Why not Just put a row of jumper pins in then if you want switch's you can connect them yourself to the jumper pins.
jumper's should come with unit.

easy and cheap should suit everyone you can buy switch,s from your local electronics store.

i would go with no case but offer one as extra put it in yourself.
you get a board and a case or just a board easy for you and anyone can place it in a case just drill your own holes to suit your own switch's.

This should suit everyone.

_________________
Have Fun!!

Dip switches are easier for the average users, especially if the case can be prepared for dip switch access. But I would prefer jumpers since I believe switching between RR-Net/TFE mode will be rarely used, since all network software is now being made RR-Net compatible. I don't know if there still exists any software that only supports TFE? Except for obsolete and outdated releases perhaps. Power users could easily add their own switches.

About interfacing the RR-Net. The RR-Net CS8900 registers ($de02-$de0f) are being enabled by setting bit 0 of $de01 which enables RR clockport access. TFE doesn't need this and use linear mapping with base address $de00. It's important that writing $de01 in RR-Net mode does not cause any troubles (should just be ignored).

It looks like jumpers are going to be acceptable. If most people use RR-NET instead of Net64, then it makes sense that the switch would be rarely used reducing the importance of easy access to mode switching.

Something that has been on my mind in the last 24 hours is: how many people that still use their C64s can be considered ordinary users? I haven't been active in the Commodore community for a very long time so I don't know where everyone really fits. Is it safe to say that most people who would want an ethernet card for their C64 would be comfortable using jumpers? If so, then there's no need to worry about making this thing as easy to use as possible. I'll still try to setup the reset button and ethernet jack to be easily accessible from a case with holes in it, but I wont worry about easy access to the jumpers through a case unless I get a firestorm on this forum about it.

And thanks for your continuous input on this, all! I really appreciate it! I'll keep you posted on my progress.

My guess is that the IDE64 lockup is caused because the SETMAC program writes something to the IDE64's memory space as it "searches", since an RR-Net can also be there.

I think anybody buying one of your cards would comfortable with jumpers. The IDE64, Retro Replay, CMD EX2+1, Turbo232, MMC64, etc all use jumpers, and that's fine. Like RaveGuru said, they won't be used all that often.

Thanks for the feedback!

Trying to transition from OrCAD to Eagle is a real pain, but I'm figuring it out. Right now, I'm still tweaking the component layout and learning Eagle's routing tools before I get knee deep into the routing. I should be done with the layout this weekend and we'll see about the turn around times of the fab houses. If we hit no snags, I should be able to get these things pumped out within 1 to 2 months.

In my efforts to make this work with a case (I am using Solakian and standard Commodore cartridge cases as templates) without major surgery, I had to make this about twice as big as I could have. It is still not big enough to completely fill out a cartridge case, but almost. Part of the size issue has to do with the constraints of the case, and part of it has to do with the fact the ethernet jack was requested to be put on the side instead of the back of the cartridge. It'll add to the cost of the PCB, and it will take a little more desktop real estate. I am also side mounting the push button reset switch and ethernet jacks so they will be easy to reach. If all goes well, the boards should come with pre-drilled center holes for the cartridge cases so all you'd have to do to fit this in a case is cut a hole in the side for the ethernet jack and drill a hole for the reset button.

I'm eyeballing a few PCB fab houses to see what I can get on pricing. Since I don't have all the numbers in yet, this is not a hard figure. But assuming the PCB fab costs are what I think they will be, I should be able to throw these bad boys out there for about $60-$70 + Shipping. I'd love to be able to get rid of them a bit cheaper, but I'll have to wait and see what I get on a fab price.

Before I get too gung-ho, if I can get them at that price, who's interested? It was mentioned earlier that some retailers were interested. Can you put me in touch with them? The production setup in the PCB fab houses usually run a few hundred dollars. I need to start working on a figure to see how many units I can spread that across to get a better idea of the price of the card.

Maybe you should contact this guy and see how he managed to produce those cheap, small cases:

http://www.prophet64.com/

This price should attract most people looking to buy a ethernet card but not interested in the extra feats and costs of a Retro Replay or a IDE64 v4 cart. Although I already own the RR+RR-net I'd still like one since it feels a little like a baby of this forum AND I'd probably do most of my network developing and playing around with it because of the more robust and compact design.


Like yourself and MagerValp already stated, there should be absolutely no reason to worry.

I will have 2
Also i would like to when the time comes pay with paypal....

Do you think you will do a second run if all goes well.

Thanks Zap

_________________
Have Fun!!

As of right now, I've got most of the layout done. I have two signals left and a few power and ground connections then I have to see what Eagle can do for me with regards to ground planes. I should have this done tonight.

I got some better numbers, but still not final. Working with a quantity of about 50, my costs will put it up there around $75 per unit assuming the first protoype works correctly and I don't have to make a second. If I can order a quantity of 100 or more, I can reduce this cost. I'm also still trying to find places to get these produced so if I find a place that can do a good job for a cheaper price, I'll use them. As I have said before, I'm not trying to make a profit on this, but I can't afford to take a loss. So, we have options.

I can sell them direct, but I can't afford to purchase the boards outright. I would have to drop a few grand that I just don't have. The only way this can be done is if I take pre-orders. The problem there is that it requires a lot of trust on the part of the customer because most of you haven't even met me and those that have have really only met me, but don't really know me well. I can promise all day that I wont screw you over, but it's really a tough call for you. Plus, if we go direct, then we can't use direct credit card purchases. I can setup paypal to accept orders or we can use money orders, bank drafts, and checks. If I don't get at least 50 orders, I have to return all the money and scrap the project.

I can (and would like to) go through a retailer, but that would boost the cost a little bit. The reason I would like to go through a retailer is to do my part to help keep the community alive and retailers are very key to this. If we go through a retailer, there may be the option of having the retailer purcahse a bulk order from me and I can go ahead and move forward with production. If we can't make that work, then we will only have the direct method.

As far as a second production run goes, that really depends. Since I am not trying to make a profit, selling all units will not give me a surplus of cash to order more units. This means I would have to be able to scope out a definite demand and have an established distribution channel. If we can make this work out, I'm willing to keep these things coming out as long as I can afford to do so.

I should also state here that my main goal was to get the thing designed and working. Once that happens, if someone else wants to produce this thing (especially if you have the means to get it done cheaper than I), I have no problem handing over the files to make this work. In fact, when all is said and done, I intend to post a tarball or zip file of all the files on my site so they can be freely downloaded and you can do what you wish with them. But, the design work is not worth it if the product never comes to fruition, so I'll still try my best to get these things out there.

Well, early in the thread I did say I would be willing to buy 20, a few for myself and a bunch for others around here...I'll still stand by that but obviously I'd prefer if the price came down some more (gulp)!

Regarding retailers, the guy from http://www.c64reloaded.com/ also posted earlier in this thread that they would be interested in some.

The current powerhouse in terms of C64 retailing is Protovision (http://www.protovision-online.com/) - They're also the main source of RR-Nets (if not directly from Individual Computers), and they seem to have some technical background. Maybe see if they're interested.

A list of (nearly?) all Commodore vendors/retailers is here:
http://www.tpug.ca/links.html

Don't underestimate the potential demand - the IDE64, MMC64, RR-Net etc. have sold out several production runs over, and are still in high demand. I'm hoping your cart will have a high 'cool' factor since it could be configured as RR-Net or TFE compatible, and not require you to buy anything extra (but will be compatible with most other carts). That makes it the most versatile of the lot.

As for production - I saw that you're on the C=PCB forum on petscii.com, be sure to ask for recommendations there.


Everyone else: I met eightbits briefly at the C4 Expo, we only chatted for an hour or so, but he seems to be on the level

Thanks for the info and the reference!

I'm seriously considering just hand soldering these suckers or the price just may be too high. The issue is with the way I've done the layout. I may have to re-work some of the cs8900a pins and the ground planes. With so many pins connected to each other and GND and VCC in 8-bit mode, and with extensive ground planes, hand soldering a TQFP chip can be a real pain. If the pins were not connected to each other, I could do a mass soldering of the pins quite easily. In fact, if that were the case, the CS8900a chip would be the single easiest soldering job on the board. But with up to 4 successive pins connected to each other as well as a ground plane, that becomes difficult without automated soldering.

So, I'm going to see if I can get a single PCB made with no assembly and try to hand solder it to see how feasible it really will be. If it turns out to be a reasonable option, I'll go that route and see if we can't drop the price.

So, I'm going to spend the next day or two talking to PCB fabs for quotes with no assembly and see if I can re-work some of the layout to optimize it for hand soldering and get some prototypes made. If that all works out well, then my initial cost will come down enough that I might be able to front the cash to get them made without dealing with pre-orders.

Just a status update.

I have found a PCB fab that can make the bare PCBs with all the right tolerances for a reasonable cost. I'm going to submit my order for a couple of prototype boards to them and I should have them sometime next week. I'm also going to order the parts I need for a couple of prototypes. Once I've got all of that in, I'll get to soldering. Assuming all goes well, I should have a working and tested prototype in about a week. If the prototype solders well, I will make these myself and I should be able to get the price reduced significantly.

And, I passed exam 2 of the GSEC so I'm now certified and done with all that business. I shouldn't have spent so much time on it. I studied for a long time but only picked up little tidbits. Had a buddy who was certified tell me to stop studying and just take it so I did and got a 93%. Sorry to waste so much time on that! Had I known I was that ready for it, I would have just taken the exams a month ago and could have had this ethernet card sooner.

Also, I have found funding to make a bulk purchase so we can skip the whole pre-order business. I'll get a final price up here when I finalize the details of the order.

If the prototyping goes well, these will be in production in a few weeks.

OK, all. I've chosen a name and I'm going to stick with it unless you think it's absolutely terrible.

C64NIC

As in C64 Network Interface Card.

For those not in the know, NIC is a standard IT acronym for an ethernet card. Actually, it's standard for any type of network card. So, I think this is pretty descriptive.

I'm wearing my flak vest so hit me with your best shot.

very good choice for the name! thumbs up! and by the way thanks a lot for all your work and effort! that's just great, really!

Ok you have your flak vest on

The name is good
Only c64web would have made me happier
gee lucky i am in oz as i can hear the slap at the back of the head coming ..

Looking forward to the finished item.

_________________
Have Fun!!