The SVP chip can move mountains. It should have been made into a separate cart like S&K and bundled it with V.R.
Printable View
The SVP chip can move mountains. It should have been made into a separate cart like S&K and bundled it with V.R.
That is so true, it's power when combined with a Blast Processor is enough to power all of China for a few years but is highly unstable.Quote:
Originally Posted by Sega Genesis
I remember reading a while back that Virtua Fighter and I believe Daytona USA were planned (or maybe worked on) but canceled for the SVP.
That would have been a GREAT idea. Make the SVP chip a cart like Sonic & Knuckles so that you only have to purchase it once. Then buy your SVP capable game and lock on to the SVP cart to save 50 bucks.Quote:
Originally Posted by Sega Genesis
Of course, I guess that's basically what the 32X is, but much more powerful - nevermind.
An SVP cart would have been a much better idea than Sega throwing all their metaphorical eggs in the Virtua Racing basket. Virtua Racing is tosh!
The 32X costs three times more than this proposed idea, it's ten times bigger, requires power and av cables, fit differently on the model 1 and 2 genesis, has some kind of plate to put under it for interference issues, so yeah, exactly the same.
Quote:
Originally Posted by oldskool
Wow, I thought I was the only one that wasn't blown away by SuperFX games. I mean, they were kind of cool, but even as a kid I saw them as a blurry novelty.Quote:
Originally Posted by Kamahl
Inaccurate emulators ruin staf fox, you need one with cycle accurate timing (or reasonably close) or it will run WAY too fast. (it's supposed to be capped at 15 FPS with occassional slowdown . . . some crappy emulators run it at about 30, and it's nearly unplayable -15 FPS with zero slowdown could be nice, but I never even noticed the slowdown as a kid, and it's still not a game changer today)Quote:
Originally Posted by Kamahl
Most of the Super FX games are of a certain taste, both in style and genre . . . if you like the old shaded polygon style and love 3D genres, you'll probably like those games. (and it's all of the above for me) ;)
As for the SVP being expensive, see my next comment for that:
The SVP probably cost about the same (maybe slightly more) to manufacture as Super FX carts, but remember that Nintendo took a different route developing that (a commissioned design from a small firm/consultant rather than a big brand name off the shelf product -SVP isn't totally off the shelf, but it uses an off the shelf ASIC set-up offered by Samsung -ie a customer-defied gate array with embedded DSP, the so-called ASDSP -aplication specific DSP). Even so, the ASDSP was designed as a low-cost embedded solution that filled the role quite well.Quote:
Originally Posted by Da_Shocker
There's also the issue with using 128k cheap DRAM rather than 32k SRAM (but that should actually be significantly cheaper on Sega's end, especially in the long run). Plus tech keeps getting cheaper to make as time goes on in general. (I think YI and Doom used more than 32k actually, but I'm not sure -that would have made the SVP even more competitive)
And then there's the fact that VR was a full 2MB game vs 1M for Star Fox, Stunt Race, and 512k for Dirt Trax and Vortex. Yoshi's Island and Doom (with 2 MB ROMs) didn't arrive until 1995. (when manufacturing/component costs were obviously significantly lower)
And, more importantly, Nintendo had much higher volumes and longer runs of Super FX games, thus making it more economical . . . though many games have prices pretty close to Virtua Racing at launch. (granted, several normal MD/SNES games reached the $90-100 mark with no add-ons)
And launch prices often drop quickly for the high priced games.
Had Sega pushed more SVP games, they likely would have been significantly cheaper than Virtua Racing's launch price and rather close to Super FX games of similar ROM sizes. (depending on profit margins, volumes, and some other things, of course)
However, they way the SVP was implemented from the start makes more sense to use as an add-on, especially since Sega had no other on-cart enhancements up to that point.
It could have been even more useful with a few tweaks (perhaps on-cart DACs or a 512k DRAM rather than 128k), but as it was, the SVP+RAM set-up would have worked great as a lock-on cart (and most such tweaks would add to cost -going to 512k probably would have added about $8 to manufacturing cost in 1994 and more to final retail price -unless Sega ate that cost with lower margins).
The SVP already gets its code from the game ROM, so it would only need to have that mapped to a lock-on cart slot rather than an onboard ROM. (so an S&K style cart)
The current SVP set-up should be very flexible for allowing compression (cramming more into smaller/cheaper games or even more into larger games), enhanced sound (DSP mixing/decompression then played to the DAC via the Z80), and of course coprocessing for various 3D/scaling/rotation/ray-casting type effects.
That would have been much, much cheaper than the 32x, ready for release in Spring of 1994 (as with VR), and while also much less capable than the 32x, it would be much more marketable and actually less conflicting with the Saturn (it was weak enough to carve a niche in the lower-end market specifically without really bleeding into the 5th gen). Such an add-on also could have been cheap enough to merge into late model MDs (sort of like the original Mars/Super MD concept).
Plus, with such low cost and potentially broad use would complement the MCD better and could mesh with the MCD's enhanced hardware rather well. (including things like using the SVP to handle 3D math with the MCD end handling texture mapped polygon rendering among other things, and conventional scaling/rotation stuff too -the ASIC's biggest advantage for 3D is texture acceleration, granted it would be limited to 16 color rendering, but still pretty useful -you could also potentially do things like drawing polygonal models as animation for sprite tiles that could then be manipulated by the VDP as normal sprites -very useful for rail-shooter type games and some others)
Yes, but if you got a good, even high sample rate (say 22 kHz) Z80 playback system going, you could use that as a pseudo DMA sound buffer that the DSP could send mixed PCM to.Quote:
Originally Posted by Chilly Willy
Depending what else you used the DSP for, it could be pretty useful. (still limited to 8-bit PCM quality, but that can still be pretty decent)
Yep, rather like the shift to DVD games a few years later. No more 3-6 disc CD games. ;)Quote:
Originally Posted by Bastardcat
I'm glad I didn't pay full price when it came out ($100ish?).
Yep, remember that GamePro issue I showed you? The ads for used games towards the back, they have it listed around $90 used IIRC.Quote:
Originally Posted by ThunderForce
@kool kitty89
SVP as an lock-on cart and in the same vein you described could have been a huge boost to the 1994-... Genesis games. Not just as a "3D games provider", but also as a "Game Enhancer". With such "small" but effective improvements it could have been in the market until 1997, easily and make good profit to Sega.
Let's take a look at a list of possible/enhanced titles (considering improved DAC and etc...):
3D games:
-Virtua Racing
-Virtual Fighter
-Star Wars Arcade
-Road Rash 3D (like the 3DO version)
-Doom (probably it would end-up been better than the 32x version considering that the developers would not have to deal with the complicated&undocumented SH2-based hardware)
-Wolfeinstein (Why not?)
2D games:
-Street Fighter Alpha 1&2 (with the extra ram and some "loading" it would be possible, improved DAC would make the voices sound much better too)
-Samurai Shodown, MK3 & UMK3,... (Enhanced DAC)
-EA Sports titles (Enhanced DAC)
-Other sports titles like International SuperStar Soccer would have made great use of the improved voice samples...
So many other possibilities... More important than that, a healthier and longer life to Genesis/MD.
How many titles could have been released until 1997/98 considering such add-on and assuming that SOJ would not have (32X) reasons to drop Genesis support in 1995? Maybe 300 new titles?
I still love me some Super FX games. Star Fox is in fact a really hard game. Haters gonna hate I suppose
The difference would be the SVP cart would be less intimidating to consumers. If Sega marketed it not as a new platform, but just an enhancement for games that supported it (think of it like the Wii Motion Plus) in a convenient adapter cartridge so any new releases would remain at their usual cost. Again, with the right marketing, an SVP Lock-on cart could have been quite successful. Sure there would have been color limitations, but in my opinion, the color deficiency was and still is too much hassle to work around. I don't even think the color issue is that bad. It gave the Genesis a gritty and dark sort of feeling which made it more appealing to the kiddies of the early 90.Quote:
Originally Posted by oldskool
Nintendo tried this with the SNES. Look how it ended up, they were left behind with the Nintendo 64 and they didn't recover until the Wii literally. No matter what you do, the PlayStation pretty much killed the fourth generation. The only thing that would have stopped it is if Nintendo didn't break the contract for making the add-on.Quote:
Originally Posted by agostinhobaroners
I did not say it would compete with Playstation or something like that. It could just give a longer life to Genesis, certainly not on top of game market, but still with good sales numbers...
Sega Saturn's design (in hardware and visual aspects) was a fail on its own concepts. Sega would fail with that anyway IMO.
My point is that they could keep making some good money with Genesis (and not losing all with 32X, ruining SOA) until 1997 with that little add-on.
It was the unexpected early drop of Genesis (and Sega CD) support + epic fail (and MEGA early drop) of 32X that ruined Sega in terms of "costumers WTB more Sega products".
Had they failed in sales numbers with Saturn but with their costumers pleased and had not financially broken (I'm considering that 32X was never released and Genesis/MD kept giving some money til 1997-98); the Dreamcast could have been a great success (big #1 until PS2 sufaced and great #2 since then).
Nintendo failed with N64 and (much more) with Game Cube, but did not completely fucked up their costumers.
N64 is a completely different story. The cartridge format was the WRONG decision that fucked up everything...
You could have an upgraded 1988 designed cartridge-based console in the market grabbing some extra money in 1997, using its big name, amazing library and millions of sold units.
You should NEVER had a 1996 released console based on very expensive cartridges, considering that many CD-based consoles were already there. Also, Nintendo fucked up completely, again, when chose the 3D-only route, letting their long term "partners" with no choices (what should Capcom and Konami do with their BIG 2D franchises???).
And, as I touched on, that "enhancement" could be pretty flexible in general (sound mixing/decompression, graphics compression, scaling/rotation, etc).Quote:
Originally Posted by agostinhobaroners
There would also be potential for games that worked non-enhanced, but had added content when used with SVP. (be it actual added levels, or better scaling, greater detail/animation -thanks to decompression- or other effects -including enhanced sound) Albeit Sega could (and should) have done that for the CD as well. (have cart games with companion enhancement CD)
Of course, the most optimized games would be exclusive to the SVP format and use the hardware more heavily (ie compressing more of the graphics, pushing sound further, pushing use of 3D/pseudo 3D more, etc, etc -of course there's trade-offs for each of those categories as resource isn't limitless -3D heavy games would have less resource to put towards other things, for example).
And there's further potential for CD games using the enhancement. (or potential for cart games with further CD enhancement support as well ;))
Granted, use of the CD was still limited to rather moderate marketshare in the US at the time, but Sega still had a chance to boost CD interest in 1994 and 1995 too. (aside from advertising and certain software development aspects, one major thing would have been getting a really low cost duo console out -something perhaps in the $200 range by fall of 1994- or perhaps roll that together with the SVP for a trio type set-up -probably closer to $250 then though, unless Sega was willing to risk selling at a significant loss)
This is getting off topic, but if they DID manage to get the CD to be really mass market in popularity, that could have made for an excellent persisting platform for the low-end/budget market moving forward (into the late 90s). It certainly would have been more expensive than the SNES, but the software would be cheaper and more flexible (attractive to developers and consumers alike) and also cater very well to compilations and re-releases. (it also shouldn't have conflicted too much . . . especially if Sega had supported tools to facilitate cross-platform development for some types of games -obviously it would be far more realistic to upgrade a MCD/MCDSVP game to Saturn than the other way around, and as the system aged more and more it would have lost most support for new games -probably cutting back by late 1996 and really getting sparse for new releases by 1997, but it could live on with re-releases in the budget market for a couple years longer for sure)
Several of those could have been done well on the Sega CD alone (Wolf 3D, Road Rash -look at Batman Returns, Batman and Robin, or Jaguar XJ220), and some others could probably have been done somewhat decently on CD alone if highly optimized. (Virtua Racing might be OK -but given some find the SVP version iffy, it might not have been worth it- and Doom probably could have been done better than the Super FX version -like the Super FX, it would have been a total custom engine, and the CD is lacking in some areas that the SFX do well but has other advantages -color would definitely be weaker, but it probably still could have looked fairly decent with careful color optimization and use of simple realtime dithering)Quote:
Let's take a look at a list of possible/enhanced titles (considering improved DAC and etc...):
3D games:
-Virtua Racing
-Virtual Fighter
-Star Wars Arcade
-Road Rash 3D (like the 3DO version)
-Doom (probably it would end-up been better than the 32x version considering that the developers would not have to deal with the complicated&undocumented SH2-based hardware)
-Wolfeinstein (Why not?)
Coupling the CD with the SVP definitely had even more potential, and would have been closer to the 32x in some regards (better in some areas even, with trade-offs).
As for the 32x's documentation. From what I understand, it was pretty good for the time (A LOT better than anything the Saturn got until mid 1995 -and even that wasn't really better than what the 32x had in late 1994), with the exception of the DMA sound support being undocumented. (that was due to the early dev units having a bug that made DMA unworkable, that was solved with all production models but never revised in the dev kits it seems -homebrewers got it working by using example code found on a special Sega diagnostic cart that developers, especially 3rd parties, wouldn't have had access to)
In fact, the SVP (being a DSP, and rather limited in flexibility and not being especially friendly to program -as many DSPs are) wouldn't have really made things easier in that respect, but it would have been more attractive due to low cost and potentially higher market share (more so once that marketshare materialized), plus it could have been on the market earlier. (so some games might not have been as rushed)
However, the SVP wouldn't be inteded to be programmed as a general purpose processor, but a specially optimized coprocessor used for certain tasks (with the MD and CD CPUs still handling the general purpose stuff), and it's actually significantly faster as some things than an SH2 (while much cheaper and also much less general purpose -it would make a terrible CPU ;)).
On top of that, you'd probably have had more CD-specific games, and in your Doom example, a CD+SVP optimized Doom could have been significantly better than the current one in certain areas (though it still would have taken a lot of work to work the game into the limited RAM available)
There's actually a lot on the sound end that's possible with the MD alone, it's just that very few developers implemented advanced sample drivers (22 kHz ADPCM playback should be possible, or lower rates for multiple channels -and potential for other compression formats, 1-bit CVSD is rather well suited to speech but less so for sampled instruments, there's also some 2-bit ADPCM formats already in use at that would offer double the compression ratio over more "normal" 4-bit stuff, and custom-tweaked formats of course -chilly Willy implemented a 2-bit CVSD derivative that works rather well in his 22 kHz Z80 demo).Quote:
2D games:
-Street Fighter Alpha 1&2 (with the extra ram and some "loading" it would be possible, improved DAC would make the voices sound much better too)
-Samurai Shodown, MK3 & UMK3,... (Enhanced DAC)
-EA Sports titles (Enhanced DAC)
-Other sports titles like International SuperStar Soccer would have made great use of the improved voice samples...
The smartest thing to do on that end probably would have been for Sega to develop such a driver in-house and offer it for licensing (like they did for GEMS). Investing in making GEMS more flexible in general would have been nice. (allowing a broader range of instruments than the current defaults -very important for composers not experienced in custom FM patches- among other things -adding more advanced sample sound support to GEMS would obviously have been significant, and as it was, GEMS was actually one of the more decent examples of PCM, nothing really special, but at least even single-channel 8-bit PCM playback at up to moderately high sample rates)
As it was, it was uncommon to even see any streaming compressed sample playback, unfortunately, and many developers seemed to have trouble getting even playback on plain PCM (or didn't care enough to invest in a dedicated/skilled Z80 programmer to write decent playback code).
But the SVP would offer far more than that. (and make doing the above much easier) Depending on how much resource you dedicated to audio, you could get a nice multi-channel sample based synthesizer using compressed samples (perhaps with interpolation, maybe software reverb effects too), though without added DACs on-cart (especially with DMA to play PCM), you'd need to rely on the 8-bit DAC port of the YM2612 and Z80 playback, having the Z80 pull the final mixed output from SVP DRAM (probably copying it into Z80 RAM to avoid contention issues) and then playing that through the DAC. (and with the Z80 simply dedicated to that, it really shouldn't have been tough to manage good/even 22 kHz playback -and a decently managed driver optimized for such a set-up would probably be a good substitute for an actual DMA sound buffer on the SVP end)
Now, 8-bit output is still limited (it's Amiga or Soundblaster quality -and mono, so more like SB), but it's also not too bad either. (and using preprocessed samples catering to 8-bit output would help even more -all Sega CD FMV uses only 8-bit PCM, though much is fairly low sample rate Sonic CD uses 32 kHz)
Oh, and there's this rather blatant example of how that 8-bit DAC can sound pumping raw 8-bit PCM (at 26 kHz in this case) :
http://www.youtube.com/watch?v=oKsUAywSyEg
Actually, aside from being mono, that 8-bit DAC actually has some notable advantages over the 32x's PWM, the big issue is that it can do sample rates of under 14 kHz without an aweful squeal (which PWM has a problem with), and it does save a bit of cost on the SVP's end.
They didn't drop support just like that as it was, though SoJ cut back on in-house development very quickly (I think all projects were shut down by the end of 1995), but their actions it certainly hampered the MD's late gen life in general.Quote:
So many other possibilities... More important than that, a healthier and longer life to Genesis/MD.
How many titles could have been released until 1997/98 considering such add-on and assuming that SOJ would not have (32X) reasons to drop Genesis support in 1995? Maybe 300 new titles?
However, that's something that could/should have been managed well regardless of such add-ons. (aside from more games, they could have marketed the MD/Genesis software/hardware much better in 1995-1999 as it transitioned from mainstream to low-end/budget) The way the 32x was handled somewhat compromised the Genesis's market by itself (ideally it could have helped extend the Genesis -but it would have been hard to pull that off on the marketing end), but what happened with the untimely shift for Saturn support on top of the corresponding 32x and PR mess left the Genesis in a far poorer managed/marketed position. (dropping the GG also was a big mistake -though not having a "game gear Jr" of sorts alone by 1995 was a notable slip-up -unlit color LCDs had finally reached decent quality by that time, so potential for a huge shift in size/cost/battery life for "jr" models)
From that respect, even if Sega DID manage to push SVP (add-on) and CD harder (and successfully) in mid/late 1994, having the Saturn launched and supported just as it happened historically probably would have ended up as just as big of a mess. (and in that respect -if Sega wouldn't shift to a more realistic schedule for the Saturn, let alone make some different design/management decisions on the system itself- releasing the SVP at all probably would have just been a waste of money in the end)
In that context (ie the Saturn being pushed just as it was without changes to design, support, regional release date, or management/marketing), Sega would have either been best to release no new hardware other than the Saturn at all (in 1994), or (especially if the Mars concept was still being really pushed -and an interim platform in general) a cut-down Saturn (like the so-called Jupiter concept) may have made the most sense. (basically a cart-only Saturn with some other cuts to RAM and such, but fully expandable to Saturn via an add-on and with games directly compatible with the full Saturn as well -ie a jupiter cart would run on Saturn, and obvious potential for companion enhancement CDs)
Something like that probably could have price-matched the Jaguar in late 1994 ($250) and been powerful enough to be a full 5th gen contender even without the add-on.
And, honestly, regardless of internal conflict or management mis-steps (or arguable design aspects of the Saturn), the 32x was really neither here nor there as a product: too expensive to really make a practical high-adoption rate (especially in the short run, or on top of the CD) add-on and also not powerful (or really efficient) as a full next-gen platform. (it also would cost a lot more for the Neptune than a simple MD with embedded SVP).
If Sega really wanted a direct compatible/evolution of the MD/CD going forward, a full standalone console (ie an alternative design concept to the Saturn with hardware directly based on and compatible with the MD+CD) would have made far more sense due to the inherent efficiencies over add-ons (and if they wanted a low baseline cost, they could have offered a cut-back bottom-end model with simple to install expansion module(s) to upgrade it -especially for RAM)
With the Saturn being all-new as it was, it made more sense to either go with a solely low-end market oriented add-on (if any at all), or an interim system based on Saturn forwards compatibility rather than MD/CD backwards compatibility.
Edit: (to avoid double post ;))
Well, it's not too hard on the easy route, though it seemed pretty hard to me when I was a kid in the late 90s . . . it seemed a LOT harder than 64. (by the time I was 14 I went back to it and played through the easy route in 1 go with no deaths and 94% finish . . . and I've done better than that since, but I still haven't pushed to try and beat the games on the harder modes -I probably could do it, but I haven't taken the time to really push it)Quote:
Originally Posted by Guntz
Oddly, SF64 actually seems a bit harder now. (easier than it was when I was younger, but harder relative to the SNES game . . . at least on easy)
Dirt Trax is rather mediocre though, it's OK, but nothing special. Stunt race is neat but lacking due to the framerate (it seems it may have been intended for SFX2 originally but the new chip wasn't ready in time), Vortex is pretty good but not as easy to pick up and play as SF (and more niche in terms of playability/enjoyment), Yoshi's Island is awesome, and Doom is OK for what it is but probably the worst version out there. (well, playing it on a 16 MHz 386SX PC probably would be worse ;))
Yes, marketing would have been key, and unlike Nintendo, there wasn't the issue of future upgraded versions (Super FX2 was more necessary and non-Super FX add-on chips like SA-1 for other purposes).Quote:
The difference would be the SVP cart would be less intimidating to consumers. If Sega marketed it not as a new platform, but just an enhancement for games that supported it (think of it like the Wii Motion Plus) in a convenient adapter cartridge so any new releases would remain at their usual cost. Again, with the right marketing, an SVP Lock-on cart could have been quite successful. Sure there would have been color limitations, but in my opinion, the color deficiency was and still is too much hassle to work around. I don't even think the color issue is that bad. It gave the Genesis a gritty and dark sort of feeling which made it more appealing to the kiddies of the early 90.
Of course, as I already rambled a bit on above, avoiding the major mis-steps with the Saturn would have been necessary for such a product to really be worth it (especially if it was to be pushed with the CD). Lacking the full expense and features (and new-gen-console marketing) of the 32x would probably help a bit with the Saturn conflicts in general, but some things (like the much too early western release and premature shift in marketing/software funding) still would have compromised the mainstream market if not done differently. (to the extent that it would probably be more wasteful than helpful to ever release the SVP at all)
Actually, it seems comments about crappy graphics in the context of Star Fox back then weren't too uncommon.Quote:
Originally Posted by beigemore
MN12bird actually mentioned that in his review:
http://www.youtube.com/watch?v=fgjGNr3pZ_0
(see 5:59)
The reason I never thought Starfox was very impressive, was because of games like X-wing on the PC.