|
Post by gannondork on Aug 28, 2008 20:14:35 GMT -5
design4dev.wetpaint.com/page/TV+Computer?t=anonSo lately I have been thinking of how silly these people are. I'm sure you have all read or heard about this project, if not the link is above. My question is, what is keeping the c64dtv from being used to fulfill this project? I mean correct me if I'm wrong but they could do the entire thing with the DTV and exceed their goals as there is already contiki for the c64, so they could get on the web with the addition of a network adapter of some sort. I realize that Jeri owns the core and would need to get paid for her work, is it just the c64 bios files that she doesn't own that prevents her from shopping it around or what? Being as the DTV was being sold for $20 at a profit, in the numbers it was selling, it is logical to assume that at a larger scale it would be even cheaper to manufacture. I mean they have network adapters and sd card hardware for the c64 already, as well as modems, it seems like this would be a no brainer. I realize that this is something that has been discussed in the past, but these people are actually trying to do it.
|
|
|
Post by David Murray on Aug 29, 2008 9:59:35 GMT -5
Well, I agree with you on a technical level. But the politics involved would prevent the DTV from ever being used. It is certainly superior to the NES platform in almost every way. But being that they would like to have internet connectivity of some sort, the extra RAM and flash RAM available to the DTV would make it a much more suitable platform.
However, I believe that the NES is now considered more or less an open platform. Its patents have expired and anyone can use the chip without paying royalties. So this is probably very attractive to the developers because it will reduce cost even more. Although personally, I'd be willing to pay a few bucks more for royalties to get the DTV platform.
Still, if this thing materializes into a real product... All is not lost for us Commodore finatics. I would say we should make it a goal to get Commodore BASIC 2.0 operating on the machine and a Commodore-like kernal (maybe even access to IEC disk drives?) I realize there would be limited software compatibility (mostly just BASIC software) but it would certainly make it easier to port existing software to the platform.
I believe it could be done and if we could boot to Commodore BASIC, then all we'd need to learn is the new POKEs and PEEKs for the new video and audio hardware.
|
|
|
Post by gannondork on Aug 29, 2008 13:46:47 GMT -5
Well if the patents for the original nes hardware have expired then would not all patents for the c64 have expired as well, since it predates the NES? So with Jeri's core one would simply need to create compatible firmware, which would probably be easier for someone to do than trying to get contiki to work on NES hardware, though I think there used to be a port I can't find any information on it anymore. It just seems to me it would be easier to work out the problems with the DTV situation and have it go into production as an enhanced c64 with network access and sdcard support since all of the r&d and software tinkering has been completed, than to re-invent the wheel for the NES. Does anyone know exactly who owns what in the DTV situation, what does Jeri own and what do the other companies own?
|
|
|
Post by gannondork on Aug 29, 2008 13:53:39 GMT -5
Oh and also what was the story about Jeri trying to get her hands on the warehouse of produced boards? Did she want to get them to make a device of her own or what?
|
|
|
Post by 1570 on Aug 29, 2008 19:23:43 GMT -5
It just seems to me it would be easier to work out the problems with the DTV situation and have it go into production as an enhanced c64 with network access and sdcard support since all of the r&d and software tinkering has been completed, than to re-invent the wheel for the NES. Producing a new batch of C64DTV chips alone costs an amount of money in the six or even seven digit range (that's what fabs charge for ASIC production), not to speak of the IP issues probably surrounding the DTV core. The NES clones are available in abundance while DTVs are sold out. NES clone devices can probably get mass-produced relatively easily to custom design without expensive chip fab runs. BTW there is no ethernet adapter for the DTV (that I know of). C64 RR-Net devices cannot be connected to the DTV (and are pretty expensive).
|
|
|
Post by Jeff Ledger on Aug 29, 2008 21:07:53 GMT -5
david, While the Propeller wasn't powerful enough to handle Commodore emulation, I suspect the next one. (To be released next year) will be. I just got a peek at the Prototype last weekend, 1 COG (cpu core) is faster than all 8 working together of the existing unit. We expect it will have 8 COGS (cpu cores), 256KB RAM, Enhanced Video ability, DAC, and other interesting toys. It also looks like they will be able to be stacked easily. The price point is expected to be under $20, so something running BASIC 2.0 might be available in 2009. Jeff/OBC
|
|
|
Post by gannondork on Aug 29, 2008 23:43:20 GMT -5
Yeah I will admit that I forgot about the DTV's lack of cartridge support and therefore precludes Ethernet support but seriously does anyone know whether or not the only thing keeping the dtv out of production again is the MSbasic? I imagine it would not be that big of an issue for a compatible version, though I could not do said work myself I imagine that others could, things like this make me wish I hadn't abandoned my CS degree.
|
|
|
Post by 1570 on Aug 30, 2008 3:34:30 GMT -5
As far as I know the IP issues around the DTV core are unresolved (see other threads in this forum; seems there's even a huge amount of core dies sitting in storage somewhere, waiting to get shredded).
Concerning the propeller, for $20 you might as well just get an FPGA and put FPGA-64 on it which is a more or less faithful hardware emulation then and not some ported software emulator.
|
|
|
Post by David Murray on Aug 30, 2008 11:42:49 GMT -5
Concerning the propeller, for $20 you might as well just get an FPGA and put FPGA-64 on it which is a more or less faithful hardware emulation then and not some ported software emulator. As far as I know, it is impossible to get an FPGA chip for $20. I think $80 is more appropriate for one capable of running FPGA-64. So the propeller is still a good alternative if somebody gets the emulation working. jeff: If speed is the only problem, why can't BASIC 2.0 be made to work now, just at a slower speed?
|
|
|
Post by Jeff Ledger on Aug 30, 2008 11:45:19 GMT -5
Memory became the problem is the first attempt. This should be resolved in the next version. It will have a flat memory model.
Jeff/OBC
|
|
|
Post by David Murray on Aug 30, 2008 17:52:18 GMT -5
Memory became the problem is the first attempt. This should be resolved in the next version. It will have a flat memory model. How much memory is required? I mean, the VIC-20 didn't exactly have a lot of it.
|
|
|
Post by 1570 on Aug 31, 2008 5:28:55 GMT -5
As far as I know, it is impossible to get an FPGA chip for $20. The Xilinx Spartan XC3S 400 costs 24 Euros here from a normal retail store, single quantity. This is used for example in the Minimig (Amiga emulation) and still has resources left for that purpose. The XC3S 200 costs about 14 Euros (I think that is used in the 1541Ultimate). Ordering some hundreds of these will probably give you quite some discount. Concerning "What about lower BASIC 2.0 speed": If you aim that low (BASIC emulation is enough, speed does not matter), you might just as well use a normal microcontroller. See for example www.electronicspit.com/users/pit/video/ for an AVR-based BASIC emulation that even exceeds the 64's speed, with very cheap chips.
|
|
|
Post by gmoon on Aug 31, 2008 8:35:34 GMT -5
So lately I have been thinking of how silly these people are. I'm sure you have all read or heard about this project, if not the link is above. My question is, what is keeping the c64dtv from being used to fulfill this project? I mean correct me if I'm wrong but they could do the entire thing with the DTV and exceed their goals..... Errr...silly it may be, but not in that sense. It is rather silly to compare this to the C64DTV: -- the Victor systems exist--as a viable product. -- the $12 dollar price tag includes a keyboard, mouse, gaming controls, etc. -- the NES-on-a-chip systems employ working cartridge interfaces (speculated about on the DTV, but never demonstrated.) -- A cart of educational/programming software already exists for the system (see one of the videos--with word processing, programming , paint, etc..) And no one designing this type of system from the ground-up is going to start with a NES chip or C64DTV--this was chosen by the MIT team as an option simply because it's currently manufactured and very inexpensive. But it is silly to expect that a few MIT post-grads will be able to add wifi without quadrupling the price (or worse) .. . Plus it's not self-contained, like the OLPC (a TV is needed.) So it's only for already-wired communities. And would those communities opt for this over more capable systems?
|
|
|
Post by David Murray on Aug 31, 2008 9:27:25 GMT -5
Wait.. I thought it had already been established that they were, in fact, using the NES. So you are saying that they will be designing this thing from ground up? There would be several advantages to that, but also disadvantages. Lets look at some of them. Pros: - Faster CPU, possibly 16 or 32 bit
- better memory map, probably more RAM accessable
- Better graphics capability
- Probably better integration with modern hardware (such as wi-fi)
Cons:- There are plenty of existing programmers for the NES that would not be able to take advantage of.
- Can't use existing programming tools that are tried-and-true.
- Can't use existing software base of games and other stuff.
|
|
|
Post by gmoon on Aug 31, 2008 10:07:09 GMT -5
Wait.. I thought it had already been established that they were, in fact, using the NES. So you are saying that they will be designing this thing from ground up? Sorry, David, I guess I wasn't clear... I'm just saying the use of the NES-based computer is "engineering by opportunity." I.E., it's choice is based on the existing "Victor technology" and it's incredible cheapness. The corollary-- designing from scratch to a set of specifications (specs probably defined by social scientists, not engineers.) If a group of EE's got together to design one, they'd never even consider the NES or the c64DTV ( maybe that's a shame.) They would go straight to ARM or some other microprocessor technology. Yes, there are some advantages to using the existing technology...the most obvious is cost, which I think is exactly the point being made by the MIT team.
|
|