@OlafS25

i wasn't taking about x1000, but the naming of orginal Commodore Amiga naming

_________________
nox's in the house!

@pavlor

you are very wrong, i have proof .. just not right now becourse i'm not home before two weeks.. patients, and i will provide proof

_________________
nox's in the house!

@Hypex

they have already said that... it would be too much work to do a JIT for the new platform. Additionally the new platform propably will be 64bit

@pavlor

Quote:

I pointed to 2000-2005 time period, long before IContain or CommodoreUSA.

Yes. And many here have given somewhat different interpretations about this naming convention and why it was chosen.

Yet it is Amiga that is the name people sought after. Yet it is Amiga that is still the name people keep inquiring about (those who still have "Amiga name" aspirations).

Now let's look at the only time it was licensed to understand why.
When questions about trademark usage arose...lo and behold...they start to license the single word version.

If you think that finally using the one word name was part of a business plan as opposed to an exercise in how to not lose the trademark, I'll have to respectfully disagree until eternity comes to an end.

#6

_________________
This posting, in its entirety, represents solely the perspective of the author.
*Secrecy has served us so well*

@Deniil715

Quote:

Deniil715 wrote: @OlafS25 68k is the nicest CPU ever but for some reason ditched early. If 68k had "survived" I think it would be better than PPC because it's more efficient in code size and no.of instructions per amount of work.

The problems of 68Ks are related to the instructions encoding and complex instructions to be executed (e.g.: double memory indirection).
Quote:

Today number of transistors is no problem either, like it was when it was ditched.

Exactly. This helps a lot, because decoders and "integer" execution units don't increase in complexity so much, whereas new productive processes gives more transistors available, so such circuitries become smaller and smaller (albeit they are the most active parts of a chip).

That's the reason why the x86 architecture was able to fill the gap with RISCs, and win the performance battle at the end.

Probably 68Ks would have been in the same position, once Motorola found some elegant and cheap solution for the above problems, but she decided to kill this beautiful processors family. However it has to be said in her favor that at the time she wasn't able to sustain the competition with Intel, which was able to better scale her x86s processors (while keeping full backward compatibility).
Quote:

It was difficult to clock fast, just like x86, but Intel and AMD poured tons of money into x86 development.

AMD was (and is) a small company compared to Intel, so she had not that much money to put for x86 development. It should be noted, also, that other much smaller companies (startups) created x86 clones, and some of them with a very good design: NexGen, Via, Centaur, Transmeta are the most well known ones.

So, the "money" argument isn't directly related to the successfulness of an architecture. What primary counts are good ideas. You can have a lot of money, but if you have no good ideas (illuminated engineers), you can do nothing.
Quote:

Quote: and "classic"? What is that? The last Amiga I bought back in the days was called A4000 and not AC4000 (Amiga classic 4000) The original Commodore Amiga machines has gotten the label Classic after the birth of the next generation Amigas: the AmigaOne/Sam/etc. with AmigaOS 4. Quite simple. You bought an Amiga back in the days, and now you have a Classic Amiga, because it got old, whether you like it or not Quite simple, happens to cars as well People didn't buy a Classic Chevy back in the '60s, but if they still have it, it has now become a Classic Chevy.

Commodore USA introduced some new Commodore 64, VIC20, and Amiga models, but after that the old ones didn't automatically became "Classic".

The same happened also when Commodore introduced the Commodore 128, Amiga 1000, 2000, 500, etc.: Commodore 64, VIC20, Plus4, etc. didn't became "Classic".

And if you take a look at the articles that were written for the 30th anniversary of the Amiga 1000, nobody mentioned the "classic" word.

In fact, such relabeling isn't common on IT, and it seems to be used by part of the post-Amiga community to give minor value to the old (but real) machines, and emphasize their new toys.

@Deniil715

Quote:

Deniil715 wrote: @cdimauro Quote: cdimauro wrote: So you excluded other amigans, like me, which like Amiga and one of the post-Amiga o.ses (AROS). So that puts you in group 2 then, with 'Amiga' meaning everything Amiga(-like): A500, A4000, MorphOS, AmigaOS 4, X1000, AROS, etc...

No. I think that the discussion with pavlor should be quite clear: I don't recognized all that stuff as "Amiga". In fact, I talk about "Amiga" and "post-Amiga" to identify the two distinct models and, in general, eras.

Amiga is simply... what legally was/is and materially defined it. So, the legal name in the first case, and the custom o.s. (ad-hoc for hardware) running on the custom hardware (custom designed chipset) in the second one.
Quote:

Quote: OK, so it excluded my Amiga 2000 and 1200, which aren't Classic Amigas, but simply Amigas. Please read my reply to OlafS. Commodore Amigas has become Classic (Commodore) Amigas simply because of their age and to differentiate them from AmigaNGs. It's just a reference term, like AmigaNG meaning PPC machines made "in the Amiga spirit" after Commodore. Maybe I should use a lower case 'c', but it looks more classy with a capital 'C'

See my answer in the previous comment.

BTW, in that case you can simply use Amiga and AmigaOne to better distinguish the two different computers.
Quote:

Quote: - To group 2 it is all Amigas (I'm in this group) which range from the original C= Amiga to the X1000, some also include MorphOS and AROS in this group. Myself I would call myself an Amiga user if I was using for example Pegasos/MOS, but I would personally probably not call the machine itself an Amiga. No problem: there was no Amiga machine sold other than Commodore or Escom.

Back to group 1; not accepting that Amiga actually did live on, and is still alive today by means of X1000/Sam/AmigaOS 4.[/quote]
See above and my previous statement for this. But, yes, Amiga isn't alive anymore. In fact, there's no machine which is can be called this way (see my previous comments with pavlor).

But I naturally recognize a link to the past ((C) by Nintendo) when we talk about AROS, MorphOS, and OS4, in this strict order of appearance, whereas there's no link regarding the "new" machines where they run.
Quote:

Quote: Group 3 complains that custom chips are not made in-house but instead bought from specialized third-party companies (AMD/Creative/VIA/whoever makes USB/whoever makes all other chips on the motherboards). "It should all have been made by Commodore and soldered onto the motherboard." Problems is: Today this is impossible because custom chips today are a million times more complex than 30 years ago. (Moores law: Double power every 18 month for 30 years = 1 million.) Only specialized companies are able to make them. Moore's law doesn't talk about doubling (processing) power every 18 months (18-24, in reality; and in the last period the rate seems that went to 30 months), but only about doubling the transistor count.

Did I talk about processing power? No. I said complexity which is pretty much equal to transistor count. And yeah, I guess the curve has flattened out a bit...[/quote]
Quoting you: "Moores law: Double power every 18 month"
Quote:

Quote: So of course the X1000 was made for AmigaOS. It leaves room for AmigaOS to expand (become multi-core, 64-bit, use more memory etc). Besides, there are no high-end 32-bit single-core CPUs on the market today, because that is not high-end. We get 64-bit and multi-core in the package whether we like it or not. Actually you got no 64-bit neither multi-core support. Regarding the first, you'll never use it. For the second, you may get an AMP solution.

I meant we got 64-bit and multi-core in the hardware as part of the package. It's just for AmigaOS to use it. [/quote]
The same with MorphOS with G5s.

The problem is that you'll not see any support for 64-bit. You can forget about them.

Regarding multicores, they will be used (albeit who knows when), but I think in an AMP model to preserve compatibility with the existing PowerPC and 68K software. SMP requires breaking the compatibility with the past.

@KimmoK

Quote:

KimmoK wrote: @cdimauro >So, please, can you report some benchmark which compares this supercool PA6T with some G4 and/or G5? https://www.ll.mit.edu/HPEC/agendas/proc07/Day3/09_Artz_Pres.pdf (page 20 etc... Already superior with just 2/8 cores. (PA was also designing towards 16 core)) ((But how one could benchmark inferior design, for example PCIe or memory performance of a CPU, when G4 and G5 does not have those controllers in the CPU chip at all.))

I saw, but take a look at some other benchmarks, with more common applications.

@noXLar

Quote:

noXLar wrote: @cdimauro don't you remember my earlier post from another tread that Amiga inc. officially renamed the original Amiga from amiga to classic amiga in late 90's? and that is in fact official. and i don't get your stubbornness, what's the problem? it was chosen just too be more easy to discern from the orginal and the new. thats why Amiga Classic was born legally and officially.

Amiga Inc. has the right to use her Amiga trademark has she wishes, but fortunately she hasn't the power to rewrite the 20 years history of Amiga computers. Which were and continue to be called Amigas and not Classic Amigas....

@cdimauro

Those benchmarks were 100% irrelevant, because AOS4 did not (does not) have compiler that can optimize for PA6T and because those mainly test JUST the core. PA6T superiority is otside the (single) CPU core.

(( PA6T is more than 5x improvement vs G5 systems, when you look what it does per Watt == superior ))

Last edited by KimmoK on 14-Aug-2015 at 07:04 AM.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@KimmoK

Quote:

KimmoK wrote: @cdimauro Those benchmarks were 100% irrelevant, because AOS4 did not (does not) have compiler that can optimize for PA6T

So what? In the meanwhile you turn off your computer and don't use it, waiting for a good compiler?
Quote:

and because those mainly test JUST the core.

Which should be a good thing too, right? Don't you like to see how the CPU performs, especially with real-world applications instead of steriles benchmarks?
Quote:

PA6T superiority is otside the (single) CPU core.

Like what? When you want to compress a file in MP3 format, you run the LAME command, and get the result. Dot.
Quote:

(( PA6T is more than 5x improvement vs G5 systems, when you look what it does per Watt == superior ))

Who cares? You, as a user, are only interested on having the final result. You don't care about the performance per watt. You only care on how long the CPU took to provide you what you requested.

@Deniil715

Quote:

Deniil715 wrote: @OlafS25 Quote: The secong group what people here call NG wants to revive amiga like it was in the past. I think that is a illusion because the OS would need so many changes that it would break any compatibility. Then you might have a modern OS but no software. Additional without drivers and expecially the software base of the other platforms I see no realistic chance there. People who are happy with shall use it but i am happy with 68k for nostalgy and fun and use my windows pc for serious working. Your second group is my third group. Those who don't want the new stuff that is available, but want the old stuff in a new shape. But that's where the paradox kicks in. It is just as you say an illusion, because one cannot modernize the old stuff without making it new and "un-Amiga-like". That's what group 3 fail to understand, and I find that strange and amusing

The problem is that having an o.s. with modern (and desirable) features means killing a lot of the Amiga o.s. "philosophy".
Quote:

I completely understand the feeling, and I also want to have the "original" Amiga in a new shape, but that's just not possible. It's an illogical thought. AmigaOS 4 tries to go this route a bit by modernizing, but then people complain that gfx isn't integrated and cannot be hardbanged like in the old days, AHI is required and sucks/low quality etc. has no AGA/Paula chipset etc etc. Well, the new way is simply not the old way. Besides, eight 32-bit channels at 192kHz is better than four 8-bit channels at 28kHz. Period.

RTG, AHI, and 3D were already available for the Amiga o.s. 3.x, well before OS4 was evenn started.
Quote:

And the almighty copper!! OMG what a master piece!!! not... With modern gfx cards you can do a lot more cool stuff that then copper and blitter ever could dream of. I would be surprised if modern 3D HW could not emulate the copper's visual effects with easy, but also vertically or in any angle, something the copper was way too slow and simple for.

You can do it, but with some price to pay, because you have to write proper shader programs to get the same result. Of course, shader processors are way better than the Copper, but they don't have the "feature" of the display that you can change in realtime with some cheap operations (a MOVE and you change the background color from here to the end of display frame, without executing any other operation; just to give an example).

They were different display logic/models, which aren't comparable. Copper made sense with a raster-based display logic. Shader processors make sense with a framebuffer-based display logic.
Quote:

Amiga OS3 with MP and SMP and seamless integration of new gfx and audio-HW (like the original Amigas) is simply what OS4 is trying to achieve. But fans aren't happy because gfx isn't planar (instead we have 32-bit 3D), audio channels cannot be individually controlled (instead we have surround capability at MUCH higher sample quality), we don't have low memory footprint (instead we have fullHD 32-bit desktop with composition and modern webbrowser), soon we can't rely on Forbid() anymore (instead we would have SMP).

MP and SMP means killing backward compatibility: it's hard to see them, especially on the very conservative OS4. Let's wait (how long? Boh) and see.
Quote:

Group 3 is simply never satisfied. They only want some dream version of OS3 on fast HW, and fail to realize that OS4 on PPC is just that. AmigaOS 4 *is* a modern version of Amiga OS3 on fast hardware. (I'm not referring to you, Olaf, here since you seem happy with current OS3 on UAE.)

It's highly questionable that it's both modern and runs on fast hardware.
Quote:

Quote: 68k is not modern or cheap? i use it with UAE on modern PC and it is fast there :) and modern? What is possible on PPC that is not possible on 68k? Well in that case you already have what you asked for The only thing OS4/PPC has over OS3/68k are a couple of more modern APIs and apps. And of course, being actively developed.

So, it's not that much modern from the Amiga o.s. 3, for which RTG and so on were already available from long time.

@cdimauro

End of discussion (about the fact that PA6T indeed was superior design vs other PPC chips of it's time).

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@OlafS25

Quote:

OlafS25 wrote: @Deniil715 Aros 68k had the advantage that it combined both Aros (OS in development) and 68k software base, that was basically how I created Aros Vision, taking a 68k nghtly and adding and testing 68k binaries.

It'll be appreciated if you provide some updated AROS/68K roms for (Win)UAE too.
Quote:

For me the most realistic next step is to create FPGA based hardware (accellerators and/or standalone) because it lifts the existing software base because I do not think that there are enough developers to create a new base.

Emulation is also a good option, even more flexible and cheaper.

@KimmoK

it is not important anyway

PA6T is now classic too so you could call it "classic NG" then

And I do not understand why people here dream of a new OS that breaks everything anyway. There will be no software for it, simple as that

Because of that a upgraded 68k platform would make more sense but it is a niche of course

@cdimauro

emulation is good but people only take real hardware seriously. That is true for both users and developers. In this sense substantial upgraded hardware would be important even if emulated environments are still more powerful.

@Hypex

Quote:

Hypex wrote: @OlafS25 I don't see why MOS on another ISA needs to drop compatibility. The fact is 68K emulated in Trance runs in a sanbdbox so all they need is a new sandbox. Shouldn't be a problem. Amuihlon proved this ten years ago.

With Trance you don't have a sandbox, but an emulation layer which let you run 68K apps as "first class citizens" like PowerPC apps, with full integration on the system.

(Win)UAE is a sandbox.
Quote:

But the question does remain what will happen to Quark and the A/Box. I pressume Quark can be compiled natively for a new ISA and then it needs an A/Box emulation layer. Since the A/Box, AFAIK, is tightly coupled with emulating the Amiga 68K ABI.

I think that the ABox will be reworked to provide a new environment and some new APIs. Of course, software has to be partially rewritten, and 68K software will run like with AROS & JanusUAE. IMO.

@cdimauro

upgraded Roms? With what? Compatibility is pretty good already

@cdimauro

"software has to be partially rewritten"

who will do it?

@pavlor

Quote:

pavlor wrote: @cdimauro Quote: Here is the same. So, after the agreement no one has the legal right to call Amiga what's labeled as AmigaOne. Except Amiga.Inc of course (again in legal sense).

Sure.
Quote:

Quote: End of legitimates successors of Amigas... Of course AmigaOne is legitimate successor of Amiga (don´t you see AmigaOne?). And up to this day only legitimate successor still sold.

A successor which cannot be called as "Amiga"?

And don't even share the same design and goals, since it's just a PC with a CPU replaced with a PowerPC one (whic is A BIT different from the Amigas), and the o.s. don't map over the full hardware features (read: the Amiga o.s. was written and offered APIs to cover the Amiga hardware features).

As I already stated, all post-Amiga experiments can be called "a link to the (Amiga) past", due to the background from which they come. But the story ends here.

@OlafS25

>I still do not see the developers creating new software for a new Hyper-AmigaOS with 64bit, SMP and MP and thus heavily changed API.

-Current APIs already have new stuff vs 3.1 API.
-There exist more developers in 64bit + SMP + MP world than on AOS3 API niche

Currently (ex?) Amigans have grown used to multiGhz, 64bit, SMP and memory protection in their programming work life. I believe some of them would like to see those on Amiga niche, before they start to see it as sane thing to program for.

>The MorphOS-Team has announced that they will make a ISA change in future and drop compatibility for it.

Yes. Because building compatibility requires years of extra work. Breaking compatibility enables one to do optimal future OS. Legacy must not be show stopper.

>In many regards it will be more like Aros with 68k running in UAE and people who are comforted to existing systems will have to change.

It should be identical to using MorphOS3.x. Except that your old SW might not run without UAE.

>But even if a system would be modernized who will create the software for it? Even worse, all devs work for fun and in sparetime so they will propably only support what they have, so you do only have NG but also NGNG.

Ony time will tell.
IMO. Staying 32bit and single core == RETRO. I personally want more.

eg. I would love to use AmigaOS5/MorphOS4 that looks and feels like AOS4/MOS3 but takes better use of a 64bit multicore HW (what ever ISA is underneath). Initially there would not be many native apps, but I could run all legacy SW via emulation or via booting to AOS4/MOS3/AROS (if no-one does the sandbox or starts to use virtualization to run multiple OSs).

((for me, x64 ISA would offer (ONLY) Elite Dangerous, via muliboot, as an extra vs some 1,4Ghz multicore PPC))

***********
Windows was originally 16bit and single core.
Then they made it 32bit and pre-emptive.
Then they made it 64bit and multicore.
Same is doable with any/every OS with enough work.
People tend to end up using the best or latest variant, same for programmers.
***********

Having super fast HW but very RETRO OS is not better than using UAE ... but at the same time ...

It would be cool if someone would develop AAA+ GFX with some FPGA.
(mainly AAA like features + chunky modes + simple 3D + DDR3/PCIe)
(but it would be grazy and very likely unprofitable)

It would be cool if someone would develop NG68kCPU with some FPGA.
(mainly 1Ghz, 36bit address bus, SoC features)
(but it would be grazy and very likely unprofitable)

It would be cool if someone could update AmigaLikeOS for that AAA+NG68k combo.
(but it would be grazy and very likely extremely unprofitable, and anyway breaks some compatibility as usual)


"just" updating AmigaLikeOS to use modern components is similarly unprofitable as business but more doable (IMO). (+it might be portable to high power AAA+NG68k as well, if one/both of those happen parallely)

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?