We redesign the STE

No topic. Everything you want to speak about. Please just stay courteous.

Moderators: Mug UK, Silver Surfer, Moderator Team

User avatar
Foxie
Captain Atari
Captain Atari
Posts: 340
Joined: Wed Feb 03, 2016 7:12 pm

We redesign the STE

Postby Foxie » Mon Jan 01, 2018 2:09 am

The DeLorean has been set to 1987. You shall use your 50 degrees in computer science, business and electronic engineering to assume a key position at Atari. Your mission: to redesign the Atari STE to become a hit product. A machine that every potential Atari ST user will buy. A machine which will compete on equal footing with the Amiga A500 for games. And a machine for which most ST games will be developed for, rather than "STE-enhanced" being a rarity.

Unlike most STE "wish-list" topics I've seen discussed before, the objective here is *realism*. What can you do to the Atari STE's design that fits within both budget, and time constrains that the Atari engineers faced at the time?

So, sorry, no 68030 for you. No 4MB of RAM as standard. No truecolour display. No hard disk. This is 1987. Nobody will ever need more than 640K!

To start the thread off, here's what I'd do...

OPTION 1

Get the damn thing released ASAP! STE should be on the market in 1988, since it seems lots of ST users bought their machines in the 87-89 time frame (myself included). It should retail for the same price as the old ST.
To that end, I'd make the following changes to the STE:

- SIMMs weren't so mainstream in 1988. So ditch the SIMM slots. Replace with either a pin header for a proprietary memory upgrade, or lose the upgradeability. Ship 512K and 1MB versions as standard.

- Ditch the LMC1992. Unnecessary. Also, ditch the switched capacitor filters in the DAC's output path. The Amiga 1200 sounds OK with the raw unfiltered sound of the DAC, so it will be fine for the STE too. Go with mono sound - most users had mono TVs anyway. This saves the cost of one DAC. It also saves the cost of re-tooling the case for the phono sockets. You could even go a stage further and make the DMA hardware write to the YM volume registers instead of using a separate DAC. This would provide all the benefits of DMA sound with none of the extra expense of a DAC.

- Fix the audio sample rates to the horizontal scan rate. Offer two options: one sample per scanline, and two samples per scanline (every other scanline in mono). This gives 15kHz or 31kHz. More than good enough (most people can't hear the difference between 31kHz and 50kHz anyway). This saves chip space since you don't need another counter. You could even ditch the 31kHz mode, but some people will hear the difference between the STE and Amiga then. Unlikely to notice on TV speakers though.

- Ditch the enhanced joystick ports. In retrospect, hardly anyone ever used them, and they do add to the bill of materials. Also required an expensive case re-tooling.

- But, I would keep the hardware scrolling. An essential feature to compete with the Amiga. I would also strongly consider keeping the blitter, since it was already in production by this time. However, if the blitter added a significant cost, then I would re-consider. Games can use pre-shifting if they want fast sprite drawing. But having a blitter is a good selling point.

- Possibly don't bother increasing the palette to 4096. Who's going to notice, really?

- And of course, the original ST should be completely discontinued and replaced with the new STE at the same price.

The net result: hopefully a machine that's ready to hit the market by 1988 instead of nearly 1990. Thus most ST users will have bought one, and most games will be STE enhanced. The most important feature by far is the hardware scrolling, followed by DMA sound.


OPTION 2

Make the machine an Amiga killer. It might be OK to take an extra 6 months, or even a year developing it in this case. Release around late 1990.
To achieve this, I'd make the following changes:

- Lose the enhanced joystick ports.

- Consider adding support for a second or third fire button to the existing joystick ports, like the Amiga supports. But no need for analogue stick support. Ship an extremely cheap and basic two or three button joypad along with the STE to encourage developers to adopt it. Should be much easier considering it works in the existing joystick ports, unlike the STE Power Pad which was useless for 99% of games. A three button pad should make the STE much more competitive against consoles later in the 90s. Widespread support of three button pads on the STE will encourage Amiga developers to do the same thing. Hence it becomes industry standard and certain types of games become much more playable on computers. Rather than the SNES and Mega Drive killing off the ST/Amiga due to easier handling in fighting games etc.

- Again go with fixed sample rates tied to the horizontal scan rate. Also ditch the switched capacitor filters. But stereo is important: keep it. Need 31kHz sample rate too.

- Yamaha made a compatible version of the YM2149 in 1987, which featured an additional FM synthesis engine alongside the PSG - the YM2203. Use this chip instead. This will provide three channels of FM (possibly usable as six, if it lets you trigger each pair of operators independently), alongside the PCM audio, and the existing three PSG channels. Even for games that don't play back MODs in software, the sound of chip music will be drastically improved. The music will be more complex than Amiga music, due to having 6-9 channels. And for games which do have a MOD player routine, you can use your four MOD channels alongside the FM channels for a massive sound. The presence of a 4-op FM synthesizer will also be of interest to musicians. This chip requires a matching external DAC. It may be possible to hijack this DAC for DMA sound playback, without needing a separate DAC chip for DMA sound. However this precludes simultaneous output of DMA sound and FM, which might be too much of a price to pay.

- Keep the LMC1992, the bass and treble boost will make FM sound better. Or, if it's cheaper, use an opamp with a fixed bass and treble boost.

- Modify the cartridge port to bring out a write line. This would allow much easier implementation of all sorts of hardware expansions, such as graphics cards. This might require the use of a register to re-purpose one of the ROM enable lines as a write line, defaulting to read-only mode for ST compatibility at boot. Or alternatively, add an internal pin header connector exposing the CPU bus. This could be used for various upgrades, even CPU accelerators like the A500's edge connector. Materials cost is extremely low.

- 256 colour "super low res" mode, by using both odd and even memory cycles to fetch display data (same memory speed as a normal ST, so same cost). The CPU will be stopped during active scan, so total CPU speed will drop to less than 50%. Not a major problem if the game is very tightly written and uses the blitter and hardware scrolling. Since 256 colour registers might be expensive to implement in the shifter, just stick with 8 bit truecolour and no palette (rrrgggbb) like the MSX2 and Archimedes used. It works well enough. This will make games look much, much better. In fact, significantly better than most Amiga games. Or perhaps you could use rrggbbxy - where x is ignored when y is 0, and when y is 1, gbbx are used for choosing a palette colour. This would allow blitter-driven parallax scrolling by drawing the single-plane background to plane x (which is ignored in the presence of foreground graphics with y=0), and transparent pixels with y set show the background. The two background colours are determined by the palette entries, which can be altered by the CPU on each scanline to give copper background effects. Think James Pond 2: Robocod on the Amiga.

- In order to make up for the lost speed of super low res mode, consider modifying the blitter so it runs on both odd and even memory cycles, like the Amiga's blitter can. But with the blitter already designed, this might not be practical. Instead, move the GLUE/MMU memory timings so the blitter runs on odd cycles during blanking, while the CPU runs on even cycles. Then during active scan, the blitter moves back to even cycles and shares with the CPU as normal. This will allow simultaneous use of the CPU and blitter during blanking, which can nearly double the amount of copying that can be done. But none of this is really essential since it's possible to get plenty of action on the screen even with 50% of your cycles stolen.

- Want to go really balls to the wall? Add an additional graphics mode, "low res parallax". Same amount of data fetched as super low res mode. Same number of shift registers. Except now, it's two separate scrolling layers of 16 colours each (sharing the same existing 16 colour ST palette). This will allow parallax scrolling in even more colours than the Amiga manages. But, it might take up a little more chip space because of the need for two address counters.

- If releasing in the late 1990 time frame, strongly consider using a high density floppy mechanism, to go with the AJAX chip. This will greatly aid PC compatibility. Also means less disk swapping compared to Amiga games = reputation of the STE being more advanced.

- Spend more time testing the TOS revision with existing software to ensure compatibility. It's not acceptable to release in this time frame with major programs like STOS and 1st Word Plus having bugs. In 1988, you could get away with it because the platform was new. But not in 1990.

- Probably ship with 1MB as standard. It was around this time that games started going 1MB only, and it would be a very bad idea to fool people into buying a 512K machine that can't run lots of games. They will feel cheated, and less likely to trust the Atari name for their future upgrades. Less of an issue if you have SIMM slots.

The net result: a machine that produces visually and audibly superior results to the Amiga, enough that it warrants a lot of existing ST users to upgrade, and future purchasers to choose the STE. It would justify a slightly higher price, albeit still below the price of an A500 with 1MB. The presence of the FM chip would make it even more attractive to the music industry. It would still be difficult to encourage developers to write STE enhanced games. But the upgraded graphics and sound would justify the effort.


Regardless of option 1 or 2, another thing I'd do is can ridiculous projects like the STE+. Who on earth buys an ST to use mainly with PeeCee software? The ST can be a serious business computer all to itself. No IBM needed. If you embrace IBM at the hardware level, you already lost. You became a faceless clone maker. In Europe we have this thing called human rights laws. That means as an employer you cannot legally subject your employees to an IBM PeeCee. Especially not one with green screen and noisy fans. You have to let them use humane platforms like the ST. But where it really counts is the home market - the big numbers. The home market doesn't care about built-in 286s, but it very much cares about the number of colours on the screen in the games.


So what would you do with the STE?

joska
Hardware Guru
Hardware Guru
Posts: 3877
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: We redesign the STE

Postby joska » Mon Jan 01, 2018 12:33 pm

Foxie wrote:So what would you do with the STE?


Hardware wise: Add hardware overscan. Other than that the STE is close to perfect IMO.

The STE did not fail because of the LMC, DMA sound sample rate or even costs. It failed because it really didn't do anything that the ST already did, and by 1989 (or 88, really doesn't matter) the ST (and Amiga) was already doomed as software support shifted towards the PC. People wanted more than just playing games on their home computers, and the ST lost because it's OS and development tool never improved. Releasing a new computer in 1989 with a 1985 OS was... stupid. Releasing the Falcon two years later with the exact same OS (only with more colours) was even more stupid.

So if I was in charge in 1987 and actually wanted to continue the ST platform, I'd spend money on adding useful features to the OS, and provide modern, efficient development tools free of charge for anyone who wanted to create software for the ST. Then do what's been done to the PC the last 35 years - make it a little bit faster, add a little bit of RAM and mass storage, improve graphics performance slightly every few months. Hardware costs are of little concern, software is everything. That is why Microsoft and Apple "won".

Oh, another thing: If I was in charge of Atari I would not design, manufacture and market a new STE, and then shortly afterwards flood the shops with a new, cost-cutted STFM at nearly half the price.
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

User avatar
Cyprian
Atari God
Atari God
Posts: 1464
Joined: Fri Oct 04, 2002 11:23 am
Location: Warsaw, Poland

Re: We redesign the STE

Postby Cyprian » Mon Jan 01, 2018 1:34 pm

+1 for Joska
Jaugar / TT030 / Mega STe / 800 XL / 1040 STe / Falcon030 / 65 XE / 520 STm / SM124 / SC1435
SDrive / PAK68/3 / CosmosEx / SatanDisk / UltraSatan / USB Floppy Drive Emulator / Eiffel / SIO2PC / Crazy Dots / PAM Net
Hatari / Aranym / Steem / Saint
http://260ste.appspot.com/

User avatar
Foxie
Captain Atari
Captain Atari
Posts: 340
Joined: Wed Feb 03, 2016 7:12 pm

Re: We redesign the STE

Postby Foxie » Mon Jan 01, 2018 2:05 pm

joska wrote:by 1989 (or 88, really doesn't matter) the ST (and Amiga) was already doomed as software support shifted towards the PC. People wanted more than just playing games on their home computers, and the ST lost because it's OS and development tool never improved. Releasing a new computer in 1989 with a 1985 OS was... stupid. Releasing the Falcon two years later with the exact same OS (only with more colours) was even more stupid.


That's an interesting perspective, and I definitely think that was the case in America where the ST and Amiga never really took off because of IBM. But in the UK, I feel like the IBM didn't really become a popular thing for home users until Windows 3.1 and Doom came out around 1992-1994. Before then, the IBM was just far, far too expensive for home users, and it was also too limited - EGA graphics in games, bleepy internal speaker, and absolutely hostile to use (MS-DOS only software, no mouse). Even basic "business" tasks like word processing or spreadsheets that the average home user might want to use were an absolute nightmare.

To that end, I think TOS and GEM were pretty much OK. Up until about 1991, its only competition was the Amiga (the Mac was WAY out of anyone's price range in the UK, and a lot of people had never even heard of it). I'd much rather use TOS 1.0 over DOS for day-to-day use.

The Falcon was a very different matter though. By that point it had to compete with Windows 3.1 coming in, soundcards coming down in price, and then Doom, which I think was really the first popular game that needed a really fast processor. That pretty much wiped everyone out, including Apple.

joska wrote:Oh, another thing: If I was in charge of Atari I would not design, manufacture and market a new STE, and then shortly afterwards flood the shops with a new, cost-cutted STFM at nearly half the price.


*nods*

That was really one of the most stupid things they ever did. That's why I focused on cost-cutting the STE design by ditching the fancy audio stuff and enhanced joystick ports. It should be possible to make an STE for a similar price to an STFM really. Either that, or you need the 256 colour mode to differentiate it as a premium product.

User avatar
mrbombermillzy
Obsessive compulsive Atari behavior
Obsessive compulsive Atari behavior
Posts: 113
Joined: Tue Sep 13, 2016 9:24 am

Re: We redesign the STE

Postby mrbombermillzy » Mon Jan 01, 2018 2:31 pm

Hmm..

People seem to have a problem with TOS. I think its pretty good for what it was meant to do. Its one of the reasons that I am here, having almost given up on trying to figure out the horrendous Amiga OS and being fed up with the obfuscated/virtualized state of things on Windows. TOS is basic, but (apart from perhaps the earliest versions) does its job and is comprehendible.

Anyhow, topic wise:

Just send Troed back in time to show Atari how easy it would be to just overclock the STF to give new 1280x200/400 and 640x480 modes:

https://www.exxoshost.co.uk/forum/viewt ... f=18&t=283

Perhaps also throw in the abandoned/retasked (TT?) STE 256 colour Shifter.

Job done. :)

joska
Hardware Guru
Hardware Guru
Posts: 3877
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: We redesign the STE

Postby joska » Mon Jan 01, 2018 2:52 pm

mrbombermillzy wrote:People seem to have a problem with TOS. I think its pretty good for what it was meant to do. Its one of the reasons that I am here, having almost given up on trying to figure out the horrendous Amiga OS and being fed up with the obfuscated/virtualized state of things on Windows. TOS is basic, but (apart from perhaps the earliest versions) does its job and is comprehendible.


TOS does what it's supposed to do, but while TOS was OK in 1985 (functionality on par with MS-DOS with a decent GUI on top), it was very quickly superseded by it's competitors. Thanks to better development tools and richer OS'es it was soon much easier and quicker to develop applications on other platforms.

mrbombermillzy wrote:Just send Troed back in time to show Atari how easy it would be to just overclock the STF to give new 1280x200/400 and 640x480 modes:


They knew. Speed was a compromise. Faster speed in 1985 would have required a *lot* more expensive components. And remember, in 1985 the ST was quick. Faster than the Macs and PC's it competed against. Not so only two years later, but when Atari caught up (hardware-wise) with the PC's and Macs again in 1990 the TT could no longer compete due to an obsolete OS and lack of software. They made an effort with the Falcon - MultiTOS and new productive software in 1992, but it was just too late. If MultiTOS came with a faster computer in 1988, and Atari had released proper, modern development tools they *might* have had a chance to establish TOS as a viable platform.
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

Wayne123
Atari freak
Atari freak
Posts: 50
Joined: Tue Jan 17, 2017 5:55 pm

Re: We redesign the STE

Postby Wayne123 » Mon Jan 01, 2018 3:09 pm

In the U.S., back in the late 80's game systems like the Nintendo NES were taking the games market away from computers. The Atari ST was mostly unheard of or thought of as a toy. Apple flooded the schools here with there Apple II computers so that is what kids were exposed to, the PC computers of the late 80's were insanely expensive and really pretty crappy for what you got for the money. I remember my neighbor bought a 286 PC and was talking about it, I showed him my ST and it blew his mind with its graphic capabilities. Atari failed here because most people didn't know about it, same with the Amiga. It was early 90's before you could get a decent PC that didn't suck without breaking the bank, Windows 3.0 and 3.1 made the PC practical for the average user, A PC pre DOS 5.0 and DOSSHELL that let you use a mouse and see where all you files were completely out of the question for a lot of computer users.

The biggest problem for Atari was they were not sold many places outside of a few Atari dealers, I bought an ST at the Federated Group that Atari owned but they disappeared, after that you had to be really looking for an Atari to find one. I used to go to the Atari conventions to see all the cool new Atari stuff, in Southern California in the 80's and 90's we had multiple giant computer swap meets every month where you could find every imaginable PC combination and add on from lots of dealers in the area.
1040 STF, MiST, FPGA Arcade, plus Amigas, NeXT and some PPC Macs.

FedePede04
Atari Super Hero
Atari Super Hero
Posts: 985
Joined: Fri Feb 04, 2011 12:14 am
Location: Denmark
Contact:

Re: We redesign the STE

Postby FedePede04 » Mon Jan 01, 2018 3:11 pm

what drove me away from the Atari, was the waiting time, if the falcon had arrived sooner i would have bought it.
for me it had nothing to do with the tos, the first many years that i had my PC i was using DOS (that i loved) it was first when win95 was release i started using windows but i still use DOS more.

why i never bought the STE was the different was to small compared to the STFM, for me to try finding those kind of money.
i think overall the stfm and the ste was good computers, and for me it did not matter that the use the ym2149 sound chip, if they had put 2 inside the ST would have been perfect when you could a have 4 channels music and 2 sound FX,
the blitter could have run faster maybe 16mhz or even 32mhz (that could have been cool) and a graphic mode with more color could also have been great.

the problem was around 1991 games for the pc start to come with 256 on screen just looking so much better, and also the cpu got so much faster, Atari and Commodore could just not keep up on the tech front compare to the PC, so all the gamer left them.
it is how i see this....
Atari will rule the world, long after man has disappeared

sometime my English is a little weird, Google translate is my best friend :)

User avatar
mrbombermillzy
Obsessive compulsive Atari behavior
Obsessive compulsive Atari behavior
Posts: 113
Joined: Tue Sep 13, 2016 9:24 am

Re: We redesign the STE

Postby mrbombermillzy » Mon Jan 01, 2018 3:40 pm

joska wrote:TOS does what it's supposed to do, but while TOS was OK in 1985 (functionality on par with MS-DOS with a decent GUI on top), it was very quickly superseded by it's competitors. Thanks to better development tools and richer OS'es it was soon much easier and quicker to develop applications on other platforms.


True. Perhaps if all the multitude of Atari R&D side project hardware had been focused on OS updates instead, it would have been a different matter! They should also have taken full advantage of their position in the music industry.

joska wrote:They knew. Speed was a compromise. Faster speed in 1985 would have required a *lot* more expensive components. And remember, in 1985 the ST was quick. Faster than the Macs and PC's it competed against. Not so only two years later, but when Atari caught up (hardware-wise) with the PC's and Macs again in 1990 the TT could no longer compete due to an obsolete OS and lack of software. They made an effort with the Falcon - MultiTOS and new productive software in 1992, but it was just too late. If MultiTOS came with a faster computer in 1988, and Atari had released proper, modern development tools they *might* have had a chance to establish TOS as a viable platform.


The OP is saying 1987.Wouldnt the cost of a 16mhz rated 68000 along with faster drams in 1987 be a better bet than 2 years of R&D plus manufacturing of a slightly improved GFX/Audio subsystem? It may of staved off the souless PC drones for a while longer! :?

joska
Hardware Guru
Hardware Guru
Posts: 3877
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: We redesign the STE

Postby joska » Mon Jan 01, 2018 3:51 pm

Wayne123 wrote:It was early 90's before you could get a decent PC that didn't suck without breaking the bank, Windows 3.0 and 3.1 made the PC practical for the average user,


And were was Atari then? At the same spot as in 1985... THAT was their problem. Microsoft, IBM, clone makers... They had all been spending and building technology whereas Atari mainly focused on cutting costs. When Windows 3 came PC hardware was still very crude and simple, it would have been very easy for Atari to come up with something with more impressive specs. But Atari never spent the hundreds of millions that Microsoft did, building software technology that by the early 90's was so much better than their competitors that they completely dominated the industry.

Wayne123 wrote: A PC pre DOS 5.0 and DOSSHELL that let you use a mouse and see where all you files were completely out of the question for a lot of computer users.


Sales figures does not agree with you :) In the late 80's PC's were outselling the ST and Amiga everywhere. By then 95% of ST's and Amigas were sold to gamers, people who were not loyal to the brand but moved on to consoles when the 16-bit Nintendos and Segas took off. The remaining few machines were sold to businesses/"serious users" who quickly moved on to platform with actual software support by 1990/1991. Windows 3 completely destroyed this marked for all it's competitors. The ST only lived on for a few more years as MIDI machine (no thanks to Atari, but Steinberg), the Amiga had some success as a graphics workstation but soon the PC took over here as well.

Wayne123 wrote: The biggest problem for Atari was they were not sold many places outside of a few Atari dealers,


And why not? Because it was not an attractive product. In Europe Atari sold an amazing amount of ST's in the late 80's/early 90's thanks to it's aggressive prizing, but it's pretty clear that prize was the ST's main selling point by then.
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

joska
Hardware Guru
Hardware Guru
Posts: 3877
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: We redesign the STE

Postby joska » Mon Jan 01, 2018 4:07 pm

mrbombermillzy wrote:True. Perhaps if all the multitude of Atari R&D side project hardware had been focused on OS updates instead, it would have been a different matter!


Not instead of, but in addition to :)

mrbombermillzy wrote:The OP is saying 1987.Wouldnt the cost of a 16mhz rated 68000 along with faster drams in 1987 be a better bet than 2 years of R&D plus manufacturing of a slightly improved GFX/Audio subsystem? It may of staved off the souless PC drones for a while longer! :?


Well, the STF was done in 1985, not 87. And even in 87 the components would be too expensive. Remember, even Apple waited until 1989 before they released a 68000-machine faster than 8MHz, the very expensive Mac Portable. The very similar Stacy was released the same year, at half the price of the entry-level Mac Portable.

Now, the Stacy could actually be clocked at 16Mhz with very simple modifications. Atari engineers most likely knew that, but also knew that it could not be done reliably without higher spec'ed components and thus a higher price.
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

Wayne123
Atari freak
Atari freak
Posts: 50
Joined: Tue Jan 17, 2017 5:55 pm

Re: We redesign the STE

Postby Wayne123 » Mon Jan 01, 2018 4:48 pm

joska wrote:
Wayne123 wrote:It was early 90's before you could get a decent PC that didn't suck without breaking the bank, Windows 3.0 and 3.1 made the PC practical for the average user,


And were was Atari then? At the same spot as in 1985... THAT was their problem. Microsoft, IBM, clone makers... They had all been spending and building technology whereas Atari mainly focused on cutting costs. When Windows 3 came PC hardware was still very crude and simple, it would have been very easy for Atari to come up with something with more impressive specs. But Atari never spent the hundreds of millions that Microsoft did, building software technology that by the early 90's was so much better than their competitors that they completely dominated the industry.

Wayne123 wrote: A PC pre DOS 5.0 and DOSSHELL that let you use a mouse and see where all you files were completely out of the question for a lot of computer users.


Sales figures does not agree with you :) In the late 80's PC's were outselling the ST and Amiga everywhere. By then 95% of ST's and Amigas were sold to gamers, people who were not loyal to the brand but moved on to consoles when the 16-bit Nintendos and Segas took off. The remaining few machines were sold to businesses/"serious users" who quickly moved on to platform with actual software support by 1990/1991. Windows 3 completely destroyed this marked for all it's competitors. The ST only lived on for a few more years as MIDI machine (no thanks to Atari, but Steinberg), the Amiga had some success as a graphics workstation but soon the PC took over here as well.

Wayne123 wrote: The biggest problem for Atari was they were not sold many places outside of a few Atari dealers,


And why not? Because it was not an attractive product. In Europe Atari sold an amazing amount of ST's in the late 80's/early 90's thanks to it's aggressive prizing, but it's pretty clear that prize was the ST's main selling point by then.


I never claimed PC sales were less than Atari and Amiga sales, you could find a PC easily, I said they were insanely expensive for what you got.
Businesses bought a lot of computers, IBMs and clones mostly, the average person of my age back then had no interest in a computer and pre Windows and pre DOS 5.0 many of them were lost as what to do with one. Commodore 64s sold like crazy when you could buy one for cheap, most were used for games and Nintendo took over the cheap game market, when the 16 bit game systems came out they made the ST or an Amiga look expensive for a game system. The Sega Genesis was a lot cheaper than an ST or an Amiga.

Once PC clones came out, anybody could make them, this drove down prices and also drove innovation, Atari and Commodore never had a chance as they could never match the spending needed to keep up. There were giant computer stores everywhere in the 90's here, even Apple almost did not survive.

I set up more than a few computers for people I worked with back then, I bought my first PC clone in 1992, it had a 25Mhz 386, a 40MB hard drive, a pretty basic video card, 1MB ram and it had no sound other than beeps, with a 14" monitor it cost me 1200 dollars. A TT030 would have been $3000 back then and you had to go pretty far to find a dealer who had one. I lived in Orange County, an area with 2.5 million people and I would drive to Los Angeles county to find a dealer that sold Atari computers.

My 386 was nothing special, it cost me $200 to upgrade the memory to 4MB I added a math coprocessor and paid $300 for a decent video card and later a monitor that could do Non interlaced video at 1024 x 768, before these upgrades it was disappointing.
1040 STF, MiST, FPGA Arcade, plus Amigas, NeXT and some PPC Macs.

User avatar
Frank B
Atari Super Hero
Atari Super Hero
Posts: 959
Joined: Wed Jan 04, 2006 1:28 am
Location: Boston

Re: We redesign the STE

Postby Frank B » Mon Jan 01, 2018 5:05 pm

The ST and Amiga didn't survive into the mid 90s because they didn't evolve. :( I'd take an STe over any PC clone in 1990. PC machines were utter junkola back then.

joska
Hardware Guru
Hardware Guru
Posts: 3877
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: We redesign the STE

Postby joska » Mon Jan 01, 2018 5:30 pm

Wayne123 wrote:I said they were insanely expensive for what you got.


Yes, if you look at hardware specs only they were expensive. But if you look at software availability and quality the PC was a much better value for money than the ST or Amiga from around 1988/1989. If all you wanted was games then the ST/Amiga was a good choice until 1991/1992. But productive software was - with few exceptions - much better and much more on the PC. The exceptions (DTP and MIDI) was not enough to make enough money to fund the needed development to keep up with the PC. Maybe if Atari had done that from the start...

One example is multitasking: It would not have been very difficult to allow GEM to multitask applications. Why didn't Atari do this in 1985/86? They could have omitted the desktop from ROM and put it on the language disk instead, making room for a better, multitasking AES in ROM.

And why did they wait until 1990 or so to make a proper GDOS? If they had spent time and money on this in 85/86/87 while they were making quite a bit of money on the ST the ST would have been a much more attractive platform to develop for. They had a GEMDOS that was virtually identical to MS-DOS - why didn't Atari's development tools make it easy for software developers to port their DOS-software to the ST?

Atari didn't make any real changes to their OS from 1985 to 1992, and by then it was much too late. Fettling with hardware details in the STE wouldn't have changed a thing. At best - if the STE was released (and the ST discontinued!) before the A500 - it would have been able to really compete with the A500 in the games market and made Atari a bit more money.
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

User avatar
Foxie
Captain Atari
Captain Atari
Posts: 340
Joined: Wed Feb 03, 2016 7:12 pm

Re: We redesign the STE

Postby Foxie » Mon Jan 01, 2018 7:05 pm

Wayne123 wrote:I never claimed PC sales were less than Atari and Amiga sales, you could find a PC easily, I said they were insanely expensive for what you got.


This was really one of the big problems with PCs back then. I remember buying a PC around the mid 90s, and I paid something like £2000 for it ($2600). It had really basic specs, couldn't even run Win95 without thrashing the swap file. I never paid anything like that for ST stuff. But I needed it for college work and the internet. Otherwise I would have stuck with the ST.

In 1990, the prices were so unbelievable for PCs in the UK that the average home user simply couldn't afford one at all without remortgaging the house. It didn't really matter one bit what the software was like on the PC, it was simply out of reach for the majority of people.

Home computer users in 1990 had pretty basic needs, but they were specific needs. Home computers usually needed to run games. They sometimes needed to run spreadsheets and word processors. And they had to be very easy to use. That's why the ST and Amiga were such big hits in Europe - they did all of these things very well. Perhaps the spreadsheets and word processors weren't as powerful as the PC counterparts, but a home user isn't going to care.

Even if you ignore the cost of a PC, it fell flat on its face when it came to running games, and it was simply too difficult for the average person to use. A lot of people forget just how hard DOS software of the time was to use. Some software even came with keyboard overlays to try and make the software a bit more accessible! There was absolutely no "discoverability" in the interface like you find with GUI software on the ST.

I've heard it said that the PC outsold the ST and Amiga even in 1990, but that surely must have been due to business purchasers. Thinking back to those days, I didn't know anyone with a PC. You just didn't see them in the home, ever. I knew several people with STs and Amigas, and lots of people with 8 bit micros. Even my dentist used STs at the practice!

I think it must have been around 1991 when I first saw a PC in someone's home. Their parents were millionaires, and lived in a mansion. It was totally maxed-out with VGA graphics and a soundcard, so it could actually do a credible job at games. But not at a price normal people could afford. By 1993, I knew two other people who had recently got PCs. One was a typical green screen IBM, to allow them to work from home. They had a Spectrum and NES for games - the PC just couldn't do it. The other was a 286 with EGA, so it could play some games, but they really sucked compared to ST/Amiga games. No soundcard of course. It wasn't until 1994 that I started seeing VGA PCs in ordinary people's homes, thanks to the popularity of Doom and the ease of use of Windows 3.1.

So I really do think the ST was relevant for its target market for a very long time. That's why I don't feel the OS needed upgrading - because it just blew the IBM to smithereens on ease of use.

I never owned a Falcon (despite really wanting one), so I can't comment much about that. But MultiTOS did come out the same year as Windows 3.1, so I would say they had feature parity at that point. Maybe MultiTOS was better since it had better multitasking. Truetype fonts were available on the ST shortly after they were on the IBM. I'm not sure what major features were missing. Perhaps when Video for Windows came out, that put Windows ahead of MultiTOS. But of course, both the ST and Amiga had been working with digitized video for years already.

Zarchos
Captain Atari
Captain Atari
Posts: 263
Joined: Wed May 09, 2012 7:38 pm
Location: FRANCE

Re: We redesign the STE

Postby Zarchos » Mon Jan 01, 2018 7:18 pm

I am surprised nobody (or only a few people) refer to the Acorn machines in this thread.
After all they are what everybody was looking for :
- Powerful CPU
- Efficient usage of memory bandwith (Please name a processor which can access 32 bits in 1 cycle (ok : after the 1st access to 32 bits taking 4 cycles)).
- Hi capacity graphics with extremely programmable VIDC chip (VIDEL is close to it, but not as programmable as the Acorn chip).
- Linear modes, which are, after all, the most logical way to organise screen layout (and perfectly suited for 3D).
- Great sound from day one.
- A very user-friendly and efficient operating system (although not preemptive, but cooperative), in ROM.
- Robust applications in the productivity area, and yes some good games, surprising for a machine Acorn only lately positioned as being a gaming machine too (with the A3010).

Be it the Archimedes with an ARM3, or later on the RISC PC (this one having the 2nd slot for a 486 card, 16 bit sound, and up to 24 bit screen modes, accelerated if you populate the V-RAM slot) with its CPU on a daughterboard card : Acorn machines, although yes more expensive than the Ataris or the Commodore machines, always delivered what people usually expect from a computer.
Furthermore, and it is certainly not the least of the strength of Acorn : the entire control of all chips inside the machine, and yes, that includes the main CPU, meaning the next machine to create would not depend on strange choices by Motorola or Intel.
See what a Titanium board has to offer today for RISC OS users using an ARM based motherboard.

I understand many people here are not from the U.K., where unfortunately Acorn only distributed their machines properly, so Acorn doesn't spring to mind ; but reading people from the U.K. not mentioning Acorn just baffles me.

Btw : the Archimedes hasn't got a fixed 256 colour palette as somebody wrote, comparing it to the MSX2.
That's urban legend most of the time spread by Amiga fanatics and Archimedes bashers.
The 256 colour modes have 16 selectable entries, from which 15 tints are automatically produced.
And yes even with this constraint, that is interesting :
http://www.stardot.org.uk/forums/viewto ... ng+palette
This has been solved with the RISC PCs, of course.

I'll add Acorn had the Phoebe ready, working (not some obscure schematics with not a single chip produced) and I let you discover its specs on wikipedia.
Falcon, Atari 1040 STE, 520ST, 800xl, xegs, Amigas, Archimedes, RISC PCs, Iyonix, Omega, BBC B, Atom, Electron, ZX 81, Spectrum 48/128/+2/+3, Russian clones, Sam Coupe, V6Z80P, QLs inc. Q68, and more !
2200 m2 museum on its way https://youtu.be/xjB6_Ez-3BA
Shorter video here : https://youtu.be/UEZisfkcN1Y

Currently porting SOTB to the Archie : https://www.youtube.com/user/Archimedes75009/featured

User avatar
Frank B
Atari Super Hero
Atari Super Hero
Posts: 959
Joined: Wed Jan 04, 2006 1:28 am
Location: Boston

Re: We redesign the STE

Postby Frank B » Mon Jan 01, 2018 7:36 pm

The archimedes was simply too expensive in 1990. My first STe was 259 GBP in 1990. How much was an arch back then? 1100?

Zarchos
Captain Atari
Captain Atari
Posts: 263
Joined: Wed May 09, 2012 7:38 pm
Location: FRANCE

Re: We redesign the STE

Postby Zarchos » Mon Jan 01, 2018 7:47 pm

Frank B wrote:The archimedes was simply too expensive in 1990. My first STe was 259 GBP in 1990. How much was an arch back then? 1100?


GBP599 ex VAT for the BBCA3000 and you could find it at GBP549 ex VAT.
http://chrisacorns.computinghistory.org ... tApr90.pdf
And with 1 Mbyte of RAM, and very often with a free extra 2nd mbyte added by the retailer.
So 'expensive' : ok ; but 'too expensive' : no. Or 'for you' and yes I can understand that.
Now you can also compare an Audi with a Vauxhall, or carrots and salads ... specs for specs it would be more honest.

The ultra high price of the Archimedes is again one of these urban legends.
See table here :
https://en.wikipedia.org/wiki/Acorn_Archimedes
and remember most Archies are not all in one systems with limited expansion possibilities, please.
Last edited by Zarchos on Mon Jan 01, 2018 7:59 pm, edited 4 times in total.
Falcon, Atari 1040 STE, 520ST, 800xl, xegs, Amigas, Archimedes, RISC PCs, Iyonix, Omega, BBC B, Atom, Electron, ZX 81, Spectrum 48/128/+2/+3, Russian clones, Sam Coupe, V6Z80P, QLs inc. Q68, and more !
2200 m2 museum on its way https://youtu.be/xjB6_Ez-3BA
Shorter video here : https://youtu.be/UEZisfkcN1Y

Currently porting SOTB to the Archie : https://www.youtube.com/user/Archimedes75009/featured

Wayne123
Atari freak
Atari freak
Posts: 50
Joined: Tue Jan 17, 2017 5:55 pm

Re: We redesign the STE

Postby Wayne123 » Mon Jan 01, 2018 7:50 pm

Foxie wrote:
Wayne123 wrote:I never claimed PC sales were less than Atari and Amiga sales, you could find a PC easily, I said they were insanely expensive for what you got.


This was really one of the big problems with PCs back then. I remember buying a PC around the mid 90s, and I paid something like £2000 for it ($2600). It had really basic specs, couldn't even run Win95 without thrashing the swap file. I never paid anything like that for ST stuff. But I needed it for college work and the internet. Otherwise I would have stuck with the ST.

In 1990, the prices were so unbelievable for PCs in the UK that the average home user simply couldn't afford one at all without remortgaging the house. It didn't really matter one bit what the software was like on the PC, it was simply out of reach for the majority of people.




I remember one of the guys I worked with buying a 90MHZ Pentium computer for $8000, In 1992 I bought a Nissan Pickup Truck for work and only paid $6588. Even Atari TT030s and big box Amigas were expensive though, my accelerated Amiga 2000 Video Toaster Setup would have been around $5000 back in 1990.

I remember a guy I worked with asking me what would be a good computer for his kids, I suggested an ST but he went and bought an Apple II because that is what the schools were using. I had even told him he could use Mac software on an ST with a Magic Sac, Apples were not cheap.
1040 STF, MiST, FPGA Arcade, plus Amigas, NeXT and some PPC Macs.

Wayne123
Atari freak
Atari freak
Posts: 50
Joined: Tue Jan 17, 2017 5:55 pm

Re: We redesign the STE

Postby Wayne123 » Mon Jan 01, 2018 7:57 pm

Zarchos wrote:I am surprised nobody (or only a few people) refer to the Acorn machines in this thread.
After all they are what everybody was looking for :
- Powerful CPU
- Efficient usage of memory bandwith (Please name a processor which can access 32 bits in 1 cycle (ok : after the 1st access to 32 bits taking 4 cycles)).
- Hi capacity graphics with extremely programmable VIDC chip (VIDEL is close to it, but not as programmable as the Acorn chip).
- Linear modes, which are, after all, the most logical way to organise screen layout (and perfectly suited for 3D).
- Great sound from day one.
- A very user-friendly and efficient operating system (although not preemptive, but cooperative), in ROM.
- Robust applications in the productivity area, and yes some good games, surprising for a machine Acorn only lately positioned as being a gaming machine too (with the A3010).
.



Too bad they never marketed them over here.
1040 STF, MiST, FPGA Arcade, plus Amigas, NeXT and some PPC Macs.

Zarchos
Captain Atari
Captain Atari
Posts: 263
Joined: Wed May 09, 2012 7:38 pm
Location: FRANCE

Re: We redesign the STE

Postby Zarchos » Mon Jan 01, 2018 8:11 pm

I'll add that's all very nice to want ultra cheap machines : no wonder the company then has no money to spend on R&D.
I bet when selling a computer, Acorn made more than 3 times the margin Atari had on theirs.
Same for Apple (here I think it is even more).

This might explain that.

You can't have it both ways.
Falcon, Atari 1040 STE, 520ST, 800xl, xegs, Amigas, Archimedes, RISC PCs, Iyonix, Omega, BBC B, Atom, Electron, ZX 81, Spectrum 48/128/+2/+3, Russian clones, Sam Coupe, V6Z80P, QLs inc. Q68, and more !
2200 m2 museum on its way https://youtu.be/xjB6_Ez-3BA
Shorter video here : https://youtu.be/UEZisfkcN1Y

Currently porting SOTB to the Archie : https://www.youtube.com/user/Archimedes75009/featured

User avatar
Frank B
Atari Super Hero
Atari Super Hero
Posts: 959
Joined: Wed Jan 04, 2006 1:28 am
Location: Boston

Re: We redesign the STE

Postby Frank B » Mon Jan 01, 2018 8:21 pm

The arch was doomed from the get go. It was not popular outside schools in the UK. I only knew one other archimedes user. I have a 5k and a risc PC but they don't mean as much to me as the Amiga or STe. I prefer the STe / TT and Falcon to the archimedes. I'd prefer smashing my head off a brick wall over using a 1990 era PC! The PC was just ... ugh.... DMA/IRQ Conflicts, 640 k limitation with incomprehensible memory "managers". Awful command line driven OS. Awful design and implementation. Video cards which displayed snow all over the screen during updates. Moving the box slightly would make a card loose and render the whole machine unstable. Beeping "sound". Horrible serial mice. Just ugh.... No GUI and when it did turn up it was unusable with windows 3.1 Ugh ugh ugh. I'm astounded the machine took off.
Last edited by Frank B on Mon Jan 01, 2018 8:26 pm, edited 1 time in total.

Zarchos
Captain Atari
Captain Atari
Posts: 263
Joined: Wed May 09, 2012 7:38 pm
Location: FRANCE

Re: We redesign the STE

Postby Zarchos » Mon Jan 01, 2018 8:25 pm

Frank B wrote:The arch was doomed from the get go. It was not popular outside schools in the UK. I only knew one other archimedes user. I have a 5k and a risc PC but they don't mean as much to me as the Amiga or STe. I prefer the STe / TT and Falcon to the archimedes. I'd prefer smashing my head off a brick wall over using a 1990 era PC!


I understand your feelings, but well for a 'doomed' computer it allowed Acorn to live longer than Atari and Commodore.
Both went bankrupt.(Which means screwing many suplliers btw).
Acorn didn't.
They stopped manufacturing computers, split their various businesses, and closed business.
Sorry to remind possibly harsh facts, I don't mean to offend anybody, but the vision of what Acorn was and what the Archie and successors were always surprise me.
Hermann Hauser, CEO of Acorn, weights in the billion pounds area.(and expanding thanks to his investment fund).
I have not read that for people at the head of Atari or Commodore ;-)
Last edited by Zarchos on Mon Jan 01, 2018 8:31 pm, edited 1 time in total.
Falcon, Atari 1040 STE, 520ST, 800xl, xegs, Amigas, Archimedes, RISC PCs, Iyonix, Omega, BBC B, Atom, Electron, ZX 81, Spectrum 48/128/+2/+3, Russian clones, Sam Coupe, V6Z80P, QLs inc. Q68, and more !
2200 m2 museum on its way https://youtu.be/xjB6_Ez-3BA
Shorter video here : https://youtu.be/UEZisfkcN1Y

Currently porting SOTB to the Archie : https://www.youtube.com/user/Archimedes75009/featured

User avatar
Frank B
Atari Super Hero
Atari Super Hero
Posts: 959
Joined: Wed Jan 04, 2006 1:28 am
Location: Boston

Re: We redesign the STE

Postby Frank B » Mon Jan 01, 2018 8:29 pm

Zarchos wrote:
Frank B wrote:The arch was doomed from the get go. It was not popular outside schools in the UK. I only knew one other archimedes user. I have a 5k and a risc PC but they don't mean as much to me as the Amiga or STe. I prefer the STe / TT and Falcon to the archimedes. I'd prefer smashing my head off a brick wall over using a 1990 era PC!


I understand your feelings, but well for a 'doomed' computer it allowed Acorn to live longer than Atari and Commodore.
Both went bankrupt.
Acorn didn't.
They stopped manufacturing computers, split their various businesses, and closed business.
Sorry to remind possibly harsh facts.


Actually I met the guy who split off Acorn from Arm once* He was the CEO of Icera. The only business with a future was the arm chip itself. The Acorn imploded just as badly as the Amiga and Atari IMHO. It was doomed because they went for the fast buck and shuttered the computing division. I worked with loads of ex acorn people at a previous job. We even had a Phoebe case in the server room! Acorn's slide to irrelevance was even more painful to watch than Commodore's implosion. At least Atari and Commodore went out fighting. Acorn just faded away.

*Take the arm and ditch the body.

Zarchos
Captain Atari
Captain Atari
Posts: 263
Joined: Wed May 09, 2012 7:38 pm
Location: FRANCE

Re: We redesign the STE

Postby Zarchos » Mon Jan 01, 2018 8:38 pm

Frank B wrote:
Zarchos wrote:
Frank B wrote:The arch was doomed from the get go. It was not popular outside schools in the UK. I only knew one other archimedes user. I have a 5k and a risc PC but they don't mean as much to me as the Amiga or STe. I prefer the STe / TT and Falcon to the archimedes. I'd prefer smashing my head off a brick wall over using a 1990 era PC!


I understand your feelings, but well for a 'doomed' computer it allowed Acorn to live longer than Atari and Commodore.
Both went bankrupt.
Acorn didn't.
They stopped manufacturing computers, split their various businesses, and closed business.
Sorry to remind possibly harsh facts.


Actually I met the guy who split off Acorn from Arm once* He was the CEO of Icera. The only business with a future was the arm chip itself. The Acorn imploded just as badly as the Amiga and Atari IMHO. It was doomed because they went for the fast buck and shuttered the computing division. I worked with loads of ex acorn people at a previous job. We even had a Phoebe case in the server room! Acorn's slide to irrelevance was even more painful to watch than Commodore's implosion.

*Take the arm and ditch the body.


Take the ARM and all the other chips, all designed by Acorn ...
The ARM alone is nothing, it is designed as an element among 4 main bricks.
And no : Acorn didn't implode like C= or Atari, leaving no technologies behind them.
There is not only the legacy of the ARM architecture, but also various IPs later used by other tec giants like Broadcom.
https://www.eetimes.com/document.asp?doc_id=1186586
Element14 was the remainder of Acorn.
Quote : 'Formed by former executives of Acorn Computers Ltd. in 1999, Element 14 is a fabless-IC design house with 68 employees. It is developing what is anticipated to be the industry's first 12-port, full-rate ADSL solution for the central office.'

Which part of Atari or C= has ever been bought for any sensible amount of money ? (Element14 : bought 600 millions dollars by Broadcom)
Or has survived to be worth something (ARM : bought 32 billions dollars by Softbank)
I won't even talk about PACE technology, who bought the rights of RISC OS, used it in its set top box, and made hundreds millions with it.
Last edited by Zarchos on Mon Jan 01, 2018 8:44 pm, edited 2 times in total.
Falcon, Atari 1040 STE, 520ST, 800xl, xegs, Amigas, Archimedes, RISC PCs, Iyonix, Omega, BBC B, Atom, Electron, ZX 81, Spectrum 48/128/+2/+3, Russian clones, Sam Coupe, V6Z80P, QLs inc. Q68, and more !
2200 m2 museum on its way https://youtu.be/xjB6_Ez-3BA
Shorter video here : https://youtu.be/UEZisfkcN1Y

Currently porting SOTB to the Archie : https://www.youtube.com/user/Archimedes75009/featured


Social Media

     

Return to “Chat forum [ENG]”

Who is online

Users browsing this forum: No registered users and 3 guests