Slashdot Mirror
AmigaOS 4
Second five-eighth writes "The Amiga is alive and sort of well (you can get the OS, but not the hardware), and Ars Technica has a review of the final version of AmigaOS 4. New features include limited memory protection, 3D display drivers, an improved suite of applications (the bounty for porting Mozilla to AmigaOS has yet to be claimed), and much better 680x0 emulation. Perhaps most telling, the reviewer was able to move his daily writing workflow from Windows XP to AmigaOS 4.0: 'Not only was it possible to do this, but having done so I feel no urge to switch back. It is nice to not have any distractions when working — there is no waiting for the system to swap out when switching between major applications, no constant reminders for updates or to download new virus definitions and even if the worst happens and the system locks up, it takes only seven seconds to reboot and get back to a functional desktop.'"
FTA: "this brings things up to ludicrous speed."
Prepare for the jump to ludicrous speed!
Interesting that he would mention not worrying about viruses. If history repeats itself that should be short lived. Amiga was one of the worst in the old days for viruses. Most of them at the time came from floppies because it had this habit of auto booting the disk the moment they were placed in the drive. Hopefully the new OS is better guarded but the limited user base is likely to be it's best defense.
..I'll be mentioning something cool in Mac OS LXVIII and some idiot will say "Why, we did that in Amiga OS 4, and we did it better!"
Now I can get ProComm to dial into those old Telegard BBSes that I still have the phone numbers for in my Apple Newton. I hope that someone ports a terminal emulator that supports the RIP protocol, because ANSI and AVATAR are just boring.
This will completely let me replace my Coco3.
Tradewars door, here I come!
Couldn't the 6 of you who are still interested just start a mailing list or something?
"I'd rather be a lightning rod than a seismometer." -Ken Kesey
The wayback machine says:
http://www.archive.org/details/Amigaand1985/
Will it run Duke Nukem Forever?
die already. the amiga's time has come and gone.
//Scuze me...
What is your problem?
I don't get all upset when somebody drives by in a 1950's Studebaker all tricked out. Yeah, it has some limitations, such as: a single-speaker AM radio, no air conditioning, cruise control, electric windows, it requires fuel additives to not die on unleaded gas, and it's hard to find parts for. Oh, and it's a death trap in an accident.
And despite all that, it's still mighty cool. I honk when I see somebody driving one.
Can you imagine what a dorkass you'd look like if you stuck your head out the window and screamed: "Dude, die already! The Studebaker's time has come and gone already!".
Oh, wait. Nevermind. You're posting O/S elitism on Slashdot. My guess is that you probably already know all about what a dorkass you look like. Never mind.
I have no problem with your religion until you decide it's reason to deprive others of the truth.
The Amiga's killer app was video production which has been trivial now on Macs and Windows XP for years. Even the Video Toaster that was cherished by Amiga users now requires a P4 or Athlon and Windows XP. It seems to me that Amiga OS doesn't offer that much when compared Linux, BSD, OS X, and Windows. Heck, I'm even going to throw WM5 in there since it has better browser choices.
'Same speed C but faster'
i'd really like to try AmigaOS 4 out.. I google'd some screenshots, and it looks fun to experiment with just for something different.. i'd like to try emulating an Amiga system.. Or possibly using something like Vmware.. does anyone know if this can be done?
*plays the Apogee theme song music*
There's something not right, here...
Something not up to Slashdot standards...
Ah... there's no "dept." caption/commentary!
What if I do the same thing, and I do get different results?
That was the first thing I noticed, too! I couldn't work out what was "wrong" with the story - some sort of disturbance in the force :)
Physicist, consultant, science communicator
no waiting for the system to swap out when switching between major applications
I hear not having any will do that for you.
Only Amiga is worth using. If you disagree, you deserve worse than death.
I suggest you read Slashdot
It doesn't help with Windows. Its *#$@! VM system is still tuned to machines with far less memory than we have today. Run anything memory intensive and I guarantee that you'll start seeing swapping and thrashing. On the bright side, at least it doesn't swap everything out to disk when you minimize the application. It used to be tons of fun working on local J2EE instances after accidently minimzing the console.
Javascript + Nintendo DSi = DSiCade
You know, that's just what they said about cuneiform. But I'm continuing to develop new kinds of clay for the tablets and to experiment with new ways of making a reed stylus- I'm working with a new kind of reed from South America which is vastly superior to the ones the Sumerians used. And cuneiform on clay tablets works fine for all my word-processing and accounting needs, plus it never gets viruses. Well, I did once have a problem with mold growing on my styluses. But I solved that by keeping them in a dry place.
The Amiga died for one reason. Closed Source on a Closed Platform.
Yep, just like Macintosh. And we all know that IBM machines survived because of Microsoft's open operating systems.
The reason Amiga died was because Commodore was completely inept on just about everything non-technical in nature - advertising, business decisions, corporate alliances, you name it.
Dan East
Better known as 318230.
If I could find an affordable Ethernet card, my Amiga 3000 would still be in active use today, mostly as an archive server for all my old stuff. Sadly, the only Ethernet cards I can find are $150 or so, and the TCP/IP stack is (usually) not included.
The way things are now, though, the only way Amiga will have a future is if A) a dedicated investor with very deep pockets and a lot of patience funds a company to look after it; or B) they Open Source the entire OS and support utilities. The latter is likely very easy from a contractual aspect, since the only "borrowed" code was from TRIPOS, and much of that was re-written in C for the OS 2.04 release years ago.
I could go on and on about what made Amiga great, but every time I even mention it, people immediately place me in the slot marked, "crazy." I'd like to see more Amiga philosophy in modern software design, but even I have to admit that light of Amiga may be irretrievably fading. Really, you people have no idea what you missed...
Schwab
Editor, A1-AAA AmeriCaptions
I remember when "Amiga" meant innovation and usability at an affordable price. One of the amazing things about the Amiga was that most of the cheesy slogans that were used to sell it (e.g. "Only Amiga makes it possible" and "The computer for the creative mind") were true. It felt good to own an Amiga, because it was orders of magnitude better than anything else out there.
Today, "Amiga" is just a trademark. Will this new Amiga-branded system compete with Mac OS X? With GNU/Linux? With Windows? If not, why should I, as an nostalgic Amiga zealot, care?
I have no need for yet more proprietary hardware running yet another proprietary OS in a time when commodity hardware and free software are where most of the interesting things are happening.
The new Amiga we dream of won't be called "Amiga". It will be something completely different---built by a small group of brilliant people that nobody has ever heard of---not the underwhelming output of some company whose only real purpose is to figure out how to extract revenue from the copyrights and trademarks for a 20-year-old technology.
From your eulogy:
Actually, OS-9 was running on 6809 based GIMIX and SWTPC systems well before the Coco ever saw the light of day. I still have working SS-50 systems that run it (and FLEX.) They also ran OS9 a lot better than the Coco could, because the Coco's hardware was uber-cheap compared to the (literally) gold-plated machines from GIMIX, not to mention that the GIMIX machines could support a lot more RAM, which, as we know, is definitely an issue in a non-VM multitasking system. :)
The Altair/S100 and SWTPC/SS50 machines started everything, pretty much.
I've fallen off your lawn, and I can't get up.
The proprietary nature of the platform had little if anything to do with the Amiga's death
a world in progress...
I'd do the sensible thing: 1) migrate to ppc, 2) put in large chunks of BSD code and 3) migrate to x86.
I hadn't the slightest objection to his spending his time planning massacres for the bourgeoisie... (P.G. Wodehouse)
See, the article author does have a point.
At the early 90's Amiga was an amazing platform, with even more amazing software. A desktop publishing software was crammed into 1.4MB, two 720KB floppies! And you had an almost perfect alternative to Word 6.0 on less than 720KB, and spell checking was only another floppy away from you.
I had an Amiga 600, with 4MB RAM and 40MB HD, and I never managed to use half of the space. Why software is so bloated nowdays? I understand that now we have multi GHz cpus, with loads of RAM... but yet, we waste too much resources using poorly optimized software. For an example: OpenOffice.
I understand that now software do a lot more, we have higher resolutions and color depths... But does it justifies the lack of performance, the bloat? I mean, OpenOffice will crawl on a machine with less than 256MB, and a average Amiga had no more than 4MB of RAM!
---- You know how some doctors have the Messiah complex - they need to save the world? You've got the "Rubik's" complex
The reason the Amiga was special was that it was a quantum leap for computers of the time for the following reasons in no particular order:
.info files for executables (a local registry for each program)
1) preemptive multitasking.
2) special hardware for graphics.
3) a unified memory architecture.
4) stereo sound with hardware-assisted mixer
5) a UNIX-like O/S with many goodies, including
6) a nice GUI that looked good on low resolutions with datatype aware drag-n-drop for every app.
7) a good DMA architecture that allowed for easy parallelization of many tasks (for example graphics not blocked by I/O)
What would it take for the Amiga to be a quantum leap today, given that the average 500$ Intel PC has much better capabilities than the Amiga of yesteryear? there are certain possibilities:
1) provide sound and graphics of 5000$ worth at the price of 500$. This is highly unlikely, because all the billion dollar pioneering research in graphics takes place in the labs of NVidia and ATI, two companies that will not be willing to sell their top technology for a mere 500$. The Amiga was the result of hardware gurus like RJ Mical that worked on their own designs...so unless a similar group of talented individuals gather up and make something unique, this possibility is less likely to happen.
2) provide a computer with a fixed hardware, like a console, but with an O/S that the users can write applications and games that hit the hardware directly. It might sell but for small numbers...back bedroom programming will certainly thrive on such a machine,
but I do not think the numbers it sells will be sufficient to sustain it.
3) do something really wild like a computer with 3d stereoscopic graphics projected either in mid air or in a special display. Now that would be a quantum leap, but only if the price is right, and it would certainly be hard to make and sell.
Overall, I do not think Amiga has a place in today's computing environment...especially when the O/S works on special hardware platforms.
Amiga turned into a three ring circus. First you have those who sort of own the copyrights (most of the patents still are owned by Gateway and are licensed out to Amiga). The sad tale of OS4, it was suppost to be owned, sort of, by Amiga Inc and Hyperion. Hyperion's orginal contract to roll out OS4 had a $25K buy back option (which I under was executed by Amiga Inc). Little did Amiga Inc know or realize, Hyperion allowed a newly coded kernel that was owned by Hyperion subcontractors (Frieds (SP) Brothers) to be used so when the buyback option was executed, Amiga Inc couldn't get the kernel since that was owned by a third party. Think it all still in the hands of lawyers and there is no licensed hardware to use for OS4. I don't expect to see any licensed OS4 products being offered for sale for a long period of time.
Second is another closed sourced called MorphOS which runs on third party PPC hardware made by Genesi (the OS and hardware are owned by seperate companies).
Third is where I think the true Amiga spirit lies, a open source version called Amiga Research Operating System (AROS). It's a community OS driven by what we loved in our Amigas. The orginal AROS coders realized that we would never see customized hardware that gave the real Amigas such power and capabilities compared to the painful window boxes of the 1980s. Common hardware (x86) was targetted as the new enviroment, it was the OS that mattered since the x86 had grown far beyond what the A4000 could have offered at the time. AROS is also being ported to PPC (and specifically Genesi's new PPC, EFIKA), x86_64 and hopefully one day, ARM. Self booting x86 ISO can be download (free as in beer) at http://www.aros.org/. AROS is a work in progress so it's not as nice as OS4. Then again, unlike OS4, it can be used on just about any old x86 that you have laying about. AROS is always looking for more developers and there is a third party bounty system setup to motivate AROS developers at http://www.teamaros.org/
Dammy
It's interesting to think whether it's a waste *not* to use the extra CPU cycles and memory we have these days, by coding efficient apps, or whether we should push a system to use every resource it can, for example by having the computer handle all memory issues instead of the programmer (I've never liked the idea of Java handling memory cleanup, when the programmer should just be doing his job properly.. not that I've done a lot of C++ coding for a few years now, and haven't used Java much either). I guess the thing is, that if you're running one application only, like a game, you want it to be using all the resources it can, but when it comes to word processors and browsers, you want them to have as small a footprint as possible. When it comes to the OS, you reaaaaally want it to hardly use any resources for its own nefarious deeds - having the system need a 128MB 3D graphics card or whatever just to run the interface as it's meant to be, seems a waste. In the future it will probably be common practice, but right now, I think Microsoft are just taking things too far... unless the interface really does improve the functionality of the OS in a useful way.
which is totally what she said
How this got modded insightful, Mods only know!
I write this on a laptop with 2GB of memory - sum total of applications running:
Outlook (yes I'm at work, we do what we have to)
Several gVim sessions
Firefox with 6 Slashdot tabs and 1 gmail tab
Acrobat Reader
VNC session
Winamp
as I alt tab to winamp, watch the hdd light flash and the delay in re-draw.
I kid you not, that with the exception of tabbed browsing, I used to do all of this on my Amiga 4000 with 16MB of ram without swapping. my old A1200 only had 4M of ram and i used that as a desktop for a couple of years and that didn't even have the concept of virtual ram!
Now maybe this is the price of progress, but seriously, how much ram do you suggest I need to buy in order to stop this swapping?
As an collery, my desktop at home at 4GB runs Ubuntu and that swaps in similar situations too. Maybe this is the price of progress, but if this article only reminds us that there is another way then I'm all for it.
"The weirdest thing about a mind, is that every answer that you find, is the basis of a brand new cliche" -
Welcome to the world of object-oriented programming. What, you thought all that crazy inheritance was free???
Applications are bloated because developers try to (and fail to, as it turns out) "optimize" for lowest development time, and they think they'll be more "productive" if they use a bunch of classes from some class library that kinda sorta does what they need (hey, no big deal, just subclass from it and reimplement the methods that don't do what you want, right?). But if everything I've seen is even halfway true, there is usually no real reduction in development time, and the resulting programs are usually even more opaque (and thus harder to debug) than they would be if they had been badly written in a procedural language. At least with a procedural language, what you see is what you get, and tracking down the flow of control is relatively straightforward. With an object-oriented program, what appear to be straightforward method calls tend to be very difficult to track back to their actual source unless you use some magic tool to do the job for you. End result: the program is harder to understand (is it really using the class method you think it is, or is it using the method of one of its ancestors?), harder to debug, and harder to maintain.
Object oriented programming is a tool, just like procedural programming is. There are certain classes of problems where it's very obviously the right tool for the job, and sometimes it's the right thing to use even in the middle of a procedural program. But it's not a general-purpose programming method.
If you think I'm wrong about all this, try justifying the 30-40 levels or so of inheritance nesting that you get from a typical Java stacktrace. Each of those levels represents an additional level of inefficiency that simply wouldn't be there if the program had been written properly (which may or may not involve writing it in an object-oriented fashion).
Use 'slashdot stuff' in the subject line in any email you send me if you want to get past the spam filter.
Er. The Amiga community strongly embraced OO programming - it had BOOPSI (Basic Object Oriented Programming Systems Interface - largely language agnostic OS-wide OO component system... - good idea, implementation could have been better, but still it was better than most), MUI, ClassAction, etc., and in fact its implementation of "shared libraries" functioned almost identically to COM objects before BOOPSI ever appeared.
Whatever's at fault, it's not OO methodology itself. It's just laziness. On the amiga, if your program was buggy or leaked memory, you crashed your machine (or got an Enforcer (memory protection add on usable on high-end machines with MMUs) hit, which meant you'd crash your customers' machines). The absence of memory protection in the presence of message-passing-by-reference pre-emptive multitasking probably encouraged careful, efficient coding. Is the lack of safety worth it? Not these days, when peopleor download random stuff off the net and run it (back in the day, the bulk of amiga users relied on filtration of the nasty stuff from the raw internet by magazine coverdisks/coverdiscs - though at one time the aminet (amiga software repository on the internet) was the largest repository for _any_ platform all the same).
...because you've never used it. In AmigaOS, the idea of assigning names to directories (not just drives) was pervasive. You'd say that "FONTS:" would comprise a list of directories where you stored your fonts files. When a program tried to open "FONTS:Helvetica.font", it'd search each of those directories in order and return the first match it found. All system libraries went in LIBS:, your command-line utilities went in C:, and so on. It was exceedingly rare to use hardcoded paths instead of named search lists for anything general.
Probably, but maybe .5% of people actually use that ability. Again, the difference with AmigaOS was not that you could do it, but that everyone universally did it. I was just something you used without making a big deal of it.
No way. You might have been able to perform those exact (poorly chosen) examples, but neither Linux nor Windows were anywhere near as good at multi-tasking in '95, let alone '85. It's like hearing someone talk about a car with great handling and not understanding; your Oldsmobile can turn corners, too, right? It was just something you had to see to really understand.
I have no illusions that AmigaOS will make a comeback, and by now I wouldn't want it if it did. Still, it did a lot of things right, even by today's standards, and you can't just dismiss it by saying that other systems can do some of the same things.
Dewey, what part of this looks like authorities should be involved?
I was the managing editor of .info magazine, which covered the Amiga exclusively until 1992; just before it died, we did.
My (admittedly high-end, for its day) Amiga 3000UX could run Windows 3.1, Unix, and AmigaOS SIMULTANEOUSLY on three pull-down screens. People would freak out when they saw me pull down and flip between three different screens running three different operating systems. And it wasn't just some cheap parlor trick - all three were running various applications in real-time.
Oh, and you could even run a Mac emulator on the Amiga screen at the same time.
This was in 1990. Can your machine do anything even remotely like that today? AmigaOS had a very different way of looking at how computers should work. There is still a lot that OS programmers can learn from the Amiga.
Serving your airship needs since 1995.