How plausible would a "16 Bit 7800" or a "Portable 7800" have been?

[AmishThrasher]

The 7800 was basically a "third generation" console (roughly equivalent to the Master System and NES) which was backwards compatible to the 2600. One thing that I've always wondered about Atari is how plausible it would have been for Atari to have followed it up with a 16-Bit console (roughly equivalent to the Sega Mega Drive and Super NES), backwards compatible to the 2600 and 7800, in around 1988 or 1989?

 

One of the problems with the Jaguar seems to be that it was released far too late to be a competitor to the Super NES and Mega Drive; by the time it came out, both Sega and Nintendo had already established customer bases. At the same time, it was released far too early to be a viable competitor, technically, to the Playstation, N64 and Saturn by the time the 32 / 64 Bit generation rolled around (when gamers were ready to upgrade to a new console).

 

There has been talk around here that Atari were looking at a 16 bit follow-up to the 7800 either in the form of the Atari Panther, or an ST based game console.

 

But how viable would - for want of a better term - a "16 Bit 7800" have been around '88 or '89? What I have in mind by this is a console with a 'more powerful' MARIA chip (i.e. 16 BIT registers, more colours on screen, higher resolution, etc.), more RAM, a sound chip, and - say - a 65C816 as a CPU. Possibly a port for a CD ROM add-on down the track (as was all the rage in the early '90s). Aside from the supercharged MARIA, all other parts would have been 'off the shelf' by that point.

 

Would such a console have been technically feasible for Atari to produce? Could Atari have created a console retaining backwards compatibility to the 2600 and 7800 while still being competitive (in price and tech specs) with the MegaDrive and Super NES? Would there have been any technical or financial hurdles which would have prevented such a console coming together? And, aside from the supercharged MARIA, would it be technically feasable for a sufficiently committed hobbyist to cobble together a working system like the one that I've described?

 

On a related note, I've read that Sega's GameGear is - in a nutshell - a portable Master System with a larger colour palette. Sega even offered an adaptor allowing SMS games to be played on the Game Gear at one point. Given this, I've wondered if, instead of the Lynx, would it have been technically and financially feasible for them to have released a 'portable 7800' or a backward compatible yet improved 'portable 7800?'

 

I'm interested in people's thoughts..?

[The_Laird]

Considering how much they had to cram into the 7800 case and the size of the cartridges etc. it would have been every hard to have built a 7800 handheld back then that wasn't massive.

[AmishThrasher]

Considering how much they had to cram into the 7800 case and the size of the cartridges etc. it would have been every hard to have built a 7800 handheld back then that wasn't massive.

 

In what ways would cramming a 7800 into a handheld be different to cramming a SMS into a handheld?

[+FujiSkunk]

Considering how much they had to cram into the 7800 case and the size of the cartridges etc. it would have been every hard to have built a 7800 handheld back then that wasn't massive.

 

In 1986 it would have been massive, yes. In 1990, when the Game Gear came out, a portable system based on the 7800 may have been more plausible. I think Atari made the right choice going with the Lynx instead, though. Pity they couldn't compete better against Nintendo.

[Underball]

Considering how much they had to cram into the 7800 case and the size of the cartridges etc. it would have been every hard to have built a 7800 handheld back then that wasn't massive.

 

In what ways would cramming a 7800 into a handheld be different to cramming a SMS into a handheld?

The SMS didnt' have backwards compatibility with another system so there were far fewer components to need to rework then the 7800, and it wasn't released in the US until 1991.

[+FujiSkunk]

The SMS didnt' have backwards compatibility with another system so there were far fewer components to need to rework then the 7800, and it wasn't released in the US until 1991.

 

The Game Gear was released in Japan in 1990.

 

A portable 7800 wouldn't necessarily have needed to have 2600 support. Just about everything other than the sound chip is independent for supporting the two systems, so ripping out 2600 compatibility would have made a portable 7800 much simpler to design. Whether it would have been marketable that way is up for debate.

 

And if you really want to get technical, the Master System did in fact have backwards compatibility with an earlier console, the Sega SG-1000.

Edited June 3, 2010 by FujiSkunk

[Atari Joe]

I like my portable 16-bit lynx just fine. I wish they would have kept the lynx going as a special controller for jaguar like a map thing for avp or whatever. and i love my 7800 too. the 2600 jr was considered portable by 1984 standards when it was supposed to have been released. it was tiny! by 1989 standards both the 2600 and 7800 were aging technology and atari was ready to jump on board by purchasing the handy from epyx and making it into the lynx. smart move! my favorite game for lynx is california games and i enjoy it very much. no need to change history.

[Underball]

The SMS didnt' have backwards compatibility with another system so there were far fewer components to need to rework then the 7800, and it wasn't released in the US until 1991.

 

The Game Gear was released in Japan in 1990.

 

A portable 7800 wouldn't necessarily have needed to have 2600 support. Just about everything other than the sound chip is independent for supporting the two systems, so ripping out 2600 compatibility would have made a portable 7800 much simpler to design. Whether it would have been marketable that way is up for debate.

 

And if you really want to get technical, the Master System did in fact have backwards compatibility with an earlier console, the Sega SG-1000.

a 7800 with no 2600 support is kind of pointless.

[Wickeycolumbus]

The SMS didnt' have backwards compatibility with another system so there were far fewer components to need to rework then the 7800, and it wasn't released in the US until 1991.

 

The Game Gear was released in Japan in 1990.

 

A portable 7800 wouldn't necessarily have needed to have 2600 support. Just about everything other than the sound chip is independent for supporting the two systems, so ripping out 2600 compatibility would have made a portable 7800 much simpler to design. Whether it would have been marketable that way is up for debate.

 

 

I don't think cutting out 2600 compatibility would really save space. The 'sound chip' also contains the VCS's video hardware, and the VCS's RIOT chip is used for 7800 stuff too (like the controllers).

Edited June 3, 2010 by Wickeycolumbus

[kskunk]

A strict evolution of the 7800 is going to be very Panther-like. The Panther strongly resembles a 7800 on steroids. That sort of design could probably do well against the Genesis, while lacking the pseudo-3D and scaling and rotation of the SNES.

 

I think the Lynx is a better inspiration. The architecture is very nice for the mid 80s. Far better thought out than the 7800, but there is a clear line of backward compatibility -- the Lynx's frame buffer is a clear superset of the 7800's line buffer. They even use the same CPU.

 

Not saying it would be a cakewalk, but it's at least plausible to add Maria's "display list" parser to drive the Lynx's blitter and achieve decent compatibility.

 

I think the SNES proved that CPU isn't everything (despite Sega's marketing). For 80s games, solid graphics and sound hardware was more important.

 

The Lynx's graphics capabilities often beat the SNES and completely surpassed the Genesis, with the availability of more advanced 3D, scaling, and rotating effects than Mode 7 provided.

 

The only reason the original Lynx design would not make a good SNES competitor is the display resolution. They could offer a lot of power in a cheap system because the number of pixels was only about 33% of what the SNES delivered. But the same architecture, perhaps with two banks of DRAM (source+destination), could be a very solid SNES beater.

 

The Lynx design also has far more limited sound than the SNES or Genesis, but that makes it a fitting successor to the 7800.

 

Given that the Lynx design was completed over 1986-1987, the timing would be right... it would just be a matter of choosing to go after the console market instead of the handheld market.

 

My biggest problem with this kind of idea is that the 7800 had only just come out in 1986. The Megadrive was out in 1988. To get a jump on Sega would require a 1988 release... but what message does that send to the 3rd parties foolish enough to support the 7800?

 

Better to release the 7800 in 1984 or not at all. 2 years is an eternity in the games industry. Atari learned that lesson -- as demonstrated when they euthanized Panther in 1990. Maybe it was fresh enough in 1988, but in 1990 the Panther was rotten.

 

- KS

Edited June 3, 2010 by kskunk

[Atari Joe]

I always thought that whatever the Panther was supposed to be SHOULD have been some kind of a Lynx home unit. TurboGrafx was on to something when they had the Turbo Express playing the same TurboChips that the TG16 played. It was the same system. The major downsides were its size, power consumption, and that most of the games were designed for a big screen not a small one.

 

If Atari could have gotten the Lynx out around the same time as another system, maybe one that was a Lynx in home system form with 3x more resolution but could also play lynx games through a card slot AND play 7800 & 2600 games AND play its own "high resolution" games that could have competed with SNES and Genesis that would have been slick as can be.

 

I bet even Billy Mitchell didnt think up an idea as great as that one.

[+FujiSkunk]

a 7800 with no 2600 support is kind of pointless.

 

Not necessarily. Nintendo got away with something similar: They released the DS with backwards compatibility only for Game Boy Advance cartridges, even though the GBA itself also boasted support for original Game Boy and Game Boy Color games. If Atari went the Game Gear route and made the "portable 7800" more than just a portable 7800, losing 2600 compatibility may have been acceptable.

 

I don't think cutting out 2600 compatibility would really save space. The 'sound chip' also contains the VCS's video hardware, and the VCS's RIOT chip is used for 7800 stuff too (like the controllers).

 

Understood. I was speculating as someone who isn't intimately familiar with everything under the hood. Still, a redesign of the 7800 using only the "bits" of the 2600 chips that it needs may still have been enough to cram it into a portable shell, especially years after the original releases. More radical redesigns of game consoles have happened before and since.

 

But I'm playing devil's advocate anyway. I think Atari made the right decision going with the Lynx instead. Too bad they didn't make very many right decisions on how to go with the Lynx.

[Greg2600]

Atari didn't have the third party support Nintendo or SEGA had, what were they going to do, port Donkey Kong, Pac-Man, and Pole Position yet again?

[Gorf]

I always thought that whatever the Panther was supposed to be SHOULD have been some kind of a Lynx home unit. TurboGrafx was on to something when they had the Turbo Express playing the same TurboChips that the TG16 played. It was the same system. The major downsides were its size, power consumption, and that most of the games were designed for a big screen not a small one.

 

If Atari could have gotten the Lynx out around the same time as another system, maybe one that was a Lynx in home system form with 3x more resolution but could also play lynx games through a card slot AND play 7800 & 2600 games AND play its own "high resolution" games that could have competed with SNES and Genesis that would have been slick as can be.

 

I bet even Billy Mitchell didnt think up an idea as great as that one.

 

 

From the specs I've seen of the Panther, the Lynx is by far a superior unit. The 7800 should have

been a lynx in console form. Maria is the only thing techically superior in the 7800. Also, removing

the 2600 parts of the 7800 would mean to remove 7800 compatibility as well as 2600 compatibility.

the 7800 games in pat still rely on the older 2600 hardware for a few things. Had the 7800 been released

at the same time as the NES, and Atari had beaten Nintendo at the 'lets exclusive all the developers'

game, NES would have been an abimal failure. The machine has so many problems...the flicker is just plain

nausiating and the cart port was a laugh.

[Curt Vendel]

When GCC was in the midst of trying to get paid by either Warner or the Tramiels (while those entities figured out who was responsible for paying GCC) GCC began laying out a proposal for the GCC1712 which would've been a version of the MARIA graphics chip to mate to a 68000, so they were considering it back in 1984. I have the document on file, I'll try to scan and post it.

 

 

Curt

 

The 7800 was basically a "third generation" console (roughly equivalent to the Master System and NES) which was backwards compatible to the 2600. One thing that I've always wondered about Atari is how plausible it would have been for Atari to have followed it up with a 16-Bit console (roughly equivalent to the Sega Mega Drive and Super NES), backwards compatible to the 2600 and 7800, in around 1988 or 1989?

 

One of the problems with the Jaguar seems to be that it was released far too late to be a competitor to the Super NES and Mega Drive; by the time it came out, both Sega and Nintendo had already established customer bases. At the same time, it was released far too early to be a viable competitor, technically, to the Playstation, N64 and Saturn by the time the 32 / 64 Bit generation rolled around (when gamers were ready to upgrade to a new console).

 

There has been talk around here that Atari were looking at a 16 bit follow-up to the 7800 either in the form of the Atari Panther, or an ST based game console.

 

But how viable would - for want of a better term - a "16 Bit 7800" have been around '88 or '89? What I have in mind by this is a console with a 'more powerful' MARIA chip (i.e. 16 BIT registers, more colours on screen, higher resolution, etc.), more RAM, a sound chip, and - say - a 65C816 as a CPU. Possibly a port for a CD ROM add-on down the track (as was all the rage in the early '90s). Aside from the supercharged MARIA, all other parts would have been 'off the shelf' by that point.

 

Would such a console have been technically feasible for Atari to produce? Could Atari have created a console retaining backwards compatibility to the 2600 and 7800 while still being competitive (in price and tech specs) with the MegaDrive and Super NES? Would there have been any technical or financial hurdles which would have prevented such a console coming together? And, aside from the supercharged MARIA, would it be technically feasable for a sufficiently committed hobbyist to cobble together a working system like the one that I've described?

 

On a related note, I've read that Sega's GameGear is - in a nutshell - a portable Master System with a larger colour palette. Sega even offered an adaptor allowing SMS games to be played on the Game Gear at one point. Given this, I've wondered if, instead of the Lynx, would it have been technically and financially feasible for them to have released a 'portable 7800' or a backward compatible yet improved 'portable 7800?'

 

I'm interested in people's thoughts..?

[carmel_andrews]

A 7800 gfx chip and a 68000....sounds like a souped up 7800 to me

 

Mind you the price for cart software (unless they used st/miggy style floppies) would have been pretty pricey

 

there again....wasn't atari trying to make their own upscale 68000 based game system (perhaps this might have been one possible configuration)

[+Mitch]

When GCC was in the midst of trying to get paid by either Warner or the Tramiels (while those entities figured out who was responsible for paying GCC) GCC began laying out a proposal for the GCC1712 which would've been a version of the MARIA graphics chip to mate to a 68000, so they were considering it back in 1984. I have the document on file, I'll try to scan and post it.

 

 

Curt

 

Is that the same one as the Maria doc I scanned and have on my site?

 

Mitch

[Gorf]

Why would you use a 65C816? Its a 6502 with a few 16 bit instructions, when you can have a 68k?

If you are going to bother at that time with a 16 bit console, you might as well go for the balls

and use a REAL 16/32 bit chip like the 68k. Long Data, short data and byte data, a true 24 bit

address space not requiring the data bus for the upper 8 bits of the address. A far more robust

instruction set and not nearly as much decoding to accomodate the mux addressing.

 

AS much as I love the 6502, I'd take the 68k over the 65C816 anyday....espcially for a 16 bit 7800.

[Curt Vendel]

Hi Mitch,

 

Yes... That's the one, I have some more details from another doc, I just pulled out a pile of 7800 docs the other night.

 

 

 

Curt

 

When GCC was in the midst of trying to get paid by either Warner or the Tramiels (while those entities figured out who was responsible for paying GCC) GCC began laying out a proposal for the GCC1712 which would've been a version of the MARIA graphics chip to mate to a 68000, so they were considering it back in 1984. I have the document on file, I'll try to scan and post it.

 

 

Curt

 

Is that the same one as the Maria doc I scanned and have on my site?

 

Mitch

[AmishThrasher]

Interesting, thanks for filling me in on the GCC1712, I'd never heard of it before. Was this console going to be backwards compatible with the 7800 and 2600? I know that obviously the 68k chip has a very different instruction set, but then again the Sega Megadrive / Genesis managed to include both a Z80 and a 68k and have adaptor compatibility with the Master System. Do you know if GCC thinking of doing something similar?

 

When GCC was in the midst of trying to get paid by either Warner or the Tramiels (while those entities figured out who was responsible for paying GCC) GCC began laying out a proposal for the GCC1712 which would've been a version of the MARIA graphics chip to mate to a 68000, so they were considering it back in 1984. I have the document on file, I'll try to scan and post it.

 

 

Curt

 

The 7800 was basically a "third generation" console (roughly equivalent to the Master System and NES) which was backwards compatible to the 2600. One thing that I've always wondered about Atari is how plausible it would have been for Atari to have followed it up with a 16-Bit console (roughly equivalent to the Sega Mega Drive and Super NES), backwards compatible to the 2600 and 7800, in around 1988 or 1989?

 

One of the problems with the Jaguar seems to be that it was released far too late to be a competitor to the Super NES and Mega Drive; by the time it came out, both Sega and Nintendo had already established customer bases. At the same time, it was released far too early to be a viable competitor, technically, to the Playstation, N64 and Saturn by the time the 32 / 64 Bit generation rolled around (when gamers were ready to upgrade to a new console).

 

There has been talk around here that Atari were looking at a 16 bit follow-up to the 7800 either in the form of the Atari Panther, or an ST based game console.

 

But how viable would - for want of a better term - a "16 Bit 7800" have been around '88 or '89? What I have in mind by this is a console with a 'more powerful' MARIA chip (i.e. 16 BIT registers, more colours on screen, higher resolution, etc.), more RAM, a sound chip, and - say - a 65C816 as a CPU. Possibly a port for a CD ROM add-on down the track (as was all the rage in the early '90s). Aside from the supercharged MARIA, all other parts would have been 'off the shelf' by that point.

 

Would such a console have been technically feasible for Atari to produce? Could Atari have created a console retaining backwards compatibility to the 2600 and 7800 while still being competitive (in price and tech specs) with the MegaDrive and Super NES? Would there have been any technical or financial hurdles which would have prevented such a console coming together? And, aside from the supercharged MARIA, would it be technically feasable for a sufficiently committed hobbyist to cobble together a working system like the one that I've described?

 

On a related note, I've read that Sega's GameGear is - in a nutshell - a portable Master System with a larger colour palette. Sega even offered an adaptor allowing SMS games to be played on the Game Gear at one point. Given this, I've wondered if, instead of the Lynx, would it have been technically and financially feasible for them to have released a 'portable 7800' or a backward compatible yet improved 'portable 7800?'

 

I'm interested in people's thoughts..?

[kool kitty89]

This is a really neat thread. I realize it's been dormant for quite a while, but it's very much in line with a topic I've been wanting to bring up on the forums. (I missed it last year, but ended up finding it in a related search by chance )

 

 

 

The 7800 was basically a "third generation" console (roughly equivalent to the Master System and NES) which was backwards compatible to the 2600. One thing that I've always wondered about Atari is how plausible it would have been for Atari to have followed it up with a 16-Bit console (roughly equivalent to the Sega Mega Drive and Super NES), backwards compatible to the 2600 and 7800, in around 1988 or 1989?

It probably could have been done, but it wouldn't be particularly cost effective. It would have made more sense to not bother with compatibility and use either the old VCS chipset alone or re-use MARIA in a new design aimed at a handheld system.

Even then, if you used the existing MARIA or 2600 on a chip ASIC, you're talking old NMOS process chips that use a fairly significant amount of power for a handheld. (re-implementing in CMOS would likely mean significant R&D investment)

They probably could have had a reasonably cost effective design that actually was 7800 compatible internally, but doing that along with a total CMOS redesign and integrating MARIA+TIA+RIOT+SALLY into a single, new CMOS ASIC would, again, mean a significant amount of R&D overhead. (and given the 7800's fast declining consumer interest in 1989, that investment would not be realized in late model 7800s either -the lack of significant consolidation in the existing 7800 design is a fair sign that production volumes weren't meriting such investment -same for the lack of embedded bank switching or embedded sound chips on carts)

 

Hmm, maybe it wouldn't have been too expensive to implement if commodity ULAs of the time had the performance and gate count necessary to implement all that hardware. (ULAs would probably be the cheapest ASIC options on the market -in term of R&D overhead)

 

 

That would also only make sense if they didn't invest in the Handy/Lynx at all. (which itself had a lot of potential for more than it was used as)

 

And honestly, a color handheld in 1989 (or any time before reflective color LCD screens at least had contrast levels capable of approximating 6-bit RGB) was a bit of a lost cause since backlighting was absolutely necessary and would thus render battery life unacceptable for most users and limit it to a niche market. (it would also add to cost and size, further detracting from it -and then there's the non tech issues of marketing/funding/management/brand recognition/software support)

The Game Gear had marketing, management, and software support close to that of the Game Boy, but was more expensive, bulkier, and (most critical by far for any portable device) the battery life was as dismal as the Lynx. (of course NEC went totally crazy and decided to not only offer only backlit models of the Turbo Express, but to only use high-end active matrix LCD screens -actually, such high end screens may have had adequate performance in reflective implementations to be worthwhile, at an obvious expense)

 

Deluxe backlit modes would have been nice (something Nintendo missed out on), but the unlit models would undoubtedly be the defacto standard due to the batter life. (and less so sue to size and cost)

Of course, both Atari and Sega could have finally met Nintendo with revised models with reflective screens (as well as generally consolidated hardware) in the mid 90s, but Atari was near collapse by then and pulling the lynx back while Sega simply (oddly) chose not to push any such radical revisions on the GG in 1994-96 and heavily pulled back support in 1996. (with almost no games released in 1997) So they both left Nintendo alone with the aging game boy as they took their sweet time to finally release the GBC in late 1998. (which was still inferior in many respects to the Lynx and Game Gear -superior to the GG's sound just as the GB had been, of course)

 

 

One of the problems with the Jaguar seems to be that it was released far too late to be a competitor to the Super NES and Mega Drive; by the time it came out, both Sega and Nintendo had already established customer bases. At the same time, it was released far too early to be a viable competitor, technically, to the Playstation, N64 and Saturn by the time the 32 / 64 Bit generation rolled around (when gamers were ready to upgrade to a new console).

Yes and no: the lack of a real 4th gen console was a very real problem, but the biggest problems with the Jaguar was a severe lack of funding and problematic management exacerbated by said funding issues as well as weak market position and brand name (less so in Europe). (all of which would have been greatly alliviated by a successful -even moderately- 4th gen game console on the market)

 

There has been talk around here that Atari were looking at a 16 bit follow-up to the 7800 either in the form of the Atari Panther, or an ST based game console.

Or Lynx based console as it was more developer friendly and/or better performing (and more cost effective) than either of the above options. (STe cut down to a console would have been weak for the time, though a hack that doubled the SHIFTER -2 hardware bitmap scroll planes- would have probably been enough to make it a realistic contender -especially with 2 separate palettes, like Dual Playfields of the amiga but 4bpp and no hardware sprites)

 

The Panther was radically different than any consoles or computers on the market other than the 7800 and while potentially powerful (with trade-offs in capability), it probably would have been relatively unfriendly to developers. (the Jaguar's object processor added features -especially framebuffer support facilitated by buffering allowign efficient operation in cheap DRAM rather than the Panther's SRAM- that made it far more programmer friendly; maybe if Atari/Flare engineers could have reworked the Panther OPL to be reasonably more like the Jaguar in a short amount of time, it would have been ready in time to meet the 4th gen market in 1991 -still a bit late though)

 

 

One other extremely interesting option that wasn't mentioned at all is the Konix Multisystem's Slipstream ASIC. Konix could not afford to license it exclusively and thus Flare had been left with full IP rights and the ability to sell it to anyone else they wanted. As such, when Martin Brennan had been consulting on the final Panther chip design and began convincing Atari management to ditch the panther and push for a new system (which became the jaguar), it would have been very sensible to suggest the Slipstream in place of the Panther for the short run. (much better for the market than the Panther and better than the lynx in some areas -256 color support, a DSP for audio and 3D coprocessing, and I think the multiplyer unit might have been faster -the Lynx's blitter might have been a bit better though, definitely in 16 color mode since it was aimed at that while Flare's was 2x as fast in 256 color mode)

One issue is that the Flare design only supported Z80 and x86 based CPU architectures, so without glue logic, they'd have to use one of those CPU types. (Atari had been buying x86 CPUs for their PC lines -8088s, 286s, 386SXs, and 386DXs iirc, so they would have had an existing supply line) An 8088 would obviously have been the cheapest option outside of a Z80. (and you did have the growing PC game market and wealth of x86 assembly programmers catering to that -the 256 color bitmap support would also facilitate VGA ports, so interesting potential for games that never got console versions historically)

 

 

 

In 1986 it would have been massive, yes. In 1990, when the Game Gear came out, a portable system based on the 7800 may have been more plausible. I think Atari made the right choice going with the Lynx instead, though. Pity they couldn't compete better against Nintendo.

I don't think there was too much they could do about Nintendo's comeptition. At best (with the funding and marketing) they could have beaten the GG in popularity (which they did in parts of Europe), but it wouldn't have been until the mid 90s with models using unlit reflective screens and with double digit battery lives (probably fewer batteries and a smaller form factor at that) that they'd have had any chance of going beyond a niche market with the Lynx. (it's a really cool machine though and stil would have been really competitive as a mid/late 90s system)

 

 

 

I like my portable 16-bit lynx just fine. I wish they would have kept the lynx going as a special controller for jaguar like a map thing for avp or whatever. and i love my 7800 too. the 2600 jr was considered portable by 1984 standards when it was supposed to have been released. it was tiny! by 1989 standards both the 2600 and 7800 were aging technology and atari was ready to jump on board by purchasing the handy from epyx and making it into the lynx. smart move! my favorite game for lynx is california games and i enjoy it very much. no need to change history.

I agree, even once they were set on withdrawing from the computers, they should have pushed ahead with the Lynx alongside the Jaguar. (especially since the Lynx should have been genuinely profitable vs the Jag which was boarderline -though useful for hype and resulting investment capital, just not enough to build up to anything remotely close to critical mass)

In hindsight, you could argue Atari could have even been healthier if they'd not invested in the Jaguar at all and kept focusing on the Lynx. (no matter how you slice it, Atari was pretty screwed after 1991 though -management problems across the board, no new home console, declining computer market, etc)

They just weren't in a position to properly market or support the Jaguar when they released it. (albeit they were in sorry shape to continue with the Lynx as well)

[kool kitty89]

A strict evolution of the 7800 is going to be very Panther-like. The Panther strongly resembles a 7800 on steroids. That sort of design could probably do well against the Genesis, while lacking the pseudo-3D and scaling and rotation of the SNES.

You mean techncially right? (sheer capabilities)

You gave me the impression that the Panther would have been much tougher to work with (or at least an alien architecture) for common PC/console/arcade developers of the time used to formal bitmap/framebuffer (blitter/CPU/DSP rendered) and/or character mode and/or sprite based system.

 

Besides, the SNES's mode 7 hardly was its defining feature; it was a neat gimmick with usefulness, but hardly the real "meat" of the system. (the sound and color capabilities -and BG/sprite to some extent- are what made it definitive, granted the sound system was pretty wasteful and cost ineffective in hindsight -a glorified interpolated 8 channel MOD player with reverb and compression; Nintendo probably could have used Ricoh's much simpler/cheaper 8 channel 8-bit PCM synth chip without much difference overall, let alone other options like simpler Amiga-like DMA sound possibly backed up by a nice yamaha FM chip)

 

I think the Lynx is a better inspiration. The architecture is very nice for the mid 80s. Far better thought out than the 7800, but there is a clear line of backward compatibility -- the Lynx's frame buffer is a clear superset of the 7800's line buffer. They even use the same CPU.

Yes, though I personally don't think backwards compatibility would have been paramount. (undesirable if it compromised cost effectiveness -by the time the 7800 finally launched, its backwards compatibility probably hurt more than it helped -given what MARIA could have been used for on its own in a new system)

 

I think the SNES proved that CPU isn't everything (despite Sega's marketing). For 80s games, solid graphics and sound hardware was more important.

It was still an unnecessary detriment to the SNES at the time and seems to have been due to poor planing and trade-offs made by Nintendo. (from what I understand, with 120 ns FPM DRAM -like the lynx- and a proper DMA interface catering to the 650x, it should have been reasonable to run the SNES's CPU at 7.16 MHz with no wait states in DRAM -aside from page breaks- vs the 2.68 MHz it does now and possibly double the ROM bandwidth of what it had -or maybe the slow DRAM is due to lack of fast page support in the interface used and ROM would have been the same 2.68/3.58 MHz -NEC had been using 7.16 MHz/140 ns glob top ROMs from the start 1987 with zero wait states -allowing code to be run from ROM with no performance hits and making the 8k of work RAM that much less of a limit -granted, the SNES's RAM was so slow that using ROM was never a detriment either and actually a boost for the many late gen 3.58 MHz ROMs)

Hmm, can you imagine a Jaguar with 140 ns ROMs? (especially if the cart ROM space was treated as a separate bank in memory: 68k reading data and running code from ROM with no wait states and no page breaks to DRAM for TOM, separate source and destination for blitter texture mapping -not as fast as dual 75 ns DRAM banks, but still ~60% faster, Jerry pulling data from ROM with no wait states or page breaks, etc)

 

The Lynx's graphics capabilities often beat the SNES and completely surpassed the Genesis, with the availability of more advanced 3D, scaling, and rotating effects than Mode 7 provided.

I wouldn't really say they surpassed the Genesis other than the scaling effects. The master palette was better, but the 16 color bitmap display is a pretty hefty trade-off against character+sprite graphics with 4 subpalettes. (and that's ignoring the resolution that made it totally outclassed even by the Master system -without modification, obviously)

It's like the STe in terms of color limitations.

 

The SNES's color capabilities really take it a step further, even beating the PC-Engine in most respects (much better master palette generally makes up for only having 16 palettes vs 32 -both far better than the MD's 4)

 

The only reason the original Lynx design would not make a good SNES competitor is the display resolution. They could offer a lot of power in a cheap system because the number of pixels was only about 33% of what the SNES delivered. But the same architecture, perhaps with two banks of DRAM (source+destination), could be a very solid SNES beater.

I agree that 256x224 (or a bit higher) would have brought the lynx into a reasonably competitive range, but it would have been massively short from the SNES with only 16 colors (even with palette swapping -which the SNES also used occasionally).

If you boosted the lynx to an 8bpp framebuffer, THAT would have totally outclassed the SNES, Genesis, and Amiga in color/graphics capabilities in general though. (it would have used double the RAM and made texture date 2x as big too without indexing with CLUTs -so having enough RAM to unpack compressed/indexed graphics from ROM would be critical without hardware to do that on the fly -depending on the price point and release date, 256-512k of DRAM might have been feasible -an optional 16 color mode would make it more flexible too)

 

Again, there was the possibility for the Slipstream as well, but of course that's 256 color mode would have the same limitations without hardware CLUT support. (and a 16 color only, high res lynx derivative would have been OK in any case, especially for 1989)

 

Aside from separate source and destination to reduce page breaks, perhaps they could have boosted the CPU to 8 MHz. (switching to a 68k would have been nice, but aside from interfacing issues -not too bad since the 68k supports 680x/650x buses, there's the cost disadvantages over 650x)

 

The Lynx design also has far more limited sound than the SNES or Genesis, but that makes it a fitting successor to the 7800.

Sound hardware would also need to be upgraded: perhaps a derivative of the STe's DMA sound and an off the shelf FM synth chip (perhaps retaining the Lynx's sound, especially if it's already embedded in the chipset and if they wanted to support playing lynx games natively), or if you had a good amount of CPU time with an 8 MHz 650x, you could use the STe DMA alone with software mixing. (FM synth gives a lot more potential for cutting back on ROM/RAM hungry sample data though -even the lynx PSG would help a bit there, and capable composers/programmers can make some awesome sounding stuff with FM -especially if they went with a 4-op chip like the YM2612 or 2151, though the lower-end YM3812/OPL2 a la PC would have been fairly decent too -though often heavily underutilized on PC games, let alone lack of PCM+FM use with SB cards)

 

Given that the Lynx design was completed over 1986-1987, the timing would be right... it would just be a matter of choosing to go after the console market instead of the handheld market.

Or both: they could have made a parallel console/handheld Lynx for 1989. (albeit, given the fundamental disadvantages of the Lynx hardware for a handheld at the time, it probably would have been much better to focus only on a home console and later -like 1994 or later- adapt it into a handheld with a reflective screen and maybe a high-end model with backlighting -and short battery life)

 

My biggest problem with this kind of idea is that the 7800 had only just come out in 1986. The Megadrive was out in 1988. To get a jump on Sega would require a 1988 release... but what message does that send to the 3rd parties foolish enough to support the 7800?

In hindsight we can see that that hardly mattered at all since the 7800 got virtually no 3rd party support at all. (almost 100% of its games were Atari Published)

 

A 1989 release wouldn't have been bad either, perfect timing with the 7800's peak in '88 and a sharp decline in '89 (and under 100k 7800s sold in 1990 in the US). Sega also wouldn't have been a real threat if they hadn't gotten such good management (a la Katz -Atari Corp's former president of entertainment ) and NEC clearly managed to screw up very badly in spite of massive advantages over Sega (and potentially Nintendo -could have been the Sony of the 4th generation with the right management). Indeed, while Sega of America seemed to be somewhat better, they were still foolishly focusing mainly on "arcade at home" marketing for the Genesis's launch, something that Katz quickly set about changing with the establishment of the "Genesis Does" campaign. (he also set about expanding Sega's influence in the western software development field, managed to turn EA's threat of going unlicensed into a favorable business relationship, pushed for celebrity endorsements to combat Nintendo's many exclusive franchises, and more)

 

Of course, as Sega's managment skyrocketed with Katz and later Kalinske (vs the mediocre mid 80s management -even Tonka was miles better than SoA had been in '86/87) Atari Corp management was going to the dogs with Sam replacing Jack and Mike Katz leaving. (apparently an odd interval of shifting management in the Entertainment division too prior to)

That and Atari had a substantially higher market share in the US than Sega up into 1989 (maybe not changing until the 1st quarter of 1990 -ie from fall 1989 when the Genesis's launch would have an impact).

That market share is almost certainly why Sega offered Atari Corp distribution rights of the Mega Drive in the US in late 1988, but that fell apart with Rosen and Jack Trameil being unable to agree on terms. (part of it being Tramiel wanting an interest -or full control- over European distribution -makes particular sense due to Atari's position in Europe- and Atari's own console plans and market position well ahead of Sega in the US would have been major mitigating factors -again, without management from the likes of Katz setting the groundwork for the Genesis's success, Sega very well may have failed -or been much weaker- in the US; they really got lucky with NEC's terrible management in spite of massive resources of such a mega corp that Sony certainly had no problems throwing around some 5 years later)

 

 

 

Better to release the 7800 in 1984 or not at all. 2 years is an eternity in the games industry.

I agree, though they definitely needed SOMETHING on the market, but given the A8 and 5200 (recently cancelled) already on the market, they could have either pushed for the A8 as the primary game platform or Atari Corp could have repealed Warner Atari's cancellation of the 5200. There's a lot of finer points on that issue I addressed here: http://www.atariage.com/forums/topic/176524-7800-what-did-atari-wrong/page__view__findpost__p__2231240

 

Atari learned that lesson -- as demonstrated when they euthanized Panther in 1990. Maybe it was fresh enough in 1988, but in 1990 the Panther was rotten.

Yes, but as above, having nothing at all was worse. They had better options than the Panther for sure, but even that even it would have been better than having nothing at all. (a system that pulls a net profit even with mediocre popularity is infinitely better than virtually vanishing from the market for 3 years and then trying to come back after things took a turn for the wost financially -ie done at least as well as the 7800, probably better due to funding and Nintendo's blocking tactics withering)

Had the Panther been more like the Jag's object processor, it would have been a much more attractive option in general with significant advantages in 2D (and fast scaling) over Lynx or Slipstream based hardware. (a blitter based system still might have been more programmer friendly, especially for home computer ports)

 

Had the 7800 been a smash hit with tons of support, programmers would have been much more accustomed to the likes of the Panther, but given that was not the case, it was a bad idea to even push forward with the Panther. (one could argue Flare shouldn't have even bothered with the OPL in the jag, but instead focused on a much higher performance blitter with line buffers, word buffers, and more flexible CLUT support -at least 4bpp textures, which would also mean much better 3D texture peformance )

[Mord]

but what message does that send to the 3rd parties foolish enough to support the 7800?

 

It would send the exact same message that Sega sent to 3rd parties I guess.

[Greg2600]

I don't think hardware choices would have made a huge difference. Systems don't live on 3rd party support alone. They live on first party games, at least until the last generation. Was Atari capable of a Sonic or Super Mario World? That was never their model. The 2600 was powered mainly by arcade ports, same for Colecovision, 5200, and the 7800. The 2600 had a huge amount of 3rd party, console-only stuff, like all those great Imagic and Activision games we love. Intellivision was an entire system of console-only stuff. Atari would have needed to showcase 4 or 5 main first party launch titles that would be very popular, and convince 3rd parties to come along. Not all old school programmers transitioned well into the higher graphics era, let alone 16 bit. Ask David Crane about that. Personally, I don't believe Atari had the people with the skills and imagination that Nintendo and SEGA had. Platformers really were king at that time, and Atari had little of them. Parker Bros. in fact wound up producing several like Tutenkham and Moctezuma's Revenge. I think Atari of 1981-1982 could have, but by the mid/late 80's, after most of their great arcade programmers had left, no.

[kool kitty89]

but what message does that send to the 3rd parties foolish enough to support the 7800?

 

It would send the exact same message that Sega sent to 3rd parties I guess.

Again, a totally different situation since the 7800 had no 3rd party support anyway and peaked in 1988. Not releasing the XEGS would have been a good move though. (a gaming bundle for the 65XE wouldn't have been such a bad idea though, but should have been pushed back in '85)

 

As for Sega, that's an extremely complex situation that I don't want to get into here (I've spent enough time arguing it on Sega-16), but that situation was somewhat more comparable to how CBM released all those overlapping and incompatible computers in the mid 80s than Atari would have done. (albeit releasing the 7800 after the 5200 also confused things -arguably making the best of the 5200 could have been a healthier move, even more so in hindsight with the mess Warner made with he split in '84)

 

Unless you mean what Sega did in the 80s by releasing platform after platform to stay competitive: then it was like the 7800 under Atari corp, no 3rd party support anyway so the only probablem was confusing consumers. (and at the time, Sega was struggling to compete and went the new hardware route to counter that -and had it not been for NEC's PC Engine in '87, they may have kept pushing the Mk.III and Master System rather than abandoning it in 1988)

Sega launched the (colecovision-like) SG-1000 in mid 1983 and the compatible SC-3000 computer the same year, they released the revised SG-1000 Mk.II in 1984 with expansion support to SC-3000 spec, they released the SG-1000 Mk.III in 1985 (same as master system with a different case and SG-1000 cart slot), released the Master System with YM2413 FM sound in 1987 (had a simple add-on for the Mk.III), released the Mega Drive in late 1988 (still compatible with Mk.III, but required an adapter which almost noonw bought given how small the Mk.III's market share was -a fraction of what it was even in the US), the Mega CD in late 1991 (really too late to jump in with NEC's fast growing CD-ROM market and hindered by feature creep -especially since few games took advantage of that coprocessing hardware).

However, all of that made sense and it wasn't until the Saturn and Mars/32x that things got confusing. (and a huge mess of management conflicts and many complex issues -some with too few details for a reasonably accurate account- and marketing/software problems all mounding up)

 

The 32x and Saturn were no more the cause of Sega's problems than ET or the 5200 had been for Atari Inc: they were symptoms of the real problems much deeper in the company. (and you had bigger issues like mismanagement of Genesis software and marketing in 1995/96, missing out on a strong late gen market, and a slough of other things too numerous and complex to begin to address here -biggest problems with the Saturn were lack of sonic games, lack of strong sports games and other Genesis staples, lack of strong marketing when really needed in mid/late 1996, and poor cost/performance ratio -heavily exacerbated by Sony's price dumping and vertical integration: sound familiar? -cough- Commodore price war -cough-)