Things the Amiga didn't get right from Day 1

[Overdoc]

C64 and C16:
- RF
- CVBS (standard colour video signal)
- S-Video (not exactly, because of slightly different chroma level, but still compatible)

Amiga 500 with A520 or Amiga 600/1200/CD32 directly:
- RF
- CVBS (standard colour video signal)
- RGB (without A520 on the A500)

So to be able too hook up a C64 to a monitor it does not necessarily need an S-Video (=separate chroma and luma cinch) input. A composite colour (CVBS) input together with an audio input is sufficient and the picture quality is only slightly weaker than via s-video.

A-600 also has this composite colour (CVBS) output, so you probably used your Taxan monitor via that CVBS input, but it could also be possible that this monitor was also RGB compatible, and you connected the Amiga via RGB, and the C64 via CVBS?

Never heard of this monitor, tough.
Here in Austria almost everyone had either C= 1084-S or Philips CM8802/8833/8833-II monitors for Amiga, and C= 1701/1702 or 1802 for C64 (but most C64 users had TV only)

[pandy71]

Originally Posted by gimbal

Interesting typo, that could either be 4 bits or 4 bytes

true, thnx, fixed

[Gorf]

Originally Posted by grond

EDIT2: and now I remember that it wasn't "Tascam" but "Taxan Vision PAL"

http://www.timlybarger.com/blog/fbroom050_monitor.JPG

sadly there is no picture of the backside

[grond]

Yes, that's the one. One of the switches on the front switches between colour, amber and green display modes. I actually liked the amber mode for programming because it was easier on the eyes.

[pandy71]

Originally Posted by Thomas Richter

Which, again, does not matter. Think about it. You just create a different color gamut. What you miss if you clip in YUV space to a smaller range (8 bit instead of 9 bit) are extreme colors, namely those where G-B overshoots, or G-R overshoots. Everything else is preserved. YUV works for "natural colors", not for extreme colors. They don't appear in natural scenes.

Nope, process you describe assume RGB source at the input but in Amiga natively working in YUV space there is no RGB source unless you explicitly working in RGB space then converting RGB colors to YUV then feed registers with converted value and later decode such values back to RGB in analog domain...

And you constantly ignore technology available in first half of 80's - there is no 8 bit, only 4 bits are available and when compared to previous technology those 4 bits per component are considered as state of the art especially in personal/home computer market.

Today with 20nm and less we may have plenty of bits and multipliers working with few GHz speed but MOS Technology/CSG use process somewhere between 5 ?m and 1.5 ?m, also wafer size was smaller - 5 inch not 12 like currently.

[Thomas Richter]

Originally Posted by pandy71

Nope, process you describe assume RGB source at the input but in Amiga natively working in YUV space there is no RGB source

It doesn't matter. Think about this. The transformation is reversible on the full domain. Hence, the points where it is not invertible on the restricted input are exactly those points in RGB space that lead to points in YUV gamut that are *not* in the restricted domain. And those are the extreme colors such that B-G and R-G are smaller than -128 or larger than 127. This is a simple math exercise. Really.

Originally Posted by pandy71

And you constantly ignore technology available in first half of 80's - there is no 8 bit, only 4 bits are available

And your point is? The transformation works in 4 bit as in 8 bit.

[Gorf]

Originally Posted by Bruce Abbott

It makes sense to produce signals in the format used by the highest bandwidth display device (ie. RGB monitor), then other (lower bandwidth) formats are not compromised by the quality of the source.

The A1000 used an MC1377, which has RGB to YUV conversion built in. This could be done relatively cheaply because the YUV signals only had to exceed the low bandwidth of its internal chroma modulator (~250kHz chroma bandwidth and 4.5MHz luma bandwidth).

Going the other way is not the same. To maintain quality going from YUV to RGB the circuit must have a much higher bandwidth than is required for composite output. This alone makes it more expensive.

You would have all the bandwidth you want, because YUV would be the native output of Denise.
And so say it with your own words, just a couple of pages back:

"Converting from RGB to YUV is relatively easy to do without significant loss of 'quality'. It just needs some precision resistors and video buffer amps with sufficient bandwidth. "

The same is true of course for YUV to RGB.
So so would have a high quality RGB output even if the native format would be YUV.
And you would have a much more useful HAM mode and better EHB modes...

And the luma signal for S-Video and composite would be great since it is just the Y signal....

The other problem was the lack of demand meant no 'cheap' off-the-shelf chips were available to do the job (plenty were available that did YUV to RGB conversion, but they were optimized for use in TVs where the source was low bandwidth composite).

maybe you would combine two of these chips and mix signals after transformation to compensate for the lower chroma bandwidth (quick&dirty approach) or you find a chip that has sufficient bandwidth or you modify the "Vidiot" accordingly ...
That problem is not really a showstopper.

It also made sense for the 'standard' format to be a simple one that was easy to implement accurately. YUV is not a simple format. It has different color spaces depending on the application (NTSC, PAL, YPbPr etc.). With RGB there is no argument over whether the signal is compatible with your equipment. With YUV it's which 'YUV'?

The one you decide to implement.
As Thomas pointed out earlier: it does simple not matter.

Two different EHB modes now? Already your idea is suffering from feature creep. But this is typical.

No this is just taking advantage of something that's already there.
There is a huge difference between "feature creep" and "not letting go to waste".

Of course Amber was 'agnostic', but it was not responsible for producing the video output. A circuit was still required to convert the digital bits to analog.
If the data was stored in YUV format then that circuit would have the added burden of converting high bandwidth YUV to RGB.

Which would still be our modified Vidiot

No, it was what kept them going.

Really: were is CBM now?

If the Amiga had been a similar price to a PC then few people would have bought it. We know this because few people bought the more expensive models.

False conclusion.

People did happily (and strangely) spent a lot of money for expensive PCs or even Macs that where less capable ...
So the problem was not that people in general could not afford to buy higher priced computers ... they could and they overwhelmingly did.

The problem with higher priced Amiga models were the missing features:
the A1000 did not provide anything (in practice) over the A500 at all
and the A2000 war ridiculously overpriced version of more or less the same model in an ugly case. It was not faster, the chips were not more capable ... the only thing where the expansion slots ... something the A500 could been modified to do as well.

The original "Checkmate" proves that Commodore could have sold a lot more A2000 at a reasonable price.
(the A1500 came too late and only in UK)

What really killed Commodore was squandering development resources trying to chase the high end (with the A3000 etc.).

The A3000 could have saved Commodore, if only Ali would have taken the deal with SUN.

The market segment for "workstations" was even faster growing than the PC market at this particular point in time ....

The A3000 was suffering from one particular flaw: still too low resolution for a workstation. Ataris counter-project the TT showed that a 64bit gfx chip (and RAM bus) was possible at this point!
That would have given the A3000 a 1280*800 resolution at 4 colors (even more pixels as the grayscale Nextstation).

So the A3000 approach was no mistake ... Commodore just did not spend enough resources on it and did not go far enough.

The video production industry was well used to paying high prices for specialist gear. For them the Amiga was an incredible bargain even without YUV (which wasn't used much). For the rest of us, having YUV instead of RGB would have just been another compatibility issue (like those parallel and serial ports on the A1000 that had the wrong gender - what were they thinking?).

Why "instead"??
We got rid of "instead" ten pages ago ...

[pandy71]

Originally Posted by Thomas Richter

It doesn't matter. Think about this. The transformation is reversible on the full domain. Hence, the points where it is not invertible on the restricted input are exactly those points in RGB space that lead to points in YUV gamut that are *not* in the restricted domain. And those are the extreme colors such that B-G and R-G are smaller than -128 or larger than 127. This is a simple math exercise. Really. And your point is? The transformation works in 4 bit as in 8 bit.

It does matter - as it is better to have 4096 colors you can reproduce on display than to have around 300 colors you can use.

Transformation works same (in 8 bit or in 4 bit) but 2 bits loss in 8 bit space is less painful than loss of 2 bits in 4 bit space...

Only limited number of colors from YUV space can be used in RGB space and this is simple math exercise. It is something else to loose 100k colors from 16.7M or loose 3700 colors from 4096.

[Thomas Richter]

You loose extreme colors, that's what I'm saying all the time. It's a smaller color gamut. If you loose two bits, that 1024 colors that remain, not 300.

[Bruce Abbott]

Originally Posted by pandy71

Please allow me to address your concerns by quoting Jay Miner (excuse for poor OCR from Byte 1985/11 - not my work) - 36 years ago Jay says:

Awesome quote there. Jay miner explains the quite sensible reasons they dropped YIQ in favor of RGB.

Originally Posted by Jay Miner

We found a Motorola chip that did a
good job of converting RGB into NTSC.
We needed the extra room on the Denise
chip for extra resolution on the color
registers. so we dropped the YIQ
NTSC completely. But we've still stuck
with the 4. 4. 4 bits. Also. you've got
a real pin limitation on a chip like this.
We tried to keep the chip simple and
low-cost to manufacture. and on-chip
ladders take up a lot of area. They're
notoriously inaccurate. and you can
buy I percent resistors external for a
penny apiece.

Kudos to Jay Miner for doing it right! This should put the matter to rest - but of course it won't because Amiga fans are never satisfied. Just imagine if they had stuck with YIQ and internal resistor ladders - poor accuracy, noise, low bandwidth, and a format that needs expensive analog conversion to work with RGB - the howls from Amiga fans would be deafening!

[TEG]

Originally Posted by Bruce Abbott

That would be the Timex Computer 2068, a later model produced by Timex of Portugal. This had analog RGB output from the ULA feeding an MC1377 (same chip used in the A1000 and A520) to produce PAL composite. It also brought all the digital RGBI signals out to the edge connector, which is probably how your friend was able to get a sharp display with SCART.

The TC2068 fixed some ZX Spectrum hardware incompatibilities, and was bundled with an 'emulator' cartridge (modified system ROM) that allowed it to run 97% of ZX Spectrum games. These machines seem to be very rare today (or at least aren't for sale on eBay).

The emulator cartridge, yes! I remember now. It was always plugged into the computer of course so I forgot about it.

Thanks for solving the mystery behind the Timex Scart connexion. Of course, being an electronic engineer means he had access to component documentation and so where to get the signals. At the time I did not realize there was such documentation for the insiders so I was scratching my head to imagine how it did that.

I can say this computer was looking very clean and more appealing than photos suggest.

[pandy71]

Originally Posted by Thomas Richter

You loose extreme colors, that's what I'm saying all the time. It's a smaller color gamut. If you loose two bits, that 1024 colors that remain, not 300.

Sorry my fault... indeed 1024 colors from 4096 makes difference but how many from those 1024 colors are valid in RGB space?

[pandy71]

Originally Posted by Bruce Abbott

Jay miner explains the quite sensible reasons they dropped YIQ in favor of RGB. Kudos to Jay Miner for doing it right! This should put the matter to rest - but of course it won't because Amiga fans are never satisfied. Just imagine if they had stuck with YIQ and internal resistor ladders - poor accuracy, noise, low bandwidth, and a format that needs expensive analog conversion to work with RGB - the howls from Amiga fans would be deafening!

Yes, YIQ was sinking ship and now there is no NTSC broadcast at all but RGB is still in use - decision to cease YIQ and put resources to the RGB was right decision.

I really don't understand people dreaming about superb composite instead superb component...

[Gorf]

Originally Posted by pandy71

Yes, YIQ was sinking ship and now there is no NTSC broadcast at all but RGB is still in use - decision to cease YIQ and put resources to the RGB was right decision.

I really don't understand people dreaming about superb composite instead superb component...

Again someone claiming "instead" ... why?
These things are not mutually exclusive.

[lmimmfn]

Lots of chat about HAM being so much better in YUV, ok so post comparisons? , should be easy to simulate with todays tech.

[Bruce Abbott]

Originally Posted by Gorf

So so would have a high quality RGB output even if the native format would be YUV.

But more expensive because you have to convert from one to the other with high bandwidth.

And you would have a much more useful HAM mode and better EHB modes...

Both HAM and EHB are of limited use. Both require 6 bitplanes which cuts CPU throughput in half, making them a poor choice for action games or rendering jobs. YUV might be better for photographs, but not if you want to mix 'true' color with palette indexed stuff. Optimizing the whole system for these two little-used modes is stupid.

And the luma signal for S-Video and composite would be great since it is just the Y signal....

That's what the A500 and A2000 had, easily derived from RGB with a few resistors. But with YUV you still need to convert digital to analog, then convert that to RGB. And it all has to be done with the highest quality, unlike going to composite.

maybe you would combine two of these chips and mix signals after transformation to compensate for the lower chroma bandwidth (quick&dirty approach) or you find a chip that has sufficient bandwidth or you modify the "Vidiot" accordingly ...

Lots of 'maybes' there...

No this is just taking advantage of something that's already there.
There is a huge difference between "feature creep" and "not letting go to waste".

You don't seem to understand the realities of custom chip design in 1985. Things you think are 'already there' might be quite difficult to achieve in the available silicon. The Amiga's custom chips were highly optimized to get the most out of limited resources - it wasn't a simple translation from discrete logic chips. Jay Miner said they had to drop YIQ because it used up too much silicon. I bet there was other stuff they thought of too that wouldn't fit, and there was no 'waste' to speak of.

Really: were is CBM now?

Where is IBM now? (they exited the PC market many years ago). Where are most of the 'big names' in the PC industry now? Where are all the other innovative home computer manufacturers? Nobody with any sense was seriously expecting Commodore to last >35 years in such a volatile market.

People did happily (and strangely) spent a lot of money for expensive PCs or even Macs that where less capable ...

Apple almost went bankrupt in 1995 due to lack of demand for their products. Now that that's out of the way...

PC's were more capable than the Amiga - not less. More capable of doing what customers willing to spend thousands of dollars wanted them to do. No.1 on the list was of course:- being IBM compatible. Simply being different was a huge fail for the Amiga.

So the problem was not that people in general could not afford to buy higher priced computers ... they could and they overwhelmingly did.

And yet they didn't buy higher priced Amigas - I wonder why? Somehow I don't think it's because Amigas didn't have YUV outputs.

The problem with higher priced Amiga models were the missing features:
the A1000 did not provide anything (in practice) over the A500 at all

Was it expected too? No, it was simply the 'pilot' model. But it did have a different form factor some people preferred, and the WCS allowed the use of different Kickstarts without changing system ROMs. If you didn't want that then the A500 was probably a better choice, and since it was a later model which superseded the A1000...

the A2000 was ridiculously overpriced version of more or less the same model in an ugly case. It was not faster, the chips were not more capable ... the only thing where the expansion slots ... something the A500 could been modified to do as well.

PC's also came in ugly cases, but people bought them in droves - often simply because a bigger box was more imposing. But the A2000 had a big case for a reason, to take all those plugin cards and extra internal drives that the A500 couldn't. Sure you could 'modify' an A500 to do the same job, but then it would effectively be an A2000! And probably about the same price.

In 1987 I bought an A1000, largely because I preferred its more 'professional' styling compared to the A500. Had I waited a few months I could have had an A2000 for about the same price. In 1991 I bought an A3000 for a ridiculous price, again eschewing the 'ugly' A2000. But I could have gotten more functionality out of the A2000 for much less money. The A3000 had a nicer styling, but was a pain to work on. It didn't have enough room to take the extra drives I wanted, the SCSI port was temperamental, the CPU slot was underneath the floppy drive tray and didn't have enough clearance to install a fan. Those ZIP RAM chips were a nightmare to install. The machine weighed a ton and the power supply fan was incredibly noisy (due to trying to suck too much air through a small space). Looking back, I should have bought the A2000.

The original "Checkmate" proves that Commodore could have sold a lot more A2000 at a reasonable price.
(the A1500 came too late and only in UK)

A1500 was just badge. In New Zealand A2000's were sold with 1 floppy drive and nothing else, but you could buy an extra floppy drive with it and have the exact same spec as the A1500 - except for the badge.

The A3000 could have saved Commodore, if only Ali would have taken the deal with SUN.

The market segment for "workstations" was even faster growing than the PC market at this particular point in time ....

You're dreaming. Where's SUN now?

The A3000 was suffering from one particular flaw: still too low resolution for a workstation. Ataris counter-project the TT showed that a 64bit gfx chip (and RAM bus) was possible at this point!

Where's Atari now?

That would have given the A3000 a 1280*800 resolution at 4 colors (even more pixels as the grayscale Nextstation).

Which would soon not be enough.

So the A3000 approach was no mistake ... Commodore just did not spend enough resources on it and did not go far enough.

The A3000 lasted a scant 2 years before being dumped at fire-sale prices.

Amiga fans couldn't afford it, while professional users didn't want it. But that didn't stop them continuing the 'A3000 approach'. The A3000+ was going to have even more expensive hardware in it, but they couldn't get it to work. How much money would it take to go far enough? More than the company had. But even if they could afford it the result would have been a money sink - and not a good reason to be in business.

Meanwhile cheaper 386SX PCs with crappy ISA VGA cards were selling like hotcakes - because they were IBM compatible. Computer companies came and went, but Taiwanese made motherboards and plugin cards kept the OEMs going. Anybody could buy the parts and call themselves a 'manufacturer', slap MSDOS on it and have a surefire winner - because it was IBM compatible. No expensive R&D to worry about, no cheapskate users complaining about the price (they were building their own PCs from parts).

Going head to head against the PC market with a computer that wasn't IBM compatible was never going to work. Commodore did eventually realize this when sales of the A500 took off, but they still wasted R&D on more expensive machines. They should have released the A1200 in 1990, the CD32 in 1992, their 'PlayStation' in 1994. I doubt they would still be here now if they did, but they might have lasted until the new millennium at least.

[Gorf]

Originally Posted by Bruce Abbott

That's what the A500 and A2000 had, easily derived from RGB with a few resistors.

No they did not since Denise had no direct Y signal.
It needed to be converted first - rather low quality with "a few resistors" as you say.
(and of course no color, and no S-Video out ....)

But with YUV you still need to convert digital to analog, then convert that to RGB.

that depends entirely on your conversion method

You don't seem to understand the realities of custom chip design in 1985.

You can see what other companies did at the same time.
The Amiga custom chip design is certainly clever but not really more complex in terms of transistor count and/or number of features..

But this is not really relevant here, since none of the changes I proposed would have increased complexity or transistor count by any significant means.

If you really want something complex - also done by a very small team at the same time:
https://en.wikipedia.org/wiki/Pixar_Image_Computer

PC's were more capable than the Amiga - not less.

Very debatable ... but there are other threads covering that...

Simply being different was a huge fail for the Amiga.

SUNs where different and survived at least the 90s
And so did Macs ... (barely)

Commodore Amiga could have among them ... things done right.

And yet they didn't buy higher priced Amigas - I wonder why? Somehow I don't think it's because Amigas didn't have YUV outputs.

No -as I mentioned much earlier here: it is the sum of all small things.
I tried to explain why people did not by the A2000 - it was CBMs own price policy and the lack of development (and design).

But the A2000 had a big case for a reason, to take all those plugin cards and extra internal drives that the A500 couldn't. Sure you could 'modify' an A500 to do the same job, but then it would effectively be an A2000! And probably about the same price.

I tried to explain it to you:
The problem was (as the Checkmate and others showed) that is was in fact cheaper to buy an A500, rip it apart, throw away the original case, but in in a new case with an expansion board ... and even provide a case for the keyboard, than buying a A2000

Hence the A2000 overpriced and did run at the same speed ...
And so nobody bought it.

And when the Commodore UK saw that small companies like the checkmate sold converted A500s in bigger boxes like crazy they finally realized they needed to drop the price dramatically - and that is what the as "A1500" relabeled A2000 did.
And it worked ... but by then it was of course already far too late.

[redblade]

Originally Posted by Bruce Abbott

In 1987 I bought an A1000, largely because I preferred its more 'professional' styling compared to the A500. Had I waited a few months I could have had an A2000 for about the same price. In 1991 I bought an A3000 for a ridiculous price, again eschewing the 'ugly' A2000. But I could have gotten more functionality out of the A2000 for much less money. The A3000 had a nicer styling, but was a pain to work on. It didn't have enough room to take the extra drives I wanted, the SCSI port was temperamental, the CPU slot was underneath the floppy drive tray and didn't have enough clearance to install a fan. Those ZIP RAM chips were a nightmare to install. The machine weighed a ton and the power supply fan was incredibly noisy (due to trying to suck too much air through a small space). Looking back, I should have bought the A2000.

The A3000 lasted a scant 2 years before being dumped at fire-sale prices.

Amiga fans couldn't afford it, while professional users didn't want it. But that didn't stop them continuing the 'A3000 approach'. The A3000+ was going to have even more expensive hardware in it, but they couldn't get it to work. How much money would it take to go far enough? More than the company had. But even if they could afford it the result would have been a money sink - and not a good reason to be in business.

Hi Bruce.

I'm interested what A3000 model your brought? Was it the 16mhz or 25mhz version?

Thanks

[Bruce Abbott]

Originally Posted by redblade

Hi Bruce.

I'm interested what A3000 model your brought? Was it the 16mhz or 25mhz version?

Thanks

25MHz.

[masteries]

Talking about A500 / A500 Plus / A600

I will add the lack of more than 2 MB of chip memory support.

This lead to an under use of the four 8 bit channels PCM audio capabilities of Paula; as well some other productivity software related to scientific or electronics I saw in the Atari ST/E family forget completely the Amiga computers.