Things the Amiga didn't get right from Day 1

[Gorf]

Originally Posted by grond

But what about the devices such as trackdisk.device and scsi.device? Aren't they pretty much what the filesystem device would be?

No: these are from the viewpoint of the system even further below.
The stack is

dos.library
filesystems
devices

where I would like to see:

filing process/task
filesystems
devices

A dos.library might still exist, but it would only provide calls to the filing-task, as the intuition.library calls intuition-task.

Like intuition takes care of all input devices, windows, menus and so on the filing-task should take care of all disk (and network?) interactions.

A program would just send a message to the filing-task which (part of) file it wants in what memory location and the filing-task sends a message back, when its done.
The filing task takes care of of possible multitasking conflicts, serializes request and delivers the requested data in the most efficient way.

A CLI or shell would also only "talk" to the filing-task, so most commands would be really tiny.

[TEG]

Originally Posted by Gorf

And, as Thomas wrote, to make it work anyways under Exec all these forbid-permit constructions became necessary.

(Maybe that is what the developers meant when the hid the message "we did it and they f*** it up" in the ROM)

Oh! Great revelation to me. I always wondered because it was not possible it was about the marketing as it was in the ROM so... before the marketing. Thanks Gorf.

[Thomas Richter]

Originally Posted by Bruce Abbott

Did you perhaps mean that it didn't have preemptive multitasking?

Yes, one needs to be more precise on all that. Thus, let me elaborate:


You find the original Tripos manual actually online (but I would need to look for it, I have it somewhere). While I wrote that "librares are the domain of exec", and this is true as you find it in the Amiga, original tripos did have "devices" that are very similar in their structure to exec devices (surprisingly similar, even), and there were also libraries, though this part is "not really used" in AmigaOs as it was taken over by exec, as written.


There is an interesting residue of this "BCPL library concept" left in the Tripos "scatter loader" aka "LoadSeg()" as the function is called from the AmigaOs side. LoadSeg under 1.3 and below (thus, the native BCPL version of it) was able to open dynamic libraries for you, and link their offsets into your binary. There would be no need to "OpenLibrary()" them as you have to nowadays, but similar to the Linux "elf" ldd loader, it would do so at load time. As said, Kick 1.3 still had this feature (mostly undocumented), but unfortunately you could not really use it. The problem is that this BCPL loader does not guarantee that the library base is in register a6, and thus, none of the system libraries except the dos.library could be opened or loaded like this. With Kick 2.0, this feature was dropped in the "arp" re-implementation of LoadSeg(), which then was native assembler.



Concerning multitasking: Yes, but... Tripos does have the concept of "Processes", which is actually what the BCPL "process table" and the "CLI number" is good for. Thus, strictly speaking, Tripos in its Amiga incarnation supports at most 10 processes, as otherwise its process table overrun. Thus, a process for Tripos is something different than what it is for AmigaOs - it is more "a CLI".



This stupid 10-processes (actually, CLI) at max changed with Kick 2.04 and extensions to this structure.



Thus, most of the multitasking was done through co-routines, native support for which you find in the BPCL "GlobVec", and its filing system (for example) is heavily multi-threaded, something that was then ported to the FFS incarnation, which was in Assembler, no longer in BCPL. I believe it was this mechanism (and not message passing) that exchanged control between handlers and the calling functions, and processes were really reserved for the "shell", and those were scheduled.


As side remark, note that this is similar how *ix handles it: BSD, Unix, Linux etc... don't run file systems in their own native process as AmigaOs does. Instead, they "call" into the file system and run it as part of the calling process, and decoupling between caller and IO happens only one level below, namely between the "file system" (actually, a collection of functions) and the "device" (the thing associated with the node in /dev). Thus, tripos had more in common with *ix in that respect. AmigaOs pushed this further and relocated the "handlers" in processes of their own.



BCPL libraries no longer exist in the Amiga incarnation, and neither devices, though Tripos had very similar counterparts. I'm not quite clear in how far this goes for processes. Tripos was somehow "bolted" on top of exec, with duct-tape and hot-glue, how it looks, and some of the Tripos concepts found almost precise counter-parts in the exec world, so similar that it may actually be that Carl copied parts of the Tripos ideas and implemented them natively rather than in BCPL, but who knows.


Thus, if anything went wrong with Amiga then this unnecessary mirroring of concepts. Tasks on exec side, processes on Tripos side, but Tripos processes did not have entries in the Tripos process table (though they should) and this process table became the CLI table (though it really wasn't). It's really an overall mess, with many inconsistencies.

[Gorf]

There is an other interesting and quite advanced for its time system from the mid 80s:

QDOS for the short lived Sinclair QL, that came out one year before the Amiga.

It featured (cooperative) multitasking and some simple memory management and an distinguished between private and public memory areas.
As well as an extensible "redirectable I/O system".
(Linus Torvalds himself wrote device drivers for QDOS...)

There is a (late) port for the Amiga, but I never tried it out:
https://aminet.net/package/misc/emu/QDOS4amiga1

[Nishicorn]

I agree about the joystick, it should have had at least two buttons and Paddles-support. However, I think it was fine for the games, we had a lot of fun, and many games worked with keyboard anyway, so it wasn't a big deal. Amiga was a COMPUTER, after all, so it always has the keyboard with it. .

I don't agree about the MIDI-port - that was a very specialiced, niche feature that most people, especially the masses would not have used anyway, and those that WOULD use MIDI, could always get an external MIDI-port anyway. It's not a problem, just a different way of implementing it than Atari, for example, who featured it so much that they managed to brainwash people into thinking it's something special.

The only 'better than' feature Atari ever had against Amiga, and you make it sound important.. come on.

With the HAM-mode thing, you are not really appreciating the era - it was a REVOLUTIONARY thing in 1987, let alone 1985! I would dare go so far as to claim it was one of the MAIN selling points of the Amiga - to be able to display a photo that looks like an actual photo, with superior colors to anything anyone had EVER seen a computer display before!

And you are ragging on it?

I mean, I see the point that HAM was slow, and it wasn't really an 'official screenmode', and in such a way, it was a bit of a 'hack' or afterthought, and unusable for anything but still pictures, but what people did with it and what could be done with it, even with the infamous 'color leakage', was still STUNNING for the time. Even now it looks pretty good.

I mean, would you rather have a computer that doesn't have that feature, or has it? It's like a free bonus, an added feature that's not needed, but nice to have. It's an extra gift on top of a marvellous creative soul's, artist's and gameplayerist's (I loathe the word 'gamer') computer. So what if it's a bit useless mode, it's not like it's the ONLY thing the computer can do, it's more like great little bonus on top of everything else.

Floppy stuff was always pretty bad for Commodore machines, ever since VIC-20 at least - but it's not a problem nowadays, and there's an atmospheric Zen-moment to be had in the middle of an autumny night when the full moon shines through your curtains in an otherwise dark room, and you see that drive light flicker, and listen to the sounds of crickets and night birds amidst the mesmerizing, trance-inducing, TRK, TRK, TRK, TRK, TRRRRRK, "Insert Disk 09".

Sure, the floppy system was non-standard, but at the same time, it was a bit more advanced than the standard stuff. It makes cross-platformy stuff difficult and annoying, of course, but it is basically compatible with the 'standard' stuff anyway with certain software. Amiga's 880k disk drives can read and write 720k DOS diskettes, so I don't see the problem. Atari ST had the same problem, but their capacity was even lower - I mean, try to write an Atari ST disk on an Amiga and .. I still can't figure it out.

However, Amiga 4000 could even read the 1.44 MB disks.

By the way, you made a mistake when you said 720 KB, it's definitely NOT KelvinBytes! It's kiloBytes, which means you have to use a SMALL "k", not a capital "K".

This, unfortunately, drops the credibility of your criticism a few points.. I mean, would you trust 'Rain Man' to give you trustworthy and well thought-out criticism? No offence, just an example of how it works.

I don't think the 'non-standard'ism was any problem, as Amiga software wouldn't have worked on PCs anyway, and vice versa. What would you have moved via diskettes that you couldn't move via modems anyway?

The hard disk stuff is always a bit of a point you can think about from many viewpoints; you have to realize this was a gameplaying machine for the masses in 1985 and 1987, and hard disks were still relatively easy to use with Amigas if you had the money and knew how. You are talking about it as if it was a huge problem - it wasn't. With Amiga 1200, hard disk installation couldn't have been much simpler; just open the case, connect, and that's it.

So when you say 'Amiga', it looks like you are talking about Amiga 500 mainly, not Amiga 1000, 2000, 3000, 4000, 1200 or CD32.

It's always going to be a money thing as well - would you rather buy a cheap, powerful gameplaying machine you can easily play Turrican II on, or a cumbersome, heavy, expensive machine that you have to install hard drive stuff just to get things to work? I mean, a typical buyer wasn't your office worker, they already had PCs.

The people that bought Amigas back in the day couldn't have cared less about any hard drive stuff, and as mentioned before, it was never really a problem anyway; you wanted one, you could always get one.

I guess you might be comparing this whole thing with the PC side, because hard drives were such an obvious thing in the PC world, but this was because it was an office machine, and it was just a practical matter, it was just lucky for the gamemakers that PCs always had hard drives (at least after a certain point), so they could count on that, and thus make the games bigger and such.

I DO remember the PAIN of swapping disks when playing my self-bought Monkey Island 2: LeChuck's Revenge (stupidly named, because the first game wasn't named 'Monkey Island', but 'The Secret of Monkey Island', so why drop the 'The Secret' part? The secret of something is completely different a thing than that something).

That was when I started to realize the value of hard drive pretty desperately.

However, from the day I got my Amiga until that day, lack of hard drive was NO problem for me, it didn't even cross my mind or annoy me at all. I saw a friend with an Amiga with hard drive, and I thought it was amazingly cool, but I never even considered getting one until I got my Amiga 1200, and realized I can't live without a hard drive anymore, and didn't cross my mind or feel problematic to not have, before the disk swapping pain.

My list of 'things that could've been better' (I don't think there were actual FLAWS in the Amiga back in the day, considering everything - Amiga was such a wonder computer back in the day, any 'flaw' could be easily forgotten or compensated for)..

1) The sound chip

I do and did love Paula, and composed many a song with it. However, why only four channels? Even 1970s technology on the Atari (Atari 400) side had four channels, couldn't a next generation, next decade miracle machine have doubled that? Why only four?! I can never forgive this.

The other problem with it, is that it's only 22kHz and eight-bit. Why such a limitation? You create a 16-bit machine, but you put ANYTHING 8-bit in it other than a whole C64, it's a damn crime. Why do that? Why even MAKE a sound chip that's 8-bit, it makes no sense! (Even C64's sound chip has 16-bit registers - so funnily, we have a 16-bit machine with 8-bit sound chip, and a 8-bit machine with 16-bit one!)

The cool thing about Amiga's sound is that it can play samples, so all kinds of 'epic sound' can be played, that Atari ST can't quite reach, and the C64 has to admit some kind of defeat as well.

I have written a lot about these sound chip things, so I will try to keep this short..(TOO LATE!)

You can also use one channel for sampled chords, and thus add the amount of sounds played simultaneously that way, and all kinds of tricks like that can be done. But it's still four channels, and for a composer, this limitation is sometimes unbearable - just give me ONE more channel at least, please! (OctaMED tried to solve this by lowering the sound quality, with mixed results)

This is one of the reasons why PC owners were able to laugh at Amigists so early on with their OPL2 with 9 channels, OPL3 with 18 channels and the Sound Blaster superior, 16-bit sample playing capacity, and of course good old Gravis UltraSound.

People did get quite a lot of amazing stuff out of that humble chip, though - Turrican II title music being one amazing example.

For practical, everyday game conversions, having four, stiff sample channels after the fluid, smoothly flowing, quirky and living OPL2 + samples-world, forced game musics to sometimes suffer terribly. I can't even look at Wing Commander without shuddering and having to listen to OPL versions of the songs just to forget the trauma.

Sometime they did a good job of converting things to the world of samples, other times, not so good. Any attempt at converting a classic C64 song to Amiga was basically doomed to fail, even with the later inventions like PlaySID and whatnot.

In MY opinion, Amiga's sound chip should've been a 8-channel SYNTH, not just based on samples, but living analogue stuff, or at least something like the smooth sounds of FM synthesis, like the OPL3 chip. OPL chips lack when it comes to noisewave and drum sounds (they don't have an official noisewave, but you can get one pitch of noise with some tricks - because the pitch can't be altered, drumsounds can't be as nice as on the SID)

Can you imagine 'living sound effects' that you just can't do with samples? (Or maybe with too much work you sort of can) I am thinking of Falcon Patrol games on the C64, where the sound constantly lives according to what's happening in the game. This kind of stuff is long gone now, sadly.

Also, Maniac Mansion's C64-version's walking sounds are very 'individualistic' - I mean, they sound 'the same', but if you capture a bit of walking, then play each step separately, you can realize they sound very different from each other, and if you just use two samples, it sounds very monotonous by comparison. But who would bother (or even think) to put like 10 slightly different-sounding walking samples to an adventure game like that? This makes samples very 'dead'-sounding in practical reality of games.

I could go on about samples vs. living synth sound (I recognize good sides of both, and they can complement each other, but I don't think anyone can properly LOVE samples as much as one can love a particular living synth, like OPL3 or SID, because after all, 'samples' is just a name for a cold, dead technique - albeit a useful one), but let's move on to other points.

2) Guru Meditation. This happened a BIT too often, and was always annoying, especially when creating a great song in full inspiration, only for it to be destroyed when trying to save it due to that damn guru.

3) Viruses - they WERE a big problem and a hassle, and you had to constantly be removing them, and they did do a lot of damage to my valuable stuff, and so on. It was very very annoying.

4) The physical design

Amiga 1000 looked cool, with the keyboard garage and everything. I still think Amiga 4000 is the second most beautiful computer in the world, I always get thrills of excitement when I look at one.

Amiga 1200 is nice and compact, easy to take with you and so on - I even remember running in John F. Kennedy airport with my trusty Amiga 1200 in my bag, trying to get to my plane in time. Good times.

Amiga 500?

WHat the... why?

I mean, it does look 'cool' in its own way, and it's by no means ugly or anything - but at the same time, the design IS a bit lackluster. Nothing quirky or personalized about it, it's very mundane and downright boring.

The big problem is the size - it's very clunky! Who could bring that anywhere without cursing about the enormity of the whole thing? Because it's so huge and the power switch is in the PSU, it can be quite a problem.

Yeah, the POWER SWITCH in the PSU! Why?

Why can't you just turn it on from the computer itself? It's not a huge problem, but to HAVE to constantly fondle the PSU means you can't just set it up neatly and hide the PSU somewhere behind the table, you have to keep it close or bend over every time you want to turn it on.

WHAT were they thinking?

5) Bad Ports

It's not really Amiga's fault, but Amiga also suffered from bad ports .. SO much.

It couldn't quite do the C64's Galway synth-freakiness so Wizball would've been better off not having been ported. What was magical, cool, beautiful, aurally superbly exciting and just amazing game all-around, was a HORRIBLE TRAVESTY on the Amiga side.

The Amiga suffered from these ports SO much, they came from all directions, and they were mangled in the process. A good DOS game was at best a 'mediocre, passable port' on the Amiga with a lot of the good stuff sucked out (no hard drive, so something has to be cut, fewer colors, so beauty has to be lost, no living sound, so everything has to be crammed to 4 channels of samples, and together with everything else, a lot of sound personality had to be destroyed, speech removed, and so on).

A great C64 game was a 'let's just use a higher resolution and make some kind of haphazard graphics - people will love it because it's higher resolution'-conversion, where, even when done relatively well, the magic was lost, and in any case, SOMETHING was almost always lost in translation (besides Turrican games, I can't think of any that were either improved, or as good - perhaps Zak McKracken, that did suffer from the sample stuff, though, listen to the wimpy doorbell sample compared to the deep and long C64-sound).

A mediocre Atari ST-game was just ported and left as-is, insulting the Amiga by making it into Atari.

Amiga had SO much more possibilities and power, more colors, animation capability, scrolling capacity and whatnot, but that was rarely used, because CAPITALISM dictates people will buy the game anyway, and it's just quicker, faster, cheaper and more efficient to just make the game on the LOWEST POSSIBLE platform, and then port THAT version to the higher platforms without taking any advantage of the higher platform's superior capabilities.

It's a lot like a Spectrum game being ported to C64 without improving it in any way - it woudl tend to be WORSE than the Spectrum original, because Spectrum has a faster CPU, and if the same technique is used, C64's slower CPU would make the game slower as well. If the game is well programmed, taking advantage of C64's capabilities properly, then it was possible to improve upon the game a lot and make it even faster than the Spectrum original.

This just wasn't done, because it's not business-effective, it would have cost time, money and effort and people wouldn't have bought the game more anyway, so it wasn't done.

Lots of Atari 8-bit ports to C64 were four-color games that only had a couple of sprites (Zorro, Bruce Lee, etc.), when they COULD have had more colors and more sprites as well, and much better sound and music as well, so easily.

When I played Paradroid '90 on my family member's Amiga 500, I realized it's not as good to play as the C64-version, because it scrolls only in two directions (I can never forgive this), and it looks like it's 15 fps or something (Atari ST, if I recall, can't do smooth scroll very easily), the screen is small and so on and so forth. It's like it was done completely on Atari ST and only lazily ported to Amiga, when Amiga could've done SO MUCH MORE.

How is no one astonished and shocked that a programmer can be so lazy as to claim that a 8-bit computer can easily do eight-directional smooth scrolling, but a 16-bit computer like Amiga can't? This game was a big crime, and those crimes kept coming. Epyx games I don't even want to look, they even removed the humorous bits from some of the games, and when they kept it in, it's just not the same.

I will always choose the C64-version of any Epyx game over any Amiga version. It's just so SO so SO much better as an experience.

Other than these points, I can't really think of anything, because Amiga was just such a wonderful, fun marvel of a computer system, and even though I still own a real Amiga 1200, I will always miss the other Amigas I have owned, and the good times I have had with Amiga.

This kind of thread is actually a bit insulting, as it's like smearing dirt on something that really doesn't deserve it.

[Overdoc]

Originally Posted by Nishicorn

Yeah, the POWER SWITCH in the PSU! Why?

Why can't you just turn it on from the computer itself? It's not a huge problem, but to HAVE to constantly fondle the PSU means you can't just set it up neatly and hide the PSU somewhere behind the table, you have to keep it close or bend over every time you want to turn it on.

WHAT were they thinking?

Well, it might be a bit uncomfortable to switch the Amiga 500 on when the PSU was positioned on the ground somwhere.
But, on the other hand, if the power switch would be on the Amiga side, then you would have the PSU on mains voltage permanently, reducing it's life span. Also the newer switching power supplies would then permanently draw some current.

I think it was not a huge problem on the Amiga, since in contrary to the C64 you could reset the machine with A-A-CTRL, and didn't often need to switch it off and back on, like you need to do on a C64 without reset button.

[drHirudo]

Originally Posted by Overdoc

Well, it might be a bit uncomfortable to switch the Amiga 500 on when the PSU was positioned on the ground somwhere.
But, on the other hand, if the power switch would be on the Amiga side, then you would have the PSU on mains voltage permanently, reducing it's life span. Also the newer switching power supplies would then permanently draw some current.

I think it was not a huge problem on the Amiga, since in contrary to the C64 you could reset the machine with A-A-CTRL, and didn't often need to switch it off and back on, like you need to do on a C64 without reset button.

Yes, the PSU of the Amiga is actually quite good. The Amiga 1000 has some Apple II alike PSU and the power switch is on the machine, but since the magic three buttons almost always worked, the computer was switched off only when not in use. Also, switching it off on the Amiga 1000, means you have to reload the Kickstart disk again. I don't know if there is such bad guru or badly written software, that would need loading the Kickstart again, even without switching the computer off.

On the other hand you can use Amiga 500 PSU on Amiga 600 and later Amiga 1200, without any issue even attaching hard drive by taking the power from the floppy drive wires. Yet another forward thinking from the later Amiga designers - standardization of the power supply.

On the other hand I hated the PSU button on the computer on my Oric clone. Once I spent several hours typing in a BASIC game, then invited a friend to play together. He wanted to place the computer more comfortably on the table, incidentally pushing the power button when moving it, switching the computer off.

Kicking the power button on the Amiga 500, Amiga 600 or Amiga 1200 rarely (if ever) happened to me, even if I placed it at comfortable distance in order to be able to switch the computer off with foot.

[Thomas Richter]

Originally Posted by Nishicorn

I agree about the joystick, it should have had at least two buttons and Paddles-support.

Actually, the Amiga does have support for two-button joysticks. Actually, even three-button joysticks, and it also supports paddles and analog joysticks. The "industry standard" back then, however, was the Atari-8-way digital joystick with a single button, so it kind of stuck. The Atari joystick system (Atari 8-bit) also supported the same type of interfaces: Paddles, analog joysticks, and 3 buttons. This is not surprising given that Paula is a "lookalike" of the Atari POKEY chip by the same designer.


Thus, the hardware can do it, just that there wasn't a market for the joysticks.

Originally Posted by Nishicorn

I don't agree about the MIDI-port - that was a very specialiced, niche feature that most people, especially the masses would not have used anyway, and those that WOULD use MIDI, could always get an external MIDI-port anyway. It's not a problem, just a different way of implementing it than Atari, for example, who featured it so much that they managed to brainwash people into thinking it's something special.

Midi is just a serial port with a special plug - that's all. It would have required a single DIN plug as add-on, but no additional hardware would have been required. It was just not in the vision of the system designers.

Originally Posted by Nishicorn

With the HAM-mode thing, you are not really appreciating the era - it was a REVOLUTIONARY thing in 1987, let alone 1985! I would dare go so far as to claim it was one of the MAIN selling points of the Amiga - to be able to display a photo that looks like an actual photo, with superior colors to anything anyone had EVER seen a computer display before!

Even though HAM was more or less an accident. It would have been more useful in YUV space, but they changed that last minute, and HAM remained in.

Originally Posted by Nishicorn

Sure, the floppy system was non-standard, but at the same time, it was a bit more advanced than the standard stuff.

The Amiga floppies are pretty much standard. These are off-the-shelve Shugart-bus DD-2S floppies, nothing special about them. The format is special, but that was a deliberate choice to increase the capacity over that a PC can write (880K vs. 720K). The hardware can do PC format just fine, but it was a delibarate choice *not* to use it.

Originally Posted by Nishicorn

The hard disk stuff is always a bit of a point you can think about from many viewpoints; you have to realize this was a gameplaying machine for the masses in 1985 and 1987, and hard disks were still relatively easy to use with Amigas if you had the money and knew how.

My A2000 manual, 4.1 board, came with a section "how to install a harddisk". So it was definitely an option CBM considered, just that HDs were quite expensive at that time. CBM just f*ck up the autoboot of the harddisk, which was fixed with the 1.3 kick.


I only got a harddisk years later, essentially saving every penny for it. It was something like 2000DM for 80MB. It was quite clear that the HD was not included in first place.

Originally Posted by Nishicorn

1) The sound chip

I do and did love Paula, and composed many a song with it. However, why only four channels? Even 1970s technology on the Atari (Atari 400) side had four channels, couldn't a next generation, next decade miracle machine have doubled that? Why only four?! I can never forgive this.

It was "tradition" (Atari had four) and it was always "good enough". Actually, the Paula sound engine is simpler than the Atari Pokey.

Originally Posted by Nishicorn

The other problem with it, is that it's only 22kHz and eight-bit. Why such a limitation?

That's simply because Agnus loads two samples (16 bit) each scan line (15khz), and it's 8 bit, not 16, because you also need to have a D/A converter that is sufficiently precise, and 16bit was high-end HIFI back then, not consumer grade. Remember, the first CD players came only with 12 bit D/A converters, and that was after Amiga. It was really hard to get such precise D/A chips back then, leave alone for a (relatively) cheap consumer grade device.

Originally Posted by Nishicorn

The cool thing about Amiga's sound is that it can play samples, so all kinds of 'epic sound' can be played, that Atari ST can't quite reach, and the C64 has to admit some kind of defeat as well.

The Atari ST sound was a huge step backwards. It was just an off-the-shelve AY chip which couldn't do much, much less than SID could do, and also much less than the 8-bit Ataris could do. Probably the only reason why they couldn't use POKEY for ST - which would have been a better choice - was because it doesn't interface with the 68K bus easily.

Originally Posted by Nishicorn

2) Guru Meditation. This happened a BIT too often, and was always annoying, especially when creating a great song in full inspiration, only for it to be destroyed when trying to save it due to that damn guru.

Unstable Os, thrown out without sufficient testing...

Originally Posted by Nishicorn

3) Viruses - they WERE a big problem and a hassle, and you had to constantly be removing them, and they did do a lot of damage to my valuable stuff, and so on. It was very very annoying.

Haven't had them on my machine, but then, I didn't use a lot of pirated stuff.

Originally Posted by Nishicorn

Yeah, the POWER SWITCH in the PSU! Why?

On the A2000, it is on the back. Very inconvenient. They should have put the switch on the front. After all, only four wires.

Originally Posted by Nishicorn

WHAT were they thinking?

Save every penny. CBM.

Originally Posted by Nishicorn

5) Bad Ports

It's not really Amiga's fault, but Amiga also suffered from bad ports .. SO much.

Probably the lack of a support program for software authors. CBM never cared about customers or the ecosystem around the hardware they built. Once you bought the machine, CBM no longer cared about you as customer. They had your money, and that's all they cared.

[Gorf]

Originally Posted by Thomas Richter

That's simply because Agnus loads two samples (16 bit) each scan line (15khz), and it's 8 bit, not 16, because you also need to have a D/A converter that is sufficiently precise, and 16bit was high-end HIFI back then, not consumer grade. Remember, the first CD players came only with 12 bit D/A converters, and that was after Amiga. It was really hard to get such precise D/A chips back then, leave alone for a (relatively) cheap consumer grade device.

And CD-Players soon went from 12-bit D/A to 1-bit D/A.

Paula would also have used the sigma/delta trick in hindsight ...

The Atari ST sound was a huge step backwards. It was just an off-the-shelve AY chip which couldn't do much, much less than SID could do, and also much less than the 8-bit Ataris could do. Probably the only reason why they couldn't use POKEY for ST - which would have been a better choice - was because it doesn't interface with the 68K bus easily.

Sadly Jack botched the other superior soundchip that was in deveopmetg at Atari at this time the "AMY"

https://en.wikipedia.org/wiki/Atari_AMY

[Bruce Abbott]

Originally Posted by Thomas Richter

Actually, the Amiga does have support for two-button joysticks. Actually, even three-button joysticks, and it also supports paddles and analog joysticks... the hardware can do it, just that there wasn't a market for the joysticks.

That's right. One button joysticks were the standard. Multi-button joysticks were rare and more expensive, so games had to support one button operation or risk losing sales.

Midi is just a serial port with a special plug - that's all. It would have required a single DIN plug as add-on, but no additional hardware would have been required. It was just not in the vision of the system designers.

Not quite right. To avoid ground loops the MIDI signal is galvanically isolated with an optocoupler at the receiving end. For MIDI output you just need a resistor to set the LED current, but for input you need an optocoupler. This can easily be included in an RS232 to MIDI cable, so why build it into the computer? The answer of course was that if the MIDI port was there then people would want to use it - so kudos to Atari for seeing the market potential of a built-in MIDI port over the 'waste' of a pair of connectors that the vast majority of owners would never use.

Even though HAM was more or less an accident. It would have been more useful in YUV space, but they changed that last minute, and HAM remained in.

More useful perhaps, but a lot more complicated. HAM, like Extra Half Brite, was a cheap trick that didn't require much silicon to implement.

The Amiga floppies are pretty much standard. These are off-the-shelve Shugart-bus DD-2S floppies, nothing special about them. The format is special, but that was a deliberate choice to increase the capacity over that a PC can write (880K vs. 720K). The hardware can do PC format just fine, but it was a delibarate choice *not* to use it.

The drive is standard, but the controller isn't. The Amiga doesn't have a dedicated disk controller chip like other computers, but implements the functionality of one with a combination of CIA I/O pins, MFM/GCR read/write circuitry in Paula, and decoding using the Blitter and/or CPU. This gives the Amiga the flexibility to produce different formats, adjust control signal timings etc.

The Amiga's TRIPOS based DOS, HAM mode, and floppy drive handling are examples of implementations that might not be technically 'the best' but make it a far more interesting machine. Would anyone have even heard of TRIPOS or Hold-and-Modify if it wasn't used in the Amiga? Would we have been stuck with boring trackloaders that were straightjacketed by a dedicated floppy disk controller? I for one am glad that Commodore didn't just throw in a mainstream DOS, basic bitmap graphics and a standard FDC chip.

[dreadnought]

One button joysticks were only a "standard" in the microcomputer world (sans MSX), quite inexplicably so. Dedicated game machines such as consoles or arcades already had 2 (or more) buttons for a long time when Amiga was released. Being innovative and all it should try to promote at least the 2+ button standard, since it makes a huge difference in gaming.

It's not really that expensive to add an extra button to a joystick/pad and should be trivial to code.

[grond]

With the bonus of hindsight, if you are developing a ground breaking games console (which then turned into an even more ground breaking computer for other reasons), you should realise that you'll also want improved games controls. As was pointed out before, the Amiga (as Atari and even C=64 before it) could deal with analogue controls and more than one button. Using a new connector form factor would have forced people to buy non-standard joysticks which then hopefully would have included the new (or rather uncommon) features. Of course, today a company would also make sure that there are at least a few AAA titles available at the product launch that make use of the new features.

[Bruce Abbott]

Originally Posted by Thomas Richter

Probably the lack of a support program for software authors. CBM never cared about customers or the ecosystem around the hardware they built. Once you bought the machine, CBM no longer cared about you as customer. They had your money, and that's all they cared.

That is not true. CBM provided plenty of support to developers, both to commercial software houses and 'bedroom' coders who might one day become famous.

If you mean they didn't directly fund software development, why should they? The general idea behind home computers was that customers bought the hardware to either program it themselves (thus the prevalence of built-in BASIC) or buy independently produced software. The manufacturer's only responsibility should be to provide sufficient information and freedom for independent developers to do their own thing.

The Amiga suffered from poor quality ports because software houses could make money from them without much effort. But so did other platforms. The PC got many poor ports to CGA/EGA from the Amiga. Even though Amstrad published many games under its own label, the CPC suffered from awful ZX Spectrum ports that emulated Spectrum graphics (including color clash!) because the publishers wouldn't spend money on doing the job properly.

Directly funded software titles were generally mediocre. An open market with no restrictions produced a lot of dross, but also more innovative and excellent products.

[Gorf]

Originally Posted by Bruce Abbott

More useful perhaps, but a lot more complicated. HAM, like Extra Half Brite, was a cheap trick that didn't require much silicon to implement.

Why would YUV be more complicated?
Everything would be exactly the same except of the CLUT ... and HAM would look better.
The output signal would be different though - but the A1000 had both anyways RGB and colour-composite.

Instead of deriving the composite signal from RGB, as it was done, you could do it just the other way around: derive the analog-RGB signal from composite.
The only thing missing would be the TTL-RGB signal (which was not used until gfx-cards pass-through in the 90s...)

For the Video Toaster/Flyer this would have been a good thing, since it is fully digital-YUV internally.

[TEG]

Originally Posted by Thomas Richter

Midi is just a serial port with a special plug - that's all. It would have required a single DIN plug as add-on, but no additional hardware would have been required. It was just not in the vision of the system designers.

Well, I remember an interview of Dave Haynie saying that the ST have buffered MIDI ports and that why the Amiga can't cope with it. If my memory serve me well, he said they made an error of conception by not considering this buffers point.

By the way I saw in another video, done by a musician, that the timing of the MIDI on the ST was very reliable. He lets the ST running MIDI a long amount of time (several days I believe) and it was still perfectly in sync.

[TEG]

Originally Posted by Gorf

Sadly Jack botched the other superior soundchip that was in deveopmetg at Atari at this time the "AMY"

https://en.wikipedia.org/wiki/Atari_AMY

Perhaps for this reason:

I remember reading that Atari just could not get a reasonable yield from the chip fab. They made two or three runs, IIRC and gave up.

https://atariage.com/forums/topic/11...omment=4845888

[Gorf]

Originally Posted by TEG

Perhaps for this reason:

https://atariage.com/forums/topic/11...omment=4845888

Maybe.
On the other hand, after they sold the design the new company had no such problems ... but Jack decided to sue this small startup and crushed it - for no obvious reason.

overall a sad story.

[Gorf]

Originally Posted by TEG

Well, I remember an interview of Dave Haynie saying that the ST have buffered MIDI ports and that why the Amiga can't cope with it. If my memory serve me well, he said they made an error of conception by not considering this buffers point.

By the way I saw in another video, done by a musician, that the timing of the MIDI on the ST was very reliable. He lets the ST running MIDI a long amount of time (several days I believe) and it was still perfectly in sync.

same is true for the Amiga: with a simple adapter MIDI is no problem and runs reliable and in realtime.
While it is true, that a buffered serial port would be easier to handle for a multitasking OS, MIDI software on the Amiga worked around this pretty well.

[drHirudo]

Originally Posted by Gorf

same is true for the Amiga: with a simple adapter MIDI is no problem and runs reliable and in realtime.
While it is true, that a buffered serial port would be easier to handle for a multitasking OS, MIDI software on the Amiga worked around this pretty well.

Actually MIDI was problematic on the Amiga in the professional field.

Pointing MIDI in my initial post is not without a reason. Since the Amiga designers never had the MIDI in mind, they didn't do buffered interface for it, which turned out to be a disaster. I will quote my post from another thread here on EAB:

http://eab.abime.net/showpost.php?p=...&postcount=140

Regarding the MIDI and why it become more standard on the Atari, than on the Amiga: There is a very good article by the author of Music-X, which explains one fundamental issue why the Amiga was never considered as serious MIDI machine:

Several years later at an Amiga developers conference, I was approached by Bryce Nesbitt, an engineer working on the next version of AmigaOS for Commodore.
I don’t remember his exact words, but the gist of it was that he had finally solved the problem of why Music-X (and other Amiga music software) seemed to have a problem recording Midi data.
Because I went to all of the Amiga expos and developer cons, I knew all of the guys who worked on AmigaOS. I had long email threads where I had pleaded with them for help on solving this bug. One of the engineers (who was eventually fired by Commodore) cynically told me at one CES, “Your problem is simple. The Amiga can’t do Midi.”
And unfortunately, the market of professional musicians seemed to agree — more and more they were gravitating towards the Macintosh as the defacto platform for doing music and Midi, leaving products like mine and Todor Fay’s Bars and Pipes behind.
But Bryce, bless his heart, continued to poke at this problem behind the scenes and eventually came up with an answer. It turned out that the Amiga’s four timer chips were interfering with the serial port. Both the timer chips and the serial hardware were interrupt-driven, and the timer interrupts were a higher priority than the serial interrupts. Worse, the Amiga’s serial chip only had a 1-byte buffer — which means that if you didn’t pick up the data before the next byte arrived, the data would be lost.
Bryce was able to mitigate the problem somewhat by having AmigaOS turn off timer chips that weren’t in use. Unfortunately, they couldn’t all be turned off — AmigaOS needed one timer, and Music-X needed another. Turning off two of the four timers greatly reduced the frequency of the bug, but didn’t eliminate it entirely. And by this point the reputation of the Amiga had been stained beyond redemption, at least in professional music circles. It was too late.

Quote taken from: https://dreamertalin.medium.com/music-x-b4abc68d6f78

So, the Atari ST even with crappier sound chip and not multitasking Operating System, turned out to be better for MIDI than the multimedia machine Amiga.

[Predseda]

Very interesting reading, drHirudo. Another part of the article that is sad, is this one:

By 1991 it was clear that the Amiga was a dying platform, and that if you wanted a large customer base you were going to have to target MS-DOS as your primary platform.

Actually I personally had no clue it is sinking until 1994 official bankrupcy announcement.