lphillips@lpami.wimsey.bc.ca (Larry Phillips) (09/22/90)
In <1990Sep22.222816.28754@zip.eecs.umich.edu>, gilgalad@dip.eecs.umich.edu (Ralph Seguin) writes: >I think that a large portion of the problem here is that people are >expecting some sort of magical solution to appear to give us >24 bit graphics. I do not think that HAME or DCTV are the answer >at all, as they are very much a kludge and are CPU intensive. Don't know about DCTV, but HAM-E is most definitely NOT CPU intensive. Nor is it billed as 'a magical solution to 24 bit graphics', ie. it is not a solution to hardware independent, full function grphics. It _IS_ a way for those involved in graphics to get capabilities previously unobtainable or very expensive. -larry -- It is not possible to both understand and appreciate Intel CPUs. -D.Wolfskill +-----------------------------------------------------------------------+ | // Larry Phillips | | \X/ lphillips@lpami.wimsey.bc.ca -or- uunet!van-bc!lpami!lphillips | | COMPUSERVE: 76703,4322 -or- 76703.4322@compuserve.com | +-----------------------------------------------------------------------+
gilgalad@dip.eecs.umich.edu (Ralph Seguin) (09/23/90)
I think that a large portion of the problem here is that people are expecting some sort of magical solution to appear to give us 24 bit graphics. I do not think that HAME or DCTV are the answer at all, as they are very much a kludge and are CPU intensive. Commodore is obviously aware of what is wanted, and I am sure that they are working quite hard at providing a solution. There is also the seeming fact that people want all of this at the same cost as the normal Amiga box. This is nonsense. If you want 24 bit high resolution graphics, YOU MUST PAY! While it would be nice to have a 32 bit chipset, I think that they are quite a ways off. In any case, I am pleased with the dedication that Commodore has. YAHWITTSD (Yet Another Here's What I Think They Should Do 8-) Someone should produce a 24 bit graphics card (there are quite a few companies doing this right now it would seem): -RISC processor, or 34020 to keep the graphics stuff non-cpu-intensive -large dual port memory -support for NTSC resolutions -support for 8 bpp, 16 bpp, 24 bpp -support for user upgradeable memory to say 16 megs of video memory -support for interlaced and non-interlaced displays (24 bit images are quite large after all) -support for high resolutions with additional memory in the board -say up to 1600x1280 (this would certainly appease quite a few people) - BNC connector, RCA connector and RGB connectors for multiple outputs - software-wise there should only be the device driver stuff, since Commodore is going to take care of the niceties of having Retargetable Graphics.
rjc@wookumz.ai.mit.edu (Ray Cromwell) (09/23/90)
In article <1990Sep22.222816.28754@zip.eecs.umich.edu> gilgalad@dip.eecs.umich.edu (Ralph Seguin) writes: >I think that a large portion of the problem here is that people are >expecting some sort of magical solution to appear to give us >24 bit graphics. I do not think that HAME or DCTV are the answer >at all, as they are very much a kludge and are CPU intensive. Huh? How is HAME and DCTV CPU intensive? Its hardware. Maybe you mean that they use up memory bandwidth like a 16color HIRES display does. >Commodore is obviously aware of what is wanted, and I am sure that >they are working quite hard at providing a solution. There is also >the seeming fact that people want all of this at the same cost as >the normal Amiga box. This is nonsense. If you want 24 bit high >resolution graphics, YOU MUST PAY! While it would be nice to have >a 32 bit chipset, I think that they are quite a ways off. >In any case, I am pleased with the dedication that Commodore has. > My question is, can those 24bit high resolution boards animate in real time? (60fps) And do they offer compression? DCTV seems to compress video real well for the amount of colors and can animate in real time 24bit without requiring 7megs of VRAM for animating or holding images. By the way, a kludge is anything that gets the job done, and DCTV and HAME provide short term solutions. Maybe I recommend a IBM+VGA or $10,000 Mac II? > >YAHWITTSD (Yet Another Here's What I Think They Should Do 8-) > >Someone should produce a 24 bit graphics card (there are quite a >few companies doing this right now it would seem): > Yes, but there is no device independant graphics library so this is no different than IBM's CGA/EGA/VGA KLUDGE! (develop code for each board instead of having the OS do it) If the Amiga is to be a multimedia machine, it needs to be fast, cheap, and reliable. How do you do a presentating with a video board than cant do real time animation? (still frames+video recorder)? >-RISC processor, or 34020 to keep the graphics stuff non-cpu-intensive >-large dual port memory >-support for NTSC resolutions >-support for 8 bpp, 16 bpp, 24 bpp >-support for user upgradeable memory to say 16 megs of video memory >-support for interlaced and non-interlaced displays >(24 bit images are quite large after all) >-support for high resolutions with additional memory in the board > -say up to 1600x1280 (this would certainly appease quite a > few people) >- BNC connector, RCA connector and RGB connectors for multiple outputs >- software-wise there should only be the device driver stuff, since >Commodore is going to take care of the niceties of having Retargetable >Graphics. The only thing Commodore should do is develop a seperate retargetable graphics library and define a standard. The rest should be left up to third parties. Why must Commodore always be the one to do everything? 'Commodore should make a DSP' , 'Commodore should make a 24bit board', 'Commodore should upgrade the custom chips to 32bit color and megapixels' I see nothing wrong with things like Ham-E, DCTV, or the Video Toaster. They are not kludges in my mind, merely hardware accesories like a Digitizer. I don't think they promise 'real' 24bit video with 34010's or anything. By the way, what is 'Real 24bit video' anyway? Still pictures like VGA? 16.7 million colors? How many have to be displayable at once to be considered 'Real 24bit'? If DCTV or HAM-E can get just as good results and save memory, at 1/2-1/3 cost then there's nothing wrong with them. If the Amiga gets something like QuickDraw and expensive 24bit boards then the Amiga system+video hardware will be around the price of an equivelent IBM or MAC system, so what would be a consumer's motivation to buy an AMiga then? Not the price, and certainly not the software base.
gilgalad@dip.eecs.umich.edu (Ralph Seguin) (09/23/90)
In article <10918@life.ai.mit.edu> rjc@wookumz.ai.mit.edu (Ray Cromwell) writes: >In article <1990Sep22.222816.28754@zip.eecs.umich.edu> gilgalad@dip.eecs.umich.edu (Ralph Seguin) writes: >>I think that a large portion of the problem here is that people are >>expecting some sort of magical solution to appear to give us >>24 bit graphics. I do not think that HAME or DCTV are the answer >>at all, as they are very much a kludge and are CPU intensive. > > Huh? How is HAME and DCTV CPU intensive? Its hardware. Maybe you mean that >they use up memory bandwidth like a 16color HIRES display does. As I understand DCTV, it uses the parallel port to pump the additional information to the device. The processor has to do all this work. HAME is more or less a kludge, but you are right, it is not really any more CPU intensive. I should not have put it as CPU intensive. > >>Commodore is obviously aware of what is wanted, and I am sure that >>they are working quite hard at providing a solution. There is also >>the seeming fact that people want all of this at the same cost as >>the normal Amiga box. This is nonsense. If you want 24 bit high >>resolution graphics, YOU MUST PAY! While it would be nice to have >>a 32 bit chipset, I think that they are quite a ways off. >>In any case, I am pleased with the dedication that Commodore has. >> > > My question is, can those 24bit high resolution boards animate in real time? >(60fps) And do they offer compression? DCTV seems to compress video real >well for the amount of colors and can animate in real time 24bit without >requiring 7megs of VRAM for animating or holding images. > By the way, a kludge is anything that gets the job done, and DCTV and HAME >provide short term solutions. Maybe I recommend a IBM+VGA or $10,000 Mac II? > The area that high resolution graphics caters to is for things like ray tracing, solid modeling in general, and several other areas where animation is not important. If you want animation, do single frame recordings. Don't get me wrong, I love the Amiga, and I do know that there are 24 bit and 8 bit boards out there. I'm more interested in seeing the Amiga taking the lead again since it has a combination of one of the best OSes around and a very good architecture. It is interesting to note thayt most of the high resolution graphics machines have very good update times. In general,they can produce adequate animation (at lower resolution). >> >>YAHWITTSD (Yet Another Here's What I Think They Should Do 8-) >> >>Someone should produce a 24 bit graphics card (there are quite a >>few companies doing this right now it would seem): >> > > Yes, but there is no device independant graphics library so this is no >different than IBM's CGA/EGA/VGA KLUDGE! (develop code for each board instead >of having the OS do it) If the Amiga is to be a multimedia machine, it needs >to be fast, cheap, and reliable. How do you do a presentating with a video >board than cant do real time animation? (still frames+video recorder)? > Commodore has set guidelines for graphics and device independence. >>-RISC processor, or 34020 to keep the graphics stuff non-cpu-intensive >>-large dual port memory >>-support for NTSC resolutions >>-support for 8 bpp, 16 bpp, 24 bpp >>-support for user upgradeable memory to say 16 megs of video memory >>-support for interlaced and non-interlaced displays >>(24 bit images are quite large after all) >>-support for high resolutions with additional memory in the board >> -say up to 1600x1280 (this would certainly appease quite a >> few people) >>- BNC connector, RCA connector and RGB connectors for multiple outputs >>- software-wise there should only be the device driver stuff, since >>Commodore is going to take care of the niceties of having Retargetable >>Graphics. > > The only thing Commodore should do is develop a seperate retargetable >graphics library and define a standard. The rest should be left up to >third parties. Why must Commodore always be the one to do everything? >'Commodore should make a DSP' , 'Commodore should make a 24bit board', >'Commodore should upgrade the custom chips to 32bit color and megapixels' > I agree completely here. There are several third party companies that are working on it. >I see nothing wrong with things like Ham-E, DCTV, or the Video Toaster. >They are not kludges in my mind, merely hardware accesories like a Digitizer. >I don't think they promise 'real' 24bit video with 34010's or anything. > By the way, what is 'Real 24bit video' anyway? Still pictures like >VGA? 16.7 million colors? How many have to be displayable at once to be >considered 'Real 24bit'? If DCTV or HAM-E can get just as good results >and save memory, at 1/2-1/3 cost then there's nothing wrong with them. >If the Amiga gets something like QuickDraw and expensive 24bit boards >then the Amiga system+video hardware will be around the price of an >equivelent IBM or MAC system, so what would be a consumer's motivation >to buy an AMiga then? Not the price, and certainly not the software base. I think that you may be getting a little too hyped here. I agree that any such system would be considerably more expensive than the current Amiga. But that is just it: If you want high resolution and high colors, it doesn't come inexpensively (just look at monitor prices for megapixel resolutions 8-( I think that there should be the option of 24 bit graphics though. Commodore is currently working on this though, so I'm not really complaining about anything. In the future I should be able to plug in company X's high resolution 24 bit board and run Workbench on it (YEAH 8-) See ya, Ralph
peter@sugar.hackercorp.com (Peter da Silva) (09/23/90)
In article <10918@life.ai.mit.edu> rjc@wookumz.ai.mit.edu (Ray Cromwell) writes: > The only thing Commodore should do is develop a seperate retargetable > graphics library and define a standard. So, when is that going to happen, eh? Oh well, let them get 2.0 up to scratch first... > The rest should be left up to third parties. Right, but Commodore has to provide this basic standard first... -- Peter da Silva. `-_-' <peter@sugar.hackercorp.com>.
david@twg.com (David S. Herron) (09/24/90)
In article <10918@life.ai.mit.edu> rjc@wookumz.ai.mit.edu (Ray Cromwell) writes: >In article <1990Sep22.222816.28754@zip.eecs.umich.edu> gilgalad@dip.eecs.umich.edu (Ralph Seguin) writes: >>.... I do not think that HAME or DCTV are the answer >>at all, as they are very much a kludge and are CPU intensive. > > Huh? How is HAME and DCTV CPU intensive? Its hardware. Maybe you mean that >they use up memory bandwidth like a 16color HIRES display does. ... > My question is, can those 24bit high resolution boards animate in real time? >(60fps) And do they offer compression? DCTV seems to compress video real >well for the amount of colors and can animate in real time 24bit without >requiring 7megs of VRAM for animating or holding images. > By the way, a kludge is anything that gets the job done, and DCTV and HAME >provide short term solutions. Maybe I recommend a IBM+VGA or $10,000 Mac II? Yes, at least DCTV can animate real-time. At the demo they put on for FAUG a couplea months ago they were: hiding their data inside a normal ILBM/IFF file. The data happened to be the same size as a normal hi-res picture file, its just that the encoding was different. using DPaint-3 (the one with animation) to read one of these files and animate it. DPaint-3 simply displayed whatever bits wuz there, magic cookie and all. All the manipulations it did were transparent on through all the Amiga software & hardware and decoded just fine inside the DCTV box. Since the amount of data is the same as normal hi-res pictures any system that can shovel that many bits around to make a hi-res animation will also be able to shovel that same number of bits around to make a DCTV encoded 24-bit animation. They are assumably going to get, or have already gotten, an officially blessed IFF type assigned to them. Since DPaint-3 was able to read their file when it was hiding inside an ILBM file, all that will need be done differently for the DCTV IFF format is to prevent the user from mucking with the pixels that do the magic cookie. >>Someone should produce a 24 bit graphics card (there are quite a >>few companies doing this right now it would seem): ... > Yes, but there is no device independant graphics library so this is no >different than IBM's CGA/EGA/VGA KLUDGE! (develop code for each board instead >of having the OS do it) ... > >>-RISC processor, or 34020 to keep the graphics stuff non-cpu-intensive Graphics is, by its nature, CPU intensive. It just helps a whole lot to have a processor that's suited to the task. An i860 for instance ;-). >>-large dual port memory helpful >>-support for NTSC resolutions >>-support for 8 bpp, 16 bpp, 24 bpp >>-support for user upgradeable memory to say 16 megs of video memory why that small?? :-) >>-support for interlaced and non-interlaced displays >>(24 bit images are quite large after all) >>-support for high resolutions with additional memory in the board >> -say up to 1600x1280 (this would certainly appease quite a >> few people) again, that small? I want a wall-sized monitor! >>- BNC connector, RCA connector and RGB connectors for multiple outputs >>- software-wise there should only be the device driver stuff, since >>Commodore is going to take care of the niceties of having Retargetable >>Graphics. > > The only thing Commodore should do is develop a seperate retargetable >graphics library and define a standard. The rest should be left up to >third parties. Why must Commodore always be the one to do everything? >'Commodore should make a DSP' , 'Commodore should make a 24bit board', >'Commodore should upgrade the custom chips to 32bit color and megapixels' There should definitely be a standard graphics library for many types of output device. That there isn't is a real big problem .. -- <- David Herron, an MMDF & WIN/MHS guy, <david@twg.com> <- Formerly: David Herron -- NonResident E-Mail Hack <david@ms.uky.edu> <- <- Sign me up for one "I survived Jaka's Story" T-shirt!
brianm (Brian Moffet) (09/24/90)
gilgalad@dip.eecs.umich.edu (Ralph Seguin) writes: >I do not think that HAME or DCTV are the answer >at all, as they are very much a kludge and are CPU intensive. I don't think that they are a kludge. They can be CPU intensive, but considering what they have to work with, this is okay. Consider... HAM-E gives a user essentially VGA graphics (not Super-VGA), for any amiga at a good price/performance ratio. The price of a HAM-E box from Black-Belt is 300 dollars list. The price of a VGA and monitor (you have to buy a new one) is about 450 or so. This seems to be fairly good. In addition, and this is where the CPU intensive part is, you get the HAM-E mode for free. This gives some huge number of colors on the screen, with the appropriate fringing effects for fun and profit. Having said that, I would like to agree with you on a few points. I personally would like to see a 24-bit (32 with Z-buffer?) which used standard NTSC frequencies. Yes, 1280x1024 is nice, but I could easily live with my 640 x 400 interlaced if I had this many colors. Until then, I will be putting in my order for the HAM-E box for my amiga as soon as the bank accounts have reached their low-water mark again. I also suspect I will not be buying another Amiga, but that remains to be seen. brian moffet