madmanwithabox:
I didnt think about a switcher or hub and would rather avoid one if possible and go direct from one ethernet board to the other.
If you're only looking to connect two of them - then a crossover cable should work fine. If you need more than two, then you'll need to use a switch or hub (I don't know of any dual ethernet shields or such for the Arduino - you might be able to do something custom like that using a Mega; the issue is the number of pins needed, more than anything - although you might be able to pull something off with a regular Arduino, software serial ports, and a couple of those serial-to-ethernet adapters).
madmanwithabox:
The coding looks the hard part to me
Network-aware comms coding takes a bit to get your head around, but it isn't too bad (I once wrote a simple web server in a COBOL-ish language (DB/C) - plus I've played around with similar things in VB, Perl, Python, and PHP)...
madmanwithabox:
Networking is very much new to me
You should probably try to understand as much as possible at a PC level if you can - it might help...
madmanwithabox:
I hope to move to FPGA one day if I can get my head round it
FPGAs, from my understanding, are a whole 'nother beast compared to microcontrollers - with an FPGA, you are essentially writing code to control logic states/output based on logic states and/or other input - which is compiled such that the FPGA being "programmed" has those inputs/outputs/gates to work as your "code" is designed. There's also a fair bit of parallel processes that are allowed (must be allowed) to go on that you have to be able to wrap your head around. The simple examples and tutorials I have seen of the coding needed, and the explanations - have left me with my head spinning a bit. I've always figured that if I needed to learn how to use and program FPGAs, I'll worry about it when the time comes, and not a moment sooner, as it is a skill which I am pretty sure I could master, but it might be something that I'll never need. I'll likely know when I'll need it if the time comes. Currently, I just don't see a need for creating custom FPGAs and the code for them for anything I am working on currently; I think that microcontrollers or other existing solutions will fill my needs at the moment.