ZX Spectrum Next Discussion

For experts to discuss very technical stuff and newbies to ask why the Spectrum they bought off ebay doesn't work.
User avatar
beanz
Microbot
Posts: 114
Joined: Wed Mar 21, 2018 1:10 pm
Location: Texas USA

Re: ZX Spectrum Next Discussion

Post by beanz » Mon Jun 17, 2019 12:13 pm

Alcoholics Anonymous wrote:
Sat Jun 15, 2019 4:45 pm
There is no update because nothing has changed
The change is the fact it wont be shipping by end of Q2 as promised, regardless of the issues that is an update in itself, and one that would take 5 mins to write and post. Not all backers are following every post made and I suspect many are still thinking it's going to ship in the next week or two.
0 x

User avatar
DouglasReynholm
Dizzy
Posts: 69
Joined: Wed Feb 20, 2019 8:38 pm
Location: North London

Re: ZX Spectrum Next Discussion

Post by DouglasReynholm » Sun Jun 23, 2019 5:34 pm

Well, latest update is here and the new membranes failed so it looks like they're going with a brute force and maybe inelegant but 'foolproof' solution by just adding a couple of extra signals to the membrane, which currently look like they need to be hand soldered onto the board. Not great, not terrible? Anyway, looks like we may finally be getting our machines, so rejoice?
0 x

namco
Dizzy
Posts: 91
Joined: Mon Dec 04, 2017 8:55 pm

Re: ZX Spectrum Next Discussion

Post by namco » Sun Jun 23, 2019 6:21 pm

Godammit Doug - you know the rules.

Bizzley has to be the first for bitching when the updates come out! :P
2 x

User avatar
Seven.FFF
Manic Miner
Posts: 320
Joined: Sat Nov 25, 2017 10:50 pm
Location: USA

Re: ZX Spectrum Next Discussion

Post by Seven.FFF » Sun Jun 23, 2019 6:28 pm

The board has a 2-tail connector socket. The hand-soldering is just awkward wording. It refers to the socket being fitted at SMS by hand in rework stations, as the assembly line board manufacture has already happened a long time ago.
0 x
Robin Verhagen-Guest
SevenFFF / Threetwosevensixseven / colonel32
seven-fff.com/blog

ZxSpence
Berk
Posts: 24
Joined: Sat Mar 16, 2019 7:29 am

Re: ZX Spectrum Next Discussion

Post by ZxSpence » Mon Jun 24, 2019 5:47 am

At last, sense has prevailed.

Then when it's done if there's still a problem the team will have to accept the problem was elsewhere all along. We've all been there!
0 x

User avatar
DouglasReynholm
Dizzy
Posts: 69
Joined: Wed Feb 20, 2019 8:38 pm
Location: North London

Re: ZX Spectrum Next Discussion

Post by DouglasReynholm » Mon Jun 24, 2019 7:36 am

namco wrote:
Sun Jun 23, 2019 6:21 pm
Godammit Doug - you know the rules.

Bizzley has to be the first for bitching when the updates come out! :P
:)

I didn't mean to sound like I was bitching, I'm really pleased about the news and glad to see common sense has prevailed.

However, I'd be lying if I said my comment wasn't tempered to being less than positive to pander to the moaners
0 x

User avatar
1024MAK
Manic Miner
Posts: 411
Joined: Wed Nov 15, 2017 2:52 pm
Location: Sunny Somerset in the U.K. in Europe

Re: ZX Spectrum Next Discussion

Post by 1024MAK » Mon Jun 24, 2019 7:55 am

Ahh, good. I was never keen on the purely mechanical approach, as I said earlier in this thread...
1024MAK wrote:
Mon May 13, 2019 2:26 pm
The problem has been and always will be that a purely mechanical membrane keyboard system is never good at making two independent switch contacts at the same time. Even during the days of Sinclair. Far too often on my ZX Spectrum+ keyboards, I get the wrong action unless I press the key squarely.

The better solution would have been to do what Amstrad did with the +2A/+2B/+3/+3B and have a normal two layer membrane and perform the additional switching in a gate array (CPLD or FPGA) chip.
Let’s hope they can get the new membrane produced quickly and without any further hiccups.

Mark
1 x

Ralf
Dynamite Dan
Posts: 1169
Joined: Mon Nov 13, 2017 11:59 am
Location: Poland

Re: ZX Spectrum Next Discussion

Post by Ralf » Mon Jun 24, 2019 7:12 pm

Ok, time for a silly (or maybe not) question ;)

Would anybody by able to explain in layman terms what's the deal with the keyboard here?

I believe keyboard manufacturing isn't a rocket science, especially in 2019. Chinese make hundreds of millions of PC keyboards
each year for a really little money. Plus consoles, plus calculators and so on. I'm currently typing on such a cheap keyboard and it
never failed me, it just works.

Why is the Next team struggling with the keyboard for almost 2 years? Don't we have a ready, working solutions how to make a keyboard?
1 x

User avatar
DouglasReynholm
Dizzy
Posts: 69
Joined: Wed Feb 20, 2019 8:38 pm
Location: North London

Re: ZX Spectrum Next Discussion

Post by DouglasReynholm » Mon Jun 24, 2019 10:43 pm

Ralf wrote:
Mon Jun 24, 2019 7:12 pm
Why is the Next team struggling with the keyboard for almost 2 years? Don't we have a ready, working solutions how to make a keyboard?
My personal guess is that they went quite far down the original +/128k look route, which in and of itself informs the mechanical design, but also wanted to improve the 'feel' of it therefore making a rod for their own backs in terms of a custom design. I don't really know. Also I was a rubber key and +2 only owner and have never used the +/128k keyboards so never had a horse in the race in terms of feel, and the first thing I'll do is hook up an external anyway to keep it in the best condition I can! :lol:

Having said that I do appreciate the teams effort to make it the best they could, but having some experience of electronic design/manufacture I'm a little embarrassed for them the 'will deffo work' solution is the addition of a couple of signal traces, when the delay has been as long as it has.

But, probably a lot of info I'm not privy to, so all hearsay and conjecture. I'm just really looking forward to getting mine now and after all the delay getting quite excited, when I put my pledge in I accepted there and then I might lose the money.
0 x

User avatar
1024MAK
Manic Miner
Posts: 411
Joined: Wed Nov 15, 2017 2:52 pm
Location: Sunny Somerset in the U.K. in Europe

Re: ZX Spectrum Next Discussion

Post by 1024MAK » Tue Jun 25, 2019 2:08 am

Ralf wrote:
Mon Jun 24, 2019 7:12 pm
Would anybody by able to explain in layman terms what's the deal with the keyboard here?
It’s not a silly question.

The problem goes back to the original design of the rubber key model. The design only allowed for a maximum of 40 individual keys (not including duplicates). The original uses a 5 by 8 matrix (hence 5 x 8 = 40). If you want a modern ‘PC’ keyboard version of this, yes it is dead simple. However, I continue with the story...

So when Sinclair wanted to add extra keys as part of the ZX Spectrum+ design, they had to come up with a cunning plan. There were four obvious possibilities:
  • Redesign the keyboard, the PCB and the ULA so that a bigger matrix could be used (more matrix lines means more keys) - rejected presumably due to cost and compatibility.
  • Find a supplier that could provide proper keyboard keys including some with two sets of contacts (so that with the extra keys, they could operate and pretend to be two separate keys from an electrical point of view - same as the user pressing two keys at the same time) - again, rejected presumably due to cost.
  • Keep the ULA as per the existing 16K/48K rubber key models, but modify the PCB to have extra circuitry to handle a bigger keyboard matrix and use a membrane with two conductive layers like in the rubber key model, but with more keys.
  • Keep the existing PCB and ULA design and instead use effectively two membranes on top of one another. So that for the extra keys, when you press them, contacts on each of the membrane layers “make contact”. Hence one physical key can operate two separate circuits and hence electrically have the same effect as the user pressing two keys at the same time. There are ‘rubber’ domes between the membrane and the hard plastic key tops rather than metal springs. This was the cheapest so that’s what they went for, despite the poor feel of the keyboard.
Why was it a requirement to electrically operate two keys at the same time? Well that’s because most of the extra keys are performing the same function as that when the user on the rubber key has to press two keys at the same time. For example to delete a character, on the rubber key you push and hold CAPS SHIFT then press the 0 (zero) key. On the plus, the delete key closes contacts that make the ROM code think that both the CAPS SHIFT key AND the 0 (zero) key have both been pressed at the same time, hence the action is to delete a character.

One of the promises with the Next, is that of a better keyboard. The thinking was that for the extra keys, one physical key could operate two sets of membrane contacts at the same time, but only using a normal layer membrane.

But alas, the reliability of both these contacts making at the same time is poor and very dependent on the angle that the key is pressed. Hence various different designs being tried to no avail.

The solution is to abandon this daft idea of trying to reliably simultaneously operate two physical contacts at the same time. And instead add extra keyboard matrix lines. Then in the circuitry, have some logic to do the conversation.

This method, is, in my humble opinion what they should have done in the first place.

Note that modern PC keyboards are membrane types, but they have a very large matrix, and have a special microcontroller chip to scan the matrix and work out which key or keys are being pressed. They then send a signal to the computer with the details.
But in a ZX Spectrum, the Z80 microprocessor under the control of software (the ROM) does the scanning of the matrix. So from the point of view of the Z80 and the software, the signals from the keyboard have to look like it is a rubber key 5 x 8, 40 key keyboard in order to maintain compatibility...

Mark
4 x

Post Reply