As myself and others have said, upgrading these old machines is turning into a huge problem. Basically because they just have so many bugs and issues right from the start that it makes upgrading these machines somewhat of a nightmare. It is mostly problematical enough to even get a stock machine working, never mind trying to upgrade it which generally lands people in a lot of trouble.
IMHO, Original stock machines should basically be left alone other than general servicing and maintenance, and maybe various tweaks and fixes to keep them operational. But as I have said a couple years ago, as soon as you start manhandling these boards something will normally break.
I put this down to bad soldering on the boards which is just simply starting to let go in recent years. I found exactly the same a few months ago, where a brand-new looking board had perfectly good-looking soldering, and yet the machine did not boot. I literally had to re-solder half the motherboard and now it works perfectly. But this is unfortunately the shape of things to come and upgrading these machines isn't not going to get any easier.
Not only that, as people know, there are just so many variations of motherboards and they all have their own quirks, that it is extremely difficult to produce a reliable upgrade which will work on every board. This isn't so much of problems with upgrades themselves, it is the fact that these upgrades should be tested on every revision of the board which just creates a considerable amount of work making it not worthwhile doing. Even two seemingly identical boards can behave totally differently. Its exactly the reason why I was motivated to create a new motherboard, to start off with something which is fixed and reliable right from the start. With original motherboards, they can be working for some months, put them on the shelf, bring them back off a couple months later and they are dead.I had exactly that happened to me multiple times over the past year.
So how things are looking to me, keeping original stock machines working is just going to be a battle in its own right over coming years. Mix in the fact that they already need various bug fixes and tweaks to even get a stock machine working reliably in the first place. It looks to me like original machines ideally need the whole motherboard re-soldering to keep them operational. So my motivation to create upgrades for original machines somewhat dwindling rapidly.
While as people know, I am moving in the direction of the H4/H5 remake boards, that will ultimately become my main area of focus. Currently I will produce upgrades which will work on STFM's and Hx series of boards as both of these can plug into a 68K socket and both run on 5V. Though I think I am not going to officially support original machines anymore with these upgrades.
For example, development will be done on the H4 board, while they can still be used on the original machines, if they "have bugs", I am not spending the time anymore in figuring those problems out. Because almost certainly, there will be issues with the motherboard itself somewhere. Mix in the fact that people need to at least do all the mandatory fixes even before fitting any upgrades, which just compounds the problem that people just generally do not do this to start with.
Another problem is just the cost of 5V parts. 3V parts are generally a lot cheaper now, which is again motivation to create future upgrades which will work on the H5 with a 3V bus. This will make development a lot faster and cheaper, and of course the price to people will be a lot cheaper. As I am generally spending about £2,000 a month to just buy parts for various kits and keeping on top of the volumes of components and the costs of things is very problematical for me. So moving to 3V parts I just see as a good thing to do.
This also backs onto the point that while the remake boards are expensive to produce, for a lot of people, by the time you have purchased various upgrades, or paid someone to do all the soldering and fixes for you, you might as well just purchased a remake board to start with everything already fixed. Then there are upgrades you can plug-in yourself without having to pay someone to do the soldering work for you. Again the problem with original machines is that they are likely not working to start with. Repairing them and fault finding them is just not a realistic thing to do any more, Unless you have lots of time and cash to throw at it continuously that is.
With more and more people supporting original machines with various upgrades, which is of course a great thing to happen, I don't see a future "market" there for such upgrades for me to keep producing them as well. I'm not getting drawn into "price wars" with people again like I was on evilbay years ago. Another problem is that some items in my store were produced a few years ago like the 4MB RAM kits. if I was to produce it again today, I could probably do it at half the price now. So newcomers who sell a similar kit at half the price, a lot of the customers will just go for the cheaper option. Which I have course don't blame people on doing, but when I have 39 of them (at the time of typing) that's currently £2,340 worth of stock on that one kit alone..If there is a downward trend of sales then of course I will just not produce that kit again. But again,£60 to buy the kit, £50 shipping of your machine to pay someone £50 to fit it, when you can plug a £10 simm into a H4 and job done.. It really is a no-brainer which direction to move in.
This does not mean I will stop producing kits which I am currently doing, if people are still buying this stuff, then of course I will still produce it as long as possible. Though as stated above, upgrades for the H5 will be by main area of focus. While some upgrades can work in original machines (5V stuff that is) I am not going to officially support those machines because of the never-ending problems which go with it. I am more than happy to let other people design and build updates and have all the hassle of getting stuff working on those machines
I have been spending likely 95% of my time debugging original hardware rather than debugging hardware add-ons. Having done this for many many years, it is just time to move on. But again, I think keeping stock machines working even as a stock machine, is a ongoing nightmare in its own right. Having read about all the issues on the forum, and seen first-hand on trends of 100's machines I've had over the years, IMHO, I think stock machines should just be kept that way from now on. Nothing wrong of course in having original machines, if you have the time and money to spend on then fixing them, there is of course nothing wrong with that. But my perspective of bug fixing and future upgrades, the only sensible direction is to move over to a new motherboard which is exactly where I'm aiming for now.
I do of course thank everyone who is buying items from my store! You guys are keeping me going. Plus of course, the great help by everyone in this community in pushing projects forward. I think moving over to our own "stock platform" aka H5 series is the only sensible direction to move in.
