Quote:

For example on AOS the only real limitation on multitasking was the available amount of RAM. We had 2Gb address space when PC had 640k.

Too bad the 68000 was limited to 24bit address space (ie: 16Mb RAM)...

_________________
http://www.warpdesign.fr/

@damocles

The right market conditions always exist for something.
+
I never said anything about desktops.

_________________
All the way, with 68k

@Mechanic

Quote:

Is that anything like 'JUNK ACCUMULATES TO FILL ALL AVAILABLE SPACE' ?

Yeah.

As an example, I'm overhauling the AROS console.device/handler these days, and one of the things I've added that was sorely missing was a scrollback/review buffer like most people would be used to from KingCON etc. since the AROS console was in a sorry state.

Then today someone asked why I don't just allocate a fixed buffer, instead of growing
it gradually, and I pointed out that a 500 line buffer worst case could need something like 250KB to keep the text, color/pen info and flags (depending on maximum width), and for someone like me that often keep 30-40 console windows open (on my main machine, anyway) that'd easily mean upwards of 10MB extra for the scrollback whether or not it's used. It feels wasteful, even though most users probably won't notice.

Then a bit after posting it, I remember that my "record" for memory usage for iTerm on my Macbook at work is ca. 1GB and I regularly get up to those levels, and normal usage is in the hundreds of MB.

Ouch.

The only reason I can think of for that kind of memory usage is that they must keep an off screen bitmap of every tab/window in memory. But at 1680x1050x32bit (which is massive worst case - a lot of the console windows are not full screen, or is the laptop screen instead of the second larger screen), 40 windows would still "only" add up to about 270MB.

So either they're doing something even more fundamentally stupid than keeping an off screen bitmap of something that takes a tiny fraction of a second to re-render from text (and that they need code to fully re-render anyway when scrolling past a whole screen) or it had leaked more than 700MB in a day...

Now this is just one app, but I have similar experiences with lots of apps. Developers just don't care on these platforms.

But to me, wasting 250KB per terminal window still seems unacceptable, and I consider it unacceptable not to ensure it handles out of memory situations gracefully...

There's just no excuse. It doesn't take much more time to properly handle redraws without a permanent off screen buffer per tab/window. Handling memory allocations properly isn't rocket science... The only reason for a console/terminal app to take that much memory is pure laziness or carelessness.

I don't mind buying more memory if allows me to do new things with my machine - I work on projects that genuinely would benefit from many GB of RAM without wasting it, - but I don't want it eaten up by bloat in apps that haven't actually added any new features I have any use for... Certainly not something as simple as a shell window.

/rant

EDIT: Yes, I realize caring about this stuff makes me an old fart. But then again, I'm in good company here

Last edited by vidarh on 18-Aug-2010 at 03:20 PM.
Last edited by vidarh on 18-Aug-2010 at 03:20 PM.

_________________
Wiki for new/returning Amiga users - Projects: ACE basic compiler / FrexxEd / Git

@Leo

"Too bad the 68000 was limited to 24bit address space (ie: 16Mb RAM)..."

Still it was light years ahead of x86 & Mac. it took 10 years for M$ to truly get over the 640k limitation. (IIRC, then there was the 64M limitation of WinNT and ....)


(...waiting for the day when they truly go beyond 8+3 letters)

Last edited by KimmoK on 18-Aug-2010 at 03:32 PM.

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

I'd have to agree with Carl, the main reason I got an Amiga to begin was simple because there no way in hell I could of afforded an PC when i was a kid and because i wanted more than just a game console, i was interested in making my own games, with programs like Amos, Bliz Basic and Delux Paint you had all the tools you needed.

It was a very easy and simple computer to get to grips with, the OS is to me the only logically OS I have worked with. Files are put into simple to follow draws and you can easily access or find that missing file and where it should be placed. Access the Ram disk, you wouldn't think it be that handy, but it just is.

Plus the computer could become anything you wanted it to be at low cost, and fairly simply, you could just use it as a gaming machine, never even see the OS, you could turn it into a music production computer (midi) low cost, or a CGI media computer (video toaster) again at much lower cost than the competition.

But then this is back in the 90s, the competition has caught up, you can now quite easily get a computer for £200 the amount I originally paid for my second hand A1200, loads of software is out there for pretty much anything you can think off, and usually you can find it free (linux), OS like Windows and Mac can multitasking, have become pretty simple to set up and use, however there OS are not quite as easy to understand if you are looking for that special file, it took me ages to work out where Windows 7 puts Saved game files for example.

So with that in mind should the Amiga of today still focus on such goals, I say yes! These are sensible goals and any computer that can meet such goals deservers to survive and prosper. Its just this time we got competition.

_________________
AmigaNG, YouTube, LeaveReality Studio

@vidarh
Quote:

Yes, I realize caring about this stuff makes me an old fart. But then again, I'm in good company here.

YUP!

@amigang
Quote:

It was a very easy and simple computer to get to grips with, the OS is to me the only logically OS I have worked with. Files are put into simple to follow draws and you can easily access or find that missing file and where it should be placed. So with that in mind should the Amiga of today still focus on such goals, I say yes! These are sensible goals and any computer that can meet such goals deservers to survive and prosper. Its just this time we got competition.

I just looked at my basic Linux/Gnome install (without OO mind you)
and peeked at the LIB and USR/LIB directories.

/LIB = 207 files + 15 folders with many many more files.
/USR/LIB = 1313 files + 119 folders + +.

How's that for competition ?

(Perhaps the Linux distros need to get together and buy a good stiff broom.)

@KimmoK

Quote:

Still it was light years ahead of x86 & Mac. it took 10 years for M$ to truly get over the 640k limitation. (IIRC, then there was the 64M limitation of WinNT and ....)

I think it is pretty much irrelevant. Most Amigas never had more than 1MB and many Amiga users had to fight with 512kB chip ram limit.

_________________
Amiga Developer
Amiga 500, Efika, Mac Mini and PowerBook

@Carl-S

Quote:

Let me post a test. I will put it in a new topic. This topic is getting quite long.

Was this new topic ever started, does anyone know? Maybe someone at the show can ask him.

_________________
"Unix is supposed to fix that." -- Jay Miner

@bison

Yes. The new thread was started and ended long ago.

#6

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

@number6

Where is it? I couldn't find anything...

_________________
"Unix is supposed to fix that." -- Jay Miner

@bison

[Poll] Carl's Amiga Test

#6

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

@number6

Thanks!

_________________
"Unix is supposed to fix that." -- Jay Miner

Quote:

Still it was light years ahead of x86 & Mac. it took 10 years for M$ to truly get over the 640k limitation. (IIRC, then there was the 64M limitation of WinNT and ....)

Sure. Now how many years (decades) will it take for the Amiga to get over DOS and Exec limitations (64bit, SMP, Memory protection) ? A lot more that it took MS to produce and have WinNT widespreaded I'm guessing...

_________________
http://www.warpdesign.fr/

@KimmoK

Quote:

KimmoK wrote: @Leo "Too bad the 68000 was limited to 24bit address space (ie: 16Mb RAM)..." Still it was light years ahead of x86 & Mac. it took 10 years for M$ to truly get over the 640k limitation. (IIRC, then there was the 64M limitation of WinNT and ....) (...waiting for the day when they truly go beyond 8+3 letters)

http://support.microsoft.com/kb/117373

Windows NT detects up to 4 gigabytes (GB) of RAM, but if you are using a computer with more than 64 megabytes (MB) of RAM, Windows NT may detect only up to 64 MB. This behavior is based on how the BIOS returns the amount of memory to Windows NT. Even though the BIOS may correctly display memory above 64 MB, it may not pass this information correctly to Windows NT.

_________________
Ryzen 9 7900X, DDR5-6000 32 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32lite/RPi 4B/Emu68)
Amiga 500 (rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

Quote:

Carl-S wrote: In the "AmiWest 2006 - Any reports?" topic, I made this statement: Quote: I know quite well what an Amiga is. To me, Amiga is well defined. Saimo asked me to elaborate, and it would be my pleasure to do so. Here is my definition of Amiga: The Amiga is a computer system that enables millions of creative people at a low cost of ownership. Are you surprised I define Amiga that way? Not in terms of CPU, MHz, bus, chips, or apps? Here's what I mean... Computer System: Amiga is a complete system including both software and hardware. You want people to be able to buy the entire package, not just the "kit". For software, from what I've seen, I think OS4 does it quite well. For hardware, I don't care what CPU, as long as it can fully satisfy the rest of the definition below. (Remember that I was one of the main 68000 supporters, but I think the CPU topic is no longer relevant.) Enables: This is the action verb. To me this word is more than just the opposite of disable (e.g. Windows), it means to empower. The computer should work for people, not the other way around. (As I've said before.) Millions: Do we want just a few Amiga users or do we want millions? I still believe in what we started back in 1985. Millions of users gives us a true marketplace where hundreds of products can afford to be developed, supported, and thrive. Creative People: Creative people solve problems. They think. They enhance. The legacy of Amiga is the amazing list of doers, thinkers, artists, musicians, programmers, and even CEOs who have owned and used Amiga over the last two decades. Those are the kind of people you want using your system - not droids. And also, I don't just mean creative users, I am also talking about creative developers and entrepreneurs who make it possible from the start. Low Cost: This has always been an important component of the Amiga dream. We want schools, families, community orgs, hobbyests, and small businesses to buy into Amiga. We want the starving artists. We want the young genius or rebellious kid. So, the barrier to entry must be low. In fact, if it were possible to rejuvenate old PC boxes and laptops to make them decent Amigas, I would be for that. (Remember how Linux got started?) There is also more to low cost than just the purchase price... cost includes support. Final Note I know that many current users will not agree with all these points (perhaps mostly the last). But, it's important to start with the top level definition. That's how we made the original Amiga. We decided what was most important. All goals are not equal. If you want to go for the big win, you sometimes have to give up a few lesser desires. I would love to see the Amiga return someday as a force in the computing arena. I actually think it is possible, as insane as that may sound. But, it's going to require a clear vision and a grassroots unity of purpose.

@Carl-S

This was true back in 2006 and if I may be so bold, is more true in 2010.

Very well put Carl.

_________________
It's not the question that's the problem, it's the problem that's the question.