Jump to content
IGNORED

Atari, 1988, and the DRAM shortage


jmccorm

Recommended Posts

I'm an old Atari fan here. I've got a few 80s arcade games in my gameroom, and I was really into Atari's 8-bit machines in their day. Perhaps I held on too long -- I never went with the ST/TT boxes. But you guys are likely the best people I can approach for this.

 

Atari, 1988, DRAM Shortage:

 

I came across some incredibly engaging documents that tell the story of Atari's colorful activities during the DRAM shortage of 1988. Yes, I will be putting my analysis of them online, as well as the original documents. No, I won't leak them in advance.

 

I've dug up quite a few articles on the late 80's DRAM shortage, but I know that you guys are likely to have a far better memory than I would as far as what was happening at the time. So I'd like to hear your perspective on the DRAM shortage on how it affected Atari users. Just your thoughts are fine, but if you have documents or articles you can reference, that's great too.

 

If by some small chance, there is an Atari employee who knows why this is such a colorful tale, I'd enjoy the opportunity to chat privately.

 

Aside from that, there are a few documents which I'd like to get my hands on, but I haven't been able to:

 

Atari's 1987 annual report to shareholders [priority item]

Atari's 1988 annual report to shareholders [priority item]

Atari's 1989 annual report to shareholders

Atari's quarterly reports from 1987 through the end of 1989

 

So, what can you relate about the DRAM shortage, and how did it apply to you?

 

Thanks.

Link to comment
Share on other sites

Can't say it really affected me - back then computer purchases/upgrades like most people's were few and far inbetween.

 

I actually bought my first ST in 1988 and IIRC the price had dropped several hundred in the months leading up.

 

I suppose for that time period, it was a case of the standard machine having 64K or 128K, or 640K and sometimes a meg or two for IBM compatibles, but suddenly jumping to higher levels, especially for the low-end machines where suddenly they had 8 or 16 times that amount.

But 1988 is a reasonable amount of time into the 16-bit era, although for many people it was a transition time and plenty were still banging about on their older machines.

On the business side though, desktop PCs were starting to become much more affordable then, and the growth of '286 and '386 clones was exponential. It was probably around the late 1980s to early 1990s that I noticed the trend emerging of a desktop per employee rather than group shared machines in the government/corporate workplace.

 

Similar shortages with price rises have occurred in recent times, fairly sure there were price hikes around the time of WinXP, Vista and possibly even Win 7.

 

With WinXP especially, you went from a situation where a home machine could cruise along OK with 256 meg RAM on the older Win OSes, but suddenly with XP you needed at least 512 and preferably a Gig to get decent performance.

 

On the other hand, price drops in RAM have occurred and probably prompted plenty of upgrades which in turn could create shortages after a time, and a surge in prices.

Link to comment
Share on other sites

Atari in the UK had a very interesting solution to the problem (commodore similarly followed suit with the amiga) Atari UK's solution was to start bundling some games with new stocks of STFM (i think it was mostly the 520, dunno about the 1040) I think that was to offset any projected price increases of the ST series (I think the total value of the games was just over the projected price increase of the ST series if there were no bundling)

 

Also I think that Atari were very smart in the games they chose (i.e over 12 months old, so obviously they could buy them at discounted wholesale prices)

Link to comment
Share on other sites

Oh the great Dram Shortage of 88, I remember it well. Actually I remember it but since I was in high school my memories may be a bit hazy.

 

I remember Atari was doing well in Europe and it seemed in the USA they were ready for a major push, third party support was going well and Amiga though a great machine was playing catch up. I remember launches for hardware products being delayed, prices rose, and finding certain models was an adventure. Wasn't it around this time Amiga came out with the 500/2000 lineup and the price difference dwindled? Regardless of the Amiga specifications I think if three people walk into a computer store to buy a new computer and see an Atari ST running almost all the same type of programs as Amiga and it's at least a hundred dollars cheaper, I bet the Atari gets two of them to buy. Take away the price advantage and then the customers are going to be a lot more critical of which machine to buy and probably go with the Amiga because of it's specs.

 

I also seem to remember that it caused Atari to delay products like the STE and caused development of future products to be put on the back-burner because of funding but maybe I'm off here. I wonder what would happened if Atari kept their pricing advantage and could of released the 520/1040ste early in 1988 and then follow up with the Mega STE/TT030 no later than early 1989 with the Falcon coming out early 1990 what could of happened.

Link to comment
Share on other sites

ST deliveries pretty much dried up in the USA I believe. Prices also rose across the board. Back then magazines had to print prices a month in advance. My buddy wanted to buy a 1040ST however when he called the mail order houses all the prices were different. He finally found 1 place that would sell the 1040 for the price listed in the magazine.

Link to comment
Share on other sites

I'm sure you probably know about, and already visited...

 

http://www.atarimuseum.com/

 

but they have some amazing Atari content there. If you've

got something they don't have, I'm sure Curt Vendel would

be interested in it - perhaps you should contact him?

 

Maybe he might even be able to help with the items you were

looking for.

 

Thanks for the post - many of us are interested in all things

Atari. :)

Link to comment
Share on other sites

I'm sure you probably know about, and already visited...

 

http://www.atarimuseum.com/

 

but they have some amazing Atari content there. If you've

got something they don't have, I'm sure Curt Vendel would

be interested in it - perhaps you should contact him?

 

Maybe he might even be able to help with the items you were

looking for.

 

Thanks for the post - many of us are interested in all things

Atari. :)

A lot of that is in need of updating though, but Curt's been delayed in doing some of that. (a few like the 3200 are updated, but a lot is still older)

Curt and Marty have brought up newer stuff in various AA discussions, but I'm not sure it's compiled anywhere yet. (their books eventually, but I assume AA will get some major updates before the full content is published in their books)

 

 

Did the DRAM price/shortage issue have anything to do with actions of the Reagan administration against overseas DRAM suppliers choking out the US market with price dumping? (an attempt that came too late to counter that competition, but not too late to put a heavy strain on imported DRAM and some related products and inflate prices heavily for US companies buying those products)

 

 

Whatever the case, it had a heavy impact on global DRAM prices for sure:

http://phe.rockefeller.edu/LogletLab/DRAM/dram.htm

 

Though also looking at that list, it seemed to mainly impact the common/lower end densities (256 kb -32kB- and lower) while the larger densities were still dropping (or less inflated) and 1Mb/128kB chips were almost down to the price of 32kB chips due to the inflation of the smaller chips. (in '87 32k chips were roughly $9 per Mbit vs $14 in '88 -compared to almost $20 per Mbit in '85, while 128kB chips had dropped to $15 per Mbit in '87 and only bumped up to $16.5 per Mbit in '88 and dropping slightly below 32k chips in '89)

 

So, maybe Atari (and others) could have partially countered that problem by shifting more towards 128k DRAM chips which, while still moderately more expensive than 32k densities, would allow much greater consolidation and reduced PCB cost -plus pay off even more as the high densities fell further in cost. (like how CBM and Atari both opted for 32k chips in '85 when 8k chips were still moderately cheaper -and more so if they'd been stockpiling since '84; 512 kB would have taken 64 8k DRAM chips vs 16 32k chips and the prices of 32k chips were dropping fast at the time) 128k chips would have allowed 512k with just 4 DRAM chips rather than the 16 previously needed. (and 2 MB machines would be down to 16 DRAM chips)

It wouldn't be until 1990/91 that 512k DRAM chips would be attractively cheap. (more expensive in '90, and possibly not enough to allow the consolidation and head start for producing consolidated models to have been worth it, maybe start in late 1990 for a sooner shift to 512k, definitely by '91 as 512k chips were the cheapest density on the market -2 MB chips probably wouldn't be attractive until ~1994)

Edited by kool kitty89
Link to comment
Share on other sites

I thought one of the reasons was a major manufacturing plant burnt down or something?

I've run across that a couple of times now. Let me see what I can find...

 

A usenet posting in 1993:

And lately, I

have heard that the price of DRAM (memory) rises dramatically from

$ US 25.00/Meg to $ 50/Meg or more in the States and Canada due to a plant

fire in Japan. Before that, the US was considering retaliation measures

against the Korean and Japanese for dumping cheap DRAM into the US market.

 

And some more discussion...

http://groups.google.com/group/comp.os.ms-windows.nt.setup.hardware/browse_thread/thread/e286304329cb9008/39bf8ee7668d472e?q=taiwan+dram+plant+fire#39bf8ee7668d472e

 

It comes up again in 2003 to explain their current events, but at least by this time someone flags it as an urban legend:

http://groups.google.com/group/comp.sys.ibm.pc.games.rpg/browse_thread/thread/5f64e95d9e0c0210/5bf82ef3ede8af76?q=japan+dram+plant+fire#5bf82ef3ede8af76

 

There isn't a lot of hard news to reference it. It looks like there may have been an event in Taiwan, but the event itself was minor. There might have been another even in Japan, can't tell for sure. 1993 seems to be the date given for these. That was, what, around the time for the Jaguar?

  • Like 1
Link to comment
Share on other sites

I thought one of the reasons was a major manufacturing plant burnt down or something?

Hmm, would that ahev been something specifically tied to Atari (ie a primary manufacturer they were dependent on), or something that would equally impact the entire industry?

 

Again, I'm not sure, but I definitely remember reading about US policies impacting DRAM prices. (at least for anything coming into the US, or by extension of that, any US export products directly tied to DRAM)

 

Maybe it was a combined effect of all that.

 

 

Looking around, there seems to be plenty of references tying the US anti-dumping policies to the DRAM shortage/inflation issue, like this:

http://books.google.com/books?id=c62TkhIZd8wC&pg=PA47&lpg=PA47&dq=DRAM+shortage+1988&source=bl&ots=2dUfWpEFbP&sig=87UqVm3ypEtKjyPuU7Sn0E7-SXM&hl=en&ei=Z19CTfqUNo34sAPXu7DlCg&sa=X&oi=book_result&ct=result&resnum=6&ved=0CC8Q6AEwBQ#v=onepage&q=DRAM%20shortage%201988&f=false

 

(see about 1/2 way down the first column)

 

 

 

And several others like:

http://books.google.com/books?id=6T4EAAAAMBAJ&pg=PA48&lpg=PA48&dq=DRAM+shortage+1988&source=bl&ots=r-GhL60dS-&sig=MTj2WFxcNlTxXGRgPVyC7tVgtcY&hl=en&ei=Z19CTfqUNo34sAPXu7DlCg&sa=X&oi=book_result&ct=result&resnum=5&ved=0CCwQ6AEwBA#v=onepage&q=DRAM%20shortage%201988&f=false

 

 

 

I don't know the details of the US anti-dumping policies, but it seems like they may have been pretty flawed: taxing imports on the DRAM components, but not tacking surcharges to ALL imported products containing said components. (which makes no sense and really defeats the purpose: if you want to prevent dumping, you need to do it across the board; doing it half assed would only put domestic manufacturers of computer/electronic products using the memory chips to be weakened on the market against favorable Japanese import products)

Edited by kool kitty89
Link to comment
Share on other sites

Many industries try that business of attributing a price increase to a disaster of some sort.... Plant exploded, etc. I remember they tried that with antifreeze (Prestone, etc) in the late 80s - "a plant blew up." They do it with gasoline - "not enough refinery capacity" similar gibberish. Earthquake in Japan - now chips cost more!!! Blah blah. Simple greed. Good for 'em if they can get it.

Link to comment
Share on other sites

Whatever the case, it had a heavy impact on global DRAM prices for sure:

http://phe.rockefeller.edu/LogletLab/DRAM/dram.htm

I'm officially notifying you that I am stealing that URL (and the parent site that it comes from).

You can thank Kskunk for that one, he pointed it out to me almost 2 years ago. ;)

http://www.atariage.com/forums/topic/143762-atari-panther/page__st__25__p__1751398#entry1751398

 

Note that those are annual averages of course and include all types of DRAM chips (slow/cheap and faster/high end stuff), hence why the 1987 price is a bti higher than the $1.50 32k chip ($6 per Mbit) price Amstrad mentioned in the first google books article I linked to. (but the $3.75 for '88 is a bit above the '88 average at

$15/Mbit vs $14.2 -albeit that was also a snapshot for '88 va a past annual figure for '87 in amstrad's case so that $3/75 may have dropped a bit later on)

 

 

Many industries try that business of attributing a price increase to a disaster of some sort.... Plant exploded, etc. I remember they tried that with antifreeze (Prestone, etc) in the late 80s - "a plant blew up." They do it with gasoline - "not enough refinery capacity" similar gibberish. Earthquake in Japan - now chips cost more!!! Blah blah. Simple greed. Good for 'em if they can get it.

Yes, but if that was indeed the case it wouldn't be "greed" alone so much as a calculated bate and switch: Japan/Asian companies had been selling at below cost prices for quite a while at that point, so it would make sense that that would have to end at some point and switch to normal/profitable prices. (an illegal business practice that the US government failed to counter, at least when it mattered) So the Japanese/Asian manufacturers get enough domestic market and assets built up that they can afford to dump the prices at or below cost for export to drive out foreign (ie domestic US) manufacturers from the market who couldn't/wouldn't endure such pricing. (not sure if there were any price war type things also fueling it)

 

 

Given the other added info above, it seems I may have been mistaken in thinking the US actually established anti-dumping taxes on DRAM. (it seems the Japanese may have beaten them to the punch and stopped dumping as they'd already won the market and chased western manufacturers out) There definitely seems to be some conflicting information on the issue.

Edited by kool kitty89
Link to comment
Share on other sites

It looks like the Reagan administration did impose a 100% duty but by the time it came into force all the US DRAM manufacturers had already folded or shifted production to something else.

 

Interestingly, from what I can find after a quick look at RAM prices in the USA at the time, it would be hard to see how Atari could have manufactured a 1MB ST at a profit?

 

I would have thought that Atari would have largely avoided the US duty by having their computers manufactured overseas (which they did) and distributing from outside the US (which they also did). From what I can see the duty only affected the DRAM as a component and not as part of an already assembled and imported computer.

 

Looking at the UK, and this is from a quick look, Commodore were talking of dropping the Amiga price during the DRAM shortage and Atari were talking about restoring prices - to re-establish the ST/Amiga price differential well before the end of the year (1988).

Link to comment
Share on other sites

I know the DRAM shortage between 1989-90 specifically (will have to check old magazines) happened between/around the time of the Amiga 500 going down £100 (CSG had no problem with RAM for Amiga production). At the same time the ST had to go from £299 to 399.

 

This catalyst was huge, and sealed the fate of the ST/Amiga war.

 

At this precise period Alan Sugar/Amstrad also went on to publicly state they were in an untenable position regarding DRAM chip availability and effectively killed the Sinclair PC200 (8088 PC in the STFM form factor with modulator built in to compete with ST/Amiga) because both this machine and their more expensive desktop PCs like the 1640 and 2000 models needed the same RAM chips and so they went with the more lucrative models and hence no home PC in the A500/520STFM styling of all-in-one machine.

 

Memorable because it is the only mistake Amstrad/Sugar made in the last century ever.

 

So there probably was two distinct instances. One was a fire in a glue factory accounting for 1/3 worlds DRAM supply and a later issue again but probably anti-monopolistic laws set by USA/EU (funny considering that piece of shit Gates never had any laws to stop him killing the home computer market!). Don't think the glue factory fire (in Osaka?) was the same thing as what killed ST pricing vs A500 price drop later.

 

I have stacks of scanned magazines covering this entire event/period of computing and read them in my lunch hour at work. There was a great article about it in new computer express weekly magazine, if I find that issue is scanned I will send you the file :)

Link to comment
Share on other sites

At this precise period Alan Sugar/Amstrad also went on to publicly state they were in an untenable position regarding DRAM chip availability and effectively killed the Sinclair PC200 (8088 PC in the STFM form factor with modulator built in to compete with ST/Amiga) because both this machine and their more expensive desktop PCs like the 1640 and 2000 models needed the same RAM chips and so they went with the more lucrative models and hence no home PC in the A500/520STFM styling of all-in-one machine.

 

Memorable because it is the only mistake Amstrad/Sugar made in the last century ever.

 

I had to Google that PC200, since I'd never heard of it. Thanks for mentioning it, damn if it didn't look similar to STfm!

 

post-16281-129635896287_thumb.jpg

Link to comment
Share on other sites

It looks like the Reagan administration did impose a 100% duty but by the time it came into force all the US DRAM manufacturers had already folded or shifted production to something else.

 

I would have thought that Atari would have largely avoided the US duty by having their computers manufactured overseas (which they did) and distributing from outside the US (which they also did). From what I can see the duty only affected the DRAM as a component and not as part of an already assembled and imported computer.

Again, this is particularly ironic: the tax came into place far too late to curb the dumping and US DRAM manufacturers had already left the market... but it came just in time to help cripple US computer manufacturers against Japanese competition -since the US companies would have the overhead from the tax and the Japanese products would be ready-made. :roll: Don't you just love politics/bureaucracies? :D

That also would have promoted US manufacturers to move more production over-seas to avoid the DRAM overhead, so another negative impact on the domestic US economy. (and in '87 there were ongoing talks at Atari to move production back to the US from overseas: not sure if that happened in '88, but boy would that have been poor timing . . . )

Link to comment
Share on other sites

I know the DRAM shortage between 1989-90 specifically (will have to check old magazines) happened between/around the time of the Amiga 500 going down £100 (CSG had no problem with RAM for Amiga production). At the same time the ST had to go from £299 to 399.

 

This catalyst was huge, and sealed the fate of the ST/Amiga war.

 

At this precise period Alan Sugar/Amstrad also went on to publicly state they were in an untenable position regarding DRAM chip availability and effectively killed the Sinclair PC200 (8088 PC in the STFM form factor with modulator built in to compete with ST/Amiga) because both this machine and their more expensive desktop PCs like the 1640 and 2000 models needed the same RAM chips and so they went with the more lucrative models and hence no home PC in the A500/520STFM styling of all-in-one machine.

Wow, I wonder why Atari couldn't shift to different DRAM sources to help fill that (or maybe the shift took too long). That, or why they didn't start pushing models with less RAM (ie baseline 512k models) with more features at competitive prices to counter the RAM issue. (faster CPUs, DSDD disk drives across the board, perhaps blitter and some sort of sound expansion)

A shame the STe wasn't out in '88, that would have added a good bit of value and retained backwards compatibility. (or even a plan ST with Blitter added to it, and short of adding DMA sound -which should have been simple, but may have still not been near production for whatever reason- they could have tacked-on a cheap FM synth chip -like the tiny YM2413, or used the YM2203 as a direct replacement for the YM2149 -maybe adding a plain 8-bit DAC in a simple IC with parallel interface to the CPU would have been a simple useful addition as well, or maybe a couple DACs wired to stereo -maybe embedded in the SHIFTER like the STE's DMA sound or 4 bare parallel DAC channels to allow Amiga format MOD without software mixing but still lots of overhead from CPU playback -not nearly as useful as even a single DMA channel, but better than only having the PSG to hack through)

 

Actually, just adding a wait-state mechanism and using a 16 MHz 68k (for not full 16 MHz performance due to contention, but a hell of a lot better than 8 MHz) would have been a big plus against the Amiga: things like adding/embedding simple direct-write DAC ports and switching to a YM2203 would further have closed the gap even without the BLiTTER or 12-bit RGB. (but having the blitter on top of that and putting that system -100% ST compatible with the fast CPU mode allowing faster plain ST games or a forced 8 MHz mode for full compatibility and potential for much more advanced sound/graphics on top of that for a lower price than a 1 MB Amiga could have been very attractive)

 

The 16 MHz CPU would have been the biggest single issue by far since it would allow all but timing sensitive ST games to be immediately faster/smoother. (and then have the STe later than it was, but with 16 MHz with fastRAM -at least optional- and video derived from the TT SHIFTER, and a 16 MHz blitter tuned more for packed pixel performance in the 256 color/8bpp modes -adding the YM2203 on top of the DMA sound wouldn't hurt either -same board space as the 2149 too other than 1 tiny 8-pin DAC chip)

Link to comment
Share on other sites

The CPC Plus range didn't last either but they looked pretty good.

The GX-4000 was a bad move too in that regard. Hell, given the much larger market share of the Spectrum, Amstrad probably should have put a higher priority on putting out a proper successor to the Speccy with backwards compatibility rather than just the modest upgrade of the 128k. (something like the CoCo III's graphics -16 indexed colors from 6-bit RGB and variable resolution and hardware V/H scrolling, a 6 or 8 MHz Z80, and the 128k's RAM and sound hardware and backwards compatibility could have been pretty damn competitive for the lower-end market in the late 80s, needed to be much earlier than the CPC+ too, like '87 or '88 at the latest -still would have been a bad idea to push it into the cart based game market though, especially as late as 1990/91, but even against the SMS it probably wouldn't have been that good of an idea)

 

Letting the Loki (Flare) team slip through their fingers after the merger with Sinclair was also a bad move. Having the Flare 1 hardware (or similar) by '87/88 would have been amazing to build onto the CPC or Spectrum architectures, or standalone for that matter. You had a chipset that could kick the Amiga's ass in graphics and sound (fast blitter, 256 color packed pixels, DSP for sound and general coprocessing, fast multiplier unit, etc), but configured such that there was enough hardware acceleration to allow a cheap 6 MHz Z80 as the main CPU (Konix pushed for an 8088 instead and the Slipstream ASIC resulting from that also was used with later x86 CPUs including a 386 based set-top box).

 

If Amstrad had pressed something like that into the role of the GX-4000 in 1990, let alone earlier (the Flare 1 chipset was completed ~1987 and the consolidated silpstream ASIC was production ready in 1989), it could have been pretty kick-ass on the mass market.

 

There were claims of Loki being vaporware, and it is in the sense that it failed to materialize (the team left Sinclair after the Amstrad buyout), but given the quick turnaround to form Flare in 1986 with a very powerful chipset completed only a year or so later, it seems like it was very real. (based on experience, but not direct designs of the Loki project -due to legal reasons they couldn't continue the exact chipset they'd started at Sinclair)

 

 

It's not like Amstrad couldn't have taken it up after the fact either: Flare never sold the IP (or not for many years at least) since Konix could only afford a non-exclusive license, so Flare was free to license/sell the IP of the Flare 1 (and Slipstream) to any 3rd party. (again, the slipstream was completed in 1989, before the CPC+ was released) For that matter, Amstrad could have taken an interest in Flare's original discrete chipset back in '87/88 (before slipstream was made), but they missed that too.

 

For that matter, Atari Corp could have licensed it as well (Martin Brennan started contracting for Atari Corp with the final chip design of the Panther after an engineer left the project; he ended up convincing Atari to push for a new design which became the Jaguar and to drop the Panther, but the Slipstream chip could have made for an attractive direct alternative to the Panther -and Atari/Flare were in a fairly favorable position to license it; OTOH, Atari already owned the IP to several other chipsets that could have been adapted to the role of a console in addition to the Panther -which proved impractical without further modification, so they might have opted to drop back to something derived from the ST/TT line, or perhaps even the Lynx chipset for a 1990 home console, or modify the Panther design -unless the Slipstream really was extremely favorable: it didn't support 68k as it was though, just Z80/x86 interfacing unlike Atari's projects or the Jaguar's 68k/x86 support -waiting for the Jaguar was definitely a bad move though, among many other mistakes under the Sam Tramiel years for both the computer and entertainment divisions)

Link to comment
Share on other sites

The GX-4000 was a bad move too in that regard. Hell, given the much larger market share of the Spectrum, Amstrad probably should have put a higher priority on putting out a proper successor to the Speccy with backwards compatibility rather than just the modest upgrade of the 128k.

 

It was called the SAM Coupé

 

http://en.wikipedia.org/wiki/SAM_Coup%C3%A9

 

It tanked.

Link to comment
Share on other sites

The GX-4000 was a bad move too in that regard. Hell, given the much larger market share of the Spectrum, Amstrad probably should have put a higher priority on putting out a proper successor to the Speccy with backwards compatibility rather than just the modest upgrade of the 128k.

 

It was called the SAM Coupé

 

http://en.wikipedia.org/wiki/SAM_Coup%C3%A9

 

It tanked.

I know about that, but that's hardly what I was talking about: the Coupe was neither marketed directly by Amstrad/Sinclair nor released early enough to matter (needed to be '86 or '87 at the absolute latest). That and it lacks many of the features I noted: namely the lack of Speccy 128 compatibility for memory map and sound (which had been standardized back in '85), and it lacks some of the critical video features as well. (hardware scrolling -a necessity for sure- and definitely doesn't seem to have the same flexibility in resolution as the GIME in the CoCo III, let alone supporting them all at 4bpp -the GIME has 16 color modes for 160/256/320/640 pixels wide and 192/225 pixels high -the coupe had a 128 color/shade palette to the GIME's 64 6-bit RGB colors, but that would hardly have been an advantage on top of everything else)

 

Or better than that would have been if the Speccy 128k had had all those features in the first place, or had been delayed until such could be added. (then again, there was also the CoCo II as the intermediate from the original models and the III's added graphics and faster CPU)

 

So basically, the Spectrum counterpart to the CoCo III with a new ASIC embedding all the old video, I/O, and DRAM logic onto a new ASIC along with the added graphics capabilities, memory mapping, and possibly added timer functionality, except that you'd also have the AY PSG rather than just the beeper (or the DAC in the CoCo's case) and basically have a machine capable of games on par (of not a good bit ahead in many cases -due to scrolling) with the Atar ST at a fraction of the price and fully backwards compatible with the Speccy. (plus the added 640 pixel graphics and 80 column text capabilities would also have allowed CP/M -and better business/professional applications in general, albeit something the CPC already offered to a more limited extent)

 

 

For that matter, Sinclair should have gone on to a more powerful Z80 based successor to the Speccy instead of the QL -similar sort of professional/business oriented design, but more conservative and aiming at bringing forward their established user base with a much more affordable and compatible design closer to direct competition with the C128 at a lower price and with an 8 MHz Z80, probably RGBI rather than 3-bit RGB, and perhaps the AY sound/IO chip as well. On top of that, a more conservative design could very well have reduced or eliminated the delays and reliability problems that the QL suffered, and they could have pushed for CPM like the C128 or maybe even invested in a GUI follow-on for it. ;) (perhaps have that totally in place of the Spectrum 128k as well, and followed up with a CoCo III style graphics update)

Edited by kool kitty89
Link to comment
Share on other sites

Hmm, aside from the possibility of offering fast CPU models at competitive prices (but still limited RAM), to counter the DRAM price/shortage issues:

another possibility might have been officially offering RAM upgrades to existing ST users (aside from MEGA models which already supported that internally).

AFIK the piggyback RAM upgrade (or any other RAM/CPU upgrade) was only an unofficial thing that voided the warranty and not something ever offered at licensed Atari service centers.

 

Atari Corp initially had the ST as a closed box design with very limited expandability, but that doesn't mean they couldn't still get around that after the fact to some extent.

 

They probably could even have assembled kits that even more easily facilitated upgrades aside from the somewhat labor-intensive piggyback RAM upgrades. (add-on boards the "clipped" onto the top of the 68k might have been possible -short of desoldering the 68k and adding a socket as many 3rd party add-ons required)

And not just RAM upgrades, but also potential CPU upgrades among other things. (at one point I think there was talk of a BLiTTER upgrade as well, and kits for service centers to add that wouldn't have been a bad idea either)

Maybe the CPU upgrade could have involved adding the necessary logic for wait states for full (non-interleaved) access modes to main memory rather than limited caching with the forced 50/50 split. (obviously, standalone fast CPU models could have that logic added easily, but as an add-on it might not have been feasible -let alone adding a fasRAM bus in addition to that)

 

 

Hell, the simplest direct response to the A500 (even before the price drop) could have been a new low-cost model 16 MHz ST (with wait states) and an added expansion port akin to the A500 "trap door" expansion port. (aside from MEGA models which really should have had 16 MHz versions from the start in '86 -and probably a fastRAM bus and FPU socket; the FPU support would have even been a jump over the A2000)

Maybe they could add the blitter (or have it optional via a socket) as well, but the fast CPU was far more foolproof and added something the Amiga lacked while improving the performance of most standard ST programs substantially. (so definitely not an 8 MHz ST with added blitter)

 

 

 

 

 

 

Commodore was really manufacturing all their own RAM? (I'd have thought they'd have outsourced for that, especially with all the other US manufacturers dropping DRAM production due to the low prices of Asian/Japanese manufacturers)

I wonder hot long it took before CSG became uncompetitive cost wise compared to direct outsourcing to the lowest bidding overseas chip vendors -not just production cost reduction either, but also accelerated R&D facilitated by fast turnaround times for prototype LSI chips. (makes you wonder why Atari Inc/Warner didn't push for a merger with one of the smaller US chip fabs back around '80/81 -or earlier if they had the funds/revenue to do so- and with vertical integration being such an attractive prospect -given the long-standing relationship with Synertek, that might have been the most immediate choice for such a partnership/merger/buyout -prior to the Honeywell buyout at least)

Edited by kool kitty89
Link to comment
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...