The Sinclair QL Forum

A lot depends on the type of socket used - push instert / push eject are the best, as it's a positive action for both operations, and I believe the socket has a switch to give 'card present' when inserted. Even someone with stubby fingers like me can get a microSD out of one of those sockets.

There is only standard SD and microSD I thought - not sure what 'mini' is ? Adaptors from microSD to SD are available, which gets around the possible compatbility on swapping cards with someone or trying to read one in a Q68

Pr0f wrote:There is only standard SD and microSD I thought - not sure what 'mini' is ?

Sorry, yes I was using logic rather than memory thinking the middle one was mini but they are called Standard, Micro and Nano. I guess people will want to stick to the size they use most rather than keep swapping adapters that probably won't take a lot of wear.

I have voted for option 2 and option 1 (in that order). I think the rom port option is more comfortable if you have some QLs, it also does not require opening or manipulating Inside the computer. (Another reason for me is that I already have the internal version ... ...).

Thanks everybody who has chimed in so far. I didn't mention one additional advantage of the external version that it will be able to self-update the ROM, which the internal version can't, but it's in the lead anyway. And producing both external versions is much easier than producing both internal versions.

Derek_Stewart wrote:Another option here is to let someone else perform the construction, marketing and sales of the QL-SD card(s).

My experience with the construction and sales of the Q68, the biggest job by far is the dispatch of the customer, this usually takes me a more time to do than the actual construction.

I will probably seek help in the actual sale part, yes. It's not the most time-consuming step for me but the most annoying.

Chr$ wrote:I went for 2 because I think there are probably re-discoverers that may not be comfortable opening their QL (even if most of them will probably have to do so for the keyboard!). Micro SD seem to be more common to buy nowadays, but it doesn't necessarily need to be 2 of them.

One or two doesn't make much difference in the end. My original design was two full-SD and that turned out too big for my taste. But if single SD is enough then one big slot might be the better option.

Get them made in China. Go for larger volume (within reason) to keep the end-user price down and then watch them fly off the shelves.

I guess I would have to order a few thousand of the internal ones for them to be completely produced in China. The external one is much easier to produce in smaller batches. Regarding the price, do not expect it to be on China gadget level. Many hours go not only in the production but also in the development of this stuff, hundreds. For the external version I do PCB design, CPLD core design and software design, that tends to add up. And halve of the price is tax in the end, we live in Germany after all...

Ruptor wrote:For an internal SD to replace a Microdrive it is better to go for that vDrive thing isn't it?

vDrive is nice but not in the least comparable to QL-SD. vDrive emulates a Microdrive, QL-SD is a hard-disc. Think 100kb versus 100MB.

Not many votes so are QL-SDs worth the bother for you to make?

The numbers aren't bad considering that I haven't publicized the poll at all. Most potential customers are on Facebook and I haven't even posted there yet.

Pr0f wrote:A lot depends on the type of socket used - push instert / push eject are the best

I exclusively use push/push mechanics, always.

I believe the socket has a switch to give 'card present' when inserted.

That isn't used as it's not part of the original hardware, so I had to do it in software anyway. Works perfectly.

stephen_usher wrote:2 would be useful if you want to do back-ups or duplicate to give to others, so it would probably be preferable.

On the other hand of course most people also have a PC and there it's even easier to just copy/backup the container file. I like two drives if I can make it fit, but if it gets too big (like with the big SD cards) then I prefer one.

What components on there are non-SMD? Just the headers? What could be left off for others to fit later if they wished, to cut down on your assembly time?

The internal one uses a through-hole EEPROM, not something you can just leave out
For the external one there are two jumpers (essential) and the 3rd SPI port, but I have ideas for a future project for that, so leaving it out would reduce potential users of that. The JTAG header you see will not be populated in any case.

mk79 wrote:I guess I would have to order a few thousand of the internal ones for them to be completely produced in China. The external one is much easier to produce in smaller batches. Regarding the price, do not expect it to be on China gadget level. Many hours go not only in the production but also in the development of this stuff, hundreds. For the external version I do PCB design, CPLD core design and software design, that tends to add up. And halve of the price is tax in the end, we live in Germany after all...

I know Marcel. If you were in it for the money you probably wouldn't be doing anything at all for the QL! Can the external ones be under 50 euros each (end user price)? What sort of volume would you have to get made for that? Were the previous ones 75 Euros each before, or something like that, from memory? It's a pity that the weedy unexpanded QL can't really benefit from a QL-SD as I think that's where more customers would be. However much they cost and whenever they come, I'm sure they'll do well, even if you don't get anything out of it especially when you factor in the amount of time you put in. Respect to you for finding the motivation.

Chr$ wrote:Can the external ones be under 50 euros each (end user price)?

50€ means 40€ without VAT. Assuming the one doing the shipping is satisfied with only 10€, that leaves me with 30€. Let‘s say I can produce it for 15€ total (the central chip alone costs 5-8€), that leaves me with 15€. Minus the income tax I‘m at maybe 8€ „profit“ (note that none of my work is paid at this point). That works out ok if we can sell 1000 or more of them.
It‘s utterly unprofitable any way you look at it and I simply cannot charge the 200€ that realistically it should cost. So 75€ or so is still a rather low compromise.

It's a pity that the weedy unexpanded QL can't really benefit from a QL-SD as I think that's where more customers would be.

I tried using an unexpanded QL recently. Amazing how utterly unusable it was, especially without TK2. I think a Trump Card or 640kb expansion is the least a QL should have.

mk79 wrote: tried using an unexpanded QL recently. Amazing how utterly unusable it was, especially without TK2. I think a Trump Card or 640kb expansion is the least a QL should have.

Funny, I've been purposefully only using an unexpanded QL and it's been going ok. Lately I've been compiling everything on it (ok, I cheat and use vDrive) since I'm not making that many code modifications weekly on my project and though it takes a bit it all works :-/

bwinkel67 wrote:Funny, I've been purposefully only using an unexpanded QL and it's been going ok. Lately I've been compiling everything on it (ok, I cheat and use vDrive) since I'm not making that many code modifications weekly on my project and though it takes a bit it all works :-/

You either have a high pain tolerance or you‘ve simply never experienced how great a really advanced QL environment can be.

mk79 wrote:You either have a high pain tolerance or you‘ve simply never experienced how great a really advanced QL environment can be.

I own a Q68 so it's got to be the former...just something nostalgic of seeing what the original beast can do.