god dammit i would like this channel a lot more if he'd just do a better job editing his videos. "hey look this game uses the genesis to display the background! hey look all the other games do it too so let's see an example for every. single. game." COME ONit's dark1x from gaf, so he's kinda like that sometimes if not quite anywhere near a black falcon...that said he does a better job of showing "same footage" of various ports of games than a lot of amateur comparison dudes do which i enjoy, like in the quake or unreal tournament videos, he made sure they were same maps, standing same place, and did split/third/etc. screens, too many fegs will just play one clip then another and not even the same levels or whatever...but that sorta is digital foundry's intention in the first place
that said, good summary of this piece of shit console. just a trash idea from start to finish and imo worse than the damn virtual boy. it should have been cancelled a long time before release when it was clear that nothing was coming together for it.
Scot Bayless: I joined SOA as a Technical Director right at the end of ’90 and, if memory serves, we started getting preliminary technical information about Sega CD very early in ’91. All the documents were in Japanese, and the guys in Tokyo just didn’t have the infrastructure to translate them, so we started hiring local translators just to get caught up on the docs. The first breadboard dev kits started showing up early in the spring. They were huge, fragile beasts, sensitive to electrical noise and prone to random lockups, but that’s the nature of prototype systems. By summer, we were in full burn, trying to get both our internal titles and several key external projects up to speed. That left us only a few months to launch; it was nuts. We literally worked around the clock all the way up to the launch in New York.
Sega-16: What were some of the technical issues? Former EA programmer Ernest Adams mentioned that developers had to check the status byte to “continually ask it whether the next block of data had arrived,” and he complained of the lack of hardware interrupt line from the CD drive. Were these major issues in game development? Was they a reason for the many cartridge games with CD soundtracks that were released (Wolfchild, Terminator, Sol-Feace)?
Scot Bayless: Ernest is absolutely right. Like him, I had to write polling code to watch that status byte. It was the only way to make sure you weren’t reading dead air. While the Genesis was an elegant design, the Sega-CD/Genesis combination was something of a Frankenstein. Add to that the schedule challenge of trying to get to the launch window with software that could work reliably, and it’s not surprising at all that many developers simply opted to use the unit as a CD player.
Scot Bayless: Well he was certainly involved at the incept point. 32X essentially started with a call to Joe Miller at CES. He, Marty Franz and a few of the senior production guys were in Joe’s suite when the call came through. Nakayama-san was on speaker phone and the word was, “We have to counter Jaguar. Make it happen.” The original concept for 32X was literally drawn on a cocktail napkin after that call.
Sega-16: That leads us to the question that has to be asked. There’s been great debate, great drama, about the origins of the 32X, so let’s settle it once and for all. Was the 32X something that began at SOJ, or was it SOA’s response to an order from Japan? There’s that famous story about you, Marty Franz, Scot Bayless and others in a hotel suite, coming up with the design for the 32X on a napkin. There’s the whole story about how SOJ basically wanted a Genesis 2, which was basically the same machine with a larger color palette, and you convinced them to go with the infamous mushroom add-on design instead. What’s truth and what’s urban legend?
Joe Miller: It’s not as dramatic at all as it’s been made out to seem. There was no palace revolt, and I don’t think there were any napkins involved, though we did have large stacks of sticky, white paper… easel paper all over the place. We were drawing pictures and diagrams (system diagrams) and doing lots of other things during those meetings which took place at all hours in rooms at our Las Vegas hotel.
Hideki Sato was right there with us, and I don’t want to rehash history – I certainly don’t want to rewrite it either, because there has been a lot said about it, about what exactly transpired there. Let me just put it this way. At CES (Consumer Electronics Show) – and perhaps we even had a little warning before CES – it became clear that there was a desire for us to take a product that was in the early design stages in Japan. It was a new platform (nobody was codenaming things “Jupiter” then, or even “Mars” at that point), and there was certainly an awareness that Japan had an idea of what they wanted to do with a Genesis platform that had more colors and was able to do 3D… take some of what we learned on the SVP chip – the polygon-pusher chip – and integrate something that was more capable and build a new platform. It was still going to be a 16-bit machine with some limited 32-bit capabilities.
When CES began, we started having discussions about the timeframe for this because there was a strong desire for it. It was January (this was winter CES), and there was a strong desire for whatever it was we were going to build to be available in the marketplace by Christmas of that year. That’s a tall order for a start-from-scratch machine. Nothing exists; no boards exist, no chipsets… a tall order for anybody at any organization to say “let’s design hardware; let’s build it; let’s get development systems, and let’s have titles that are compelling enough to actually gain the attention of our customers… not alienate them but actually cause them to be excited about it,” in basically a six to nine-month timeframe.
Joe Miller: No, it didn’t have an effect on what we were working on. It obviously made it more important that we resolve thorny challenges – and we had many. In all the things that have been written about the transition from the 32X to the Saturn I don’t think this one has been told. One of the rationales for the 32X was to give developers a step-wise way to get into 32-bit and multiprocessor programming. Saturn had eight processors, and the set of tools, the documentation, training, sample code – to get that platform to do what it was designed to do was a very complex and painful learning process for developers, including the best and sharpest minds that Sega had to bring to bear on it, in both Japan and the U.S.
Our challenge, our focus was to make sure we had documentation, we had compilers that were producing correct code, that the SH-2 was optimized, that we had debuggers, hardware emulators and a variety of other things. I don’t think very many people understood the layers of infrastructure required to actually get developers up to speed on a new platform.
One of the stories that hasn’t been told is that the 32X actually helped development teams that had been successful on Genesis make the step up to the Hitachi SH-2 – dual SH-2s in that case – and there were a bunch of routines, a collection of tools and compilers that were created to support 32X development that were adapted and ported over to Saturn. The early launch meant that we had to accelerate all of that in the pipeline as well – all the cross-development tools, all the documentation – and get developers ramped up 6, 8, 12 months earlier than we had expected or intended.
It's really amazing how quickly Sega squandered all the success they'd spent nearly a decade building up during 1994-1995.
Fuck I'm really curious as to what that SVP Genesis version of VF looked like