Perhaps this is why the Sinclair computers were crushed by Commodore early in the UK?

[jotd]

All Bitmap Bros games were first coded on the ST, then ported to the Amiga.

On Gods they added copper skies, but overall it's a 16 color game (I can't imagine what it could have looked like with 32 or 256 colors!).

Also 32 colors were really slower, so it was another reason to choose 16 colors.

[TCD]

Originally Posted by jotd

All Bitmap Bros games were first coded on the ST, then ported to the Amiga.

Chaos Engine was developed on the Amiga first and then ported to ST

[jotd]

Ouch! I should know as I read the excellent book about the Bitmaps. They still used 16 colors for speed (and also because the ST version would have looked so washed up or they'd have to rework the palette completely).

But at least there's an AGA version (that some people hate, but I find it a good effort)

[Megalomaniac]

All other things equal, 16 colours would be faster than 32 on the Amiga, but Xenon 2, Gods and Magic Pockets aren't exactly fast are they? I know Gods is designed to be methodical and puzzle-oriented, but the other two arguably suffer for it.

[saimon69]

About Xenon 2 i remember sometime ago mcGeezer did a dissection of it

[desiv]

Originally Posted by Megalomaniac

... but Xenon 2, Gods and Magic Pockets aren't exactly fast are they? I know Gods is designed to be methodical and puzzle-oriented, but the other two arguably suffer for it.

Have to admit, as someone who isn't great at games, not being "exactly fast" has always been something I considered a plus! ;-)
That probably explains why I have a much more positive view of Xenon II than a lot of other people apparently do... ;-)

[TCD]

Originally Posted by desiv

Have to admit, as someone who isn't great at games, not being "exactly fast" has always been something I considered a plus! ;-)
That probably explains why I have a much more positive view of Xenon II than a lot of other people apparently do... ;-)

Hehe, I used to think that too There is a difference between a game that is intended to be slower and one that just runs poorly though. Not sure which one is the case for Xenon 2

[Bruce Abbott]

Originally Posted by TCD

Hehe, I used to think that too There is a difference between a game that is intended to be slower and one that just runs poorly though. Not sure which one is the case for Xenon 2

Being faster isn't always better. If the game doesn't need or want the speed then not being able to do it isn't a problem. The problem with Xenon 2 isn't the speed. It's just a boring game.

In the typical fashion of Amiga fans, games get trashed for 'only' running at 25 fps. But many of us find it hard to tell whether a typical game is running at 25 or 50 fps, which proves that it's usually not that big a deal.

However there are games that run 'poorly', like Sim City 2000 for example. The mouse pointer freezes while rendering the terrain. That's just bad coding, a fault that should have been corrected or the game canned for not meeting a reasonable standard.

[Megalomaniac]

I massively prefer the earlier Hybris to Xenon II, and it baffles me that most reviewers preferred Xenon II (the Bitmap Brothers' brash rockstar image contrasting with the anonymous US-based Scandinavians behind Hybris, and perhaps the apparent ST bias in the UK media at the time), however while I assume Hybris is 50fps and Xenon II is 25fps tops, I'm not sure that reversing that would change my opinion. If Xenon II was designed to be methodical, they did a poor job of marketing it that way, and never expressed any regret over that. It's not just the speed that puts me off Xenon II, I prefer Plutos too, and that has slowdown. Blood Money has much better design for a methodical slightly strategic shooter (and plays slightly better on the ST due to being a bit slower and thus less fearsomely hard, ironically).

Sim City 2000 was just poorly coded, apparently the Mac version runs faster through a Mac emulator than the Amiga version does natively. They probably just figured that people had nothing to compare it to that required such a high spec.

[TCD]

Originally Posted by Bruce Abbott

Being faster isn't always better.

I think that is what I said in my post.

[Megalomaniac]

The A3000 did well within its intended niche, a 7Mhz processor was impressive for 1987 but looking a bit sluggish for serious work compared to PCs by the early 90s. Maybe the A1500 should have had a faster processor, to make it a bit more of a worthwhile model compared to the previous A2000.

[Bruce Abbott]

Originally Posted by Megalomaniac

The A3000 did well within its intended niche, a 7Mhz processor was impressive for 1987 but looking a bit sluggish for serious work compared to PCs by the early 90s.

Perhaps you are forgetting that the A2500/020 (A2000 with A2620 accelerator card) was released in 1988? The A2500/030 was released in 1989, a year before the A3000. By 1991 (when the A3000 was actually shipping) there were several 3rd party 030 and 040 boards available for the A2000, some with on-board SCSI or IDE.

For those who already had an A2000 when the A3000 came out, it was cheaper to add an accelerator card. The Video Toaster didn't fit in the A3000 either, another niche it couldn't fill.

Maybe the A1500 should have had a faster processor, to make it a bit more of a worthwhile model compared to the previous A2000.

The A1500 was just an A2000 with 2 floppy drives. Stephen Jones liked to complain about Commodore using the A1500 to undercut his 'A1500' conversion box for the A500, but if true that just proves that the A2000 was better value.

But the real attraction of the A2000 was what you could put in it. Start with the base model, then add an accelarator card of your choice, a hard drive and controller, RAM, I/O ports, RTG etc. or even make your own cards for it. It was a system that grew with your budget and needs.

The A3000 had a lot more in it to start with, but also came with a much higher price tag that put it well outside the budget of the average Amiga fan. Furthermore expansion was limited by the cramped space inside it - you couldn't even add a 5.25" floppy or CDROM drive internally.

I bought an A3000 because I thought it would be a good replacement for my A1000, but looking back I probably should have bought an A2000 like my friends did - instead of choosing form over function.

[grond]

I guess the important aspect in upgrading the CPU from a 68000 was that this implied exchanging the 16bit architecture and bus width by a 32bit architecture. 32bit boards and subsystems were rather expensive in 1987-89. A 16 MHz 68000 wouldn't cut it either with 020 and 030 accelerarors available. It would have meant more money wasted for those who wanted to upgrade the CPU and thus had no use for the 68000. I think the A3000 in 1990 was the right point in time for a 32 bit machine but it should have made better use of the doubled bandwidth by having AGA. Also, the wedge computer to accompany the A3000 should have made the same step up.

[Megalomaniac]

Originally Posted by Bruce Abbott

Perhaps you are forgetting that the A2500/020 (A2000 with A2620 accelerator card) was released in 1988? The A2500/030 was released in 1989, a year before the A3000. By 1991 (when the A3000 was actually shipping) there were several 3rd party 030 and 040 boards available for the A2000, some with on-board SCSI or IDE.

For those who already had an A2000 when the A3000 came out, it was cheaper to add an accelerator card. The Video Toaster didn't fit in the A3000 either, another niche it couldn't fill.

The A1500 was just an A2000 with 2 floppy drives. Stephen Jones liked to complain about Commodore using the A1500 to undercut his 'A1500' conversion box for the A500, but if true that just proves that the A2000 was better value.

But the real attraction of the A2000 was what you could put in it. Start with the base model, then add an accelarator card of your choice, a hard drive and controller, RAM, I/O ports, RTG etc. or even make your own cards for it. It was a system that grew with your budget and needs.

The A3000 had a lot more in it to start with, but also came with a much higher price tag that put it well outside the budget of the average Amiga fan. Furthermore expansion was limited by the cramped space inside it - you couldn't even add a 5.25" floppy or CDROM drive internally.

I bought an A3000 because I thought it would be a good replacement for my A1000, but looking back I probably should have bought an A2000 like my friends did - instead of choosing form over function.

Especially for a business, having to buy and install upgrades inside an expensive computer before it reaches the potential you need from it would be a major issue - imagine if you get it wrong and destroy the machine. The A3000 had all of this out of the box, and for less than buying an A2000 (or A1500) and upgrading it. Plus it had Workbench 2 out of the box, not sure when (or if) that was available as an upgrade for the A2000.

[Bruce Abbott]

Originally Posted by Megalomaniac

Especially for a business, having to buy and install upgrades inside an expensive computer before it reaches the potential you need from it would be a major issue - imagine if you get it wrong and destroy the machine.

Yes. That's why people shunned the PC (which needed constant upgrading to stay relevant). It's also why you never saw anybody but highly trained electronics engineers build PCs out of parts.

In those days most business people weren't buying their computers from Woolworths. There were specialist computer stores which installed stuff for you and provided ongoing support (that's how they made most of their money). I was doing this in New Zealand in the 90's. At this time the big department stores started selling name-brand PCs such as Compaq, Acer and HP like they were TVs. But they had no technical staff and their after-sales support was abysmal. I picked up a lot of business from that oversight.

[Bruce Abbott]

Originally Posted by Megalomaniac

The A3000 had all of this out of the box

Actually it didn't. I only had my A3000 a short time when I realized it didn't have enough RAM and the hard drive was too small. I managed to squeeze an extra drive in there in there but it was tight, and getting the SCSI bus stable was tricky. Installing those ZIP chips was a Pain!

Later on I installed an 060 card, and that was even more painful. The card barely fitted under the drive shelf, and didn't have enough room to install even the lowest profile fan. I ended up cutting a hole in the drive shelf, which took ages because it was thick steel. Every now and then the machine would get flaky and I would have take it all apart to reseat the accelerator card.

Plus it had Workbench 2 out of the box, not sure when (or if) that was available as an upgrade for the A2000.

I ordered my A3000 when they were first announced, but didn't receive it until the 2nd quarter of 1991. It came with KS1.4 in ROM and softbooted KS2.0, which used up some of the precious RAM. Even this version wasn't totally stable.

[grond]

Originally Posted by Bruce Abbott

Actually it didn't. I only had my A3000 a short time when I realized it didn't have enough RAM and the hard drive was too small. I managed to squeeze an extra drive in there in there but it was tight, and getting the SCSI bus stable was tricky. Installing those ZIP chips was a Pain!

Later on I installed an 060 card, and that was even more painful. The card barely fitted under the drive shelf, and didn't have enough room to install even the lowest profile fan. I ended up cutting a hole in the drive shelf, which took ages because it was thick steel. Every now and then the machine would get flaky and I would have take it all apart to reseat the accelerator card.

But, you see, this made perfect sense. Commodore would have had to use a much more expensive case if they had made it bigger and thus the A3000 would have become even more expensive than it already was because every dollar they spend more costs the user at least 10 dollars extra. And all this for the few users who weren't happy with a 16 MHz 030 and the RAM the A3000 came with. And why wouldn't you be happy with a 16 MHz 030? It was several times faster than a 7 MHz 68000. PC envy?

Also, you could have removed the drive that already was in the A3000 rather than adding a second one. SCSI-drives were so cheap in 1998 that it didn't matter.

[Daedalus]

Originally Posted by Megalomaniac

Maybe the A1500 should have had a faster processor, to make it a bit more of a worthwhile model compared to the previous A2000.

There was also the A2500 released by Commodore, which was an A2000 with an accelerator already fitted. So a faster A2000 version was already available, though a model with a faster CPU is always going to be more expensive than a 68000-based one.

[Bruce Abbott]

Originally Posted by grond

you could have removed the drive that already was in the A3000 rather than adding a second one. SCSI-drives were so cheap in 1998 that it didn't matter.

This was 1991 not 1998. The Quantum 120LPS cost NZ$999 wholesale. I could have removed the original 40MB drive but I had no other use for it so better to have the extra capacity.

Commodore would have had to use a much more expensive case if they had made it bigger and thus the A3000 would have become even more expensive

I'm not so sure about that. The A3000 case was very heavy and the power supply fan made a lot of noise trying to keep it cool. I bet they were producing the A2000 case + power supply pretty cheaply, probably not much more than the A3000. And it was worth it for the extra room and quieter operation.

In 1990 the A3000 cost US$3999 for the base unit with 25MHz 030, 40MB hard drive and 2MB RAM (1MB Chip + 1MB Fast).

A GVP 28MHz 030/68882 board with 4MB of 32 bit FastRAM and 40MB hard drive for the A2000 sold for $2399. A 16MHz 030 board with 4MB RAM only cost $999.

The A2000 base unit cost less than $1500. If you already had one it was much cheaper to upgrade it than buy an A3000, but even the combo of A2000 + 28MHz 030/68882/4MB and 40MB hard drive was cheaper.

In 1992 the A4000 with 25MHz 040, 6MB RAM (2MB Chip + 4MB Fast) and 120MB hard drive cost $3699. That's 3-4 x more computing power, 3 x more RAM and 3 x more hard drive space, and of course AGA graphics - all for $300 less than the A3000. By 1993 the street price had dropped to $2299 - and Amiga fans complained that it was too expensive!

[grond]

Come on, Bruce, I was pulling your leg by replying to your comment in a way that I think seemed similar to what you often write. That being said, you do add a lot of perspective to these discussions which I appreciate. It's reassuring to see that you do find some faults in what Commodore did.

My Amiga friend (who "infected" me with the Amiga virus) was very happy to upgrade from his A500 to a used A2000 with a 25 MHz 030 and SCSI controller when AGA was already around. So yes, A2000s were good value a few years after they were introduced. In 1987 they seemed to offer too little in the base package (my opinion) when compared to the A500 but, as I stated above, there were good reasons why it was still too early for a full 32bit architecture which pretty much ruled out any processor above the plain 68000 it came with. This makes me wonder why there wasn't a 68020 with a 16bit data bus.