Jump to content
Keatah

The reason the Amiga failed.

Recommended Posts

Perhaps the Amiga was too complex for its own good. It's always been the simple and generic which succeeds, PCs are no exception. There was a whole different feel to a word processor on the IBM as opposed to the Amiga. It somehow seemed typing a character was hard, crude, and simple. Down to the bare metal. Keyboard, Ram, CPU, and display registers. No complex intervening, black magic in custom chips slowing down the works.

 

It was a simple box, much like the Apple II and CP/M rigs. Just enough to do the job, but do it well. Believe me I wasn't interested in multi-media word processing. Nor was running demos and watching animations while conducting word processing an important thing.

 

Even the 1st MAC was simple and shared far more with the PC and Apple II. It had no graphics chip, essentially, all the functionality came from software routines. Routines that would easily scale in speed as the platform evolved. Something not possible with fixed-function custom chips. Ironic, the very thing that gave the Amiga some wows also prevented it from growing.

 

Besides not doing what I wanted, the Amiga was hard to upgrade.

 

Expanding my 486 was simple, call gateway and ask them to send me out an 8MB card. Drive to Computer City or Comp-USA to get a 2nd parallel port or sound or graphics card, even hard disks. The stuff was proliferating everywhere. And those 6-inch thick ComputerShopper phone book catalogs? They were the department store wishbooks of the 90's!

Share this post


Link to post
Share on other sites

I looked up the date all those things were introduced, feel free to do the same.

I'm well aware of the dates they were introduced since I sold, spec'd, and supported them. And I'm very aware of what the majority of PC users purchased and how good (or bad) the software support for those features was. The truth is that it wasn't until the release of Windows 95 that the PC really shot ahead of the Amiga. Prior to that it was only in certain areas (like productivity, flight simulators, raw data processing, first person shooters, ...). As for smooth scrolling graphics, multi-channel audio sample playback and manipulation, serial transfer speeds, multi-tasking, video playback, memory management, and jumperless plug-and-play technology, the PC was lagging behind the Amiga.

 

Again going back to audio, the Sound Blaster 1.0, 1.5, and 2.0 all were 1-channel 8-bit audio. The audio software available for the PC at the time was garbage. The DACs on the Sound Blaster 16 were so noisy that you could get far superior results sampling on an Amiga 3000 using a DSS8 interface. Then there's the accelerated graphics of the ATI Mach 8. That one was pure garbage (one of ATI's own engineers even flat out admitted to me that it was an inefficient hardware accelerator compared to the Amiga's graphics hardware). Even the Mach 32 was underwhelming. Yes, I know the 90s PCs very very well. They were far from elegant, suffered from numerous hardware and software bugs, were mired in third-party support issues, plagued with driver problems, and varied dramatically in build quality.

Share this post


Link to post
Share on other sites

..and whole industries were built around enhancing and fixing up the PC. This generated even more exposure and pervasiveness as a by-product. Seems to me that the Amiga excelled in areas of insignificance (at the time).

 

To upgrade the amiga one typically had to go through specialty shops. But the PC had the equivalent of real full-size supermarkets going.

 

I'd gladly accept a noisy DAC in exchange for having all those "warehouses" and "support". I'd gladly tolerate some stuttering in Windows 3.1 in exchange for being able to use my old Epson MX-80 straight out of the box with it. I'm happy to have Flight Simulator and FPS as long as they were nicely done.

 

IDK, to me, there was too much abstraction going on in the amga for it to have a clear way forward.

Edited by Keatah

Share this post


Link to post
Share on other sites

..and whole industries were built around enhancing and fixing up the PC. This generated even more exposure and pervasiveness as a by-product. Seems to me that the Amiga excelled in areas of insignificance (at the time).

 

To upgrade the amiga one typically had to go through specialty shops. But the PC had the equivalent of real full-size supermarkets going.

 

I'd gladly accept a noisy DAC in exchange for having all those "warehouses" and "support". I'd gladly tolerate some stuttering in Windows 3.1 in exchange for being able to use my old Epson MX-80 straight out of the box with it. I'm happy to have Flight Simulator and FPS as long as they were nicely done.

 

IDK, to me, there was too much abstraction going on in the amga for it to have a clear way forward.

Heh. Yes. The PC's quirks certainly gave me job security (and paid for my university education). So I do owe it that at the very least.

 

As a musician, I'm not sure I could settle for the noisy DACs. However, I have to admit that the sheer variety of add-ons for the PC was amazing. I was just thinking about comments regarding the Amiga hardware falling behind. And then I spotted an ad that reminded me that people could easily go out and buy a really nice hardware-accelerated 24-bit graphics card for their Amiga. Problem was, it would only really be useful for a few apps. As you mentioned, there was the issue of both economies of scale and lack of wider software support for the Amiga hardware enhancements (and only a handful of places nearby that could repair an Amiga).

 

Now, having said that, let's not forget that the PC's real ace was clones. If it wasn't for the clone manufacturers making PCs cheaper and abundant (while making third-party manufacturer's lives much easier), who knows where the very expensive IBM PC would have ended up. It's ironic that the thing that IBM hated the most (clones) ended up propelling that platform to the top.

Share this post


Link to post
Share on other sites

Yawn.... ofcourse it was. , But the Amiga 500 was there in 1987.... the 1000 in 1985... a suitable 486/33MHz around 1992, using ISA Bus for communications and Datatransfer to the Soundcard and the VGA.... You had to overclock the ISA Bus, if you wanted to have a "fluid" animated Video in WC2. With the change to VLB graphics one had something like "speed" on the PC. A 386 was in now way able to show Wolf3d fluidly playable. You had to use a very small windows for that!

AMIGA was able to do multitasking for real, and you don't need several CPUs of the same kind, to have multiprocessing, btw. ... and there were no bottlenecks like ISA Bus, or memory limitations as on the PC.

 

So my question again: How would you name what Amiga did in advance to - particular - the Atari ST.

 

One hint: The Copper is inbetween a DSP and a CPU. in both cases, the "P" stands for "Process...." ...

 

Printing a document, writing a document, and a demo is running in a window.... fluently... the same experience of an ECS Amiga, you got after the Appearance of Windows 2000 ...

The Copper as between a DSP and CPU in capability?

I've worked with DSPs and the copper is much more narrow in what it can be used for.

Try performing a fast furrier transform on the Copper and let me know how it works out for you.

You could run a computer entirely with a DSP alone if it were designed for it. Good luck doing that with the Copper.

DSPs can seem kind of weird to program, but they are a form of microprocessor, not a coprocessor.

The Copper has a very specific number of functions. It is strictly a coprocessor.

 

The Amiga hardware definitely helped with the speed of graphics and sound, but much of the fluid nature of the Amiga was due to the OS, not the hardware.

Try multi-tasking on a 6809 under OS-9, it was pretty fluid without coprocessors and was available years before the Amiga was released.

 

Windows used cooperative multitasking, and was 16 bit through Windows 3.1.

It wasn't until Windows 95 that Windows switched to preemptive multitasking, protected mode, and 32 bit.

The early Mac OS and GEM were also cooperative multitasking.

 

The 486 was introduced in 1989.

They were 20MHz to 33MHz at first, but the 486DX2 was out in 1992 and clocked at up to 66MHz. The first DX2s were probably 40MHz to 50MHz but 66 MHz wasn't far behind.

The 40MHz 68040 benchmarks much faster than an equally clocked 486, but Amigas shipped with a 68030 until 1992, and those are slower than the 486 by something like 20%.

The 25MHz 68040 based 4000 came out in 1992 and it was slower than 486DX2 machines by that time.

Then the Pentium came out in 1993 making the gap even wider.

 

High color came out on SVGA cards in the 80s.

The S3 911 graphics card had hardware accelerated 2D graphics and high color in 1991.

The higher speed VESA local buss (up to 50MHz) came out in 1992 giving PCs significantly higher bandwidth to graphics RAM.

Share this post


Link to post
Share on other sites

...

Again going back to audio, the Sound Blaster 1.0, 1.5, and 2.0 all were 1-channel 8-bit audio. The audio software available for the PC at the time was garbage. The DACs on the Sound Blaster 16 were so noisy that you could get far superior results sampling on an Amiga 3000 using a DSS8 interface. Then there's the accelerated graphics of the ATI Mach 8. That one was pure garbage (one of ATI's own engineers even flat out admitted to me that it was an inefficient hardware accelerator compared to the Amiga's graphics hardware). Even the Mach 32 was underwhelming. Yes, I know the 90s PCs very very well. They were far from elegant, suffered from numerous hardware and software bugs, were mired in third-party support issues, plagued with driver problems, and varied dramatically in build quality.

Until AC97 there wasn't really a widely accepted standard that produced decent sound on a PC.

 

I worked on an update to a program for radio stations in the early 90s. The guy that hired me flaked, so I stopped work on it before getting very far.

I think the card had 12 bit audio but I'm not certain what it was. Adlib Gold maybe? It was 1992 or 93 so that's about the right time.

It definitely had better sound than the Soundblasters.

 

Share this post


Link to post
Share on other sites

Don't neglect the fact that Commodore built the PET/CBM computers years before the VIC-20. While the PET may not have been a market leader on business computers in the US, where I understand Apple ][, TRS-80 and other CP/M systems had significant shares before the IBM PC, the PET line was very strong in some European countries and used for business well into the 1980's.

 

According to the book, "Commodore: A Company on the Edge", by Brian Bagnall, the Profit Margins on the PET in Europe were much higher than in the US and Canada, so most ALL of the Production went there..

 

So yes, I think Jack Tramiel knew what to do with a "real" personal/business computer. After all, that is where the company came from before testing the waters with home computers.

"Computers for the Masses, not the Classes".... Jack Tramiel

 

 

MarkO

Share this post


Link to post
Share on other sites

The Amiga hardware definitely helped with the speed of graphics and sound, but much of the fluid nature of the Amiga was due to the OS, not the hardware.

Try multi-tasking on a 6809 under OS-9, it was pretty fluid without coprocessors and was available years before the Amiga was released.

I wonder if you know what we're talking about. Even ONE of the "3" Tasks, you could have run on the Amiga at the same time, would pull ALL CPU time on an ST. Multitasking cannot add CPU Speed , and a single Task is always faster than several tasks , using one CPU, do we get that in common?

 

To make it more clear: I'm still at OCS/ECS AMIGA at ~7.2MHz and an 8MHz ST.

Edited by emkay

Share this post


Link to post
Share on other sites

I wonder if you know what we're talking about. Even ONE of the "3" Tasks, you could have run on the Amiga at the same time, would pull ALL CPU time on an ST. Multitasking cannot add CPU Speed , and a single Task is always faster than several tasks , using one CPU, do we get that in common?

 

To make it more clear: I'm still at OCS/ECS AMIGA at ~7.2MHz and an 8MHz ST.

 

I used to compile on my Amiga while continuing to edit code all the time. But then I had extra RAM, a hard drive, and sometimes I'd lower the task priority of the compiler so I'd never even notice to compiler was running if it weren't for the flashing LED on the hard drive.

The compiler got every clock cycle the editor didn't need and the editor went to sleep until wakened by an event like a keypress, mouseclick, etc...

With ny 33MHz 68040 A3000 I could compile, download and edit at the same time without even changing task priorities.

It has little to do with the custom chip speeding up the system since none of those tasks required more than occasionally updating some text on the screen.

 

Don't get me wrong, on an 8MHz machine running 3 tasks that want all the CPU time, all 3 will slow down.

But most software other than games stop to wait for user input, and don't need to keep eating CPU time. That frees up time for another tasks.

Timeslicing also means your program may run slower, but it's still running.

With cooperative multitasking, a program can hog every clock cycle and there's nothing you can do about it.

Share this post


Link to post
Share on other sites

Here's an interesting video on the making of the Amiga. The breadboards were just amazing and how they were developing things without having the hardware ready. Later in the video they even had an emulator showing Lotus 123.

 

The Amiga's lack of acceptance in the business world as a serious machine I believe was a huge detriment. Nebulon mentioned the stigma of being a game machine was definitely something that had a negative impact on success with the business market. Word Perfect eventually made its way to the Amiga but to late. The Amiga had emulators for PC software in both software and hardware forms for ages however they did not prevail. It was a time when the Mac, ST, and Amiga had chances to change the course of history. The Mac had the benefit of having MS write software for it and of course the desktop publishing revolution which were huge for business. The ST and Amiga were mainly viewed as gaming machines even though each had their own niche markets... ST - MIDI / music, Amiga 3d and Video - Lightwave and Video Toaster(real time video switcher)-Newtek amongst other programs. http://www.amigareport.com/ar134/p1-12.html has a nice list of innovative uses of the machine that only started to appear in the mid 90s on a PC in terms of multimedia that existed since the late 80s early 90s on the Amiga. I recall when time magazine declared multimedia has arrived in when Win 95 showed up when in reality it was around for quite sometime. I'd blame lack of effective marketing ,dealerships and acceptance in the US market for both the Amiga and ST while in Europe they did very well. This also was a time where IBM PCs were king in the business world and if it wasn't a PC there was this stereotype against the 68000 based systems even though they ran circles around PCs. I used to show folks who worked in the government or consultants at its capabilities and would only here "where is the PC software" even though a hardware version was available. They were blown away by the multimedia aspect but viewed it as a toy and wanted their monochrome business apps even though they were options still would not budge. So effective marketing was a big problem and saying no to something from Big Blue was just unheard of. Computer magazines at the time for the Amiga were decent sized however compared to the PC side were tiny in comparison. Computer Shopper was like a decent sized phone book in comparison. It would be interesting to see how different things would have been if the Amiga, ST, or Mac were viewed more as serious machines than they were at the time. All were ahead of the PC for ages with the Mac having the largest support in order to succeed in the larger markets. At least that is my take on it.

 

Video was pulled by some clowns in Italy. WOW

post-3709-0-09500400-1456374065_thumb.jpg

Share this post


Link to post
Share on other sites

 

In the beginning and at around 20 minutes they show some of the boards. The CC guys could learn a thing or two!

Share this post


Link to post
Share on other sites

The comments in where Dale Luck used a beep to tell him when something was done made me laugh.
I used a beep to tell me compiling was done when I kicked something off at night.
I stopped doing that once I had a hard drive so I had forgotten about it.

Share this post


Link to post
Share on other sites

Here's an interesting video on the making of the Amiga. The breadboards were just amazing and how they were developing things without having the hardware ready. Later in the video they even had an emulator showing Lotus 123.

...

The Amiga Transformer.

If I remember right, it didn't work on later versions of the OS and it still had a couple bugs.

Too bad they didn't update it and just give it away with the machine... not that we didn't give it out to some people anyway.

 

My first thought now would be that they should have made it open source, but the concept of open source didn't even exist yet.

 

Share this post


Link to post
Share on other sites

Don't get me wrong, on an 8MHz machine running 3 tasks that want all the CPU time, all 3 will slow down.

But most software other than games stop to wait for user input, and don't need to keep eating CPU time. That frees up time for another tasks.

And, ofcourse. The sheer CPU "power" only works at one Task on a single "Task" computer. What I'm writing about is that you can have a full demo running in one window and without a glitch and write and print documents on the Workbench. When I bought my Amiga 2000 with and ICD Card (4MB RAM and an SCSI HD ) , it blew everything away , called PC . HD installed Games were marvellous, as you never had to think of turning any floppy. Just click an icon and play "immediately" . Working was such a bless... but, well the Interlace reduced the working resolution to "640x200" while "640x400" had to be standard. Also, after half a year , no new "Workbench" software was working on the "new" Amiga. I watched that for some year , and read about compatibility problems using Kick 2.x. Which made me change the view of things and buing a 486/33 , changing from Amiga to PC.

 

To compare it: It's like on a 100m Dash, Commodore had the starting position at 99m and got overrun by the PC.

Share this post


Link to post
Share on other sites

And, ofcourse. The sheer CPU "power" only works at one Task on a single "Task" computer. What I'm writing about is that you can have a full demo running in one window and without a glitch and write and print documents on the Workbench. When I bought my Amiga 2000 with and ICD Card (4MB RAM and an SCSI HD ) , it blew everything away , called PC . HD installed Games were marvellous, as you never had to think of turning any floppy. Just click an icon and play "immediately" . Working was such a bless... but, well the Interlace reduced the working resolution to "640x200" while "640x400" had to be standard. Also, after half a year , no new "Workbench" software was working on the "new" Amiga. I watched that for some year , and read about compatibility problems using Kick 2.x. Which made me change the view of things and buing a 486/33 , changing from Amiga to PC.

 

To compare it: It's like on a 100m Dash, Commodore had the starting position at 99m and got overrun by the PC.

And that last part is pretty much what I was talking about.

Commodore killed pretty much any intermediate project that would have kept it competitive until they moved to the PA RISC architecture and new chipset.

 

The thing is, Amiga wasn't getting shut out until about 1992. The first signs were SVGA, but it was nowhere near a knockout punch.

There was no standard API yet.

The first hardware acceleration for graphics required different drivers.

But once you have 486 processors and VESA local buss... it really didn't matter so much. Everything ran faster even without a custom driver.

When Windows 95 was introduced, the biggest advantage the Amiga had, the Amiga OS, was suddenly behind. I suddenly heard "Virtual Memory" all the time.

If AGA had hit with the 3000 in 1990, it would have at least been a partial answer to SVGA until the VESA local buss hit.

If the 3000+ with DSP had come out a year later, it would have required special software but it would have still been the fastest for AV work by far.

Lightwave 3D would have had a huge speed boost once the software was updated.

If Multi-processing hits by 1992... a couple 68030s would be faster than a 486 and a couple 68040s would stomp a 486DX2 or first Pentium for any software that required a lot of cpu calculations. Maybe make the switch to virtual memory along with multi-cpu changes to the OS.

2 or more 68060s keeps the Amiga competitive with anything processing wise until the end of the 90s.

I'm afraid a graphics chipset upgrade would have had to hit between '93 and '95 to stay competitive though.

 

Share this post


Link to post
Share on other sites

I'm afraid a graphics chipset upgrade would have had to hit between '93 and '95 to stay competitive though.

No fears there, as the "basic" AMIGA was lightyears ahead of every PC ... But they had to do their task... and didn't.

Some things would be that easy : If a double clock (32MHz) VGA is detected, just put the chips into the double clocking with some adjustments. Back in that time, it really wasn't .... technically ... a problem. And As I wrote above: The AGA Chipset and the Kick OS suffered in the most consequential part: Compatibility to prior Computers! . People lost the trust in the computer line development at all.

Really: The PC only won by the fact that you could use "your old software" even after 2 or 3 updates on the Hardware and for several years. So people usually bought new Software , when "new Software" was needed, for own purposes, not just because a new computer had to be there.

Today an Intel based PC with Windows running on it, is the best solution. They made SO MUCH RIGHT, no other competitor even can scratch their socket of the surface.

Edited by emkay

Share this post


Link to post
Share on other sites

Im a bit late to the party and will likely be repeating what some have said but did not feel like reading all 10 or so pages.

 

Commodore was poorly managed in the US and Canada. Their customer service was awful and they did not market the Amiga very well. In Europe in the early 90s the Amiga out sold the PC nearly two to one and was number two next to the Mac in sales and to this day has a very strong following. When the A4000 was released in North America (I still have one) it sold thousands particularly to Television studios looking for a cost effective production switcher and editor using the Video Toaster and Flyer.

The AGA graphics was second to none at the time and the Amigas ability to do true multitasking gave it alot of power even though it only had 16mb or ram and 2mb chip.

Bottom line is if Commodore had the sense to market and support the Amiga it would have survived.

Edited by Tonyvdb

Share this post


Link to post
Share on other sites

The US was over 90% through the end of the 90s. There was a thread discussing this.
I think Amiga briefly outsold the Mac, but it never outsold PCs market wide.​
PCs were something like 2/3 of the market in the 90s.

Share this post


Link to post
Share on other sites

I didn't think the Amiga ever outsold the MAC. But considering I had a 1000, and later a 500.. Maybe..?

 

I bet I still would have gotten into the PC with a 486 machine at about same time I dumped the Amiga. But.. BUT!! Had I had reasonable expectations and understanding of the machine I would have kept it alongside parallel to the 486 - effectively using 2 computers.

 

Damnn snotbag salespeople. Lying bastards for all I care..

Share this post


Link to post
Share on other sites

The US was over 90% through the end of the 90s. There was a thread discussing this.

I think Amiga briefly outsold the Mac, but it never outsold PCs market wide.​

PCs were something like 2/3 of the market in the 90s.

That was supposed to say the US was over 90% of the market through the end of the 90s.

Share this post


Link to post
Share on other sites

Commodore put out some decent stuff but the A600 was probably their biggest POS.

 

It was supposed to be cheaper than the A500 but ended up costing the consumer more whilst it offered them less. They intended to aim the sales of the machine at the lower consumer end but some genius decided to go ahead and market it as a replacement for the A500 when they realized it would cost more to produce.

 

So they go ahead and cancel the A500; which was still their number #1 selling machine at the time.

 

Users were duped into upgrading ( or downgrading, which ever way you want look at it ) from an A500 to an A600. Commodore would then release the A1200 only a few months later & would stop support for ECS machines shortly after.

Share this post


Link to post
Share on other sites

I think public perception of the Amiga's lead over it's competition was as important as the hardware itself, which is part of the reason I think they needed to release higher end systems.

But even certain smaller upgrades to the Amiga line would have made Amiga look like it was continuing to be a leader.

 

Hard drives were a big deal to a lot of people in the US by the late 80s.

Commodore should have introduced the built in IDE interface sooner with updated versions of the 500 and 2000.
Maybe add it during ECS development. Still offer upgrade chips for existing systems but intro updated systems a few months down the road.

IDE came out in 1986 so that could have been possible even in the 80s, and the parts shouldn't have added to the retail price of the systems.
That would have made hard drive systems much more cost effective since SCSI interfaces and drives were expensive..

A lot of our customers that bought PCs may have gone with the Amiga just because of that.

If Amiga moved to a 14MHz 68000 CPU on the 2000, it adds more bang for the buck even if the higher speed isn't available when accessing CHIPRAM.
This would only add a few dollars to the cost of the base machine since the difference in CPU price is minor and the faster RAM is on a separate card.
Since the OS ROMs are on the FASTRAM buss, it's still going to make the machine feel faster even without a 14MHz RAM upgrade.
It might even make sense to add the speed upgrade to the 500 if it's a cheap upgrade.

I think those two upgrades alone would have made a noticeable difference in sales.
The development time and costs for those upgrades should also be less than many other hardware projects Amiga engineers worked on, and it's a nice intermediate step between ECS and later machines.

 

Commodore still has to follow through with higher end machines but it makes it look more like Amiga is being regularly upgraded.

If they had even added a 14MHz 68000 to the 600 it would have actually made more sense to release a new OCS/ECS based machine.
If the CDTV had come with a 2x CD-ROM drive, IDE hard drive interface, and 14MHz CPU, it's a much more impressive multi-media platform.

Edited by JamesD

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...