Quantcast

Page 5 of 7 FirstFirst 1234567 LastLast
Results 61 to 75 of 92

Thread: More Alternate Reality Stuff or Did the 32x Make Sense?

  1. #61
    Hero of Algol kool kitty89's Avatar
    Join Date
    Mar 2009
    Location
    San Jose, CA
    Age
    30
    Posts
    9,725
    Rep Power
    64

    Default

    Some more thoughts pertient to this thread:

    http://www.sega-16.com/forum/showthr...l=1#post607329

    Quote Originally Posted by Chilly Willly
    SVP
    One processor - a standard Samsung SSP1601 DSP. This is a 16-bit DSP with a multiplier. As with most DSPs, it has separate instruction and data spaces. The instruction space is mapped to part of the game rom, while the Sega chip holding the DSP also has embedded ram for the data space. It uses external bank pointers to read or write the rom and DRAM. From tests, it seems to be clocked at about 13MHz.
    I'm betting it's using 7.67 MHz x2 (15.34 MHz -or 15.2 PAL), given that it derives its clock from the MD CPU clock line, and I also assume it's not using a fractional multiplier. (though 11.5 MHz would be even closer to what emulator results show)
    In any case, this should mean SVP running at a full 23 MHz would have around double the polygon count of VR. (framerate or screen size are more fixed limits, of course)
    The DSP itself is rated at 25 MHz though . . . but heat or power consumption may have been factors (especially given it uses the MD power and especially considering what the added ASIC logic might pull beyond the DSP's rated amperage). As it is, it doesn't run that warm (cooler than the MD VDP ASIC) and I'm not totally sure why the US/EU versions got that heatspreader. (I'm thinking it has more to do with the added RF shield . . . and given how sloppy the thermal paste is, I'll bet the "bare" JP ones actually run cooler )
    GIven the nature of the SSP-1601 though, Sega definitely COULD have added a lot more "extra" stuff using a larger ASIC, and pretty fast and dense logic too if they'd gone with standard cell (albeit more engineering effort than gate array). Would have been interesting if they'd done the 32x that way instead. (beefed up SVP ASIC with superVDP-like functionality built-in along with some audio DACs, and dual 128k DRAMs instead of the one -added blitter hardware would have been great, but given they avoided that with the existing 32x, I rather doubt it would be any more likely there, given a similar timeframe)
    Hmm, actually 1x 512kB DRAM might have been better. (you'd have contention to deal with, but double the work RAM, less board space, and similar cost or cheaper overall -512k chips were around 3/4 the price/bit of 128k in 1994) Hell, you could probably still cram the whole thing into a slightly bulkier S&K style cart (with bulk mainly due to A/V in/out and AC ports) and save much more on internal complexity and packaging/shipping/etc costs over the actual 32x. (still the hassle of an added power supply and av cable set-up though -more so for model 1 users, unless they provided adapters for output as well as input -rewiring TV and/or audio receiver inputs can be a pain . . . especially if you've got it all embeded in an entertainment center AV cabnet )


    I commented a little in this thread before on middle-ground between SVP and 32x, but mostly just on an SVP cart with either 128k or 512k DRAM. Using a beefier ASDSP than the existing SVP, and including SuperVDP type functionality (with 2x 128k or 1x 512k DRAM) and added audio DACs would still have been much cheaper than the existing 32x, especially if they crammed it all into an S&K style cart. (still less convenient to set up due to the AC in and AV mixing stuff and potentially different AV cables on the output end for model 1 users)

    The existing SVP is bottlnecked in rendering speed by the 4-bit tilemap pixel format combined with DMA bandwidht issues (color limits aside), but it also seems to run at a much lower clock speed than the rated 25 MHz and originally assumed 23 MHz (3x7.67), closer to 10 MHz (so it's probably actually doing 2x 7.67 MHz or perhaps 1.5x -the latter would best fit the experience emulator programmers have experienced with VR running ar full 15 FPS speed with SVP emulated at ~10 MHz, and no gains from going faster).
    So it's fair to say that polygon counts could have rougly doubled over what VR on the MD shows.
    Super VDP style line fill should speed things up a fair bit too. (or, rather, free up more DSP time for other things)

    Also, clock for clock, the SSP-1601 DSP is actually faster at some things than an SH2 . . . more primitive and limited (and not good for general CPU stuff), but still significant.


    With the much cheaper and more minimal components, smaller board, smaller/lighter packaging, etc, I think it's far to assume that something like that could have hit around half the price of the existing 32x in 1994 . . . that's still about 1.5x what I was suggesting for a vanilla SVP add-on, but again this is middle ground here. (and, honestly, I think most of the existing 32x games could have been done quite well with hardware like that . . . somewhat tougher to program perhaps, but mostly similar in overall performance and features)
    Doom would probably have been among the toughest to do compared to 32x though . . . you don't have a nice fast CPU to push the game code onto, so you'd have to split things up between the 68k and DSP a lot more. (and other trade-offs depending on the 2x128k or 1x512k DRAM arrangements)
    Last edited by kool kitty89; 09-09-2013 at 08:27 PM.
    6 days older than SEGA Genesis
    -------------
    Quote Originally Posted by evilevoix View Post
    Dude it’s the bios that marries the 16 bit and the 8 bit that makes it 24 bit. If SNK released their double speed bios revision SNK would have had the world’s first 48 bit machine, IDK how you keep ignoring this.
    Quote Originally Posted by evilevoix View Post
    the PCE, that system has no extra silicone for music, how many resources are used to make music and it has less sprites than the MD on screen at once but a larger sprite area?

  2. #62
    Road Rasher
    Join Date
    Feb 2013
    Posts
    423
    Rep Power
    16

    Default

    Quote Originally Posted by sheath View Post
    Well that is interesting. [URL="http://youtu.be/HPuL90VENzk?t=7m"]
    With this breakdown I just do_not see how two add -ns in five years for the Genesis was a bad idea from a marketing perspective.
    Unlike your personal experience, most game consumers during that time (children, as you were) were gifted their consoles by parents and/or relatives. They were not things that the core users themselves bought, were not something that the gifters themselves would keep buying. Most game console owners are single console owners for that very reason, with very few being multiconsole owners (I list myself among the latter, but then again I had very few software titles growing up, didn't own a computer until I was in my junior year of high school, etc.).

    More importantly, though, we run into the question of "why do these add-ons exist/for what type of consumer are these things for?". With Kinect, MS had a ready answer: the mainstream consumer/newbie gamer interested in the sort of motion sensing and local multiplayer as those who went out and bought a Wii. That's why Kinect Sports, and even stuff like Zumba Fitness, were such important software titles for Kinect: it sold that add-on in the same way that Wii Sports sold folks on Wii (and, btw, this is one reason why WiiU hasn't really taken off: there's no "killer app" to sell folks on the Upad). The Wii Balance Board? Wii Fitness sold that, and it was clearly targeting what Nintendo called the "non-gamer market" (a misnomer, as anyone who plays games are, by definition, gamers; the more proper term for the Wii Fit market should've been "fitness minded game consumers" or something along those lines).

    Now, let's turn our attention to Sega CD. For whom was Sega CD originally intended? Well, it was originally intended for game consumers in Japan who were flocking to NEC's CD-ROM add-on for PC Engine, which made it, for a time, the home of pretty high selling "JRPGs" (another misnomer, as "JRPGs" should really be labeled "console-style RPGs"). Well, that experiment didn't work in Japan. Mega CD didn't sell. So SoJ tasked SoA and SoE to try to sell it in the West. But...eh. These were high priced add-ons, TGCD hadn't caught on earlier, and the TG itself was a pimple on the ass of Sega that had been popped on day one and long since had its tiny scar cleared up. "JRPGs" weren't major sellers in the West, and sure as hell weren't system sellers, so the few Sega CD RPGs there were weren't going to sell the add-on. So what then? FMV! Yeah, that's the ticket. More tech! Awesome! Except, y'know...the market for that was pretty small, especially for the price of that add-on.

    32X? I think history has spoken on it being a solution to a problem that really wasn't all that apparent. Jaguar turned out to be a paper tiger. 3DO could've been a contendah but its fight plan was so faulty that it ended up knocking itself out early in the fight. And, again, 32X was $150 at a time when Sega had several high profile Sega Genesis releases that year. Remember that the Western markets are more price sensitive than others, so each new add-on hardware sale and one or two 32X software sold to accompany it could've meant a bunch of canabalized Genesis software sales, or a cannibalized Game Gear sale, etc.

    And from a marketing perspective, it's far easier, and less costly, to sell something with existent, ongoing, and still relatively high demand (Genesis, possibly Game Gear) than practically create demand from scratch (Sega CD, 32X). Every dollar spent on 32X marketing could've been better spent on marketing a handful of Genesis software titles, or something along those lines. Every dollar spent marketing 32X was a dollar less on marketing Genesis and/or Game Gear software.

    Lastly, from a basic business perspective: Genesis won the West. We know that now. SNES won the world-wide race. But here's the thing: Nintendo may have been making higher profit margins on hardware sales than Sega was at the time. Genesis had the price advantage (a big reason for it outselling SNES in the West), but Sega being so price aggressive cut into some of their potential profits on each unit sold. Nintendo was more aggressive with SNES than they had been with NES (and wouldn't be more aggressive on price than the GC days, and the 3DS price drop from $250 debacle), but still priced their hardware to maximize profits as much as possible. End result: Nintendo sold less than Sega, but may have been earning more overall.

    So why, in light of that, would anyone think Sega overextending their cash to sell more hardware at razor/blade principle a "good idea"? It wasn't. It simply wasn't. That's from a marketing perspective, from a business perspective. Nakayama thought Sega was in the hardware business, but he was half right, which made him half wrong. Sega was in the hardware and software business, specifically the video game business. K.I.S.S is the best practice in any business, and doubly so in the games business.

  3. #63
    Road Rasher
    Join Date
    Feb 2013
    Posts
    423
    Rep Power
    16

    Default

    On topic, and better case scenario with 32X:

    No 32X released. Maybe SVP pushed on a lock-on cart, and I'm thinking Sonic & Knuckles would've been the best release for that sort of thing. Had it been built into that game, with it being used for mostly eye candy on some stages and paricularly 3D polygonal special stages, and maybe sold at slightly above cost it probably would've achieved greater penetration than the $100 Virtua Racing did.

    Even better case scenario:

    No 32X. Maybe some SVP chip releases here and there, maybe not. Regardless, Sega focuses on Genesis and Game Gear alone, slowly preps for a '95 Saturn launch behind the scenes.

    Best case scenario:

    Same as above, except somehow Sega adds Genesis VDP emulated instructions to a revised Saturn VDP, sets up the 68k processor in Saturn similar to how they set up the Z80: can be used as the main system CPU with appropriate sofware, specifically previous gen software, specifically Genesis software. Put that Saturn cartridge slot to greater use, enable Genesis BC out-of-the-box. In addition to a consolidated and less costly to manufacture Saturn set up (single VDP handling what VDP1 and 2 ended up doing, simplified RAM set up, etc.), that, alongside a competitive price point without having Sega lose so much money on each unit sold, makes for a VERY good competitor against PS1 and N64. It's been stated that BC "doesn't matter", that folks buy new systems to play new games. That's true enough, but having BC allows for a longer tail, and longer revenue potential, for the previous gen software as well, and potential for less "mental blocking" of desire for the new system ("and it can play my older favorites!").

  4. #64
    Hero of Algol kool kitty89's Avatar
    Join Date
    Mar 2009
    Location
    San Jose, CA
    Age
    30
    Posts
    9,725
    Rep Power
    64

    Default

    Quote Originally Posted by spiffyone View Post
    On topic, and better case scenario with 32X:

    No 32X released. Maybe SVP pushed on a lock-on cart, and I'm thinking Sonic & Knuckles would've been the best release for that sort of thing. Had it been built into that game, with it being used for mostly eye candy on some stages and paricularly 3D polygonal special stages, and maybe sold at slightly above cost it probably would've achieved greater penetration than the $100 Virtua Racing did.
    I agree to a point . . . or more in the general sense that a better "killer app" than VR would be good (and the point that you coudl use it for a LOT more than jsut polygon games -which I've argued in detail already).

    However, I'm not 100% sure it would be practical to cram SVP+RAM+ROM onto a S&K sized cart module (and much taller and you start to run into balance/stability issues -then again, Game Genie wasn't too bad, and using 1/2 height carts for SVP specific carts would help). Meh, that probably would have worked OK, and definitely would have been a better selling point too.
    Hell, that would also give more reasoning to splitting Sonic 3 into 2 parts in general. Though it also would mean a later release of the SVP (fall rather than Spring), but still in tine for the holiday season and also potential for manufacturing costs to come down somewhat.
    Having S&K built-in would nix the whole idea of a ~$50 bare bones SVP add-on release (following an initial release with a VR ROM cart bundled), but I suppose the marketing value of a killer app like S&K would be well worth that trade-off too.

    Again, I think it would be worthwhile to give SVP a small (~.4 amp 9V) external PSU . . . for the modest cost that would add (or force Sega to take a loss), it would be a major boon to performance since the chip could run at 23 MHz rather than 1/2 that (emulation testing shows VIrtua Racing hits full speed at just over 10 MHz with no gain beyond that, and the 7.67 MHz clock is used for the source, so 3/2x that would make sense). Additionally bumping from a single 128kB DRAM to a 512 kB DRAM wouldn't add too much to component cost (significantly cheaper at that density, and Sega was already stocking large numbers of FPM DRAMs in both densities for the Sega CD). That and the audio DAC thing.
    With all that it still probably could have been able to manage $100 with built-in game for fall '94. (the larger amount of RAM would make it MUCH easier to port over compelling PC games of that time without a ground-up rewrite -like SNES Doom needed, plus the more general advantages it would have for all games in general -more buffer space to decompress into in additon to the DSP resource for more complex streaming decompression -in cases where DSP sound/graphics processing isn't super demanding)


    Heh, another interesting possibility would be having a special eddition Genesis with S&K and SVP+RAM built-in. You could argue they could have switched over to that as standard, but that might risk the budget-market sector too much, since SVP games weren't going to appeal to everyone and perhaps not even the majority of Sega's market either. (or, rather, you'd really want to look at who was buying the MD/Genesis in '94 up through the budget-market years of the late 90s, and see whether that would mesh with the added cost and performance SVP would provide) It probably could have kept up with SNES pricing, but with less of an advantage and/or lower margins.
    Probably worth marketing as a deluxe option, but not as standard at that point IMO. (so a good deal for new prospective genesis owners who also want SVP -a better deal than buying it separately, but still giving the standard MD models for users who don't care about that -you'd also probably have the odd existing MD owner who'd prefer the sleak form factor of the all in one model over the add-on, and would be willing to invest in that and/or trade/sell their old console, but that wouldn't be common)

    Even better case scenario:

    No 32X. Maybe some SVP chip releases here and there, maybe not. Regardless, Sega focuses on Genesis and Game Gear alone, slowly preps for a '95 Saturn launch behind the scenes.
    For on-cart enhancements, I think Sega really should have started exploiting the potential much earlier than 1994, and also focused more on cheaper/simpler add-on hardware that would still have a significant impact on graphics/sound/etc. (I have a separate thread detailing that though )


    Same as above, except somehow Sega adds Genesis VDP emulated instructions to a revised Saturn VDP, sets up the 68k processor in Saturn similar to how they set up the Z80: can be used as the main system CPU with appropriate sofware, specifically previous gen software, specifically Genesis software. Put that Saturn cartridge slot to greater use, enable Genesis BC out-of-the-box. In addition to a consolidated and less costly to manufacture Saturn set up (single VDP handling what VDP1 and 2 ended up doing, simplified RAM set up, etc.), that, alongside a competitive price point without having Sega lose so much money on each unit sold, makes for a VERY good competitor against PS1 and N64. It's been stated that BC "doesn't matter", that folks buy new systems to play new games. That's true enough, but having BC allows for a longer tail, and longer revenue potential, for the previous gen software as well, and potential for less "mental blocking" of desire for the new system ("and it can play my older favorites!").
    Not realistic.

    Sega definitely COULD have built a console around integrated (and efficient) backwards compatibility, even one that included Sega CD compatibility (32x would be more of a wasteful mess though), but they simply didn't have such design goals used with the Saturn. (side note: aside form having no CD making that easier, my "Sega CD should have been its own console" scenario would also lend itself to considerably more efficient/straightforward compatibility and evolutionary upgrades of the architecture)

    Had Sega actually taken the existing MD+CD and focused on a design that made the most of the existing components while adding the necessary hardware for a full next-gen platform, it could still have been MORE cost effective than the Saturn if they streamlined it enough. The Saturn goes overkill on cost (relative to useful performnace) in several areas that could have been moderated a lot more, and would have made plenty of sense to do so with the MD+CD hardware in mind. We've had some very long-winded discussions on this topic too.

    ---{rant on tech specifics of such a console}
    The Sega CD's CD-ROM subsystem is significantly cheaper than the Saturn's and already supports a 2x data rate mode (just needs a 2x speed drive mech to match ), the MD+CD CPUs could have been bumped to 16.67 MHz (from the CD's 50 MHz master clock) to maximize potential performance given the relative low cost of 68ks in general by that point. Bump up MD work RAM, enhance the MD VDP somewhat and have it mesh with a new VDP (or just a VDC with built-in digital video mixing, palette expansion, and maybe alpha effects for the MD layers) and an enhanced blitter compatible with the MD's ASIC capable of rendering at more color depths (and to a lienar framebuffer) and rasterizing distorted quads and/or triangles (or trapizoids like the Jaguar 2 . . . and given the Sega CD blitter works a lot more like the Jag blitter than the Saturn/3DO sprite warping engine, that would probably make the most sense) and added shading/lighting/blending effects (possibly saving cost/complexity using custom color models like the Jag did -using 4-4-4-4 RGBY would work too). A single SH2 would have been fine for handling the main game engine (more crafty programmers could make heavier use of the 68ks, but you wouldn't NEED to) and maybe a dedicated DSP for 3D coprocessing (a 25 MHz SSP-1601 would be faster than an SH2 for 16-bit matrix math, and faster than the SCU in the Saturn too, though it doesn't do 32-bit . . . plus it would be programmable for use in added sound or other effects, not just 3D math . . . re-using the SSP-1601 could also mean built-in SVP compatibility). The MD+CD sound hardware was decent, especially with added sound ram for the Ricoh chip. (though probably better with some DMA stereo the DSP/CPUs could work with for decoding/mixing added SFX and/or streaming audio without having to load in chunks into PCM RAM) Various considerations on how to configure all the system buses, how many discinct chips there would be, etc, but I've rambled enough here.
    ----{end rant}



    In the end though, backwards compatibility is neat, and a useful marketing tool and convenience, but it's never been critical for the success or failure of a console. (computers, yes, but not consoles)
    I'd go so far as to argue that Atari's lack of compatibility in the 5200 would have been a non-issue in the long run with the right marketing/management . . . though there's other areas that make the 5200 stupid. (like the fact that they had a cheaper AND 2600 compatible design in the 3200/SS-1000/Sylvia but abandoned that, and then, rather than just marketing a new cost reduced Atari 400 derivative as their top-end console platform, they go and make something just as expensive -technically more expensive for consumers- and virtually identical internally, but make it incompatible . . . and for no good reason -no added lock-out or such)

    Actually, there's the argument that the less sucessful a console is, the more important backwards compatibility becomes, due to the importance of maintaining loyalty of the small consumer base you have managed to establish, and to expand available software with the last-gen stuff. (though, from a Japanese PoV, that should have made that significant for the MD and MCD )
    6 days older than SEGA Genesis
    -------------
    Quote Originally Posted by evilevoix View Post
    Dude it’s the bios that marries the 16 bit and the 8 bit that makes it 24 bit. If SNK released their double speed bios revision SNK would have had the world’s first 48 bit machine, IDK how you keep ignoring this.
    Quote Originally Posted by evilevoix View Post
    the PCE, that system has no extra silicone for music, how many resources are used to make music and it has less sprites than the MD on screen at once but a larger sprite area?

  5. #65
    Hero of Algol kool kitty89's Avatar
    Join Date
    Mar 2009
    Location
    San Jose, CA
    Age
    30
    Posts
    9,725
    Rep Power
    64

    Default

    Quote Originally Posted by spiffyone View Post
    Now, let's turn our attention to Sega CD. For whom was Sega CD originally intended? Well, it was originally intended for game consumers in Japan who were flocking to NEC's CD-ROM add-on for PC Engine, which made it, for a time, the home of pretty high selling "JRPGs" (another misnomer, as "JRPGs" should really be labeled "console-style RPGs"). Well, that experiment didn't work in Japan. Mega CD didn't sell. So SoJ tasked SoA and SoE to try to sell it in the West. But...eh. These were high priced add-ons, TGCD hadn't caught on earlier, and the TG itself was a pimple on the ass of Sega that had been popped on day one and long since had its tiny scar cleared up. "JRPGs" weren't major sellers in the West, and sure as hell weren't system sellers, so the few Sega CD RPGs there were weren't going to sell the add-on. So what then? FMV! Yeah, that's the ticket. More tech! Awesome! Except, y'know...the market for that was pretty small, especially for the price of that add-on.
    There's a relatively vague explanation for SoJ's design methodology (or vague as far as anything definitive).

    PART of it was obviously due to NEC (and the entire industry's) interest in CD-ROM at the time, part of it was the general interest in low-cost bulk storage media (hence the floppy disk add-on planned and scrapped prior to the CD), and then there's some issues of general hardware upgrade to the MD architecture.

    If it was JUST about competing with NEC's platform, even assuming Sega had some knoledge of NEC's plans for the Super System card upgrade, the Sega CD goes WAY beyond the scope of that. Had they just wanted a "better than CD-ROM^2" add-on built around the MD architecture, a basic CD-ROM drive with minimalistic MCD for drive handling, 256 kB of DRAM, and basic DMA stereo sound (using the MCU or MD CPU for software mixing as needed -though simple 8-bit stereo PCM +MD FM+PSG is already better in most respects than the PCE's mono ADPCM cannel -which cannot be software mixed or multiplexed- along with the onboard PCE sound). Doing that, you'd actually have more usable RAM and overall functionality than even the Super CD or duo. You wouldn't improve the MD's color limitations, but neither did the MCD.
    Not only would have have been cheaper, but also should have been simple enough to engineer and push out earlier than the MCD. (a late 1990 release would have been significant for the Japanese market) Then again, Sega may have already hit other roadblocks preventing an earlier release, so that might not have been possible.

    The much larger chunk of added RAM, the scaling/rotation capable blitter, the 12.5 MHz 68k sub-CPU, and the 8 channel PCM sample synth chip all point to MUCH more than just wanting to compete (or eve one up) NEC's paltform.
    It seems much more like a combination of competing with NEC, and one upping Nintendo (both the SNES and the looming SNES CD), and adding functionality to cater to Sega's high-end arcade titles of the time (and/or similar genre stuff done as home-console specific games), and quite possibly to cater to some of the computer game market as well. (FM Towns and 68000 stuff, or even stuff starting to emerge in the western PC game markets at the time)

    They made a system so powerful and complex that it's petty close to being a standalone console already . . . all that's missing is a VDC and I/O ports for the controllers. (which kind of leads into this thread: http://www.sega-16.com/forum/showthr...-Reality-stuff


    And with all that in mind, it's really hard to say exactly what sort of consumer base they were targeting. Given the price point tied to all that, I'd say it's the high-end bracker of the mainstream consumer, at least potentially. It would depend on software and marketing to directly cater to that bracket, and the hit and miss software and marketing (SoA's valiant efforts included) didn't really pan out in that sense.
    A sub-section of that would be a throw-back to the floppy disk add-on idea though, for lower cost game media potentially being attractive to more logical consumers . . . the same sort who would buy a game computer for several times the price of a console to take advantage of the cheaper games. (with the added potential of disk+RAM based game content being an added marketing point) And I mean the likes of ST and Amiga in Euope not early 90s multimedia PCs. (especially brand-name ones) Thoguh 1993 gaming PC vs 3DO might be a more valid comparison.


    Additionally, when the Sega CD was being developed, Sega had no solid idea of how bad NEC would botch things in the US and Europe . . . and even after that point (by the time the Sega CD came to the west), you could easily argue that Sega was trying to fill a role that NEC left completely open due to their incompetence. (which is true . . . though I'd argue NEC could have pushed CD MUCH harder than Sega due to their in-house advantages and resources -again, more like Sony later did)


    32X? I think history has spoken on it being a solution to a problem that really wasn't all that apparent. Jaguar turned out to be a paper tiger. 3DO could've been a contendah but its fight plan was so faulty that it ended up knocking itself out early in the fight. And, again, 32X was $150 at a time when Sega had several high profile Sega Genesis releases that year. Remember that the Western markets are more price sensitive than others, so each new add-on hardware sale and one or two 32X software sold to accompany it could've meant a bunch of canabalized Genesis software sales, or a cannibalized Game Gear sale, etc.
    Yeah, Nakayma's apparent concerns were a no go in reality. Howver, the expanded market target SoA ended up molding Mars into did make a good bit more sense. It still was a bad idea considering everything else Sega had to offer for the time (and obvious alternatives), but it at least made a hell of a lot more sense then "we have to counter the Jaguar".

    In hindsight it's much worse still with the mess they made after the fact in '95/96 and the related Saturn debacle, but my arguments are more centered on practical points of view from the time . . . and even then there were much more practical alternatives to achieve the goals of the 32x. (which is the main set of points I argue in this topic)

    Lastly, from a basic business perspective: Genesis won the West. We know that now. SNES won the world-wide race. But here's the thing: Nintendo may have been making higher profit margins on hardware sales than Sega was at the time. Genesis had the price advantage (a big reason for it outselling SNES in the West), but Sega being so price aggressive cut into some of their potential profits on each unit sold. Nintendo was more aggressive with SNES than they had been with NES (and wouldn't be more aggressive on price than the GC days, and the 3DS price drop from $250 debacle), but still priced their hardware to maximize profits as much as possible. End result: Nintendo sold less than Sega, but may have been earning more overall.
    Not capitalizing on software production and sales on the MD (and even CD) is on eof the major areas the 32x hurt Sega, and would have hurt them even at best.

    The revenue and market share figures from the 32x sales may have looked good from a corporate investor's point of vide compared to lower key revenue from pure software sales, but the profits would be much better and they company would be more stable. It would be up to the PR guys to explain that to investors, and how volumes going DOWN could very well be a good thing! (Nintendo was criticized by market analysts and investors in 1990 for the same thing, but that was one of their highest profit years in the NES's life)

    Sega had already taken a big risk with the Sega CD (which is a point to argue that it shouldn't have been released either), but the 32x was just ridiculous to add on top of that. You could even argue against SVP for that reason, though it's a more modest compromise that might have lent itself as a powerful marketing tool and useful for expanding the MD's late gen software potential. (and was much cheaper than the Sega CD while offering some similar potential hardware enhancement and beyond in some respects -namely better for flat shaded polygonal 3D)

    So why, in light of that, would anyone think Sega overextending their cash to sell more hardware at razor/blade principle a "good idea"? It wasn't. It simply wasn't. That's from a marketing perspective, from a business perspective. Nakayama thought Sega was in the hardware business, but he was half right, which made him half wrong. Sega was in the hardware and software business, specifically the video game business. K.I.S.S is the best practice in any business, and doubly so in the games business.
    Yep, and Sheath's main argument towards this points out how market investors fail to understand the razor and blade marketing strategy.

    Hardware sales of consoles FASLELY inflate volumes and revenue for corporate worth/success . . . which is why healthy console makers delay the release of new hardware as long as possible. (ie until it's neceassary for means of competition -which could include pre-empting a known competitor too)
    The stagnation of the 7th gen consoles is a prime example of this . . . none of them wanted to sacrifice their existing (successful) consoles and had little incentive to replace them . . . they were doing well enough on there own, and there was no new compelling competition to theathen their market positions.

    In that respect, Sega's hardware push for the home market up through the Sega CD did make a lot of sense, they were at a major disadvantage in all markets but Europe up to late 1991, and responding with new, more powerful hardware was their main response. (granted, with SG-1000 vs Famicom, that was an obvious necessity, and with SMS still failing against FC/NES -outside Europe- and the PCE arriving, the MD made a fair amount of sense too . . . albeit without the PCE there's a stronger argument that marketing was their main need given the SMS is very competitive with the NES hardware -and better in a good number of areas- . . . had there been no PCE, a renewed marketing effort mid-gen with the SMS would have made sense, and the MD's actual release being a year later would have made sense too -more time for refinement and added features in that added year too)

    Hell, Sega got pretty lucky with NEC in general, their total incompetence in pushing outside Japan was a big stroke of luck considering the massive resources NEC had at their disposal (and vertical integration advantages too). Hell, their management in Japan was only so-so as well, successful to a point, but not really outstanding. (nowhere near what Sony managed to leverage by comparison, including attracting key Famicom publishers to jump ship -and even in the late 80s many of them were already fed up with Nintendo, Namco in particular)
    6 days older than SEGA Genesis
    -------------
    Quote Originally Posted by evilevoix View Post
    Dude it’s the bios that marries the 16 bit and the 8 bit that makes it 24 bit. If SNK released their double speed bios revision SNK would have had the world’s first 48 bit machine, IDK how you keep ignoring this.
    Quote Originally Posted by evilevoix View Post
    the PCE, that system has no extra silicone for music, how many resources are used to make music and it has less sprites than the MD on screen at once but a larger sprite area?

  6. #66
    Road Rasher
    Join Date
    Feb 2013
    Posts
    423
    Rep Power
    16

    Default

    I dunno if having a PSU, even a small one, with SVP would've been a good idea. I mean, hell, as is there were already two possible PSUs plugged into outlets as is (Genesis and Sega CD). Those Sega branded power strips were in small supply lol. Jeez...I recall a day in my youth when I spent hours worrying that my TV didn't work anymore after I tried to power it up to no avail, only to realize it wasn't plugged in because all of Sega's stuff took all of the power outlets! First time I mumbled "f'n Sega...." under my breath, but wouldn't be the last (more recently, their apparent decision to localize that Project Diva game, while they overlooked so many other games they could've localized this gen).

    As for Sega CD, yeah, Nintendo was part of the thinking (particulary in light of the better sprite scaling on SCD), but it's pretty clear that NEC's PCE CD add-on spurred Sega of Japan in that direction first and foremost. NEC may have had a chance, but when SoJ tossed SCD onto SoA's lap, it was '92. By that point it was clear that NEC was a non-factor in the West, and I think SoA (and SoE) had reservations about even selling SCD in their regions. SoJ should've listened to that. Kinda speaks to the issue with the company as a whole, quite frankly: there wasn't enough shared input from each region as to the direction of the overall global entity. I think had SoJ kept the Western offices in the loop during the entirety of the Saturn research and design phase, the Western offices would've at least made it known that "hey...our internal teams and 3rd party partners over here in the US and Europe kinda think having a simple set up, and good dev kits using this thing they call 'C language' might be a good idea to keep in mind".

    On a more tech related note (and forgive my admittedly lamens understanding of these things), but isn't part of the reason Saturn's CD-ROM subsystem so complex (in comparison to other game console hardware using CD-ROM) due to both potential piracy prevention/protection as well as shorter load times?

    Oh, and on a different topic: I think you asked about 3DO's VRAM in previous topics. Something about it being able being able to be read/write destination? Looked at the spec docs, says the VRAM can be "also capable of holding/executing code and data". Don't know if that specifically sheds light on what you've asked, though. Also oddly enough the sound processor/DSP has an on-chip cache.

  7. #67
    I remain nonsequitur Shining Hero sheath's Avatar
    Join Date
    Jul 2010
    Location
    Texas
    Age
    42
    Posts
    13,313
    Rep Power
    130

    Default

    I do not recall citing myself about why Sega decided to go with the 32X. If I recall I cited Stephen Kent's book, Kathleen Morris "Nightwmare in the Funhouse" and a couple of other sources that showed Sega was in the worst place with the market slump in 1994. The 32X solved that temporarily and then for reasons unknown, still, was discontinued in favor of the Saturn.
    "... If Sony reduced the price of the Playstation, Sega would have to follow suit in order to stay competitive, but Saturn's high manufacturing cost would then translate into huge losses for the company." p170 Revolutionaries at Sony.

    "We ... put Sega out of the hardware business ..." Peter Dille senior vice president of marketing at Sony Computer Entertainment

  8. #68
    Hero of Algol kool kitty89's Avatar
    Join Date
    Mar 2009
    Location
    San Jose, CA
    Age
    30
    Posts
    9,725
    Rep Power
    64

    Default

    Quote Originally Posted by sheath View Post
    I do not recall citing myself about why Sega decided to go with the 32X. If I recall I cited Stephen Kent's book, Kathleen Morris "Nightwmare in the Funhouse" and a couple of other sources that showed Sega was in the worst place with the market slump in 1994. The 32X solved that temporarily and then for reasons unknown, still, was discontinued in favor of the Saturn.
    In my comment above, I meant more that you're the one who pointed out the coprorate/market reasoning/logic in those areas that could (not necessarily WAS) applied to the 32x at the time . . . and the same things that Nintendo accomplished by (mostly) focusing on the SNES, and to lesser extent GB. (and the Virtual Boy . . . though how much a wasted investment that was compared to the 32x I'm not sure -Nintendo was in much better shape to AFFORD such screw-ups though, same with their mistakes in the N64 )



    Quote Originally Posted by spiffyone View Post
    I dunno if having a PSU, even a small one, with SVP would've been a good idea. I mean, hell, as is there were already two possible PSUs plugged into outlets as is (Genesis and Sega CD). Those Sega branded power strips were in small supply lol. Jeez...I recall a day in my youth when I spent hours worrying that my TV didn't work anymore after I tried to power it up to no avail, only to realize it wasn't plugged in because all of Sega's stuff took all of the power outlets! First time I mumbled "f'n Sega...." under my breath, but wouldn't be the last (more recently, their apparent decision to localize that Project Diva game, while they overlooked so many other games they could've localized this gen).
    Yeah, but there's the difference of 2x the DSP clock speed for no added manufacturing cost beyond the PSU. OTOH, it may have been more the MD's internal voltage regulator that was the limit and not the 9/10V 1.2/1.0/.85A (depending on model) AC adapter itself. And, if that's the case, they might have implemented a piggyback/splitter type arrangement for using the same PSU . . . except the model 2 MD uses a different connector, so you'd need 2 cables there . . .

    Actual power consumption of the SSP-1601 is only .04A at 5V, the 128 kB DRAM should have been ~.12A at 5V, plus probably a bit more due to the added gate-array logic in the SVP ASIC. (DRAM controller and MD interface . . . plus added 1kB of scratchpad over the default SSP-1601's 512+512B arrangement)
    http://notaz.gp2x.de/docs/SSP1601.pdf
    http://pdf1.alldatasheet.com/datashe...1664BJ-80.html


    Hmm, interesting:

    I was going to say switching to 512kB DRAM would add a bit more still, but actually looking at the datasheet for the 256kx16-bit DRAM used for the Sega CD's program RAM (or at least one of a few chips used in different models), it actually uses a good bit LESS power than that used for word RAM (and BTW it seems Sega only ever used that one model Toshiba chip for word RAM, SVP, and SVP -unlike CD program RAM and the 512kB DRAMs in Saturn, which had a few alternates from NEC and Fujitsu).

    See: http://www.datasheetarchive.com/dl/S...IH00209246.pdf

    So switching to the modestly more expensive, higher density DRAM might have actually solved this problem entirely . . . since the combined power usage of the full (40 mA) 25 MHz SSP-1601 (or 23 MHz rather -so slightly <40 mA) and the 256kx16-bit DRAM would be LESS than VR's 128kB+SVP ASIC. And BTW, that DSP at full speed still consumes vastly less power than an SH2 or SH1 (which is actually slightly MORE power hungry than an SH2), a fraction of the power in-fact. (at 23 MHz and 5V it's around 1/2 the wattage of an SH2)
    Hell, the cost difference in that DRAM chip would likely be far less than the total cost of an added PSU, plus power jack, plusvoltage regulator, plus shipping (and marketing/instructions for consumers).

    Granted, that problem assumes the limit has anything to do with power consumption and the tiny (few dozen mA at most) power savings for underclocking the DSP, and not heat concerns. (but given the fact that the JP release has no heat sink -or RF shield- and the chip barely gets warm -much less so than the MD VDP or an SH2, I rather doubt that heat/temperature was the problem here)

    That, and we're also assuming that the findings of emulator programmers (of VR maxing out in framerate/speed at ~11 MHz DSP clock rate) is entirely accruate too.


    Edit:
    Hmm, combine that with the S&K integration idea and you've got something really interesting.

    Say something like: with SoA management and engineers confronted with the product requirement which historiclaly resulted in the Mars project, they instead chose the lower cost route of exploiting a direct derivative of the SVP technology (which they were fully aware of and had been demoing at the Winter CES where that teleconfrence had taken place ). They take the SSP-1601, modify the gate array specifications slightly (namely added audio DACs -or do that externally initially, and later redesign the ASIC block of the SVP), switch to the 512kB DRAM and 23 MHz DSP clock, and there you go.

    They decide to cancel the release of the standalone VR cart outside Japan, and add that game to an array of SVP titles to be ready for the introduction along with the (now SVP enhanced) Sonic & Knuckles that October, with a nice marketing campaign (and a few later releases) leading into the holiday season.

    Pretty much all the existing 32x games should have been very possible on a 512kB 23 MHz SVP cart at similar polygon count, perhaps at somewhat lower resolution or framerate (probably more up to CPU logic type tasks bottlnecked by the 68k along with VDP DMA limits), and maybe some loss in BG detail for games using tile backgrounds and a framebuffer layer (VRAM space taken by render buffer), and of course the color limitations. But overall that could be pretty reasonbly worked around for the most part. (dithering in mode H40 blends/blurs very well to fake 64-136 12-bit RGB colors . . . at least in NTSC though there's some choma/luma moire artifacts to consider -varying by video encoder)
    It would go way beyond what the Sega CD or Super FX could practically do too, and the added RAM would allow some flexibility beyond what even the 32x allowed. (including making some PC ports easier . . . though programming the SVP+68k would be tougher than porting games over to run mostly on the SH2s . . . far better than the Super FX+65C02 though, and probably less painful than Jaguar programming in some respects )


    As for Sega CD, yeah, Nintendo was part of the thinking (particulary in light of the better sprite scaling on SCD), but it's pretty clear that NEC's PCE CD add-on spurred Sega of Japan in that direction first and foremost. NEC may have had a chance, but when SoJ tossed SCD onto SoA's lap, it was '92. By that point it was clear that NEC was a non-factor in the West, and I think SoA (and SoE) had reservations about even selling SCD in their regions. SoJ should've listened to that. Kinda speaks to the issue with the company as a whole, quite frankly: there wasn't enough shared input from each region as to the direction of the overall global entity. I think had SoJ kept the Western offices in the loop during the entirety of the Saturn research and design phase, the Western offices would've at least made it known that "hey...our internal teams and 3rd party partners over here in the US and Europe kinda think having a simple set up, and good dev kits using this thing they call 'C language' might be a good idea to keep in mind".
    Yes, NEC was certainly major inspiration there . . . probably spurred the FM Towns design too to some extent. (NEC announced the add-on prior to the 1987 launch of the PCE iirc, or at least at the time of launch -Tomaitheoush has mentioned it in detail before) Heh, in that respect, that's another reason Sega could/should have had a simpler/cheaper CD add-on out earlier. They knew about NEC's end of it since before the MD had even entered mass production . . . so unless Sega had really serious problems finding a reliable vendor for CD drive components and the CD-ROM data decoder (ended up using a Sany part), it must have been down to Sega dragging their feet on the issue and/or trying to add more and more to the design, extending development time. (bare minimum for "better than PCE CD" should have been basic 128-256kB DRAM, DMA sound, and a cheap MCU to handle CD-ROM control/data transfer . . . 128k DRAM would put it slightly weaker than the Super CD RAM-wise, but better than the original CD, though 256kB probably would have been a better trade-off for the long run, especially if we're talking a 1990 JP release -1989 might have merited the single 128k chip more . . . either example could mean flat addressing of that DRAM block within the limited 17 address lines on the expansion port)

    On a more tech related note (and forgive my admittedly lamens understanding of these things), but isn't part of the reason Saturn's CD-ROM subsystem so complex (in comparison to other game console hardware using CD-ROM) due to both potential piracy prevention/protection as well as shorter load times?

    Oh, and on a different topic: I think you asked about 3DO's VRAM in previous topics. Something about it being able being able to be read/write destination? Looked at the spec docs, says the VRAM can be "also capable of holding/executing code and data". Don't know if that specifically sheds light on what you've asked, though. Also oddly enough the sound processor/DSP has an on-chip cache.
    Not sure about CPU access to VRAM (aside from DMA paging between the 2), or using VRAM for textures for that matter (much more significant as it would avoid CPU contention). I've asked and looked into it before, but couldn't make out anything specific in the documts I've seen (which, granted, isn't much), or any of the tech guys I've asked on the issue.

    If it's anything like the Lynx's SUZY chip, it SHOULD be reasonably efficient at working on VRAM alone, presumably with some added features specific to VRAM's capabilities. But it was my impression that the primary way rendering was accomplished is rather like the Saturn's VDP1 using separate source and destination bus (albeit using 32-bit buses rather than 16-bit, and slightly slower than 1/2 the bus speed/timing, VRAM instead of 2 dedicated framebuffer banks -also done on the 32x, albeit with 64kx16-bit DRAMs rather than 128kx16-bit SDRAMs)

    Crazyace did mention that the full 1MB of VRAM might also allow for more VRAM-specific speed optimization (mask load operations among other things) where having 512kB of VRAM wouldn't allow (or 512kB of FPM DRAM for that matter), but it seems to me that at least part of the reasoning was to allow high res 24-bit color stuff to be displayed for high quality still images among other things. (maybe FMV, but probably not . . . given we're talking 2x bitrate cinepak style stuff -and even the VCD decoder used 352x240 or 352x288)
    Or maybe it had something to do with the interpolated/upscaled rendering. (interpolating to 640x480i . . . which a few N64 games did too, actually -particularly later Factor 5 games at low detail, like Episode 1 Racer)
    Last edited by kool kitty89; 09-23-2013 at 01:50 AM.
    6 days older than SEGA Genesis
    -------------
    Quote Originally Posted by evilevoix View Post
    Dude it’s the bios that marries the 16 bit and the 8 bit that makes it 24 bit. If SNK released their double speed bios revision SNK would have had the world’s first 48 bit machine, IDK how you keep ignoring this.
    Quote Originally Posted by evilevoix View Post
    the PCE, that system has no extra silicone for music, how many resources are used to make music and it has less sprites than the MD on screen at once but a larger sprite area?

  9. #69
    ESWAT Veteran Da_Shocker's Avatar
    Join Date
    Apr 2009
    Location
    Cashville,TN
    Age
    38
    Posts
    5,250
    Rep Power
    69

    Default

    @kool kitty89 even with all of the massive fuck ups with the N64 they still sold 20 mlilion units in the states. Hell had Sega sold 20 million Saturn's in the states, they'd probably still be in the hardware bizz.
    Quote Originally Posted by Zoltor View Post
    Japan on the other hand is in real danger, if Japanese men don't start liking to play with their woman, more then them selves, experts calculated the Japanese will be extinct within 300 years.

  10. #70
    I remain nonsequitur Shining Hero sheath's Avatar
    Join Date
    Jul 2010
    Location
    Texas
    Age
    42
    Posts
    13,313
    Rep Power
    130

    Default

    Speaking of the N64's success in the State, I wonder about an alternate history where Sega just kept on pumping out games for the Saturn and kept it on shelves, while maybe repairing some retailer relationships, and just didn't release the Dreamcast until 2001 or something. I bet in this scenario Saturn's overall sales in the US would have been better than the Dreamcast's was and it would have been more profitable. Hell everything that worked for Nintendo would have worked better for Sega right?
    "... If Sony reduced the price of the Playstation, Sega would have to follow suit in order to stay competitive, but Saturn's high manufacturing cost would then translate into huge losses for the company." p170 Revolutionaries at Sony.

    "We ... put Sega out of the hardware business ..." Peter Dille senior vice president of marketing at Sony Computer Entertainment

  11. #71
    Hero of Algol kool kitty89's Avatar
    Join Date
    Mar 2009
    Location
    San Jose, CA
    Age
    30
    Posts
    9,725
    Rep Power
    64

    Default

    Quote Originally Posted by Da_Shocker View Post
    @kool kitty89 even with all of the massive fuck ups with the N64 they still sold 20 mlilion units in the states. Hell had Sega sold 20 million Saturn's in the states, they'd probably still be in the hardware bizz.
    Quote Originally Posted by sheath View Post
    Speaking of the N64's success in the State, I wonder about an alternate history where Sega just kept on pumping out games for the Saturn and kept it on shelves, while maybe repairing some retailer relationships, and just didn't release the Dreamcast until 2001 or something. I bet in this scenario Saturn's overall sales in the US would have been better than the Dreamcast's was and it would have been more profitable. Hell everything that worked for Nintendo would have worked better for Sega right?
    Yeah. Well . . . in terms of "repairing relationships" they really should have been doing carefully managed damage control as soon as they realized the 32x wasn't going to pan out as they hoped, and carefully managed/marketed/positioned the Saturn, etc, etc all in 1995.
    Plus the 32x was a non-starter outside North America, so only one region to have damage control for . . . though the Saturn launch would be just as important in Europe. (and the rushed NA launch impacted Europe at least to some extent, including compromised software quility due to rushed/accelerated development for games originally planned to release that fall)

    Then they might have had a decent chance of matching Nintendo's position in the US . . . sure they had more problematic funding and infrastructure issues than Nintendo, and one generation less of market saturation, but a decent position and other advantages. (CD-ROM advantages of multimedia and manufacturing cost and less painful licensing policies for 3rd parties)

    In Europe they were the oldest mainstream console maker in existence and still had good PR up to the early Saturn years, and the Saturn had made a big splash in Japan.

    Even with the 32x being released, and some of the bloated corporate structure Sega was dealing with in 1995, if they'd really refocused management efforts in a comprehensive, decisive, and tactful manner, they had a good chance of pulling ahead of Nintendo on the mass market and managing a decent (if distant) Second place to Sony in the worldwide market. (not to mention the potential the Game Gear might have had if pushed in the mainstream through the late 90s)




    As for DC . . . a later release might have made sense, and/or a more tentative development scendule including reservations for cancellation if the Saturn didn't hold up on the market. (if they didn't have the position or resources to sustain the Saturn, then it wasn't realistic to invest in launching a new system . . . new hardware is nice, but solving a market situation like Sega had is not something you do with new hardware alone, and especially not the way they tried to do it)
    As far as "tentative" in design terms, it could have meant having the general system and hardware fonfiguration solidified, but leave the final mass production units somewhat in flux due to a variable launch target (depending on specific levels of Saturn success) and keep things like main RAM capacity, RAM clock speeds, and CPU/GPU clock speeds somewhat variable depending on improved yields and reduced costs for a delayed release. (if the DC got pushed back about a year, there should have been decent potential for 8 or 7.5 ns SDRAM -125 or 133 MHz- and a 250-266 MHz SH4 along with 125-133 MHz GPU, and maybe more RAM)

    While a definitively later target (non tentative) could have meant for more solid planning around specific later, more advanced hardware. (like a series 3 PowerVR GPU and maybe an SH5) I'm ot sure when the SH5 endered mass production though, but the original roadmap was for samples to be avilable in the 2nd half of 2000 with full scale prodction in 2001, so probably more in line with the Xbox and GC release schedule.
    http://www.telecompaper.com/news/hit...c-chip--187697
    http://www.iuma.ulpgc.es/~nunez/proc...whitepaper.pdf

    But holy crap does the SH5 have a small die area . . . the fully featured version with FPU is only 14 mm^2 . . . that's like 1/3 the size of the GC's CPU or 1/8 the size of a 180 nm Celeron. Even given it's 150 nm and has no L2 cache (64kB L1) that's pretty small. The DC's SH4 is close to the size of the GC's CPU too (both 42~43 mm^2 range), though that's on a much larger process. (SH4 launched on 350 nm around 1996, but the DC used a 250 nm version)

    Given the 2001 date for the SH5, an up-clocked SH4 probably would have been a better idea. I'm not positive what speed grades the SH4 was avilable in by 2000, but I'd assume it was well beyond the 166/200 MHz grades of the original 350 nm part. (more so if a die shrink beyond 250 nm became available between 1998 and 2000 . . . Hitachi's 150 nm process wasn't ready yet, but they probably had a 180~200 nm process -going straight from 250 nm to 150 nm seems a little odd, and most major chip manufacturers had 180~200 nm fabs going by 1999/2000 -several were using copper interconnect too)
    Or, if they really wanted to but up against Nintendo and MS for launch, and lose a year to Sony, the SH5 would have made some sense. (except for a 2001 console, (and I guess an SH5 and PowerVR series 3 GPU with 32-48 MB total SDRAM would have been OK, though a little weak in some areas perhaps . . . then again, lower cost combined with the timing should have made DVD realistic, maybe with optional DVD-video licensing like MS did)
    6 days older than SEGA Genesis
    -------------
    Quote Originally Posted by evilevoix View Post
    Dude it’s the bios that marries the 16 bit and the 8 bit that makes it 24 bit. If SNK released their double speed bios revision SNK would have had the world’s first 48 bit machine, IDK how you keep ignoring this.
    Quote Originally Posted by evilevoix View Post
    the PCE, that system has no extra silicone for music, how many resources are used to make music and it has less sprites than the MD on screen at once but a larger sprite area?

  12. #72
    I remain nonsequitur Shining Hero sheath's Avatar
    Join Date
    Jul 2010
    Location
    Texas
    Age
    42
    Posts
    13,313
    Rep Power
    130

    Default

    Quote Originally Posted by kool kitty89 View Post
    The 32x, at least on the SoA's end, is one more example of this, both in terms of the actual market goals they applied, but more so in the amount they invested into it. Beyond that, the way they handled it (and the Saturn) in '95 and '96 was a total mess that ended up wasting more money and screwing up PR as well. (they did a terrible job of damage control with the 32x . . . and good damage control was needed at the very least, on top of tactful transition to Saturn as their "go to" next-gen marketing platform . . . though moderating 32x support BEYOND damage control could have made sense too -ie actually exploiting it for some of the intended merits, at least in the NA marketplace where it had more potential, and actually recoup some of the investments on top of maintaining PR and market share)
    I have said this before and I still don't think it is hyperbolic. It is only "bloat" if it doesn't succeed at growing the company. It is only "mismanaged" if sales don't take off. That is the way people think, that is the way the masses excuse bungled products. That is why people just love to point to Nintendo as the shiny example of what Sega should have done. Same goes with Sony, they can screw up all of their products simultaneously and sell a manufacturing plant and headquarters to cover it all up and their fans/investors are none the wiser.

    The fact of the matter is that growth is always extremely risky, staying the same is safe. This is why megapublishers today simply research which genres sell the most and stick to popular franchises within those genres.

    Quote Originally Posted by kool kitty89 View Post
    Indeed, this makes the most sense, from SoA's perspective. That doesn't change that the original product specification that led to the Mars project was a direct result of Nakayma's concerns over the Jaguar (Jaguar specifically too, I haven't seen 3DO mentioned in that context). Naturally that wouldn't have made a whole lot of sense to SoA, who had a more realistic understanding of Atari's market position at the time (even with Atari's despirate PR stunts in late 1993 confusing matters), and it thus made a lot of sense to run with that basic idea in a direction that made more realistic sense for a target product.
    I am fairly certain that the Saturn was SoJ's answer to 3DO and PS1.

    Quote Originally Posted by kool kitty89 View Post
    Now, even so the 32x, both conceptually and in real-world execution, still has major flaws (especially given Sega's position, resources, and alternatives at the time), but there's at least SOME logical explanations for why it existed beyond "countering the treat from Atari." (I maintain there were much better ways to do it, given Sega's position at the time . . . be it doing "nothing" on hardware terms at the time to things related to SVP -and/or add-ons much cheaper than 32x, all the way up to a Jupiter-like system -if they really HAD to have a lower-cost "early next gen" platform to "complement" the Saturn -which the Jupiter should have done much better than the 32x at least, as far as being genuinely next-gen mainstream level hardware and being directly forwards compatible with the Saturn, etc, etc)
    And I maintain that a 1995 Neptune and even a trickle of software support through 1996 would have solved all pressing problems that the 32X created while keeping the Genesis and Sega CD relevant as budget options. That is as opposed to launching two new systems and abandoning all former products even while they were still selling (Sega CD, Pico and Game Gear excepted).

    Quote Originally Posted by kool kitty89 View Post
    See above . . . or in short, both of you are right. (including SoJ . . . or Nakayma being dumb with the whole Jaguar thing, or at least out of touch with the western market -which isn't THAT surprising )
    That is entirely possible, it is also likely that the head of a multinational corporation had a bigger picture than we have speculated in mind. We can only intimate what that big picture might have been, but it looks like keeping Sega as a premiere developer of new games and having a presence in every consumer electronics market was one phase in addition to creating Amusement centers to boost the flagging Arcade industry.

    Quote Originally Posted by kool kitty89 View Post
    Yep. Hardware wise it could have been a real threat, but Atari was in no position to support it (in terms of funding, management, retail channels, consumer PR, 3rd party developer PR, internal software development resources, etc). And as much as their 1993 test market stunt may have shocked people (at least by Atari standards of the time ), it's really hard to argue that they should have been perceived as a major threat . . . maybe a modest concern at best. (the fact that they cancelled the European test market combined with the pitiful test market sales of 1993 should have made competitors dubious . . . not to mention Sega had 1/2 of 1994 to reconsider any 1993 perception, same for 3DO too)
    I haven't found one financial magazine from the time, much less game magazine, that reported Atari's Jaguar was dead on arrival due to lack of capital. We can arm chair CEO with limited hindsight today, but at the time neither we nor the management side of any of these companies had that benefit.

    Quote Originally Posted by kool kitty89 View Post
    Though there's little evidence that SoA actually persued the 32x because they were "scared" of the Jaguar. They had many other incentives that made more sense (if still questionable overall). OTOH, had they really not liked the idea, I wonder how/if they could have rejected the idea outright. (not that they really needed to, since there's some much less risky -and in many cases potentially much more successful- alternatives for a 3D-ish add-on for the MD at the time )
    Why did they have to be scared or threatened by the Jaguar to see it as something that might erode their revenue? The Jaguar very well might have eroded Sega's revenue further in 1994 and for all we know the 32X's successful launch and revenue boost helped kill the same.

    Quote Originally Posted by kool kitty89 View Post
    Such as I've addressed here:
    http://www.sega-16.com/forum/showthr...ke-Sense/page5
    (I'm particularly fond of that realization on the DRAM power consumption thing . . . 512kB DRAM and a double speed SVP -ie 23 MHz rather than 11.5- seems pretty nice, that and the S&K+SVP idea is interesting on a number of levels)
    I still say 16-color games weren't going to cut it in 1994 onward and SVP wasn't going to improve basic audio flexibility for lazy cross platform developers like the 32X did.

    Quote Originally Posted by kool kitty89 View Post
    I agree to a point, but see above. (I really doubt SoA seriously bought into the Jaguar thing -but they had plenty of other reasons to do 32x, not all GOOD reasons IMO, but much better at least )

    Also, Atari was not "legendary" as such by 1993, they had some brand-recognition in Europe thanks to the ST (and lesser extent Lynx), but they were nobodies in the mainstream in North America at the time. (I mean people actually cared that Atari was abandoning the ST -and possibly Lynx- in 1993, incluidng the Falcon designs in the pipeline -like the '040 microbox- while in the US you didn't see that sort of sentiment remotely near the mainstream)
    I think we must be using two different dictionaries. Legendary doesn't mean they had all the power, money and management to make an instant hit mass market product. Legendary means they had the prestige to launch a product and stir up all of the major players in the process with or without all other ideal resources.

    Quote Originally Posted by kool kitty89 View Post
    Yeah, in 1988 the Atari name still held some water in the US, and more so Atari Corp themselves actually managed #2 in market share with consoles (a distant 2nd, but a 2nd nonetheless, and ), so Sega's very real attempts to partner with them for distribution of the MD in late '88. Also note that 1988 was their peak hardware sales year for the 7800 at just over 1.42 million consoles, and following the 2nd best year of '87 at 1.31M, not to shabby, especially given their limited funds for marketing and software.
    Things deteriated rapidly from 1989 onward though, between the slipping sales and slipping management (and stagnating technology with questionable or no replacements -and the problems with marketing the Lynx) they entered a downward spiral that continued until the liquidation in 1996. (1991 was probably the point of no return for both the console and computer ends of things, beyond that they lacked the PR, market share, and funds to really push anything new . . . at least without some miraculous shift in management combined with good -or exceptional- products arriving on the level of what drove Atari Corp out of debt and to mass market success from their meager beginnings of 1984 -except the mid 90s was a MUCH more hostile and competitive time to do that sort of thing than the mid 80s, in both consoles and computers, NA and Europe alike)
    Really though, 1989 is when Atari really needed something new (and really good), both on the console and computer ends of things. (had the STe been a really powerful, definitive, generational upgrade and one applicable for both low end up through the MEGA models on the level of power and efficiency for 1989 as the ST had been back in '85, that could have been huge, but the STe was nowhere near that, and neither was the TT chipset -let alone the TT itself)
    Yup, and this is all hindsight. My point was that in 1993-4 there was very little indication that the Jaguar would false start in the game industry, including retailers, at least.
    "... If Sony reduced the price of the Playstation, Sega would have to follow suit in order to stay competitive, but Saturn's high manufacturing cost would then translate into huge losses for the company." p170 Revolutionaries at Sony.

    "We ... put Sega out of the hardware business ..." Peter Dille senior vice president of marketing at Sony Computer Entertainment

  13. #73
    Raging in the Streets Yharnamresident's Avatar
    Join Date
    May 2013
    Location
    British Columbia
    Posts
    3,873
    Rep Power
    62

    Default

    I bet if they had a model 2 of the Dreamcast in maybe 2003, they could have the SuperH-4 running at 210 Mhz, like what Sony did with the PS2.
    Certified F-Zero GX fanboy

  14. #74
    I remain nonsequitur Shining Hero sheath's Avatar
    Join Date
    Jul 2010
    Location
    Texas
    Age
    42
    Posts
    13,313
    Rep Power
    130

    Default

    Quote Originally Posted by azonicrider View Post
    I bet if they had a model 2 of the Dreamcast in maybe 2003, they could have the SuperH-4 running at 210 Mhz, like what Sony did with the PS2.
    Do you mean 300Mhz? The Dreamcast's SH4 is already 200Mhz. I do recall SoA spouting nonsense about the Dreamcast being "modular" and designed to be upgraded over the years with little effort. I don't know about the CPU or VPU, but the boards in the system are super easy to unplug and replace. They might have thought they could do a mail in upgrade program for more RAM or whatever down the line. They were also talking about using the Dreamcast as a dumb box for streaming gameplay though, and that was WAY before anybody had fast enough broadband for cloud gaming.
    "... If Sony reduced the price of the Playstation, Sega would have to follow suit in order to stay competitive, but Saturn's high manufacturing cost would then translate into huge losses for the company." p170 Revolutionaries at Sony.

    "We ... put Sega out of the hardware business ..." Peter Dille senior vice president of marketing at Sony Computer Entertainment

  15. #75
    Raging in the Streets Yharnamresident's Avatar
    Join Date
    May 2013
    Location
    British Columbia
    Posts
    3,873
    Rep Power
    62

    Default

    Quote Originally Posted by sheath View Post
    Do you mean 300Mhz? The Dreamcast's SH4 is already 200Mhz. I do recall SoA spouting nonsense about the Dreamcast being "modular" and designed to be upgraded over the years with little effort. I don't know about the CPU or VPU, but the boards in the system are super easy to unplug and replace. They might have thought they could do a mail in upgrade program for more RAM or whatever down the line. They were also talking about using the Dreamcast as a dumb box for streaming gameplay though, and that was WAY before anybody had fast enough broadband for cloud gaming.
    Model 1 PS2 is 295 Mhz, model 2 PS2 is 300 Mhz.

    The model 2 Dreamcast could be running at 210 Mhz, to keep it more relevant with the Gamecube and Xbox.


    I bet for Grand Theft Auto 3 on the Dreamcast, they'd probably need a RAM expansion, and a revised controller with more buttons.
    Certified F-Zero GX fanboy

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •