Re: [Jmri-developers] Fast Clock Proposal

[David D. <djd...@ma...>]

Ken,

Sounds fine.

My layout is both Loconet and C/MRI.  I'm unaware of any commercial C/ 
MRI clock implementation.  Do you know any?

Dave

On Oct 23, 2007, at 10:16 AM, Ken Cameron wrote:

> David,
>
> At the time I started this I had asked Bob if some 'generic'  
> approach was the better way to go. The result was that the current  
> implementation was still quite a bit short of doing things that  
> way. I hadn't been doing Java in a few years (like 10 years) but  
> lots of JavaScript and other similar languages. So I felt safer  
> taking the LocoNet version and approach and making it into a NCE  
> version. So far that's getting closer as I think I've learned the  
> tricks around the NCE clock.
>
> So I agree I'll finish out this way first and once I get the clocks  
> matching (how good is good enough anyway??) then I'll pickup your  
> 'generic' base and make the right NCE modules for that. From what  
> you have been saying it shouldn't take that much once I get these  
> tricks right.
>
> BTW I also have a club Digitrax layout available besides the NCE at  
> home and another club. So if we get to the point where we want to  
> have one clock sync the other system (think rails ran by one system  
> and the turnouts/signals by the other system) I'll be able to setup  
> a test base for that too. But I don't think that's a very important  
> combo. However between either and CMRI is going to be an important  
> version combination. We have a sizable layout in the area that is  
> Digitrax and CMRI so I may have that as a test bed if nobody else  
> does.