I'm getting curious enough to think about actually programming something on the PC-FX to see how bad the limits are.
128 32x32 sprites completely fills the screen ... twice. That seems like a fairly good starting point.
It's not the worst imaginable, but you've also got limits on the number of sprites per scanline, made worse by the fact that it's difficult to optimize two chips drawing 64 sprites each in a way that prevents flickering in all situations. So not only are bullet-heavy shooters out because of the 128 limit, but beat-em-ups and run-n-guns are fairly restricted as well.
No rotation/scaling is definitely a good point ... IMHO, I'd personally prefer to see a hand-drawn frame of a sprite turning than the mess that the 5th-gen hardware makes of the job. But I guess that consumers did reasonably expect that any 5th-gen console should be able to do it.
You do have the rotating/scaling KING background for the obligatory huge boss monster.
Yeah, but that's just one solid layer. Something like this guy would not have been possible because there's more than one object rotating.
I agree, low-res rotation often looks chunky and awkward. However, I do think it can look nice when applied appropriately. Given the amount of RAM the systems have, it's impractical to try for a lot of hand-drawn frames anyway.
I expect that they thought those capabilities would be enough as a starting point until they released the 3D expansion which would have added (according to the Japanese guys that tested the PC-FX GA), up to another 500 sprite capability.
That's really interesting. Would you mind sharing your source for that? I love reading about PC-FX stuff like this.
Yes, it was pretty lame by that time as a music synthesizer ... but from my memory, everyone was expecting CD soundtracks at the time, and it can certainly do that. So the sound processor would be just sitting there doing sound effects. If you limit the CD's ADPCM to mono, you've even got another ADPCM channel for digitized sound effects.
So, definitely not as capable as something like the Saturn's dedicated 68000 sound processor, but you should be able to get some reasonably decent results out of the system.
True, it's not that bad. The thing is, besides the obvious problem of not being able to load levels dynamically, PC-FX games were also very voice-heavy. There are a lot of games that wind up using the PCE sound chip for music because they want to access the CD for voice at every turn.
Software developers at Sega criticized the company for designing the Saturn purely in the hardware department without consulting any software devs. This seems like a similar cause/effect.
NEC just didn't have the video-game-hardware resources to fall back on to make similar last-minute changes ... and weren't willing to take the loss-leader approach of including the 3D-chip as standard.
Well, that's logical, but are you sure that's what happened? NEC is an utterly massive company, and a few million dollars more on R&D is nothing if they have any faith in a product.
That brings me to my own speculation: We all know that the PC-FX was based on the 1992 Ironman design, minus a 3D chip and plus an FMV decoder. I think the reason why they didn't build a new system from the ground up in 1994 isn't so much about resources, but about NEC's lack of passion for consoles period.
This article mentions at one point how NEC started balking at investment in the early 1990s when the Super Famicom came out and PCE sales took a dive. One result was the abandonment of of the US market by NEC and the establishment of TTi. Another result, I suppose, was the PC-FX and the next generation not being taken seriously.
There were probably some people within the department who knew what needed to be done, but they weren't given the resources to do it, so in that sense, what you say is true. But if NEC had wanted to take a risk and release a really powerful console with the loss-lead model, they definitely had the capacity to do so. Unlike Sega, Nintendo and even Sony, they even had the capacity to manufacture the system entirely within the company using entirely their own parts.
The Duo-RX price should definitely have been cheaper by then ... their production costs in 1994 should have been pretty low on that system. I'd bet that they were just being greedy and perhaps attempting to defray the R&D costs of the PC-FX.
While the common wisdom is that console manufacturers lose money on the consoles (particularly in the first few years), and then make it up on licensing fees ... that's not always the case. For instance, I was told by someone at Nintendo that they have never lost money on a console sale.
The same certainly can't be said of Sega when it comes to the first version of the Saturn ... they must have cost a fortune to manufacture!
Yep. There are endless hypothetical possibilities for NEC and Sega in 1994, but I personally think that they best option would have been to drop the cost of the Duo/MegaCD as low as possible, wait through 1995, and release a system with a CD drive and powerful dedicated 3D hardware in 1996. The PS1 may not be perfect in every way, but its price and features were too good to take on so directly IMO.
Alternatively, for NEC, I think they just might have succeeded in making a niche anime/digital comic system in Japan, since those games do in fact generate significant sales. However, they would have had to have gone all out, and done some very unorthodox things.
One thing that I often read from Japanese gamers, which I don't hear often at all in English discussions, is that Windows 95 really killed the PC-FX's chances at being the digital comic system. With PCs in that era, you've got no restrictions on content and price, and no licensing fees. You've also got higher resolution, which is great for digital comics, and you can use the RAM and hard drive to reduce load times.
If NEC had been really forward thinking, and if they really didn't want to take the 3D route, I think they should have been radical. Give the system lots of RAM and strong 480i capabilities, a modest CPU and GPU to keep the costs down, and court digital comic devs as much as possible with cheap, easy dev tools and flexible licensing terms and content restrictions. The 3DO actually had some OK sales in Japan after adult games started coming out on it, and NEC might have cashed in similarly had they quietly allowed a real "adults only" line of PC-FX games.
Also, after seeing Lunar 2 and Urusei Yatsura on the Mega CD, they should have gotten on their knees and BEGGED Game Arts to make software for them.
Nice summary!
Thanks! You seem like a great fellow, and I'm glad to see you around the forums and trade posts.
