@clebin

Quote:

clebin wrote: @olegil If I went too far with my language, then I'm sorry jkirk.

don't worry bout it. i understand your viewpoint of not setting yourself up for dissapointment.

Quote:

I do think even dreams need to have some grounding. I've asked for details of new ideas, features, advantages niches to go into, and nobody can give me a *single good reason* to believe in their dream.

this is my grounding. amiga is still alive. if all we had was the classic line i would believe the amiga was dead but would dream of someone resurrecting it but admit it was not possible.someone did so the dream is for a comeback. this won't happen near term imo but i hope that long term we will get a decent number of users back to the platform.

Quote:

As as happened for the past 10 years, they just repeatedly tell me to dream, be ambitious and so on. How is this ambition supposed to manifest itself, exactly?

amigaone, sam, peg 1&2, aros, etc. this is how it is manifested currently. you cannot expect to fly when you haven't walked yet. in 94 the amiga was devastated by the collapse of commodore. we(users & companies) currently are picking up those pieces for the next chapter. will it succeed? i don't know but i have my belief that we could rebound. it just will take more money and time that has been invested up to this point.

Quote:

I'm not sure how we can work together on a strategy for a private company, unless you're talking about another Amiga-like platform. That's just an aside - give me an idea of these ambitious goals and how we set about achieving them.

ok try this goal
amiga revival- done (could have been done better but at least she is here)
now we as a group must create a foundation so the amiga can grow. the private company can create the next amiga. it is up to developers(and users to a lesser extent) to create the reasons for people to come here and use them. features in an os is pointless if the computer won't do what you want it to do. productivity apps, games, etc must be created to lure more people to the platform as we grow as a group so does word of mouth which will bring more people if/when we reach a point to where the smaller developers create games.apps that is more well known this will help since people will recognize the brands. then we can grow more till EA,wordperfect, and other big name companies take notice.

so no i don't believe in the magic bullet but with hard work and diligence we can at least move forward into a self-supporting marketplace.

_________________
Win•dows: n. A thirty-two bit extension and graphical shell to a sixteen-bit patch to an eight-bit operating system originally coded for a four-bit microprocessor which was written by a two-bit company that can't stand one bit of competition.

@Frek

Quote:

UNIX today isn't backward compatible with UNIX in the 70s; It moved on- adopted new ideas dropped legacy support etc. that's why it's alive and kicking today. Not to forget that it's prove of how well designed the system was.

but think did they change all at once or was the changes filter through all the years of development time? even windows keeps its legacy aspects. so no just because aos is backward compatible(to an extent). does not mean it is retro only.

now the backwards compatibility did suit the retro crowd but this had more to do with the fact of the library of available software. this way they could release the os and not have to worry about coding the other apps up front. this saved development time and money. as we go forward we will see those "legacy" apps fall by the wayside now that we have some apps built for os4. in essence weaning us off the old system while creating the new one.

_________________
Win•dows: n. A thirty-two bit extension and graphical shell to a sixteen-bit patch to an eight-bit operating system originally coded for a four-bit microprocessor which was written by a two-bit company that can't stand one bit of competition.

Amiga f"n ruleZ!!!

@jkirk

Quote:

but think did they change all at once or was the changes filter through all the years of development time? even windows keeps its legacy aspects. so no just because aos is backward compatible(to an extent). does not mean it is retro only.

Windows did major changes- totally new API (Win32) etc, which it pretty much stuck with since. But the Win32 was written with modern features in mind.

UNIX changed pretty slowly in some areas, but you have to remember that UNIX run on computers that was far more advanced than the home computers we're talking about, virtual memory, memory protection etc was already known in the 60s and 70s.

It just disappeared for home computers to save money, that was the mistake many manufacturers made- they never assumed they would need to support those features.

@cv643d

Quote:

Amiga f"n ruleZ!!!

Yeah mate! You do the ruleZ? I take care of the rest. C'mon, tis weekend!

OldFart

_________________
More then three levels of indigestion and you're scroomed!

@Frek

Quote:

Frek wrote: @jkirk Windows did major changes- totally new API (Win32) etc, which it pretty much stuck with since. But the Win32 was written with modern features in mind.

not really NT was a total rewrite of the api. win 9x was a mishmash (32 bit, 16 bit, and 8 bit coding,] slowly replaced over time but stability was not a hallmark till xp which is actually NT made years earlier. the reason they waited so long to do this was to create the legacy compatibility to keep from breaking every program written for windows.

Quote:

UNIX changed pretty slowly in some areas, but you have to remember that UNIX run on computers that was far more advanced than the home computers we're talking about, virtual memory, memory protection etc was already known in the 60s and 70s.

yes but as i said legacy compatibility is foremost in their mind when creating something like that. and this is why.
company b buys ultimate database program 1,
unix decides it is time to break off and do something new revamps unix from scratch,
ultimate database program won't work now,
company decides os must be changed for fear of hackers,
the choices are to go with new unix and not have a database prog available at launch or go to another os and buy a proven database program that runs on it.

the devs of unix don't want their os to be useless overnight. they make minor changes while making a compatibility layer so at least the newer software works. this way they can make changes without putting their customers in the lurch.

an os without apps is useless. therefore supporting the previous gen software is just common sense. just so happens our previous gen software is 10 yrs old.

Quote:

It just disappeared for home computers to save money, that was the mistake many manufacturers made- they never assumed they would need to support those features.

i really don't understand what you are trying to say here.

_________________
Win•dows: n. A thirty-two bit extension and graphical shell to a sixteen-bit patch to an eight-bit operating system originally coded for a four-bit microprocessor which was written by a two-bit company that can't stand one bit of competition.

@OldFart

lol

Quote:

And then I'm talking about both the processor architecture and system architecture, because MacOS X has absolutely nothing (well except Carbon) to do with the Classic MacOS- still Carbon binaries for example run happily on both systems.

I agree: especially both the Classic => OSX and the PPC => x86 were really smooth: that's impressive.

Quote:

Anyway my point is, some systems just aren't worth trying to extend because they're fundamentally broken with modern features.

What do you mean ?

_________________
http://www.warpdesign.fr/

@jkirk

http://en.wikipedia.org/wiki/Win32

"One of the largest changes the Windows API underwent was the transition from Win16 (shipped in Windows 3.1 and older) to Win32 (Windows NT and Windows 95 and up)."

@Leo

I mean the API they currently expose can not be patched to support modern features without breaking legacy applications, so what's the point retaining it?

Something like Carbon seems like a good move to me, retain the general layout- but don't fuss about BC because its simply not going to work, you can solve BC with sandboxes.

Quote:

I mean the API they currently expose can not be patched to support modern features without breaking legacy applications, so what's the point retaining it? Something like Carbon seems like a good move to me, retain the general layout- but don't fuss about BC because its simply not going to work, you can solve BC with sandboxes.

I agree. Question is: do they really want modern features ?

Seems like everyone whine about backward compatibility and complain about adding modern (new) features is making AmigaOS a new Windows/Unix clone.

I think you can have modern features while keeping AmigaOS its feeling though...

_________________
http://www.warpdesign.fr/

@Frek

Quote:

Wasn't the custom chipset that made Amiga different and good? And that most computers picked up these days- you won't find a modern computer that don't have alot of hardware to unload the CPU.

For a while that was true but Intel has slowly dragged everything back onto the CPU, only graphics and (sometimes) video are done by a custom chip now.

To give an example, you used to have mini-synthesisers on sound chips to do audio. These days other then the conversion to an analogue signal, audio is all done by the CPU.

_________________
Whyzzat?

@olegil

Quote:

And I know the SAM doesn't use the MPC5121e, but if the OMAP3530 at 500MHz isn't faster than a PPC at 400MHz, how on earth can it be faster than a PPC at 600MHz (and higher)?

You really have to ask???
Even the 5121e and 440 are not going to be the same at 400MHz, they use different PPC cores. There'll also be differences in cache, bus interface, memory controller etc...


However, I strongly suspect the OMAP will outgun the 440 in some areas:

The OMAP has a number of things the 440 or 1521e doesn't have:

An L2 cache, this'll give it an advantage in anything that is cacheable.
It also has a vector unit so anything media based will have an advantage.
The OMAP also has a DSP (actually so does the 5121e), it can do 720p video decode and there's one company that has used it and the ARM vector unit together to do 1080p decode. Yet the 440 appears to have trouble playing DVDs!


However...
I suggested ARM because it'll be in cheap netbooks, They'll be using things like the i.mx515. Same ARM core as the OMAP but it's running at 1GHz. Still think a Sam will be faster?

BTW The ARM netbooks are expected at around $200

_________________
Whyzzat?

@minator

Yes, video playback is actually faster when using the DSP. That's a very good point.

I keep ignoring the DSP because I won't need it. And for a normal Linux application it isn't very easy to offload to a DSP.

Actually, the only thing I can think of that we could offload in my system is PNG decoding and encoding. We draw block diagrams of the signal paths which change with different alarms etc. So decode PNG, modify, encode. But now that I think about it, predecoding the PNG would probably help MORE. And preloading the image into memory when you suspect you might need it rather than the "oh my god, we have to decode it NOW" which they probably implemented would mean a great deal of time saved.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@minator

Well. I don't think the SAM will beat an i.MX51 at full speed, no.

But it's still impossible to find out if the i.MX51 actually survives full speed. As I mentioned, the i.MX31 does not. It'll last you 1 year and 3 months at full speed. To make it last longer, you need to make sure it doesn't run too much at full speed. And to make it last for the durations Amigans classically keep their hardware, you would need to make sure it was powered down much of the time.

But the i.MX51 has a lot of cool features, I'll grant them that.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@Darth_X

Quote:

How about we all put $1000 into a fund to develop PearPC to run OS4 on standard PCs?

PearPC could already OS4, you just have to wrote the UBoot emulator!

In actual fact, not to try and play devil's advocate here, but emulating an OS4 machine is a whole lot easier than emulating a real Amiga. No fancy chipset or particular hardware to emulate with a CPU. The AmigaOne is almost a PC with a PowerPC CPU! It's so easy by comparison, I'm surprised no one has had a crack at it. I can "emulate" OSX on my A1, but I can't do the opposite?

Quote:

I'm surprised no one has had a crack at it.

Guess no one cares...

_________________
http://www.warpdesign.fr/

@Frek

Quote:

UNIX today isn't backward compatible with UNIX in the 70s; It moved on- adopted new ideas dropped legacy support etc.

What about it's more popular clone, Linux? I notice that even modern Linux sources still look at argv and figure out all the strings. Well the Amiga passed that sort of old fashioned string parsing with OS2 and template keywords. And still being able to implement the old "-s" switch way; yet on Linux they expect you to type "--help" for help. What this business with needing to enter two minuses?! Are Linux bash commands controlled by X11 all the time and the experienced user never enters all that?

Quote:

Why on earth do you save the most flawed part of the amiga (the OS) that's basically impossible to bring uptodate,

What else is there left? The hardware, though it still could have been used for certain uses, has been left for dead by Amiga more than AmigaOS was.

Now if you want to remove the OS and just have a modern computer, how is that different to a PC? What so special about a PC today? Nothing. You might as well run Vista and get all you can from it.

What are your thoughts on the OS4 idea being a better approach?

@Frek

Quote:

It did however retain memory zones, which was inconvinient (having to specify the max/min amount of memory the app was allowed to use.)

Classic MacOS also had pointer typing. The upper 8-bits of a pointer were used to store info. This caused a major problem!

AmigaOS also has some pointer-typing. The highest bit is used as a flag in relation to memory handling, ergo, it has a maximum range of 2GB. But, I don't see it as a problem, since it is only for specific routines and wouldn't normally be used.

Quote:

The rest was handled in the sandbox approach.

I thought the Mac sandbox was silly, it had to boot an older version of itself! And then it seemed to merge into the system.

I don't see any reason why the OS4 task based approach couldn't go on. Tasks could run as they do now as tasks and a separate stack of 68k libraries could be used to protect the newer OS4 system.

Quote:

I notice that even modern Linux sources still look at argv and figure out all the strings. Well the Amiga passed that sort of old fashioned string parsing with OS2 and template keywords.

This has nothing to do with beeing backward compatible...

What's more old fashioned is the way any Amiga program can bring the system down. Every other systems passed that old fashionned shared memory (and this includes unix, windows and macosx).

Amiga still has it. Because any program requires it... I'm sorry, but that's far more of a problem than keeping "--" to specify command line arguments ;)

That being said, I agree that Amiga's template command line args parsing is nice. So this is something I would keep in a modern OS...

_________________
http://www.warpdesign.fr/