Dreamcast was absolutely the better system to have in 2000 and parts of 2001. I was very content with it. PS2 launched in Japan, a store had it on demo and it was sort of a relief to me; the games I saw were TTT, RR and SF EX3. They didn't really push beyond what the DC offered already. While PS2 was still a no show in the west, we played DoA2, JSR, Virtua Tennis, RE CV and Crazy Taxi. Wouldn't want to miss those evenings. Sega got it on since the western launch day, they kept pushing and pushing games.
In retrospect though, I don't think i've seen many DC games looking more impressive than Ridge Racer, SSX and a few other PS2 launch games. DC games looked very clean and colorful, but its geometry was rather simple. But PS2 had this immense hype going on, I just expected way too much.
TBF, budgets play a part in that. A lot of the big AAA 3P developers, especially Japanese but some Western ones too, they didn't put their biggest teams or budgets onto Dreamcast games. Yes companies like Capcom supported DC a lot but most of that were arcade ports outside of the odd RE: CV (which at least some RE fans will say didn't have the "A-team" insofar as the writing and scenario planners compared to the PS1 titles). Other companies like Namco just gave it one (maybe two) games and then dipped.
Others like Konami, basically relegated it to an enhanced port machine for PS1 titles (Capcom did this as well), but you'll also notice during that 1999-2001 period many of these
same devs were obviously saving their bigger (and therefore bigger-budget) games for PS2: Devil May Cry, MGS2, Silent Hill 2, Final Fantasy X, etc. were all PS2 games and had much larger budgets than most of the 3P Dreamcast games of that period outside of Sega's own.
Also FWIW, 3P devs probably had more market incentive to put more resources towards those PS2 games compared to Dreamcast efforts, even in the first couple of years, considering where Sony and Sega were brand-wise by then. So a lot of them probably left more performance on the table WRT their Dreamcast games, which combined with smaller budgets and team sizes, likely left their PS2 efforts looking better even if Dreamcast was capable of more. You did have some 3P devs make the bigger investments and efforts, like Tecmo with DOA2 and Game Arts with Grandia II, but if you wanted most of the bigger AAA Dreamcast games you more or less had to count on Sega themselves (and keep in mind Sega did do smaller efforts too, obviously, like Chu-Chu Rocket, Samba de Amigo etc.).
The Mega CD was maybe not a failure financially but seeing how the majority of the Megadrive library stayed on cartridge, I'm not sure Sega achieved what they wanted with this add-on. The Megadrive sold 42-47 Millions of units while "only" 6 millions of the people who bought it also bought the Mega CD worldwide. That's a ratio of about 12 to 14%. Not even a 1/6th of the public bought it. I was obviously not working at Sega back then but I'm sure they aimed higher than this. Yes it's the third best selling add-on after Kinect and the Balance Board, but these were only accessories for certain games (well for kinect it's a bit more complicated but that's not the subject). This thing was a hardware upgrade for the Megadrive and I'm sure they were planning to try and go all out on CD-Rom games but had to backdash when they saw this. It's just like Nintendo with the N64 and the Gamecube: They made profit with these 2 consoles, but they obviously wanted more market shares than what they got on these generations.
It is complicated WRT Mega CD because that's really something Sega of Japan were pushing to revitalize MegaDrive in Japan. So your point, it holds true, but I think it holds
much more true for SoJ than the Western divisions, because in the West MegaDrive/Genesis was doing very well. Due to that the Sega/Mega CD for them might've seemed more like a value-add to the ecosystem compared to being a wholly new hardware push, especially when you consider they sold it for a profit from the get-go while (IIRC) the Genesis/MegaDrive were sold for a slight loss (I don't think Sega particularly pushed the "razor & razorblades" model that Sony (and later Microsoft) did, but we know they didn't sell Genesis/MegaDrive for hefty profit the way Nintendo did their hardware).
12%-14% may seem like a bad install base ratio but it's not that much further off than what PS4 Pro was to PS4 sales, and I think that would serve as a better analogy. Like PS4 Pro, Mega CD wasn't intended to outright replace MegaDrive, but to expand revenue opportunities and expand the ecosystem offering. It was also a testing bed for CD technology experience that they went on to apply to the Saturn.
The problem there is opportunity cost and their war chest. It takes a lot of time and money to R&D, launch, and distribute an add on and all the supporting documentation and tooling and getting developers onboard (or trying to). For genesis they did not one, but two big add-ons (amidst company infighting).
Imagine if they had let Genesis live a little longer with full support and invested the resources in better Developer tooling and HW components for the Saturn (as well as their distribution channels/retailers), and game studio resources (or simply saved some of that away). A better Saturn generation would have positioned themselves better for the Dreamcast project (where avoiding to piss of EA would have left Madden, FIFA, NHL, and other EA games stay on the platform…).
One thing I do think they should've done with Genesis/MegaDrive is push the SVP add-on chip a la Virtua Racing vs. what they actually did in 32X. 32X actually caused a shortage of SH2 processors for Saturn, since both systems used that processor and both used two of them in parallel. Apparently Sega's goal was to have 32X serve as a training bed for Saturn development, but the issue there is 32X used the 2nd SH2 for graphics while Saturn used
completely different VDP1 & VDP2 chips for that task.
Basically a dev making a 3D game on 32X might've came up with a wholly unique pipeline and rendering technique done in software on the 2nd SH2, that would not be easily translated to Saturn because it used fixed/dedicated graphics hardware built to function its own way. So I don't really buy that line of thinking from people who put it out there. That aside, the Saturn didn't really have "bad hardware"; it was challenging to develop for yes, but if you learned the hardware you could make it sing. Sega managed to, certain 3P like Treasure did as well.
Hardware-wise Saturn's problem was Sega relied on the old documentation method; they provided all the hardware documentation necessary but it was a lot of stuff to read up on and learn and game development was about to increase drastically in complexity thanks to 3D. The early Saturn SDK was pretty poor, that didn't help things, and then Sony came in and built a very friendly SDK that emphasized programming in C while still getting almost 100% out of the hardware (to get best results from Saturn, you needed to code in Assembly language).
Sony read the changing winds of game development behind-the-scenes better than Sega or Nintendo, partly thanks to them being an outsider (in terms of being a platform holder), which allowed them to have a fresh perspective. Then, they were able to capitalize on that thanks to having a LOT more money and resources to them than Sega and Nintendo combined. There's a reason Nintendo took a different path shortly into N64 that wasn't directly competing with Sony: they
literally couldn't justify it financially.
Sega would've came to that same realization if they weren't up in their own ass fighting among the different divisions, and that would've allowed them to lean in on their specific strengths (arcade gaming, merging/synergizing their arcade and home console efforts more cohesively), perhaps even stem the gradual downsizing of the arcade market in the West while at it.
Sadly, that's not what ended up happening.
The Mega Drive came out in 88. SEGA supported the Mega Drive too long. SEGA should have dropped all In-House software support in 1994 for the MD and moved all In-House production to Saturn, just for example a 32-bit Comix Zone with 24- Bit visuals,, 2 player mode and a CDDA soundtrack would have been something special
If we never had the 32X SEGA would have been in a far better position too, but what really killed SEGA
Agreed that 32X being out of the picture would've been for the better, but I don't know if dropping MD in 1994 would've been for the best, either. Remember, MD/Genesis was doing very well in the West, it's just in Japan where they had done poorly.
By 1994/1995 SoJ had more or less left MegaDrive behind anyway, but with no 32X I think Sega could've pushed the SVP more. You mention Comix Zone for example: picture that as a SVP-enhanced Genesis cartridge with an accompanying audio OST CD that could run the game music and even some sound effects in CD-quality audio if the player had a Sega/Mega-CD unit attached as well.
This way, instead of a 32X in 1994, Sega could've released a merged MegaDrive/CD unit (which they already did in a way prior with CDX, and partners like JVC had done with WonderMega) at a cheap mass-market price, and saved the 3D enhancements for per-game SVP cart releases. The MegaDrive/CD unit wouldn't of been viewed as
"yet another new add-on", and going with the SVP solution would have similarly avoided such an image, while still providing very decent (for the time, and for that generation) 3D (SVP was more capable than the SuperFX chip and IIRC not that far off from SuperFX2).
Meanwhile, with the Western market taken care of for a while (even into late 1995) via that solution, SoJ doesn't feel rushed to release Saturn in America too early, and they have more time to think of a response to PlayStation.
Maybe they realize they can't outdo PS in terms of 3D, but they integrate something to a benefit in adding MegaDrive BC, and include enhanced SVP cartridge games to run via BC on Saturn as well. Keep the Saturn's internal design simple, and since you have the MegaDrive BC built in as well as functionality with SVP cartridge games, Saturn games wanting more graphical power could just do so either combined SVP & RAM expansion carts (Capcom-style), SVP & ROM expansion carts (KOF-style), or just have a future enhanced SVP-type graphics cartridge that can potentially be included with a new game release (a-la the RAM Xpansion cart with DK64 on N64), or sold separately at a cheap price ($40-$50) and then included in a new system SKU at a later date.
That way, you get extended MegaDrive/Genesis support in its best markets for 1994-1996 or even into 1997 (people forget the SNES still held majority marketshare worldwide even into 1997, where PlayStation finally overtook it) with pseudo-5th gen offerings thanks to MD/CD combo unit and/or expanded SVP-cartridge game releases, you get a 2D powerhouse/modest 3D Saturn for Japan in 1994 that also has MegaDrive & MegaCD BC support and easy expansion support including for enhanced 3D & 2D graphics capabilities, and to that effect, a Saturn that has a much bigger draw in the West thanks to BC support built-in from Day 1, which can potentially offset some of the momentum of PS1 thanks to still being capable of 3D but having more time to polish the launch lineup, MD/Genesis BC and being a more affordable system.
And in terms of Sega unifying the arcade & home initiatives better, well doing things like allowing for save transfers in arcade & home versions of games, maybe an early version of what we know today as Achievement & Trophy systems (probably built into the games themselves and on a per-game basis, but it'd of been a nice start), focusing on bringing unique arcade-style peripherals to the home market for fuller arcade-style immersion at home, etc. Giving them a unique identity from Sony and Nintendo that could've helped them continue their own path to success, instead of the nightmare that actually transpired.
Well, we can always dream can't we...
16-bit sales were declining in 1994.
Maybe, but they still accounted for the lionshare of industry sales even into 1996. It seems like 1997 is when the 16-bit market finally collapsed, because that's around when Nintendo ceased official support for the SFC/SNES.
Sega didn't need to cease Genesis support in 1995, much less 1994, in the West. Not at all. There were more affordable options for them to still capitalize on that market and ease in better to Saturn without alienating fans. Their business mindset was just completely too borked to see, let alone contemplate, any of those options clearly.
