On Thu, 15 Jun 2000, Spehro Pefhany wrote: > I've just been looking over the documentation on their web site, > some demanding requirements as to putting their name on the front > of *your* product and they are a bit cagey about the gateway > licencing fee. Yeesh. Makes that Seiko chip look better all the time, doesn't it? > I presume if you intend to actually produce an internet- > enabled product using, say, a PIC an gateway you have to > licence the gateway, correct? More to the point, you now have a PIC device that has to have a PC gateway... so whatever you're doing, you probably could just as well do it on the PC to begin with. It might work OK for a situation where you'd have a whole BUNCH of PIC based devices all using a common gateway, but I suspect there's a better way than that. And it would take a lot of Seiko chips to equal the cost of a PC and the emWare gateway software, I suspect. > And finally, anyone have a ballpark on the ROM footprint > of the firmware portion that goes in the PIC? No, but I'm most of the way through a very basic (read stripped-down) IP/TCP/UDP/ICMP/SLIP stack for the PIC. I was using CC5Xfree until I hit 1024+ instructions, and have moved to C2C to see how well it does. The code bloats WAY up (1600+ instructions, after manually stripping out a BUNCH of stuff in the .ASM file). When it's done, I think it will be a good fit for a 16C73b or an '87x processor. Forget about doing it in an 'F84, unfortunately. I am currently trying to decide what to do about my compiler situation, since I'm not about to port this to assembler. I don't think C2C is going to work well for me, so I'm now looking at CCS, since the cost of entry for CC5X just went from $90 to well over $200. This is a hobby project -- albeit a big one -- and I just can't see plowing that kind of money into it when the dog needs kibbles. Dale --- The most exciting phrase to hear in science, the one that heralds new discoveries, is not "Eureka!" (I found it!) but "That's funny ..." -- Isaac Asimov