Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!aplcen!haven!decuac!bacchus.pa.dec.com!granite.pa.dec.com!mwm From: mwm@raven.pa.dec.com (Mike (My Watch Has Windows) Meyer) Newsgroups: comp.sys.amiga Subject: Re: 24 Bit Video .. Message-ID: Date: 24 Sep 90 19:16:33 GMT References: <26f592d2-2c05comp.sys.amiga@tronsbox.xei.com> <406@tlvx.UUCP> Sender: news@wrl.dec.com (News) Organization: Missionaria Phonibalonica Lines: 89 In-Reply-To: sysop@tlvx.UUCP's message of 23 Sep 90 19:31:20 GMT In article <406@tlvx.UUCP> sysop@tlvx.UUCP (SysOp) writes: > Yes, we should push for a better entry model. But at the right time, > which isn't now. What makes NTSC important at the low end is that it When is the right time? I haven't seen a post explaining just what needs to be done to do things's "right." Yes, I want things done well, so we can move forward with our software base, but what is the real problem here? Time and money. I gave (one) answer to "when" further down. The first thing that has to happen is an OS that's been uncoupled from the graphics chips; or at least decouple to the extent that the vendor of the silver bullet hardware can provide an alternaitve library to use that hardware. Printer drivers are a good way to look at this. Take HAM-E for example. What's the worst thing you can say about it? "Oh, for the clones, you must program different code for each supported video mode." HAM-E by itself isn't to bad. Neither is DCTV, or the Firecracker, or so on. But they're each different. That means someone wanting to do software for one of the deep boards must decide which one to support, or spend time supporting other boards instead of adding features. The device independent grahics (DIG) implementation solves these problems. > keeps monitor cost down; possibly to nothing. The right time to > upgrade the low end is when higher-resolution monitors start being > cheap. SuperNTSC (roughly double the resolution of stock NTSC) is > supposed to start showing up in about two years. That will drive the > price of high-resolution monitors down, and would be the right time to > introduce a new low-end Amiga to take advantage of that resolution. Shoot, I got a great multisync for about $500. I'm ready for high-res, now; I don't need to wait for it to get cheaper. You don't need to wait. But how many of the people who paid $500 for their machine would have bought them if they'd had to spend $500 extra for the monitor? And if all those people vanished, how many Amiga software vendors would vanish? Would C/A still be around? It seems to me that as soon as a new standard is known (I haven't heard of the term SuperNTSC -- does this imply that someone is working on a new standard?), then you should go ahead and try to implement it. There are _lots_ of people working on new standards to replace NTSC, usually under the buzz-phrase high definition telivision (HDTV). In the US, it's hung up with the FCC trying to decide what the new format should be. The people doing SNTSC decided they could make it work inside the current standards. To get the full benefit, you'll have to by a monitor with roughly double the NTSC resolution. That's true for most HDTV formats (though some are using wide-screen movie formats as well). After these become popular, expect the price on your $500 monitor to drop to more like $100. Expect most people to already own suitable monitors. You could always sell the A500's without additional modes, if there was such a price difference. ... Yes, but you've either got to have the DIG already mentioned or fix the custom chips and software to support the new mode. For anything serious, the latter is probably as much or more work than the former. Once you've got the DIG, then people can do real magic (I've as yet to see anyone asking for a Z buffer). That's the route that's been discussed here, and that's what seems to be going on inside of CBM. You mean, Lotus to create for the same machine that, above, you said was supposed to be a game machine? ;-) ;-) Yup. If Lotus can produce software that runs on a 512K system, then all those boxes sold as game machines are potential customers, and can be used to convince Lotus to do a port. If they can't produce software that runs on a 512K system, do we want them in the market in the first place? :-) Given the specs of the '040, I really won't be lusting for a RISC chip any time soon. Given that the best that Moto quotes for the '040 is 24 MIPS - someday, when its available - and that you can buy R3000 based systems today that get those same numbers, you may wind up wishing for one. However, the '030 in a 3000 unencumbered by Unix makes me wish for something faster than a 15 MIPS R2000 at work.