Virtua Fighter 2 should have been a 32X game and not a Genesis game

[Team Andromeda]

Except it was SOJ who pushed project Mars in the first place... and SoA who'd already pushed at least 1 substantial proposal for a nextgen console (even if it wouldn't have necessarily matured until '95/96 -which would have been fine by SoA's standards )

SOJ also pushed for Jupiter at the start, but ideas and projects can get canned very early in. It was clear to anyone very early in, SEGA Japan were backing the Saturn .

he's tried many times to get interviews with a number of Japanese staff/former staff, but none have any interest at all

Sad, but the only man that can answer all the Consumer hardware questions would be Hideki Sato. He and he alone could clear everything up.

Having a company that just makes hardware and software and then throws it out to marketing to sell is hardly an efficient or realistic business model.

It happens in Every Company, So I really don't get your issue with SEGA at all. Some companies just have better PR machines like Apple.

That's something I'd like to have an answer to, there's quotes and paraphrasing from Joe Miller that points to SoJ being siad "Super MD" to the table in the January of 1994 meeting

The Super Mega Drive was just rumour, though I don't doubt SEGA Japan may have thought of one to counter act the Super Graff , But after that proved to be a horrible and expensive flop, I would imagine SEGA Japan went cold on the idea

That's a big fat assumption right there, but I agree that you can't simply take Tom's interview at face value

Tom's a typical American sales man: Can't take losing. I'll have far more time for the man if he just came out and admitted he made a bad call in backing and believing in the 32X. There is no shame, many thought it was a great idea at the time (developers and 3rd parties ECT)

I rather that to this bullshit that SONY had a chipset ready for SEGA

Nope, nothing to do with Sony Japan at all, totally an American thing. The whole thing was (supposedly) done directly between SoJ/STI and Imagesoft and there was never

Seeing as SONY America didn't have a R&D computer/console division, that would have been mighty clever for one, and 2 the Japanese Paymasters would have been pissed.
So it was never a goer at all was it, or are you making out SONY America had a Prototype Chipset with a CPU and GPU ? . I just don't believe

OK, so SoJ was the first to make that mistake.

SEGA Japan quickly dropped the 32X and Jupiter idea.
All Corps have multi projects and Back up plans don't they? Didn't Microsoft have at least 2 console projects in development , before going with X-Box Team

Like I said as Soon as SEGA America saw the full official unveiling of the Saturn in April 1994 , was the day they should have dropped all work onthe 32X.

[Chilly Willy]

I'll have far more time for the man if he just came out and admitted he made a bad call in backing and believing in the 32X.

It wasn't a bad call, it was SOJ's call. If it was bad, it was their fault for supporting it. They were the ones to green-light the 32X. Yes, SOA pushed for it, but that's their job - to push what they think is best for the American market. The 32X did great that first xmas, and would have continued to do great if SOJ hadn't pushed the devs to drop it to throw all resources behind the Saturn - which was also SOJ's call.

Like I said as Soon as SEGA America saw the full official unveiling of the Saturn in April 1994 , was the day they should have dropped all work onthe 32X.

So they should have abandoned a cheap add-on for the existing market of tens of millions for a $500 new console that was scheduled to be released almost two years away? You crazy, man!

[kool kitty89]

@kool kitty89

I really like SVP, the chip itself and the concept of addon enhancement chips. I just point there was lot of potential in MD+CD, and maybe even able to acomplish similar tasks than MD+SVP.

Yeah, but my point was that there was even more potential for MCD+SVP to build on. (the ASIC in the CD is already useful for helping with polygon rasterization and texture mapping, but the 68k is inefficient for completing the 3D capabilities while the SVP -or other DSP- is very fast and efficient at 3D math relatively speaking and the ASIC could offset a fair chunk of other overhead the SVP is charged with when used with the MD alone -especially drawing tasks where the ASIC would actually be better anyway)

That and they could probably have added the SVP to a duo system (or even later model MDs) without a big price mark up. (still should be cheaper than MD+CD alone, let alone MD+CD+SVP cart)

But it would complicate the market, but much less so than something more expensive and comprehensive like the 32x. (and would conflict less with the Saturn too as it would be less powerful -or at least lack the color enhancements- as well as being a much smaller investment)

anyway, CD launch was too early from my point of view. if it was done just to compete with Nec CD / Duo in Japan then it wasnt needed. but the real competition was the SNES, and from that aspect, I think SEGA was aware of that. but SEGA underestimated MD capabilities. the fact is GEN/MD sold pretty well in EU and US from 91->93. So a CD add on was not the priority in 91, at least in market dominance terms.

Actually it could have been too late too. Any later and it would have conflicted even more with the Saturn and it would have meant delaying Sega's foray into CD-ROM (a lot of the early multimedia stuff was a huge learning experience that could have paid off in the next gen -it did for Sony for sure). Unless they pushed back the Saturn for a streamlined and more powerful alternative. (not really a bad idea )

But looking from the Japanese perspective, a less powerful Sega CD (say, slightly better than the Super CD, maybe 512 kB program RAM bank switched to the 68k in the MD and instead of the added 68k, simple DMA audio -maybe just 8/10-bit stereo or something closer to 32x PWM- a minimalistic 8-bit MCU -maybe 6800 based to mesh more easily with the 68000 interface- managing the CD-ROM/cache and maybe also useful for doing software mixing of PCM -maybe ADPCM software decompression) and released in 1990 instead of '91. (so a full year ahead of the Super CD and still better in some respects)
That would have made it significantly cheaper/simpler (more feasible to release earlier too), and while it would lack the nice coprocessing of the MCD, those features were largely wasted anyway, especially early on. (FMV likely wouldn't be any worse either as the best compression schemes tended to rely on tile formatting catering to the MD VDP)
Maybe they could have added a low-cost off the shelf DSP to handle some of that too, but that could also have been added as a combo MD/MCD enhancement cart later on anyway. (like the SVP or predating the SVP for that matter)
Hell, instead of the poor man's mode 7 clone in Sonic CD, they could have had a CD FMV quality successor to Sonic 2's choppy/low res special stage.

So not only a critical earlier lead in Japan, but also a potentially earlier release in the US but at a fairly close price point (maybe $300-350 in 1991 -the CD drive would still be a major cost, though maybe if they switched to a lower cost flip top form factor earlier, that would help somewhat) and consistently cheaper than the historical MCD in parallel time (maybe $250 in 1992, $150-200 by mid/late '93, etc).
Not only that, but imagine how much more impressive Night Trap and Sewer Shark would have been in 1991 in the US (remember both were completed/edited in the late 80s and halted due to the cancellation of the planned platform -I do wonder if they ever considered going laserdisc prior to the MCD though).
And the whole FMV/multimedia thing could have started sooner and evolved sooner, not just getting to the point of increasingly better quality compression formatting, but also the use of FMV in games in general. (getting over the awkward/gimmicky hump sooner)
Again, FMV should have been little to no worse than it was. (maybe weaker for cases that relied on lossless compression -though if it was just RLE that wouldn't be too bad for the MD CPU anyway, or more simultaneous use of both 68ks, but for cutscenes it should be too different)

And not only that, but the limited interface of the side expansion port would have been much better suited to that sort of MCD anyway. (albeit a cart mounted CD unit would have tons of great possibilities including 32x-like functionality, and that's probably the direction they could have gone with a later release of the MCD as you say)


A lot of this got discussed here too (among other places)
http://sega-16.com/forum/showthread.php?t=13697 (mostly on page 1 though 4 with some on 5 mixed in with off topic stuff and a lot of rehashing within the same topic and a lot of rambling from me you should skim through if you decide to read it at all)
Both discussions on earlier and later MCDs, either could have been better as such as the existing MCD was neither here nor there in some respects. (albeit, rather than waiting longer, they could have just made it more expensive and powerful from the start, or even cut back in some areas to balance out some cost -definitely make it cart mounted and more compact though, and add a 256 color framebuffercoupled with the ASIC -maybe still retain support for the ASIC to render to MD layers too, but it would probably be cheaper to have it solely optimized for the 256 color bitmap layer)

One thing that came up there was doing both that simpler CD unit (or even simpler/earlier -like 256 kB in 1989 with slightly greater cuts in other areas too -not totally unrealistic since NEC had been promoting the PCE CD well before release and Sega could have responded more quickly).
But then also have a 32x-like cart module using hardware somewhat like the MCD (more powerful PCM chip, graphics ASIC coupled with a 256 color framebuffer indexed from 15-bit RGB -like 32x and SNES-, word RAM buffers, added fast CPU -though maybe a DSP coprocessor would have been more significant than another 68k)
Say released in 1992 in Japan and 1993 in the US for the booster module.

That last option (even cheaper CD and earlier and more hardware oriented 32x alternative) is maybe my personal favorite. Not only do they maintain an early launch of the CD unit in Japan and the west, but they later add more substantial graphics enhancement in general via the 32x-like booster unit. Hell, maybe include the full 512 kB+256 kB word RAM of the original MCD (word RAM would be needed in either case for framebuffers) in that as expansion to the 256 kB in the CD (and all of that could be flat mapped to the MD, no bank switching -aside form the framebuffers). I especially like the idea of omitting the added CPU (aside from a cheap MCU for the CD) in favor of a fairly general purpose DSP coprocessor (like SVP/Super FX, or various commercial DSPs -the motorola 56000 was pretty nice, but I don't think it was comparatively cheap, maybe one of TI's older DSPs or even of the graphics oriented DSPs like the TMS34010), not just for 3D stuff but also decompression or other things.

Now, you could potentially split the market with 2 add-ons, but also consider this: it could also have been more foolproof as if one add-on failed to catch on but the other did, they could push more for that one (and either would be cheaper than a CD incorporating all of that) though it could be tough if one region favored a different platform than another in an extreme manner. And, even if one took off faster, the other may have still been significant in the long run, let alone the potential to make a fully consolidated trio system (could have a CD duo prior to that, maybe even a neptune like boosted MD, but that's getting a bit complicated -though all of those could still be fully upgraded to trio standard).

And going back to the Saturn, they could either push the Saturn back in favor of a more powerful system (say more like a 1995/1996 version of the Dreamcast) and relying on the capabilities of the trio system (and plain MD/CD/booster) in the intrim, or if they did push for a full 5th gen console in '94/95, they could have directly built on the trio system too (not in an add-on manner mind you, but a far more efficient and comprehensive internal modification) making it into something closer to Saturn/PSX in overall capabilities.


For that matter, that's EXACTLY what NEC could/should have done except it was even dumber of them to not do it as, unlike the MD (or SNES even), the PCE's expansion port was extremely comprehensive and flexible with both the main and video buses (and full CPU address range unlike the MD sidecar bus), digital video output, analog RGB out, and stereo audio in/out, so even better than the MD's cart slot. (if the MD had the digital video bus from the VDP connected to the cart slot it would be almost as good -from a cost perspective, maybe they should have removed the side port and added SNES style outboard expansion pins on the cart slot)

NEC easily could have made the Super Grafx an add-on rather than only a whole new system and it might have been smartest to make it in a PCE CD-ROM tray interface form factor (ie so existing CD users could buy it and simply swap it for the original CD tray -or whatever it's called- mating the CD to the PCE -the only thing wasted would be the 64kB of DRAM of the original tray). Hell, maybe they could/should have released that in place of the Super CD in general (of course the super CD was in '91 vs '89 of the SGX). So maybe the SGX alternative could have added everything that the real SGX did but instead of the 32kB CPU SRAM, use 256 kB of DRAM (or 128k if they were really skimpy, but 256k should have paid off hugely in the long run -and wasn't that expensive in '89 either) to replace the 64k of the original CD as well. (so overall, you'd have 2 VDPs with 64 kB each, 8 kB internal PCE work RAM, 256 kB DRAM program/work RAM on the CPU bus, and 64 kB slow DRAM for ADPCM -and they probably could have afforded to leave ADPCM out of the original CD entirely for that matter and opted for more RAM instead, but that's a separate issue and actually meshes somewhat better with the replacement tray idea as you only lose that 64k and ADPCM+RAM is inside the CD drive unit itself)

And then release the duo with the SGX hardware built-in as well. Maybe add the arcade card too (or a more practical 512k or 1MB alternative to the arcade card), but that might not even matter much anyway.
The big thing would be well beyond that and more like the same idea with Sega:
Evolve the duo into a full nextgen platform that's backwards compatible as well. Honestly that's pretty much what the PCFX did except they ripped out the CPU and compatibility at the last minute (so to speak) for whatever reason and they didn't have a SGX userbase to build on. (the PCFX included the SGX hardware -sans the CPU- and added 2 critical features: 1 it bypassed the PCE color limitations and indexed using 24-bit RGB rather than 9-bit and 2 it allowed combining of layers to 8-bit 256 color tiles/sprites -though for practical purposes using the 16 color tiles/sprites would have been more practical and that's what the PSX/Saturn did for 2D to save space) The PCFX also added the somewhat Saturn VDP2 like 6 BG planes on top of all of that and that was probably overkill and going in the wrong direction in general. (if they'd added a single 256 color/highcolor/truecolor framebuffer layer and something like the Saturn's VDP1 -let alone something closer to the PSX GPU, the PCFX probably would have been a pretty nice 5th gen competitor and fully backwards compatible as well as smartly using the old hardware for enhanced 2D and the new hardware for further enhanced 2D and 3D and video decoding hardware that apparently kicked the PSX's ass too)

But that's way off topic still, though it's still interesting that you could potentially parallel Sega and NEC developments as such. (Nintendo could have gone that route too, but they didn't show remote interest in backwards compatibility -plus something based on the SNES efficiently would have been very different than the N64)

agree about the 32x/Sat in parallel. and it also confuses the market, in that time I remember my friends thinking that MD+32x+CD = Sat
from what I read, SEGA didnt know about the PSX power, at least when the first Sat was designed arround 1994 (the version with just 1 SH1 CPU). but as kk89 stated, SEGA did know PSX HW specs by 32x launch.

No, the earlier Saturn (or whatever code name) consoles would have been in '92 and '93, the final redesign of the Saturn began at the end of 1993 but we don't know what the state of the System was by that point. (some magazine articles in addition to internal industry speculation -in programming/engineering circles- seem to point to it maybe using an SH1 CPU and possibly lacking VDP1 or having a simpler predecessor to VDP1 -lacking VDP1 entirely would have made it more like the PCFX.
And assuming that premise (could be totally off but there is evidence for it) the final revision of the Saturn would have been pushing the SH1 to CD-ROM duty only (presumably originally shared for CD like the MCD 68k), increasing RAM, and adding or enhancing VDP1 with 3D quad rasterization (if it had been more like the MCD ASIC originally, quad rasterization would have been much easier to add than triangle -and made sense from Sega's arcade stuff too given they went the odd route with quads -3DO did too), and added dual SH2s and maybe the DSP in the SCU as well. (that might have been there originally)
Things that support this premise would be the transparency bugs of VDP1 being limited to 3D mode only, the SH1 being in the system at all (very wasteful in its role), among other things. (hardly definitive though, just speculation, but it makes a lot more sense than Sega intentionally designing the Saturn to be as inefficient as it is)

so in that scenario, SEGA already had a complete Sat redesign, at least on paper. they KNEW the battle was going to be against SONY with high end new technology. so why to mantain the pretty old tech with all the flaws of the ancient MD design?

It would have never have been a complete redesign. Engineering a system that complex takes years from paper to a workable production design even with big budgets (with smaller budget you have cases like the Jaguar going from paper in 1990 to mass production -with bugs- in 1993 -preproduction in late 1992 iirc).
Sega probably had parallel designs going on in the mid 90s (I wouldn't be surprised if at some point they'd pushed for MD/CD compatibility), but if the redesign thing IS true it probably took place somewhat like the above premise of:

Weaker mid 1993 Saturn prototypes with fully completed audio, CD-ROM, and VDP2 subsystems, no SH2s (dual duty SH1), weaker or no VDP1 (ie VDP2 would be the only VDP), and maybe no SCU DSP. (if they had both the SCU and weaker VDP1, it probably wouldn't have been terrible at 3D as the DSP could help with both 3D math and triangle rasterization alongside the SH1 and VDP1 would handle line filling of textures and flat or smooth shading and probably managed unscaled/scaled/rotated sprite objects similarly to the final VDP1; but with no VDP1 or DSP it would have been really weak in 3D, worse than the 32x)
And probably less RAM too.

And such a late and somewhat hacked together redesign really would explain the low cost efficiency of certain aspects of the Saturn, albeit they probably could have pushed both for more 3D power and a low price point by tempering what features were added and especially how much RAM was used. (cutting back to under 3 MB would have been substantial, getting it down to 2.5 MB would have been great, and maybe dropping an SH2 or preferably the SH1 -or if they had to use the same CD interface, maybe hack in the slave SH2 in place of the SH1 and allow it to still help the main CPU too) -And push for more RAM expansion after the fact.
Lots of other options, but it's pretty clear that Sega wasn't concerned enough about cost efficiency in general, regardless of 3D emphasis. (even if they wanted to push 3D, time was really limited by late '93 for the planned fall of 1994 release date -albeit on a cost saving side they could have scrapped VDP2 in favor of integrating simple bitmap/framebuffer functionality to VDP1, but that would have made it much weaker than the PSX in 2D)

my preferred sceranio is a late CD release in 1993, with decent DSP (a 10Mhz SVP was possible on that time) + RAM. drop the 32x thing and go for a fairly polished Sat in 1995: 1cpu (sh3 was available) + enhanced DSP, similar style to what SONY was doing.

See above, I've given that a lot of thought and there's a number of alternate scenarios. (I really tend to favor the early/cheaper CD and separate 32-x like booster in '92/93 as such, or skipping the booster in favor of a full 5th gen console in Japan in '93 and US in '94, but that's more extreme) Incremental upgrades may have been rarely executed well on consoles, but that hardly means they're a bad idea.

Edit: As for the 10 MHz SVP alternative, not necessarily that slow either, it would really depend on the case (you could rather easily have a 25-50 MHz DSP that was still an older/less efficient design and not doing as much per clock cycle -some of the older/cheaper TI examples would be that), clock speed can be a bit ambiguous. (the SSP-1601 used in the SVP was rated for 25 MHz at launch)

And on the SH3 thing, yeah there's been a few discussions on that too (especially a dreamcast like system on mid 90s tech -dreamcast in terms of streamlined and focused architecture), actually that came up a few times in the above linked alternate realities thread.

PD: didnt know about the posibility of a SEGA+SONY partnership, interesting to know. if you have more info please share

Read:
http://www.sega-16.com/forum/showthread.php?t=1410

[Team Andromeda]

It wasn't a bad call, it was SOJ's call. If it was bad, it was their fault for supporting it. They were the ones to green-light the 32X.

At the start it would have been, but after the failure of the 3DO, the lack of love and support for Jaguar and entry of SONY into the market, it was a mistake by SOA to continue with the 32X, and a mistake by SOJ to think SOA/SOE knew best

So they should have abandoned a cheap add-on for the existing market of tens of millions for a $500 new console that was scheduled to be released almost two years away? You crazy, man!

1) It was far from cheap : 32X cost over £170, came with no game, and each games cost £60 each: how anyone can make out that is cheap, I do not know (that's with out adding the cost of the Mega Drive)

2) SOA had 1 development studio, SOJ had multiple and SOJ were backing the Saturn (and like or not SOJ tented to make the best games ) . Not hard to figure out which machine would enjoy the best games and the most Arcade conversions is it ?, Never mind CD-Rom was quickly becoming the standard medium for games .

[kool kitty89]

The difference was that SEGA was trying to maintain their high-price arcades. If they made a cheap GOOD 3D home console, they may have felt it would cut their income from the arcades. Sony had no such reason to hold back on a home console and didn't.

I'm not sure if that was the case.
It seems much more likely that they started with a much more 2D oriented design and then kept adding things on with a particularly heavy push at the end for more power in 3D (and general processing) without sensible regard to cost or ease of development. (I mean in some cases you can have lower cost and more power at the expense of ease of development, and that has some merit -though with the shift to high level programming in the mid 90s, not such a good idea- but they didn't even push for good cost effectiveness)

I mean if they were stuck in a position where they (or SoJ at least) was totally adamant about a 1994 Saturn release and thus the only options were rapid redesigns/hacks to the existing mid/late 1993 design(s) being developed, there's a lot of areas they could have pushed much more for low cost in general. We've already gone through all the unused/underused/wasteful design aspects and possibilities to temper RAM use better, but one other thing (if they really recognized both the sheer priority of 3D and low cost) might have been to chop out VDP2 entirely.
However, that would have really made the Saturn almost totally inferior to the PSX but with the possibility to undercut the price even without the funding and vertical integration of Sony (albeit inferior is how the mass market ended up seeing the Saturn in the west anyway and only a hardcore niche sector of the market really cared about the 2D advantages, so it might not have even been that extreme of a difference).

And that's assuming that they couldn't have actually beefed up VDP1 some more (beyond obviously adding the VDC logic needed to replace VDP2 -ie something close to the Super VDP) with buffering/caching for higher bandwidth in general. (I think it worked on a 16-bit SDRAM bus -both for source and destination, so adding larger phrase buffers for 32-bits or 64 bits could have been the start, and the nest step would have been line buffers or an actual GPU cache and/or texture cache to allow it to work efficiently in a single unified bank of video RAM and cut cost further -no need for the separate framebuffer bus/banks and could have a unified 1 MB of RAM more like the PSX)
But I doubt they had time to do that for a fall of 1994 launch. (especially assuming even the warped quad feature was hacked in at the last minute)
Plenty of options well beyond any of that if they pushed the release back another 6 months to a year. (including using an SH3 with FPU and/or using triangles in hardware -FPU would mean much less headache for quads as you mentioned though)

It wasn't a bad call, it was SOJ's call. If it was bad, it was their fault for supporting it. They were the ones to green-light the 32X. Yes, SOA pushed for it, but that's their job - to push what they think is best for the American market. The 32X did great that first xmas, and would have continued to do great if SOJ hadn't pushed the devs to drop it to throw all resources behind the Saturn - which was also SOJ's call.

It was also SoJ's call to push the Mars project in the first place before greenlighting the alternative SoA compromised with.

And also supported the development of the Saturn apparently without any realistic cap on development costs for the mass market, then again the JP market tends to be less price sensitive so they may not have cared... unless they'd taken the westerm markets as the main priority.

So they should have abandoned a cheap add-on for the existing market of tens of millions for a $500 new console that was scheduled to be released almost two years away? You crazy, man!

Or better yet, no add-on at all and a fully backwards compatible console that was similarly powerful (as far as the public was concerned) but much more affordable than the Saturn, or simply more affordable in general regardless of compatibility. (or hold back the Saturn for a more comprehensive redesign that made it more suited to the worldwide market in general -in terms of capabilities, cost, and programmability- at the expense of a later release and higher R&D costs)

And all the while supporting the existing MD/CD userbase correspondingly. (maybe releasing an even more practical/affordable add-on than the 32x, less powerful but more realistic to expect a mass market -and not niche- adoption rate and also conflict less with the Saturn, or only release on-cart enhanced games. (if they could manage the right marketing, an add-on should have been far more popular though -they just needed to make consumers realize that it would pay for itself compared to tacking stuff onto carts -aside from really low cost enhancements more easily embedded in carts alone, though those too could have been relegated to the add-on platform instead)

Hell, if such a cheap add-on had been powerful enough to pull of Virtua Fighter reasonably close to what the 32x (or original Saturn game) managed, that might have held the fort in Japan too. (especially if embedded into a new console as Mars was originally posed as)

I'm not sure exactly how much SoA understood about the Saturn's release timeline (or whether it was SoA or SoJ who had control over when the US launch occurred for that matter), but it would seem to me that they'd have pushed for a simpler/sooner/cheaper alternative to 32x if they fully understood what was going on with the Saturn. (assuming they couldn't do anything to impact SoJ's finalized design configuration and the impractical cost -technically not a massive production/material cost difference from the PSX, but Sony's vertical integration and greater ability to sell at a loss and remain stable inflated that problem substantially -and Sega should have realized what they were up against, unless they were banking on Sony making the sort of mistakes NEC did in the west, which would not have been proactive at all)
Hell even if they couldn't push for hardware changes (cut down RAM would be the easiest thing before actual production started gearing up -also comething they already had OK provisions to expand too), SoA at very least could have pushed for software solutions to make development more attractive (but that would assume SoA noted that problem) and push SoJ for better hardware docs. (as well as investing in expanded SDKs based on the comprehensive low-level docs -improving high level support as best they could with the tight architecture, good low level docs and workarounds for any bugs or oddities would also be significant -one thing I think may never have been used is a SCU DSP program to aid in triangle rasterization coupled with VDP1 line rendering)

[kool kitty89]

SOJ also pushed for Jupiter at the start, but ideas and projects can get canned very early in. It was clear to anyone very early in, SEGA Japan were backing the Saturn .

Really? Where did you find this out? (were there direct interviews from SoJ and SoA staff regarding the Jupiter -let alone proper descriptions of the Jupiter hardware -most magazine clips I've seen are disjointed and boarderline nonsensical in what Jupiter really was to be)

Jupiter would have been FAR better than 32x/Mars IMO, especially as it could have alleviated the mad dash with the Saturn.
However, after furhter thought in some recent threads (including your own points about the bad PR of switching to carts after pushing the CD), it would have been smarter to have a jupiter CD instead, maybe even more cut back in some areas than the plain Jupiter, but similarly expandable to Saturn spec. (albeit one problem would be the fundamentally inefficient/exorbitant CD-ROM interface/subsystem used in the Saturn that would have needed to be corrected for such to be feasible -and by extension also making the Saturn cheaper)
There might have been some decent last minute (ie started by the beginning of 1994 while the Saturn was undergoing its final revision of hardware before being frozen for production), maybe swapping the slave SH2 into the SH1's role and removing the SH1 entirely (making the slave dual duty). They should have had a much more cost effective CD interface from the start (maybe based on the MCD but clock doubled for 2x CD-ROM), but at the 11th hour other hacks would probably have been the only option.
Then there's various areas where RAM and unnecessary features (like the audio DSP) could/should have been cut out too, again not all necessarily feasible given time constraints. (cutting down RAM at least should have been feasible -if they extended the cart bus to 32-bits that would have made it even better for expansion, maybe tweak it more for cheaper/easier addition of 28.6 MHz SDRAM)

If no such hacks could be completed in the desired timeframe, a cart based jupiter would have been preferable. (maybe cut to 512 kB 32-bit main SDRAM RAM and keep audio and video RAM alone -for a grand total of 2.5 MB to the Saturn's 4.5 MB -not counting the SRAM- and then there's the SH1 and such)

Sad, but the only man that can answer all the Consumer hardware questions would be Hideki Sato. He and he alone could clear everything up.

I doubt he could answer all the questions that were SoA specific, let aloen those that were more management related.

It happens in Every Company, So I really don't get your issue with SEGA at all. Some companies just have better PR machines like Apple.

No, soem companies actually have consumer product divisions that focus on trying to formulate success based on market trends and predicted demands. Successful application of such depends on a number of factors including accurate information about the market itself. (that's what hurt Atari Inc -and the early 80s NA game market in general- more than anything else, and arguably Sega as well -both had heavy division with dual management issues and conflicts of interest as well, along with infighting)

Some companies can get away with strong PR alone, others can't. (especially when they come up against a company with advantages in pretty much every category -not only good PR, but good management all around, a killer product, and tons of money to back it up, like Sony -NEC screwed up in that sense as they were not in a dissimilar position to Sony in 1988/1989 with the PCE, and had Sony screwed up like them, Sega and Nintendo would both have been much more competitive -even 3DO and Atari could have been)


Kalinske (and Katz before him) were the ones who provided that critical PR, but they also had the fortune to have products that fit the market well, unlike the Saturn

The Super Mega Drive was just rumour, though I don't doubt SEGA Japan may have thought of one to counter act the Super Graff , But after that proved to be a horrible and expensive flop, I would imagine SEGA Japan went cold on the idea

Funny, I got more the impression that the Jupiter was a rumor than the Mars project. (Sato's initial design before 32x was created by SoA to counter the less desirable option)
Pettus's article is too poorly cited to be definitive, but it does explicitly claim that Miller stated in an interview:
http://www.sega-16.com/feature_page....%20a%20Failure
The system that would be known as Project Mars was given birth on 8 January 1994, the night before the opening of the 1994 Winter CES in Las Vegas, Nevada, in a hotel room during a conference among top-level Sega executives from both Japan and America. Those present at this meeting included Sega CEO Hayao Nakayama, Sega of America president Tom Kalinske, his special assistant Joe Miller, Sega of Japan's Hideki Sato, Sega of America's Paul Rioux, and a couple of other Sega of Japan personnel. Surprisingly enough, Nakayama was the one who first broached the subject at this meeting. As such, it is he and not Sega of America's Joe Miller who should be given credit as being "the father of the 32X." Miller remembers this meeting well.

Quite simply, Nakayama-san had directed the company to design and produce a cartridge-based 32-bit platform and bring it to market before the Christmas selling season of 1994. This was a lengthy, somewhat heated meeting - but in the end there was no question that Sega of Japan (in the form of a classic Nakayama mandate) had determined that this was what we were going to do. It was [now] up to the senior team to figure out and go execute. The difference, this time, was that Sega of Japan was actually inviting Sega of America into the process - instead of creating new platforms in a vacuum and throwing them over the ocean at us when it was too late to have meaningful input ....Sega of Japan was completely committed and was [ready to] mobilize whatever internal resources were require to finish the design and produce it in quantity for Christmas.

As first presented by Hideki Sato and his team of engineers, the original concept for Mars was little more than a Genesis with an extra 32-bit processor (a Hitachi SH-1, according to some reports) and an expanded color palette (128 out of 512 possible colors on screen). Joe Miller, who was in fact chief technical wizard at Sega of America, was appalled at the suggestion. "That is a horrible idea," he told them. "If all you're going to do is enhance the system, you should make an add-on. If it's a new system with legitimate software, great. But if the only thing it does is double the colors ...." There was some grumbling about this, but in the end Sega of Japan conceeded the point. They had several other hardware projects in the works, so this one was to be left up to the Americans. Mars was to be Sega of America's baby, although senior management staff from Sega of Japan would be present and oversee it through to production. By the time all was said and done that could be accomplished at that meeting, Nakayama was so excited at the prospect of Project Mars that he wanted its "core senior design team" to leave CES before it had even started and get started working on the new system right away. Miller, Sato, and the rest wound up attending the rest of the show, but went ahead and began the process during a series of late night meetings in Miller's hotel room over the next four days.

I know there's inaccuracies in that article, and this may be no exception, but it's certainly something that Joe Miller could confirm or deny. (though it might fall under the "no discussing 32x" condition of any interviews -apparently he's tired out of rehashing things)
Or maybe Pettus has the original interview on hand still and could properly cite it and post it in its entirety.

Tom's a typical American sales man: Can't take losing. I'll have far more time for the man if he just came out and admitted he made a bad call in backing and believing in the 32X. There is no shame, many thought it was a great idea at the time (developers and 3rd parties ECT)

Sure, but that doesn't mean it was even his fault as such.
He could be BSing or he could be rightfully frustrated over it depending on the actual context.

I rather that to this bullshit that SONY had a chipset ready for SEGA

No one ever said anything like that at all, you're not reading the interview properly or my summary of it.

Seeing as SONY America didn't have a R&D computer/console division, that would have been mighty clever for one, and 2 the Japanese Paymasters would have been pissed.
So it was never a goer at all was it, or are you making out SONY America had a Prototype Chipset with a CPU and GPU ? . I just don't believe

Nope, try again, no solid hardware, jsut specifications, or even a looser wish list of features compiled by software developers and hardware engineers (possibly with some input from programmers who had hardware engineering experience too -it wouldn't be the first time a software house came out with hardware -the tiny garage company GCC pushed out the 7800 in under a year in the early 80s and Hudson designed the PC Engine, but that's not even what Kalinske was claiming).

The premise was the partnership was between Sony's US consumer SOFTWARE entertainment division (Imagesoft and Sony Interactive Entertainment, Inc -who'd collaborated with STI on some MCD projects, and that's the main factor that spurred the hardware proposal).

Again the premise was to have a multimedia machine built on the fundamental learning experience of early multimedia as well as rising industry/software development trends and to apply that to a list of features desirable in a nextgen console. (have that compiled from a programming and engineering perspective and also put in "plain English" so that management and marketing could understand what it meant)

The proposal would then be sent to SoJ and Sony headquarters along with the corresponding proposal for a joint console development project. THEN the 2 companies were to begin actual hardware design in collaboration for such a system and share R&D expenses and later marketing/distribution costs and profits while keeping licensing and publishing deals totally separate. (it's unclear if Sega would actually have had rights to license 3rd parties or only act as a publisher for the platform -given what Sony did to Nintedno, I wouldn't be surprised if it was the latter)
Had such gone through it would more than likely have been the case that such specs would be applied to existing hardware in development at Sega and/or Sony, but with shifts in the design based on the input. (and whether or not this had anything to do with it, I'll bet Sony was working on the PSX hardware -may have been several separate projects that later merged too- long before they scrapped the SFC based CD unit in early 1993)

SEGA Japan quickly dropped the 32X and Jupiter idea.
All Corps have multi projects and Back up plans don't they? Didn't Microsoft have at least 2 console projects in development , before going with X-Box Team

Yeah and I bet there were canceled alternative designs to the MD too, or Saturn, or in-house alternatives to the N64, etc. (Atari's 3200 is one case where the design was better than what they actually replaced it with by the looks of it -but the 5200 was chosen to accelerate development and rush to meet mounting competition)

However, when a project is canceled, it's usually canceled, not allowed to continue. And don't tell mean SoJ had no control over it either, if plans had changed for the Saturn, they danm well should have notified SoA immediately and forced the cancellation of the 32x. (or maybe compromised on a cheaper/simpler design that could not only be released earlier, but have a higher adoption rate and conflict far less with the Saturn due to the low cost -less investment out of pocket- and weaker capabilities -making the Saturn more attractive, with the lower end add-on mainly catering to users who would definitely not be getting a Saturn until 1996 or later regardless)

Like I said as Soon as SEGA America saw the full official unveiling of the Saturn in April 1994 , was the day they should have dropped all work onthe 32X.

Yes, assuming that was an actual surprise and if they thought anything had changed from January when the 32x was initiated.
If SoJ didn't explicitly state a change in priority (if there really was one -and they hadn't planned to push 32x alongside Saturn all along), then they needed to make damn sure SoA understood that and forced them to drop the design or make another compromise to better mesh with the change as mentioned above.

SoJ made other bad decisions with the design and management of the Saturn (including software direction, documentation, and design cost -let alone streamlining for better programmability), but the conflict with 32x exacerbated all of that.

And then there's the apparent event of SoJ taking over after the fact and pushing SoA management to make odd choices regarding Saturn (and western software and hardware in general), but that's another thing that's not totally verified. If it is true it's quite ironic that SoJ decided to not push the 32x's cancellation (or potential less conflicting compromise) back in April of 1994, but then came in and made a bigger mess by not allowing SoA to make all the decisions on when and how to market the Saturn and what to do with the 32x, CD, and Genesis during the transition. (there's absolutely no reason they needed to cut the 32x off so abruptly, or the GG, or shift the MD like the did -or even the CD to some extent-, let alone not launching the Saturn in September as planned -about a week ahead of the PSX)


Edit: the Kailinske discussion moved here:
http://www.sega-16.com/forum/showthr...750#post335750










EDIT

At the start it would have been, but after the failure of the 3DO, the lack of love and support for Jaguar and entry of SONY into the market, it was a mistake by SOA to continue with the 32X, and a mistake by SOJ to think SOA/SOE knew best

None of that would have been clear until fall/winter of 1994, if not later. (3DO wouldn't be clear until after the PSX was released, Jaguar probably by spring of 1995 -1994 was the Jag's first holiday sales season, 1993 had only been an extremely limited test market -not unlike Nintendo in 1985 and early 1986, so basing anything on that would have been premature)

1) It was far from cheap : 32X cost over £170, came with no game, and each games cost £60 each: how anyone can make out that is cheap, I do not know (that's with out adding the cost of the Mega Drive)

OK, so SoE either screwed up or it was simply unavoidable due to the cost inflation of Europe in general. (what was the 1995 Saturn launch price in the UK?)

In the US the 32x launched at $160 US with a $60 value coupon book giving $10 off the first 6 32x games released. The price was dropped to $100 at the beginning of 1995 and that's the price it was at when the Saturn made it's bungled launch in May at $400 for the bare bones system and often $500+ for many bundles retailers were offering. (I wouldn't be surprised if the 32x continued to outsell the Saturn up until the discontinuation, or at least until fall of 1995 when the Saturn dropped to $300 in the cheapest configuration)

Even without the 32x the Saturn would have been a mess in the US, albeit not quite as big of one.

2) SOA had 1 development studio, SOJ had multiple and SOJ were backing the Saturn (and like or not SOJ tented to make the best games ) . Not hard to figure out which machine would enjoy the best games and the most Arcade conversions is it ?, Never mind CD-Rom was quickly becoming the standard medium for games .

If you consider STI one studio, then yes, but I'm not sure that's really accurate unless the many different internal segments (larger than just separate dev teams) were subdivisions of the same studio. (I also wouldn't doubt STI's staff was smaller than SoJ's 1st party console only staff -let alone the arcade divisions- but they were growing all the time and it would certainly be interesting to compare the sizes with departmental breakdowns)

It also seems to be the case that STI (and several 2nd party Sega published/collaborated examples) were making up a huge segment of SoA's sales, if not the majority of 1st party sales in the US (or the west in general) in the latter part of the MD's life. (ie after 1993) Then again, Sonic 2, 3, and Knuckles were all developed at STI, so that's a massive chunk inflating things right there. (they may have been heavily developed by former SoJ staff, but it was all done within STI)

[Team Andromeda]

Really? Where did you find this out? (were there direct interviews from SoJ and SoA staff regarding the Jupiter -let alone proper descriptions of the Jupiter hardware -most magazine clips I've seen are disjointed and boarderline nonsensical in what Jupiter really was to be)

Yes there were: though it was strictly off the record. Jupiter was dumped and worked pushed ahead on the ST-V Cart basted system.

Jupiter would have been FAR better than 32x/Mars IMO, especially as it could have alleviated the mad dash with the Saturn

It would have bee a Saturn minus the CD Drive, Less Total Ram and SH-1 chip, kind of like the ST-V (just on a lower scale) . It's not that hard to figure out , is it ?

I doubt he could answer all the questions that were SoA specific,

What are you on about. He was hardware chief of all SEGA Consumer Hardware products, and you can bet your bottom dollar, that any hardware development (even those with only for/or in America) would have all had to been approved and green lighted by SOJ.

No, soem companies actually have consumer product divisions that focus on trying to formulate success based on market trends and predicted demands.

You are just being silly.... No developers makes a game to flop, they all set out to make a winner/ and the best game they can make and one they think people will want to play.; All developers with have their own PR divisions and various focus groups data , trying to predict what the next big 'thing' we be

That happens in more or less every gaming corp would could ever care to mention .

Funny, I got more the impression that the Jupiter was a rumor than the Mars project

The Super Mega Drive had nothing to do with the Jupiter project.

I know there's inaccuracies in that article

Yes they are and quite a telling one . SEGA America were the 1st to show off the Saturn hardware and games (panzer Dragoon, Victory Goal, Clockwork Knight ECT) at the start of 1994 in January on the same day as Mars went into the development. Kind of blows SEGA Japan keeping its 32bit ace up its sleeve for as long as it could, don't you think ? Never mind the Mars project used the Saturn Twin SH-2's

Sure, but that doesn't mean it was even his fault as such

He backed the 32X he said it would sell more units than the PS and Saturn and that it was the 'Only Mass Market 32 Bit system'
You can bet your life it's Tom Fault, just like to be fair it was Tom's fault that SEGA did rather well in the 16 bit days

No one ever said anything like that at all, you're not reading the interview properly or my summary of it

I'm reading you are just not being realistic at all. For 1). SONY America had no Hardware R&D (for consoles) and SONY Japan is harldy going to say ..oh look SONY America is spending millions , but they'll not telling us what on.

2) no-one invests in a Major chipset with out some prototypes or at least some sort of Track record.
Console chipset R&D runs in the millions (then, never mind now)

So no I don't buy TOM's bullsh8t at all.

OK, so SoE either screwed up or it was simply unavoidable due to the cost inflation of Europe in general. (what was the 1995 Saturn launch price in the UK?)

SEGA America screwed up too. What was the cost of a Launch 32X game Vs a Launch Saturn game in the USA ? , what was the Launch Price of the Saturn in the USA Vs that of the Launch price of the 32X ? .

I bet the price differences between the games and Hardware are similar levels to that of the UK . So don't even try and play the Inflation game

Nope, try again, no solid hardware, jsut specifications, or even a looser wish list of features compiled by software developers and hardware engineers

vapour ware ? Oh sure SEGA was going to invest millions in that, never mind if the team could deliver the Chipset on time and to cost ECT

Yeah and I bet there were canceled alternative designs to the MD too, or Saturn, or in-house alternatives to the N64, etc

No doubt there all Corps do it, even Atari (Jaguar came from the Panther) . So what's the issue with SEGA then ?

Yes, assuming that was an actual surprise and if they thought anything had changed from January when the 32x was initiated.

Are you for real?. I'll say again for the very last time. SEGA America showed off the Saturn at the Consumer Electronics Show at Last Vegas at the 6th of January 1994 .

[IMG][URL=http://img717.imageshack.us/i/satn.jpg/]



Surprise my Foot !

[Chilly Willy]

The surprise was how fast SOJ rushed the Saturn to market. It came out with a couple games late in 94 in Japan, and mid-95 in the US with virtually no software at all! THAT was the surprise.

[kool kitty89]

Yes there were: though it was strictly off the record. Jupiter was dumped and worked pushed ahead on the ST-V Cart basted system.

A damn Shame SoA didn't keep pushing for it, a much better idea than Mars, at least considering the Saturn timeline and how the 32x ended up a bit rushed and underpowered itself. (or neither here nor there even if the Saturn had been later, too expensive to make a really cheap accessory, but too rushed and weak in certain areas to really make a good competitor in the 5th gen market -if they'd optimized it a bit more, made it fully asynchronous from the MD with its own clock like the MCD, chucked and SH2, and added VDP capabilities at least on par with the MCD, better if more like the Saturn's VDP1, and added more RAM -and used cheap DRAM on a 32-bit bus- it could have been something, but still moot if the Saturn was launched as it was -again in that case a much simpler add-on or none at all would have been better) OTOH, a cost cut Saturn or a new system built on the MD/CD (and compatible) with cost effective evolution to saturn-like spec would have been better than any of that. (the latter implies a totally different design goal than what SoJ's Saturn was, or similar goals but with a different concept and the added goal of compatibility)
Sort of like this: http://www.sega-16.com/forum/showthr...847#post324847 (he left out the part about using dual MD VDPs, but that was implied due to previous discussions detailing that)

That's one thing I really wonder about Sato's supposed original Mars design (or "Super MD"), whether it was actually closer to that in some respects or not.

It would have bee a Saturn minus the CD Drive, Less Total Ram and SH-1 chip, kind of like the ST-V (just on a lower scale) . It's not that hard to figure out , is it ?

Well no, but that's a lot of misinformation (confusion of Jupiter and Neptune for one), and the actual configuration is still vague.
I mean how much less RAM and what RAM are we talking about? Did they just remove the Low RAM (slow 1 MB DRAM) and SH-1 RAM leaving the system with 3 MB (2.5 MB SDRAM 512k DRAM for audio), or did they cut main RAM down further to 512k or other things? (the more they cut, the more practical it could have been as a low cost complement to the Saturn)

OTOH if they'd just made the Saturn architecture less bloated and wasteful (especially the CD-ROM interface) and cut back on RAM, that would have made more sense as you've still got CD-ROM but at a price not too far above what the Jupiter could have been. (and instead of a CD+RAM+SH-1 add-on, they could just have a RAM add-on)

What are you on about. He was hardware chief of all SEGA Consumer Hardware products, and you can bet your bottom dollar, that any hardware development (even those with only for/or in America) would have all had to been approved and green lighted by SOJ.

Yeah, but I wasn't just talking about hardware stuff, but political/bureaucratic/management level stuff as well as some of the software stuff. (Miller was involved on multiple fronts in that respect at SoA and could confirm claims kalinske made that Sato couldn't)

OTOH, Sato could definitely answer all questions about any canceled prototype designs and revisions to the Saturn. (and almost certainly on the SGI thing too)

You are just being silly.... No developers makes a game to flop, they all set out to make a winner/ and the best game they can make and one they think people will want to play.; All developers with have their own PR divisions and various focus groups data , trying to predict what the next big 'thing' we be

Nope, but they could cater to the wrong market... which is what many claim to be a major problem with the Saturn, way too much emphasis on the Japanese market and not meshing with the west -especially the US, the market they should have had the highest priority on bar none -from a practial business perspective. (and arguably stifling STI as well)

The Super Mega Drive had nothing to do with the Jupiter project

Yet, in some respects it could have been a viable alternative. (solid point nonetheless)

Yes they are and quite a telling one . SEGA America were the 1st to show off the Saturn hardware and games (panzer Dragoon, Victory Goal, Clockwork Knight ECT) at the start of 1994 in January on the same day as Mars went into the development. Kind of blows SEGA Japan keeping its 32bit ace up its sleeve for as long as it could, don't you think ? Never mind the Mars project used the Saturn Twin SH-2's

Which is why I ignore those parts of articles, if I discounted every article based on partial inaccuracy I would have totally discounted almost every single video game related magazine and online article released and most reference books for that matter. (including every single one you've posted on this site)
It's trying to pick things out that seem like relatively valis paraphrasing or direct quotes that's worthwhile.

He backed the 32X he said it would sell more units than the PS and Saturn and that it was the 'Only Mass Market 32 Bit system'
You can bet your life it's Tom Fault, just like to be fair it was Tom's fault that SEGA did rather well in the 16 bit days

And he was right, or could have been if they followed through. (for the timeline the 32x was to be pushed at least ie 1994-1996)
The PSX and Saturn were niche until the end of 1996, then again, selling better than the PSX and Saturn up to late 1996 could still have meant selling worse than the Genesis or SNES.
Likewise that doesn't mean there weren't better options than either Saturn or 32x. (both being pretty flawed and undesirable for the market at the time, 32x being too limited and underdeveloped and Saturn being too sloppy/bloated and correspondingly expensive -and less than convenient to program for)

I'm reading you are just not being realistic at all. For 1). SONY America had no Hardware R&D (for consoles) and SONY Japan is harldy going to say ..oh look SONY America is spending millions , but they'll not telling us what on.

Nope, you still aren't reading my posts. It would have been the Japanese companies developing the hardware, but with western input. (too bad Sega didn't do that with the Saturn, would have been a far better idea than spinning off mars)

2) no-one invests in a Major chipset with out some prototypes or at least some sort of Track record.
Console chipset R&D runs in the millions (then, never mind now)

Hardly contradicts what I said, the idea was to BUILD a new chipset based on the desired features (based on the noted weak points of current hardware for multimedia and other growing market trends). Sony and Sega of Japan most likely wouldn't have started from scratch either, but used existing designs as the starting point. (I bet the PSX hardware didn't start out as a game console originally, or at least not parts of it -wouldn't be surprised if it was a hybrid of 2 or 3 different projects, maybe more, especially given the very short time period from when the finally scrapped the SFC based CD system towards the end of '92 -maybe early '93 not sure- and when the PSX hardware was frozen for production -especially given how clean and bug-free it was)

So no I don't buy TOM's bullsh8t at all.

Seeing as you obviously haven't even read through it properly (or my own summaries) that's hardly a valid conclusion anyway. (could still be BS but perfectly believable -maybe not the point about Sony of Japan being all for it, but the rest- and that's something Joe Miller certainly could confirm)

SEGA America screwed up too. What was the cost of a Launch 32X game Vs a Launch Saturn game in the USA ? , what was the Launch Price of the Saturn in the USA Vs that of the Launch price of the 32X ? .

And? The 32x's price was fine for what it was (it was flawed and underdeveloped though and could have been much more for the same cost, the Saturn was heavily flawed as well though -as above).

The Saturn price was really impractical though and they squandered hype with the early release.
I'm also unsure of why they didn't quickly push ongoing 32x development over to Saturn when priorities shifted, let alone not porting existing 32x games over to the Saturn with enhancments. (hell, even the porting process to Saturn would enhance many games due to the sheer bandwidth available with the RAM and faster CPUs, let alone using added video hardware and audio -in the latter case even more extreme given the lack of use of the DMA mode on the 32x's PWM)

I bet the price differences between the games and Hardware are similar levels to that of the UK . So don't even try and play the Inflation game

If that's true, I don't get it. The prices seemed reasonable enough in the US. ($160 suggested retail -apparently $150 at some stores- with $50 for the first 6 games released with rebate coupons used -nominal price of $60 otherwise)
I mean they could have managed a better band for the buck with other things, but the same is true for the Saturn.

vapour ware ? Oh sure SEGA was going to invest millions in that, never mind if the team could deliver the Chipset on time and to cost ECT

well yeah, anything is vaporware if it doesn't make it to market.
And it was SEGA (or STI/SoA) that pushed that in the first place coupled with Imagesoft's input.
And again, they wouldn't start from nothing, but probably (sony and/or SoJ) would have existing hardware to build on as they both obviously did. (ie could have taken parts of the Saturn and/or PSX designed and molded them to fit the findings/features of SoA/Imagesoft's report/list, or even combine parts of the PSX and Saturn architectures, though as it was the PSX was pretty much perfect for nextgen multimedia)
OTOH, the SGI chipset would also have meshed extremely well with that premise, though I'm not sure if that came before or after the Imagesoft thing (both must have been in 1993, SGI definitely early in the year given they joined Nintendo by late Summer).

From that perspective, if they were actually going to outsource a design, let alone to the west, the best mass-market option probably would have been ATi's new RAGE chipset (though that wouldn't come to the consumer market until 1995) with a very nice balance of 2D and 3D acceleration as well as MPEG-1 acceleration also aimed at a practical consumer price point (unlike the slightly later 3DFX Voodoo chipset, or the less versatile S3 ViRGE released in '95, or the ill-fated quad based Nvidia NV1 that Diamond ironically marketed specifically for playing Saturn ports for the PC -built in controller ports and audio hardware in their EDGE 3D card of 1995 -that and the 2 others above were quite literally the only consumer 3D accelerators on the market by 1995).

However, there's no indication that the proposed Sony deal would have involved western hardware development directly. (ongoing feedback and contributions from STI/Imagesoft, but not any outsourcing to US companies let alone STI headign up development)

No doubt there all Corps do it, even Atari (Jaguar came from the Panther) . So what's the issue with SEGA then ?

Not quite, the Jag's history is rather complex and the Jag came more from the Konix Multisystem (the 1988 Slipstream ASIC more or less a slightly enhanced integrated version of the FLare 1 chipset developed a couple years prior).
http://www.konixmultisystem.co.uk/in...content=martin
Some is in there, but other stuff I've onyl gotten from the exceptional Atari historians Marty Goldberg and Curt Vendel.
The Panther itself came after the scrapping of an ongoing late 80s Atari ST/STe based console design (and also after Atari Corp turned down North American distribution of the MegaDrive in 1988 -Mike Katz was all for it but Rosen and Tramiel couldn't come to an agreement as Jack kept pushing for European distribution rights as well -to tie in with their computer market among other things- so it fell apart -Katz ended up taking a break from the industry in early '89 before cutting that vacation short and joining Sega of America slightly after the launch of the Genesis in late summer -he of course shifted the "arcade experience home" campaign to the well known "Genesis Does" advertisement -pulling a trick he'd used many times before, direct competitive marketing -even some of the Atari Corp adds show that though were pretty limited funding wise and fairly few and far between -especially TV ads)

Anyway, the Panther itself was an in-house Atari Corp design (more or less) and the custom chips comprised only the graphics chip. (they planned to use a 16 MHz 68k 32k of 32-bit fast/expensive SRAM shared with the Panther and 68k -68k would have much of its bus time eaten by the Panther- and an off the shelf sound chip -likely FM synth)
That's where Marin Brennan comes in: Flare Technologies (Brennan, John Matheison, and Ben Cheese) had just completed the Slipstream ASIC for the Konix Multisystem (note Flare retained whole ownership of that ASIC and would later go on to license it to several set-top box manufacturers) and all 3 of them split up to do various consulting work (Cheese would go on to Argonaut and do the Super FX ASIC, or MARIO chip as it was originally named) and Brennan got picked up by Atari Corp to complete the chip design (silicon) of the Panther video chip (formally and Object List Processor) in 1989 after one of Atari's key engineers left the project.
Brennan then convinced Atari that the Panther would end up being a bad idea and that 3D was the way to go for a future design, so Flare 2 was formed under Brennan and Mathieson and they started work on what would become the Jaguar. The core logic doccuments were set in 1990 and some additional stuff (namely the JERRY sound/IO/DSP ASIC) was put on paper in 1991 before the "real" engineering started (actual TTL and LSI design and prototyping) and it was finally frozen some time in late 1992 I believe, or at least preproduction units were made ready at that time. (may have tweaked a few things before the final production versions -which unfortunately still had some significant bugs due to time/funding constraints)

The Jaguar Design from the start was EXTREMELY cost-oriented to be as affordable and mass-market acceptable as possible and aimed at a $150 price point. (had RAM prices not stagnated like they did in the early/mid 90s, it might have met that goal too -it didn't drop to that price until early 1995 as it was) But in spite of that, and the bugs it was also exceptionally powerful and the same chipset was flexible enough to be used in much more powerful designs and other configurations... like ones closer to the Saturn or PSX, except it could have been more cost effective than either of those with such tweaks. (put an SH2 or an R3000 and 26.6 MHz EDO/SDRAM and/or dual 2MB RAM banks with a CD-ROM drive and see what happens and it would still be much cheaper than either of those or maybe even the N64 -in terms of production cost and retail price given similar circumstances, the Jag itself was necessarily sold for profit due to Atari's financial situation- -the bugs wouldn't have been nearly as problematic with a decent CPU, as it is the only way to really push it is to use the GPU for most CPU operations and that's really not designed for it -only attractive due to how weak the 68k is for pushing more advanced stuff)

The Jag was inspired by the multisystem and Brennan's idea of smooth shading being revolutionary for 3D. (in '89/90 it would have seemed far more important than texture mapping, especially due to memory constraints, as it was the competition pushed that too, but less significantly than the custom CRY colorspace allowing direct interpolation to 24-bit truecolor shading with 16-bit pixels -256 colors with 256 shades)
It did also draw on the Panther's design (perhaps at Atari's insistence as Brennan really didn't care for it) and incorporated an evolution of the Panther Object Processor that removed a few featured and built much more on others as well as adding a blitter (handling texture mapped objects for scaling/rotation -the OP could only scale "sprites" but was much faster- as well as adding gouraud shading; the TOM ASIC also added a Z-buffer and the 64-bit RISC GPU microprocessor -a DSP like chip with CPU like programmability aimed at transform and lighting operations but flexible enough to be pressed as an audio DSP -as in the hacked clone in JERRY- and even as a CPU though inefficient -and pnly preferable to the 68k due to the latter being really weak and a huge hog on the shared bus due to the lack of caching)


There's so much misinformation and myth in such history that you'd hardly believe it. (the very reason I don't believe anything Sega related too easily, let alone Nintendo -albeit I already know a lot of the truth behind common myths in that case too)

Are you for real?. I'll say again for the very last time. SEGA America showed off the Saturn at the Consumer Electronics Show at Last Vegas at the 6th of January 1994 .

So what, unless SoJ was explicitly stating a planned release date (or even a solid timeframe) for Japan, let alone the US. None of what I said has been countered.

I never claimed SoA didn't know about the Saturn's development, just that they may have been misinformed on its release plans.

[kool kitty89]

The surprise was how fast SOJ rushed the Saturn to market. It came out with a couple games late in 94 in Japan, and mid-95 in the US with virtually no software at all! THAT was the surprise.

That's EARLY 1995. (May) It was SUPPOSED to be released in mid 1995 (early September, a week ahead of the PSX), but for whatever reason (some claim forced by SoJ) they decided to change those plans without notifying any of their 3rd party developers or most of their key retailers and launch at E3 1995 at select retailers with limited supply and a hefty price of $400 for the bare-bones system.
That not only led to problems with developers (not just frustration at being left out, but leading to some games being rushed to completion in light of the launch) as well as retailers (even though included in the launch were apparently frustrated by the limited supply and weak sales the system saw), but Sony also pretty much thwarted any advantage that launch would have had by announcing the $299.99 launch price just after Sega's announcement. (who knows if Sony would have even pushed that price point if Sega had kept things vague until just before the September launch -rumors were for around $500 up to that point, the Japanese system had launched at the equivalent of $450 US iirc while the PSX launched at roughly $410 though so those savvy to that would have a better idea)

By the time the Saturn would have launched, it had a very respectable library (though oddly missing several games on the 32x, not even slightly enhanced ports rushed over to bulk up the library) including Virtua Fighter Remix (maybe could have tweaked Daytona too if the US launch had been delayed to September) and soon after that Sega also slashed the price to $300 on the baseline system to match Sony.


The conflict over the 32x may have messed with things regardless, but I don't see how they thought releasing a $400 ($500 in some bundles) console with a handful of games (lacking some critical US-centric genres as well as a Sonic game) in SPRING of 1995 was a good idea at all. In fact, the very presence of the 32x should have favored delaying the Saturn's US launch if anything to minimize the shock and make for a smoother transition. (even with no 32x a fall release would have made far more sense, and they needed to focus on some things like getting a sonic game out -or even an enhanced Chaotix port maybe compiled with an enhanced Sonic CD- and an American Football game, though maybe they were banking on Madden '96 -not proactive if nothing else, should have had a Joe Montanna/Sports Talk Sega Sports release to be sure)
Not only could they have started pushing 32x games over to Saturn, but also some late Sega CD games and possible Genesis games too. (both canceled projects and new games that could have parallel enhanced released on the Saturn)

Hell Core was making a kick ass remake of Soulstar for the Jaguar CD and a dumbed down version for the 32x, THAT would have been awesome on the Saturn. (or Batman and Robin -especially if adding the sidescrolling game on top of the driving game, or plain MD games like Ristar and Vectorman)
I know 3D was big, but no reason not to pad out the library with 32-bit quality versions of hot 16-bit games and take advantage of the 2D power of the system (Toy Story would be nice too).
2D may not have been the next big thing, but there was at least a sizable niche to fill with that, and if there were already good games being developed for parallel platforms, why not take advantage of that?

Hell, with tactful marketing they might have even been able to pull something off pointing out how they could manage good 3D as well as awesome 2D and have the best of both worlds. (regardless of whether the Saturn was technically inferior to the PSX/N64 in 3D -or 2D even in some specific respects- market perception rarely actually gets built on specs... but the marketing campaigns and software on the systems -and things like price point etc and early on, Sega had enough brute force to at least put up a decent fight against Sony in direct market competition, maybe even enough to keep it up without going into debt -quite possibly managing far lower profit margins than Sony, but still in the black unless Sony really went all in to specifically try and crush the competition outright and not aim more at profitability -they did do that to an extent as it was, but tempered by practical business sense as well)


Sega certainly would have had an easier time with a cheaper console though, let alone the gimmick of backwards compatibility (and simplified situation of no 32x), let alone one with comprehensive development tools.
Every additional variable in Sega's favor would have helped considerably. (or in more minor ways for some things)
With a consistently cheaper system (that they could afford to sell), Sega could have had an additional selling point even if they system got labeled as "inferior" as the Genesis did at a certain point against the SNES by the masses. (seems to have been some time in 1994 that that happened to a definitive extent... and 1994 is also when the suggested retail price of both finally met at $100 for the baseline unit, prior to that it had been $150 vs $130 and before that $200 vs $150 -and earlier on you also had more of a software library advantage too whereas the SNES caught up more and more in that area as time went on)

A shame 3D blast didn't come out in 1994, that could probably have stolen a fair amount of DKC's thunder. (at least on a pure graphics basis)

[gamegenie]

Kool kitty, I'm going to raise the bar on my hypothetical.

Not only in 1994 SEGA now confirms rumors that the Saturn's cartridge support will support 32X. But in 1995 the early launch announcement is delayed. SEGA adds a 64-bit bus to the console and begins to heavily tout it as 64-bit, they launch the Sega Saturn world wide in August 1996 with a demo disc.

In October 1996, Sonic the Hedgehog 64 is released.


Nintendo's Super Mario 64 praise is quickly over shadowed by Sonic 64. The Nintendo vs SEGA wars continue and the rest is history.




Sony that is, PlayStation gets destroyed by 3DO M2.

[Chilly Willy]

That's EARLY 1995. (May) It was SUPPOSED to be released in mid 1995 (early September, a week ahead of the PSX),

Early = 1 to 4 (Jan to Apr), Mid = 5 to 8 (May to Aug), and Late = 9 to 12 (Sep to Dec).

but for whatever reason (some claim forced by SoJ) they decided to change those plans without notifying any of their 3rd party developers or most of their key retailers and launch at E3 1995 at select retailers with limited supply and a hefty price of $400 for the bare-bones system.

Which was a HUGE mistake. The other was to freak out that the software wasn't ready and to command all licensees to shift everything to Saturn only.

That was SEGA's biggest problem at the time - they tried to rush too much. They tried to rush a number of 32X titles out for xmas-94 (notably, Doom). They tried to rush Saturn to the market. They tried to rush Saturn games to meet their rushed Saturn launch (notably, Virtua Fighter). Rush, rush, rush... and they wound up paying for it.

[kool kitty89]

Kool kitty, I'm going to raise the bar on my hypothetical.

Not only in 1994 SEGA now confirms rumors that the Saturn's cartridge support will support 32X. But in 1995 the early launch announcement is delayed. SEGA adds a 64-bit bus to the console and begins to heavily tout it as 64-bit, they launch the Sega Saturn world wide in August 1996 with a demo disc.

As I said... if they weren't overly focused on a 1994 release they could have scrapped the Saturn and selectively taken parts of the design (or ven most of it) and evolved from there with much more buffering/caching for a unified 32/64-bit video bus with shared framebuffer a la PSX (if they didn't go 64-bit, definitely allow SDRAM to be clocked at 2x the internal clock rate of the video chips to allow comprable bandwidth -givne the SDRAM used was all 66 MHz rated or higher, that would have been smart... or do both ).
Use SH3 as the main CPU, maybe shared on a unified 64-bit bus (especially if video could use 2x speed and 64-bit), so somewhat more like the N64 in that respect.
1 CPU, 1 combo video ASIC (maybe integrating I/O) with triangle rasterization perhaps simultaneous quad/trip support, one audio ASIC using the 32 DMA channels of the Saturn without the DSP (maybe combine I/O and audio if the video ASIC was full), maybe ditch the SCU's DSP as well (since you'd probably be using the SH3's FPU more, especially for any quad stuff), maybe strip out the FM synth feature (neat, but pretty much never used -almost as bad as the audio DSP's underutilization), perhaps add 4-bit ADPCM decoding in addition to plain 8/16-bit PCM (12-bit might be nice too), though even without that the added CPU resource could help there too.
So probably less RAM at 4 MB (plus CD buffer), but unified and more flexible while being higher bandwidth (66 MHz 64-bit would be 528 MB/s, similar to the N64's peak bandwidth -but with less latency issues- or a bit less if it was 2x 28.64 MHz at 57.28, still pretty damn good though)

-Caching not only facilitates shared bus with high bandwidth, but also much less headaches for developers to optimize for bottlenecks. (one thing that make the Jaguar 2 miles better than the Jag in programmability and the main thing that made the PSX friendly -in additon to tools to match)

Oh and a cheap MCU driving the CD-ROM/cache (and a much smaller cache, maybe 128k tops -still 4x that of the PSX/3DO) and probably 4x CD-ROM.

And then make the DC fully backwards compatible (adding better caching, faster GPU internal and bus, etc) and if the SH4 didn't emulate the SH3 with 100% compatibility, retain SH3 as coprocessor and switch to something else (power PC would be nice) or SH4+SH3 coprocessor on a slower/audio bus. (not sure if Hitachi was offering faster SH4s though, even by 2001, so PPC might be better, maybe dual SH4s if that was cheap enough -or just better support the SH3 as a coprocessor and maybe bump it to 100 MHz in DC mode -though 66 MHz is what Sega used for the audio bus on the historical DC with 100 MHz for the main bus only)
Lots of possibilities to imagine.

That's one bad thing about the proposed MD/CD compatible stuff, it probably wouldn't have been cost effective to make a compatible successor to it. (at some point a fresh start just makes more sense, so the DC could have been that, or the Saturn could have and been designed clean enough to make it attractive to build onto -PS1 to PS2 did that OK, but the PS2 architecture itself was a mess for other reasons)

In October 1996, Sonic the Hedgehog 64 is released.


Nintendo's Super Mario 64 praise is quickly over shadowed by Sonic 64. The Nintendo vs SEGA wars continue and the rest is history.

I doubt it would kill SM64, even if Sonic Adventure quality (with weaker graphics) I doubt that.

The very fact of a more powerful (yet cost effective and reasonably developer friendly -probably closer to PSX than Dreamcast) system with CD-Drive would have crushed the N64 though. A nice killer app for sure though.

Sony that is, PlayStation gets destroyed by 3DO M2.

Nah, I doubt Sony would loose out, they'd still have the cheaper system with excellent tools and tons of money (and vertical integration). They'd have had much better competition though. (let alone if the N64 used CDs/proprietary CD derivatives )
Pannasonic+3DO would have to correct all the mistakes made with 3DO (especially market model -but also allowing low level programming) and then it would be more expensive than the PSX or even Sega's new streamlines system.




In any case, from SoJ's "must release in 1994" issue and the existing Saturn, the best thing to do for a more practial western release was to put the remaining efforts in final revision (assuming that's starting late 1993) not only to tweaking things for more power, but more towards making it affordable and cutting back on wasteful areas (CD-ROM interface is really obvious, even without hindsight) as well as selectively making said cuts to allow easier expansion, and also adding a bit more provision for expansion. (cutting back on main RAM was probably the easiest in both regards, then expand the cart slot -or a separate expansion port- for 32-bit CPU bus with easier interfacing eith 28.6 MHz SDRAM). The more time they had to hack things down, the cleaner it could have been. (I really don't see why they didn't at least use a derivative of the MCD's 68k+CD interface with a clock doubled CD chipset for 2x CD-ROM and use the 68k double duty with audio)
But a late hack could still probably have put the SH1 in dual duty with audio, have the CD-ROM buffer double for audio, scrap the sound DSP (just the DMA/PCM/synth part of the SCSP left) and maybe switch to plain 16-bit DRAM for the SH1 rather than SDRAM. (and either share for the CD-ROM cache or have a separate small chunk of SRAM/PSRAM for that buffer -might not have been a bad idea to use the same 16-bit 64k PSRAM chips in the MD2 -SH1 could have been powerful enough to do multi-channel software audio decompression too, and not limited to ADPCM either, you could do 2-bit, 4-bit, CVSD, etc, maybe even stream MP3 or MP2 audio)

As it is, the SH1 was already on bus A shared which the 68k/audio block also shared, so that might not have been out of the question.

They could have dropped the SCSP entirely in favor of somethign off the shelf (like Yamaha's OPL4 plus a bit of interface logic to allow DRAM use rather than SRAM), but by the late design stage, the SCSU was probably more solidly stuck in. (and they had R&D/contracting costs for its development, the OPL4 could have been better to go with from the start though)


Then drop to 1 MB or maybe even 512 kB main SDRAM (32-bits full speed) and maybe drop down to a single SH2 (especially if they supported the SCU DSP more, not just for 3D math but maybe coupled with VDP1 for true triangle rasterization or more advanced engines using VDP-1 quads and DSP trips filled with line writes from VDP1 for texture/shading).
And either have an SDRAM RAM add-on, or maybe a couple steps with a later add-on having RAM+coprocessor (DSP, FPU, or whole fast CPU, maybe an SH3 even depending on cost and timing), though a single definitive add-on could have been better, or just transitioning to the Dreamcast anyway. (maybe designed around backwards compatibility even, more like the super buffered Saturn above -the simpler the Saturn was, the easier that would be too)

Early = 1 to 4 (Jan to Apr), Mid = 5 to 8 (May to Aug), and Late = 9 to 12 (Sep to Dec).

OK, so early-mid 1995. Opposed to early late as it was supposed to be. (early September given the original press release)

Which was a HUGE mistake. The other was to freak out that the software wasn't ready and to command all licensees to shift everything to Saturn only.

That was SEGA's biggest problem at the time - they tried to rush too much. They tried to rush a number of 32X titles out for xmas-94 (notably, Doom). They tried to rush Saturn to the market. They tried to rush Saturn games to meet their rushed Saturn launch (notably, Virtua Fighter). Rush, rush, rush... and they wound up paying for it.

Yeah, they had mistake layered on top of mistake making into a really harmful mess that snowballed and contributed to even more mistakes.

From a historical standpoint, I'm really interested in who's call the Saturn's launch date change was and how it was managed.

I mean a head start sounds good on paper, but as soon as you take real-world limits into account for the date (both 1st and 3rd party wise, software, hardware volume, price, and retailer wise), it really seems obviously unattractive.

I could see maybe moving it up a week or two to give a bit more head start over Sony, but there were better options entirely, like pushing for a gradual hype build up peaking in early September and keeping enough cash in reserve to afford dropping the price below cost if necessary. (especially if lower cost production revisions were pending)
In hindsight at least, it seems like it would have been a good idea to keep feeding the vague price point until much closer to the actual release and thus not give Sony such an opening as they did. (hell, Sony might have announced a higher price even, or kept their's vague too)

They definitely needed to avoid a price dropping game of chicken with Sony as Sony would win that game no contest. (more money, cheaper hardware, vertical integration, etc) Well, unless the Saturn was cheaper.


Plus, if they WERE going to discontinue the 32x (and probably CD -not plain MD as it had a much larger and stable base and potential as a budget system) earlier than originally planned they would still have been best to delay as long as possible and make every effort to soften the PR blow and sell off stockpiles at profit. (maybe work on dev kits specifically favoring 32x to Saturn conversion for 32x ports and ongoing 32x projects as well as transition from 32x to Saturn development in general -by extension that might have meant facilitating MD to Saturn transition as you'd need to take the MD part of the 32x into account, maybe even MCD though no CD32x games really pushed the CD-specific hardware much as such)

[Team Andromeda]

The surprise was how fast SOJ rushed the Saturn to market. It came out with a couple games late in 94 in Japan, and mid-95 in the US with virtually no software at all! THAT was the surprise.

Anyone that brought system on Import will say that Japanese launches were always short on games, be that the SNES, the Mega Drive, the PC-Eng, the Saturn, the PS, The PS2, the N64 . Also the PSX was shown latter than the Saturn (to the press & Gaming World) ) and in the Toyko Game/toy show of 1994 SONY was not able to have any Software running on units or Videos of PSX software (unlike SEGA with the Saturn) So it was even more of miracle that SONY was able to make its late 1994 date.

What about that Surprise ?

A damn Shame SoA didn't keep pushing for it, a much better idea than Mars

SOA didn;t pushed for Jupiter, SOA wanted to hold on for the 16 bit Market for as long as they could.

Well no, but that's a lot of misinformation (confusion of Jupiter and Neptune for one), and the actual configuration is still vague

I don't know how ? Jupiter was a Saturn Minus the CD-Drive and Main system RAM and SH-1, Very much ST-V; though that used a different sound processor and a better I/O to the Saturn

Neptune was a Mega Drive with the 32X built in . Really what is so hard about that ?

but political/bureaucratic/management level stuff as well as some of the software stuff

Oh lets get a bit real. That happens in every Corp and it never be a clear picture, because everyone will have different memories looking back.

Nope, but they could cater to the wrong market... which is what many claim to be a major problem with the Saturn, way too much emphasis on the Japanese market and not meshing with the west -especially the US, the market they should have had

???. You'll always look to focus on your home Market 1st and foremost. SEGA Japan just let SOA go their own way with the 32X that was the mistake, and getting no Sonic for Saturn's launch., And you can focus far too much on the USA too, just ask MS which as never been a success in Japan or Europe (bar the UK) and has always played 2nd fiddle to SONY and NCL in those markets.

Which is why I ignore those parts of articles

Then stop going on about them and referencing them, when they're clearly wrong.

And he was right, or could have been if they followed through. (for the timeline the 32x was to be pushed at least ie 1994-1996)

Try he was hopelessly wrong , the 32X sold like shit after Christmas and the Market was moving to 32bit CD-Rom system's . Better luck Next time Tom...

It would have been the Japanese companies developing the hardware, but with western input.

ONLY then SEGA would have known the exact spec's of the PSX (coming from SONY only R&D Console tech Team) and what it would take to beat them, and guess what ?, SEGA didn't. So no need for reading lessons , just use your common sense

And? The 32x's price was fine for what it was

No I want to know the Price Difference between the Saturn and 32X launch prices in the USA, and the same for the Launch Software . If the diff is similar levels to that of the price of the 32X VS Saturn in UK. It blows your inflation and High tax to bits

So what was the difference in the Launch prices ?

$160 suggested retail -apparently $150 at some stores-

160 dollars in your back pocket , is like me having 160 quid in my back pocket , It real is that simple.

The Saturn price was really impractical though and they squandered hype with the early release.

It didn't help, but the High Price never really hurt SONJY with the PS2 or PS3. And unlike the 32X or PS, you had a free game, you didn't need to buy a Memory card (unlike the PS) So when you start to factor in those the price difference was that huge VS the PS.

And it was SEGA (or STI/SoA) that pushed that in the first place coupled with Imagesoft's input

Who said that TOM .

Not quite, the Jag's history is rather complex and the Jag came more from the Konix Multisystem

Dear GOD All corps have multi Projects on the Go (well most) Listen to Jeff Minter Atari had Jaguar, Panther and plans for the Jaguar II all in development at the same time

So what, unless SoJ was explicitly stating a planned release date

So what ?. In Jan of 1994 one would have known the Saturn was for real, it was close to completion and SEGA Japan were spending millions on it and putting their Top teams to the task of Making software for it .

No surprise at all

[kool kitty89]

Anyone that brought system on Import will say that Japanese launches were always short on games, be that the SNES, the Mega Drive, the PC-Eng, the Saturn, the PS, The PS2, the N64 . Also the PSX was shown latter than the Saturn (to the press & Gaming World) ) and in the Toyko Game/toy show of 1994 SONY was not able to have any Software running on units or Videos of PSX software (unlike SEGA with the Saturn) So it was even more of miracle that SONY was able to make its late 1994 date.

What about that Surprise ?

The weak Japanese launch lineup wasn't that big of a surprise (I'm sure Chilly WIlly understands the nature of JP launches... especially with that example since he had a Japanese PSX prior to the US launch that his boss had brought back from a business trip to Japan ).

THe REAL surprise was the poor/weak US launch lineup and the rushed mess the Saturn was in the US with the idiotic decision to launch it so early (32x made it worse, but it would have been horrible either way)

However, that's assuming SoA didn't have a choice in launching it at that date, something that needs to be properly established on a factual level. (ie whether SoJ set a deadline to have the Saturn out by spring or not)

It's blatantly obvious that the Saturn should not have been launched until later. (and even for the perspective of Sega in early 1995 it made almost no sense at all, the 32x may have made little sense in some respects, but it still made a hell of a lot more sense than what they did with the Saturn -the very existence of the 32x made the rushed Saturn launch more nonsensical)

SOA didn;t pushed for Jupiter, SOA wanted to hold on for the 16 bit Market for as long as they could.

Didn't you state earlier that SoA favored Jupiter early on?

Oh lets get a bit real. That happens in every Corp and it never be a clear picture, because everyone will have different memories looking back.

That's the point: we need proper historical doccumentation that cuts through false memories and gets to the truth. We're not talkign slightly skewed stuff either, but things that build up to the point of being nowhere close to the truth.
That's what the Atari Historians have done, albeit Curt Vendel has the benefit from working from 1984 up to today building that stuff as well as digging deeper and uncovering the most astounding things recently.

At very least more fact checking and cross referencing needs to be done before any real conclusions can be made. (especially since Sega's situation seems to have been almost as sloppy/confused as Atari Inc/Corp -if not more so in some respects)

???. You'll always look to focus on your home Market 1st and foremost. SEGA Japan just let SOA go their own way with the 32X that was the mistake, and getting no Sonic for Saturn's launch., And you can focus far too much on the USA too, just ask MS which as never been a success in Japan or Europe (bar the UK) and has always played 2nd fiddle to SONY and NCL in those markets.

That's bad business though... they most definitely should not have put Japan over US or EU, they should have solidly pushed for all 3 and put the emphasis on solidifying their existing success. (and expanding it if that could be done without compromising sustained strengths in existing markets)

At very least they should have been putting big support behind SoA/STI to push the western/US friendly stuff, especially since Sony most definitely was doing that (and in spades).

Then stop going on about them and referencing them, when they're clearly wrong.

Read the rest of that statement, if I was to ignore them, I'd ignore every such article including all that you've posted and everything in Kent's book and only go for directly factual documents and untainted interviews.

Try he was hopelessly wrong , the 32X sold like shit after Christmas and the Market was moving to 32bit CD-Rom system's . Better luck Next time Tom...

Saturn sold even worse. (albeit we don't have any proper breakdowns to really compare the 2)
We'll never know how the 32x could have done without the Saturn mess (not that I think 32x was preferable, but to reach its true potential, conditions would have needed to be different).
It's obvious that 32x sales were going to taper off in early 1995 and wouldn't pick back up until fall (it's a strongly seasonal market), the Saturn started selling half decently in the fall (along with finally building up a decent library and dropping to $300), and by that point the 32x had been effectively killed off so there's no valid comparison.

Again there seems to be far too little factual information on just how and why Mars was brought about and how those goals evolved and differed between SoJ and SoA. (and why it wasn't canceled)

I don't think the 32x should have been released, but I also think the Saturn was poorly thought out and executed all around, especially in the context of what the US market needed.
Hell, no 32x and no Saturn in the west until mid 1996 would have been better than what they did. (assuming the MD got good support in the interim and the Saturn launched with a very strong library, massive hype, a highly competitive price point, and at least a few killer apps)
Albeit it would have been even better under those circumstances to push back the JP launch of the Saturn in favor of fixing the hardware or even enhancing it. (for the short term they obviously couldn't enhance it more, but they could have stripped it down, and assuming the late 1993 initiated final design revision of the Saturn is true, they should have been pushing heavily for streamlining the system, NOT adding more features -only the latter when facilitated by the former)

ONLY then SEGA would have known the exact spec's of the PSX (coming from SONY only R&D Console tech Team) and what it would take to beat them, and guess what ?, SEGA didn't. So no need for reading lessons , just use your common sense

Heh, you're still not even reading what I summarized or what Kalinske claimed. There's nothing in there about Sega having access to Sony hardware, they never got that far. (it wouldn't have gotten to that point until SoJ greenlighted the proposal, which never happened -at least according to Tom's interview)

160 dollars in your back pocket , is like me having 160 quid in my back pocket , It real is that simple.

Nope, but it was affordable for people in 1994/1995 who wouldn't be buying Saturns or PSXs until 1997 or later. (especially given the later price drops)

Again, I don't think the 32x was a good idea as it was (and a cheaper add-on definitely could have made more sense), but I AM trying to take things from a reasonable overall perspective.

With the Saturn out there a it was: the 32x was too expensive, as was the Saturn, and had there been no Saturn, the 32x was too rushed and simple to compete in the long term. (then again, that could have meant a push for something between Saturn and DC released ~97)
If they were playing it smart, but still locked-in on pushing to cater to Japan foremost, the best option by late 1993 was the push for a streamlined redesign of the Saturn that trimmed off all the unnecessary crap and reconfigured it for high cost/performance. (I already detailed the technical possibilities and oddities of the design, so repeating that is moot)
And it's not like this is pure hindsight stuff, cost is a HUGE issue for the console market, hell, they'd have been better off with the Jupiter alone than the mess with the Saturn in that respect. (better would have been a system that kept CD-ROM, but tempered to a more sane level of cost)

It didn't help, but the High Price never really hurt SONJY with the PS2 or PS3. And unlike the 32X or PS, you had a free game, you didn't need to buy a Memory card (unlike the PS) So when you start to factor in those the price difference was that huge VS the PS.

PS2 wasn't really that expensive, it was significantly cheaper than the PSX had launched at... and the DVD feature along with Sony's massive PSX success and saturation marketing clinched it.
OTOH the PS3's launch price was a HUGE issue, had it been cheaper than the 360, it probably would have been far more competitive. (they made a lot of mistakes, the late release probably least of the bunch... they made many of the same mistakes as the PS2, except they had stronger competition to face without the gimmicks and -to some extent- luck/coincidence that helped the PS2 transition so well -the PS2 was reasonably priced, higher than the DC but still acceptable for what it was, it had DVD -especially critical to Japanese adoption- and was slashed to the point of being cheaper than many mass market DVD players, it had backwards compatibility -a gimmick but a useful one- and it had massive hype)
Had the PS3 launched at $300 for the lower-end model and had near complete backwards compatibility as well as comparable technical capabilities (at least visibly -ie they could cut the design back heavily in some areas and have mot games look just as good, hell they could even do better for some things and save cost, like adding more, but cheaper RAM and possibly using a unified bus).

Being smaller/sleeker would have helped too.

Dear GOD All corps have multi Projects on the Go (well most) Listen to Jeff Minter Atari had Jaguar, Panther and plans for the Jaguar II all in development at the same time

Yeah, but that's just is, Atari didn't have multiple projects on the go, other companies did (or rather an independent small company -former Sinclair engineers from the Loki project- had the design and got contracted by Konix before splitting up and doing contracting which led to 2 of the 3 teaming back up at Atari to design the Jaguar).
It's a complex progression, but they only had 1 being pushed at any time, they had brief transitions between them, but in most cases they scrapped one before moving on to the other. (it makes sense given the limited funding -and computer developments also eating up other R&D funds) The Panther was abandoned before the Jaguar really got underway, if it lasted at all, that was due to Atari Corp management being unsure of the viability of the Jaguar.

The Jag II didn't start until the Jag chipset was pretty much solidified, they probably had some stuff floating around on paper for potential Jag 2 designs, bt I highly doubt anything solid before the end of 1992. (by which point the jag was pretty much frozen for production and preproduction dev units had already been sent)

So what ?. In Jan of 1994 one would have known the Saturn was for real, it was close to completion and SEGA Japan were spending millions on it and putting their Top teams to the task of Making software for it .

Never claimed otherwise, I only claimed that SoA and SoJ may not have understood eachothers plans as such... let alone that SoJ may have forced SoA to take certain actions. (it's unsubstantiated, but there are considerable claims for SoJ forcing the Spring US launch of the Saturn)

The JP launch may have been less of a surprise, that would have depended on what SoA had been told (or interpreted) in terms of specific plans for the Saturn's Japanese release. (and more importantly, just WHEN they were made aware of such) But beyond that there may have been perceptual differences in how SoJ had planned Mars to be handled alongside Saturn and what SoA envisioned for it. (there's conflicting information on whether SoA ever even wanted to make Mars/32x or if SoJ more or less forced it on them -by some accounts, a compromise from a less desirable proposition from SoJ)

And beyond that there's the US launch of the Saturn.


The US is obviously the most critical issue since Sega managed pretty well in Japan and even Europe was OK with Saturn by comparison. The North America had accounted for more than 60% of Sega's overall 4th gen sales, and having that crumble like it did was pretty substantial. (that's one thing Nintendo at least managed to hang onto a fair bit, the N64 may have been a bit of a mess too, but it did manage a pretty strong market share in the US and had Sega been even that successful, they might still be in the hardware business today )