Latham/Miller/Sutton/The Late 16-bit Era

**chessage** · 04-28-2011, 05:54 AM

Since the 'SNES Versus Genesis: Year After Year' thread is gone, I thought I'd make a thread to continue the discussion. Below are the cached pages of the old thread, which might expire with time:

**kool kitty89** · 05-01-2011, 07:07 PM

Cool, that's one I forgot about. (I've got those pages saved as firefox compatible htm files now, so I could post those if/when google dumps them)

**Christuserloeser** · 05-01-2011, 07:29 PM

Thanks a lot for these links! Below a repost of my questions to Miller:

-1991 to 1993 -

When were successors to Mega Drive first discussed ?

Kalinske mentioned a debate on how Sega would approach the transition from Mega Drive to the next generation. What options were discussed aside of 32X and Saturn ? What options did Sega of Japan propose ? (There was this "Giga Drive" rumor)

What was the mood at SOA in regards to a 32-bit successor at that time ? - I can imagine that in 1993/1994, when all of these discussions took place, things were very different to the situation in 1995.

When was (what we now know as) Saturn first being discussed ?

-32X & SVP-

Did Sega of Japan first approach SOA or did SOA approach Sega of Japan first in regards to the idea for a MD successor in addition to Saturn ?

According to what little we know, one of Sega of Japan's proposals to SOA seemed to have been a minor update to the Mega Drive. (Maybe there are some details about Hideki Sato's concept that Miller still remembers ?)

Why wasn't the SVP used for more games or hardware aside of Virtua Racing ?

Did the idea for 32X evolve from the rumored SVP lock-on cartridge ?

The 32X was redesigned by Sega of Japan's engineers to lower production cost and retail price. What was the original design?

Where there ever thoughts or plans to skip the release of Saturn in North-America and focus on 32X entirely ?

Why was Neptune delayed and eventually scrapped ? It seemed like the hardware was pretty much finished? Were 32X sales too low to warrant a production ? - I imagine a 32X+MD-in-1 combo from day one could have helped the 32X concept.

-Saturn-

What is the story behind Saturn ? How did it evolve to what it eventually became ? Did SOA had any input on Saturn's design ? or are any of SOA's proposals reflected by Saturn's final design?

Are the rumors true that Saturn was changed at the very last minute ? (i.e. Dual CPU + VDP setup as is often cited as proof for this change)

The Saturn was Sega's very first console that was not compatible to its predecessors. Were there ever any plans for Saturn to be MD compatible ? (perhaps during the earlier development stages, before branching out Sega's next gen plans into Saturn / Mars)

Releasing two incompatible platforms (Saturn and 32X/Neptune) instead of just one led to a split in the customer base. We assume Saturn was released earlier than initially planned (in Japan). How much earlier ?

What was the financial situation at SOA at Saturn's launch - i.e. in comparison to the 32X' launch ?

Were there really plans for a Saturn based cartridge-only platform (I think its name was rumored to be "Jupiter") ?

-Sega of Japan and Sega of America-

Where there any real problems (tensions) between Sega of Japan and Sega of America before the 32-bit era ? (Kalinske said that SOJ didn't interfere with SOA's business until 1995, that'd be post 32X/Saturn)

If SOJ did interfere, when and where did they do so ?

Had SOJ any influence on what Japanese software would be published in the West ? - Did they push for certain titles that had to be published (which I imagine they did - see Sonic), or did they refuse to allow localizing certain Japanese titles in the US? and for what reason ?

What was Sega's policy on 2D games during the Saturn days ? There seem to be many early 2D games like Golden Axe and Shinobi, but fewer and fewer during the later years. Was a 2D Sonic game ever considered ? or was there no interest in a 2D Sonic for Saturn after Knuckles Chaotix failed to spark any real interest in 32X ?

There were rumors of people at Sega of America discussing leaving hardware and focusing on software as early as 1995 (or possibly even earlier). Is that true ? What did the Japanese side think of that at the time ?

**Christuserloeser** · 05-03-2011, 08:16 AM

So, as it turns out, Latham was not correct when he claimed that Saturn's 'surprise' launch was "definitely" ordered by Sega of Japan:

http://www.sega-16.com/forum/showthr...l=1#post355751

This does make me question his other statements as well. How much truth is there to his statement that Japan decided which games SOA could localize, and how much of it is speculation on his part ?

**kool kitty89** · 05-04-2011, 08:10 PM

Originally Posted by Christuserloeser

So, as it turns out, Latham was not correct when he claimed that Saturn's 'surprise' launch was "definitely" ordered by Sega of Japan:

http://www.sega-16.com/forum/showthr...l=1#post355751

This does make me question his other statements as well. How much truth is there to his statement that Japan decided which games SOA could localize, and how much of it is speculation on his part ?

Yes, that's old news. (I was sure you were present for the past argments over the validity of that article)

I'm not positive, but it seems that those quotes came from statements made by Kalinske while he was still at Sega. (and thus under obvious obligation to boost PR and say nothing that would compromise Sega's image, let alone go as far as exposing evidence of internal conflicts and management problems)

His statements also seem to be a bit odd and contradictory in general, also indicative of twisting for the market. (not sure if the "fall" comment was even intentional)

Now, he may also have been covering his ass (ie selfish reasons for twisting things) in the various interviews after he left Sega (including Melfs), but those at very least are more likely to be truthful than anything he said while at Sega. (it's pure speculation to think he's lying like that . . . it's certainly possible, but there's no conclusive evidence for such -certainly not like there is for Nolan Bushnell

)

Now, that alone made me drop the issue in the context of "too little evidence", albeit with more support for SoJ forcing the early launch, but far from definite.
Lantham changed that and I'm not all but totally convinced that SoJ made SoA do things they didn't want with the Saturn's launch.

Still, it would be interesting to get a followup interview from Kalinske with more details on how that all went down. (how soon did SoJ mandate the early launch, how much time did SoA have to prepare that launch, why 32x software -perhaps some late-gen MD and CD games too- wasn't up-ported to Saturn or canceled 32x projects moved over to Saturn, etc -he already mentioned the obvious issues with severe shortages)

And then all the stuff for miller:
-details on the SGI chipset situation (and when they came to Sega with it, if he remembers)
-details on the collaboration with Imagesoft (confirmation that it happened like Kalinske said and any details on the resulting hardware concept/feature set as far as he can recall)
-details on Sato's Mars design and alternate 32x designs (including SoA's before it was streamlined for production by SoJ)
-maybe some info on the delays in Saturn development kits and even further delays in getting reasonable tools out beyond the preliminary dev kits (ie how much was limited by SoJ making documentation available, how much by difficulty of the hardware, and how much by sheer resources put into building the tools)
-any information on whether SoA was ever allowed input on the Saturn's design and how up to date SoA was with ongoing development (and plans for release in Japan)
-why were some SoA staff (especially engineers involved with the 32x) so shocked by the Saturn's 1994 launch in Japan if it has been set in stone and promoted for months (in Japan) by that point?

**Christuserloeser** · 05-04-2011, 09:04 PM

I doubt there's anything new that Kalinske could add to the debate. I'd love to hear the story told from other people's perspective and see if we can learn something new.

**kool kitty89** · 05-05-2011, 07:42 PM

Originally Posted by Christuserloeser

I doubt there's anything new that Kalinske could add to the debate. I'd love to hear the story told from other people's perspective and see if we can learn something new.

That's your prerogative.

I for one think that a follo-up interview from kalinske could be enlightening (after all, there's few better to ask on certain subjects).

Of course, cross-checking for confirmation is always useful as well, and there are other topics that others could address far better. (aside from Miller, there are some former SGI people to look into, Imagesoft staff -especially Olaffson- among others at SoA . . . lots at SoJ, but none seem interested in interviews and there's also major issues of culture clash when dealing with that side of things)

There's a few things that would be interesting to ask Katz about as well. The timing and nature of his transition out of SoA is a bit vague. (seems like Kalinske was there as a PR person in late 1990 before fully transitioning in as president in 1991)
And whether he ever considered alternate pack-ins (aside from special bundles) with the Genesis in 1990. (probably a few other things like the price drop -though there already had been a $30 rebate dropping the price to $160 in late 1990, or more details on how he was expanding western software resources -maybe whether he ever considered something like STI-)

**16bitter** · 05-15-2011, 06:59 PM

Originally Posted by kool kitty89

-any information on whether SoA was ever allowed input on the Saturn's design and how up to date SoA was with ongoing development (and plans for release in Japan)

Kalinske: The specs for our next-generation console, the Saturn, didn't look very good, and it was way too expensive - Sega Japan told us it was going to retail at $549. Then Jim Clark, the chair of Silicon Graphics, says, "I've got this chipset that's a derivative of the MIPS chipset that would be perfect for your system." We call Sega Japan and say this thing will be cheaper than Saturn, and that it will move polygons 50 percent quicker. The Sega hardware group comes over and says that the chip is too big, it won't be efficient to manufacture. Forget it. When I tell Jim Clark this, he says, "What do I do with this now?" And I say, "Well, I'm sure there are a few folks who might be interested in buying it." And he says, "Yeah, I've already talked to Nintendo." The rest is N64 history.

-why were some SoA staff (especially engineers involved with the 32x) so shocked by the Saturn's 1994 launch in Japan if it has been set in stone and promoted for months (in Japan) by that point?

There's a lot of talk, as antecedent and fallout, of the Saturn being such a mess that many developers and many within Sega found it hard to believe that it would ever be released.

Some, such as Trip Hawkins (obviously a very particular and particularly biased source), stated that the Saturn was a machine for the Japanese market, period.

The Saturn is a very troubled project. Many don't expect it to ship outside of Japan.

EDGE10

Sounds nuts, right? But that's what defined the project itself: ineptitude that became market insanity. The machine's few virtues matched better, specifically software bias, to the Japanese market; perhaps also explaining why SOJ were so intent on launching the flawed hardware at all.

**kool kitty89** · 05-21-2011, 02:25 AM

Originally Posted by 16bitter

Kalinske: The specs for our next-generation console, the Saturn, didn't look very good, and it was way too expensive - Sega Japan told us it was going to retail at $549. Then Jim Clark, the chair of Silicon Graphics, says, "I've got this chipset that's a derivative of the MIPS chipset that would be perfect for your system." We call Sega Japan and say this thing will be cheaper than Saturn, and that it will move polygons 50 percent quicker. The Sega hardware group comes over and says that the chip is too big, it won't be efficient to manufacture. Forget it. When I tell Jim Clark this, he says, "What do I do with this now?" And I say, "Well, I'm sure there are a few folks who might be interested in buying it." And he says, "Yeah, I've already talked to Nintendo." The rest is N64 history.

Yeah, like it went down anything remotely close to that . . . and even if it had:
As if the SGI chipset would have been foolproof . . . The Saturn didn't turn out like it did because of engineering ineptitude (at least I highly doubt that was a factor), it was all about management. (both problems on the hardware design end -R&D management guiding direction of development and evolution of the design, not to mention all the other marketing/management decisions made)

If they'd made the same sort of shitty management decisions, the SGI chipset could actually have been worse. (less favorable production contract, more R&D costs, etc, etc -or compromises with special deals with SGI to avoid R&D overhead -I'll bet Nintendo was compelled to use those ridiculous SGI workstations rather than cheap N64 derived development systems)

Nintendo made some major mistakes managing the N64 and they were generally in a better position under better international management (or at least less conflicted management).
Given how badly Sega screwed up, I doubt the SGI chipset (even if better than what Nintendo ended up with in terms of time/cost/performance trade-offs) would have been remotely close to being foolproof, especially with Sony sending a tsunami over the entire industry. (not just putting up fierce competition, but shocking the market and inducing panic in some players -NEC, Sega, and Panasonic/3DO seemed to overreact to Sony, though Nintendo underreacted and possibly made the biggest mistake -especially since Square was a massive catalyst for Sony's success)

And in any case, it's very likely that the Saturn WAS significantly cheaper (and much closer to market ready) than the SGI design was when proposed to Sega (don't have a date, but it must have been 1992 or early 1993 as Nintendo was partnered with SGI by mid Summer of '93). It obviously would have been much less powerful, so weak in that respect (and perhaps totally aiming at the wrong feature set -total hypotnetical speculation, but not unreasonable), so there would have been a realistic complaint.
SoJ also would have had to consider throwing out existing R&D or continuing with it. (or a good compromise: reviewing existing development, sort out the most useful elements to continue building on and discard the rest -also costly, but less wasteful and a lot better than adding unnecessary/inefficient features and cost to a design -let alone trying to add to that with new hardware to address other areas . . . at least the Jaguar's 2D "sprite" engine -the object processor- didn't take up a ton of silicon -relatively speaking- and was quite powerful and flexible -no good for any "real" 3D, but it can scale large numbers of 2D objects at high speed)

However, there's one very important similarity between the jaguar and the Saturn, neither one was made or broken because of its hardware . . . both had some technical problems (some not related to hardware but SDK support), but the real overall issues were on the management end . . . Atari's much worse than Sega. (or rather, their situation was much, much worse right off the bat -from funding to management to public image, etc, etc- while one could argue that Sega's problems were actually worse as they had that much more to fall from coming off of their massive success of the Genesis -NEC was probably even worse, but they'd already been screwing up with the PCE for a while . . . it just took a lot longer to ruin them in Japan -their mistakes on the Japanese market weren't nearly as extreme as in the US, but they did catch up with them . . . and the combined panic from losing their PC monopoly in Japan and Sony's entry in the market really ruined their position in the games market -if they'd held it together and held off on the PC-FX until it had been properly revised, things might have been different)

Both the Jag and Saturn (as is) probably would have done pretty well under Sony management and funding . . . probably not the PCFX though. (unless they brought out the 3D add-on really early in)

There's a lot of talk, as antecedent and fallout, of the Saturn being such a mess that many developers and many within Sega found it hard to believe that it would ever be released.

Nah, it's not THAT bad compared to what was already on the market, it's disappointing compared to what Sega's engineers should have been capable of and it's overbuilt and thus unnecessarily expensive, but there were a lot of other bad to mediocre examples out there. (3DO was arguably more cost effective on the raw hardware end, but it was manufactured in a high-end multimedia form factor and sold under a totally unrealistic business model, that alone made it far worse than the Saturn . . . and actually, it's only more cost effective than Saturn because it omits features, the actual design is very conservative and not particularly advanced for the time -if you didn't even improve the Saturn, but stripped out some things, it probably could have been cheaper to manufacture than the 3DO, but still significantly more powerful -that's also assuming that the 3DO chipset didn't get a cost-cut redesign by the time of the Saturn's release -like newly consolidated custom chips on a smaller board in a low-cost form factor)

The Saturn was a great design compared to the PC-FX

(talk about rushed releases . . . dusting off shelved hardware and rushing it to market as-is and not even including backwards compatibility -in spite of already having 90% of the necessary hardware)
Or CD32 for that matter. (that was even worse than the PCFX . . .)

Aside from cost effectiveness, it was better than the jaguar too . . . albeit "better" in terms of as-is performance (the jag is a very tight low-cost optimized chipset with a super low-cost configuration used on top of that -with associated bottlenecks- while the Saturn has resources just thrown at it haphazardly to boost performance -OK, an exaggeration, but a decent comparison). Of course, taking configuration into account, it's a totally different comparison.

Sega was obviously in a position where cost effectiveness was extremely important, so that issue with the jaguar really can't be ignored.

Some, such as Trip Hawkins (obviously a very particular and particularly biased source), stated that the Saturn was a machine for the Japanese market, period.

And 3DO was a machine for NO market, period.

(well, to be honest, the 3DO was also better suited to Japan than the US . . . they probably should have launched it in Japan in '93 and held off until fall of '94 for the US release -aside from the faulty, albeit experimental, market model)

The Saturn is a very troubled project. Many don't expect it to ship outside of Japan.

I wonder if they said the same thing about the PCFX. (which, of course, never did make it out of Japan

)

Sounds nuts, right? But that's what defined the project itself: ineptitude that became market insanity. The machine's few virtues matched better, specifically software bias, to the Japanese market; perhaps also explaining why SOJ were so intent on launching the flawed hardware at all.

LOTS of stuff in the industry sounds "nuts" in hindsight. It's finding and understanding the "WHY?" in detail that's the really interesting and important area of such historical issues.

(the 32x itself makes a LOT more sense when you understand the situation, albeit only part of that situation is even clearly defined)

With that context, you can start to make some arguments over which things were actually bad decisions for the time, and what are only bad in hindsight. (the Saturn's emphasis on 2D -especially in a mutually exclusive/inefficient manner -ie not using flexible blitter logic that could double for 2D and 3D drawing performance- is obviously a mistake for the time and not just in hindsight . . . even back in 1992 it wouldn't have made that much sense -the Jaguar designers had to make guesses on market demands based on what they knew in 1989-91, so much of that at least makes sense) One could argue the Flare guys were smarter/more capable/more insightful than Sega's best engineers . . . but they were still only 3 guys (only 2 guys doing most of the actual hardware design) while Sega had dozens of talented engineers and at least an order of magnitude more R&D funding . . . so there's no way they should have done better than the combined efforts of Sega's resources. (aside from misdirected R&D management -there was some meddling in the jaguar's design too on the management end, but there always is for better or worse -the "better" end would be keeping engineers in check and staying within practical cost constraints and staying on deadline)

And then the much more serious business decisions gone wrong. (the early Saturn launch being one of the most definitive examples)