disagree on what we don't want.  We don't want the Amiga relegated to history.

The different visions are exactly why people are tearing at each other. I wish that everyone would move to Amiga OS 4 (ignore the hardware availability issues for a second please) and help push that forward; MorphOS fans would prefer it if everyone switched to MOS and pushed that forward (also ignore limited hardware availability please, more is coming); Bloodline & co wish that everyone would join the AROS initiative; and, there are also those who would insist that a real Amiga has to have AGA compatible hardware.

Now we seriously have to think about what each project or rather what each group of people really want to achieve! please see below...

In fact, there's also a group that want Amiga to remain a retro machine.

And if we want to ensure that the Amiga's place in history is not forgotten forever, may be the best way to preserve what little dignity there is left. Please see below.

What would be neat is if a next-gen API could be collectively agreed upon, and then implemented on all the next-gen Amiga systems. That is, however, unlikely to happen.

But we already do have a common API, that is the collection of technologies that grew up over the period between 1993 and around 2001... We are now stuck with that forever. There is no reason to build a new "next-gen API", as it couldn't offer anything that an off the shelf solution could not provide... much like Apple in the late 90's it would make sense for the three exisisting clones to become the legacy layers on top of a new system. But while that is the only sensible solution, there is really no point since there is no vital Amiga only software anymore... no resaon for this legacy layer. :-(

[bloodline]

Hans_ wrote:

What would be neat is if a next-gen API could be collectively agreed upon, and then implemented on all the next-gen Amiga systems. That is, however, unlikely to happen.

But we already do have a common API, that is the collection of technologies that grew up over the period between 1993 and around 2001... We are now stuck with that forever. There is no reason to build a new "next-gen API", as it couldn't offer anything that an off the shelf solution could not provide... much like Apple in the late 90's it would make sense for the three exisisting clones to become the legacy layers on top of a new system. But while that is the only sensible solution, there is really no point since there is no vital Amiga only software anymore... no resaon for this legacy layer. :-(

Whether you design a new API that is based on the existing one, or steal someone else's (e.g., use BSD as a base OS), it's still a new API as far as the Amiga is concerned. What needs changing is nowhere near as much as was required with Mac OS (which didn't even have pre-emptive multitasking).

-Edit- I would say they had an easier task, since it's very easy to see that MacOS Pre OSX was not going to cut it... AmigaOS has the problem is that it can keep up with the big boys... just as long as it doesn't have to carry very much, and there are no hills, and there's no cross wind, and no one jumps up and shouts "Boo!", etc...

Basicly if conditions are good... AmigaOS is a great system, it just doesn't have to ability to deal with problems... and it is the robust nature of modern OSs that give them their strenght!

I'm personally in favour of keeping what's good in the Amiga OS API, discarding the rest, and creating a new API that has all the modern essentials such as full memory protection, SMP, etc. I do not see the point in becoming another Linux distro.

I've often considered this... but think about the implications of this... Once you give exec memory protection, SMP, and a nice new scheduler, some modern data sstructions not just endless linked lists... etc, etc... it's going to end up looking like a pretty ordninary Microkernel... and one that has taken quite a bit of work to get to and won't have the proven track history of modern Microkernels... and then quite a lot of the old Amiga subsystems are either outdated; graphics.library, layers.library, audio.device or wouldn't work with a modern MP/SMP exec; intuition.library and I think dos.library too...

So all that has to be replaced... leaving you with really very little reason to have chosen AmigaOS in the first place :-(

I also think that enough people want to continue to use their old Amiga software to warrant having a legacy layer.

I have the problem that 95% of the old Amiga software I would want to run needs the Amgia hardware... so even a legacy software layer is of little use.

[bloodline]

wawrzon wrote:
bloodline, if you want to have 68k amiga compatible hardware then you should actually support such an approach as the natami team. they consider replacing amigaos kick and libs with 68k aros substitutes. in long therm this might be the solution for a-clone too at least to get rid of legal dependencies and to be freely distributable.

This idea that I don't like the Hardware clones can only come from people not reading my posts... Please reread them.

[bloodline]

cicero790 wrote:
I love Amithlon too, but that will not be sufficient for slaying the dragons Windows and MacOS. A new OS, a new more powerful sword is needed, for the cutting edge hardware of today.

These Dragons can't be slayed with AmigaOS... Any more than a longbow can fight a Challenger2 Trank...

[bloodline]

cicero790 wrote:
How would you do it Bloodline? What would it take? :-)

Hmmm, I'm not really sure you can topple these top systems... their budgets alone make that task virtually impossible.

A computer platform is defined by the available software, Windows and Mac have a software history that cannot be matched. I would say the only time you ever get a chance to make a new platform is when a new technology emerges... that renenders the old software obselete. Then you can quickly jump on to that new technology, and if you are more innovative than the competition, you can dominate the market.

The Personal Computer (I mean the Apple1, CommodorePET, etc...) platform was the first... The Second was the GUI, the Third was the internet... the Last was the mobile computing device...

Microsoft won the first one.
Microsoft and Apple won the second.
Google won (is winning at least) the third.
And we shall see who wins the fourth, and youngest platform.

[bloodline]

cicero790 wrote:
Isn't this the time then? with windows shutting down and AROS available on pc that is surprisingly almost A1200 like, as you showed in the fpga forward thread.
Is this the time to develop AROS into a better Midori?

Remember that the three Amiga platforms are basicly each a new platform, they have no meaningful software history anymore.

But AROS, like its Amiga clone Breathren, is a an old GUI platform. This platform has been won.

If something new came along... I don't know... like brain implanted computers, then a new platform would need to be developed to fill that void... Would any features of AmigaOS have any relevance on that platform?

[bloodline]

persia wrote:
So let me get this straight, a group of people that can't even break simple boot code to get AmigaDos 4 up and running on a Mac-Mini are now going to redesign AROS around a complex memory management model?

This is the reason why I get annoyed :-)

The only possible solution is to take Zune and the WOrkbench clone that has been built in AROS and get it to work on top of a QNX microkernel and run classic apps in the QNX version of UAE.  

A rather resonable idea, and the only sensible idea I've heard in a long time. Zune even started life as a MUI clone on Linux, so there might not take very long to get it building in a POSIX OS.

Even then it's probably too late, but it might be worth a shot.  

It is too late. But still sounds like fun.

A small Posix compliant OS with an interesting and small GUI might just turn a few heads.

I'm sure some of them do... but one more in the long list isn't going to make a splash.