* Posts by JohnLH

15 publicly visible posts • joined 28 Feb 2014

Meta's AI safety system defeated by the space bar

JohnLH

Intelligent? Ha!!

ZX Spectrum, the 8-bit home computer that turned Europe on to PCs, is 40

JohnLH

There was an upgrade kit for the ZX80 to ZX Spectrum, basically a new plastic box with proper keyboard. I had it on good authority that WH Smith had so many problems with 80s being returned that they just put 'em in a box and sent them to Sinclair to sort out and gave the customer a new one. Sinclair soon found they were getting returned 80s that were just the plastic box and rubber keyboard...

JohnLH

Re: Where it all began...for some

Did you enjoy the Dragon's weird colour ghost on every vertical line in the display? Early Dragons in the name of saving a few pence used an NTSC frequency crystal in the video encoder, in the belief that it wouldn't matter if the line time was 63.5 usec rather than the correct for PAL 64 usec. They forgot that every PAL TV incorporated a 64 usec chrominance delay line...

Apple seeks patent for 'innovation' resembling the ZX Spectrum, C64 and rPi 400

JohnLH

Well if you look at the actual document's claims they take the usual US approach of throwing in the kitchen sink into the application in the expectation that they will be whittled down to the actual innovative meat by the examiner. The actual claimed innovation is probably to do with design for cooling and appears from claim 13 onwards. Even then it looks pretty obvious.

Nothing's working, and I've checked everything, so it must be YOUR fault

JohnLH

I once flew to Melbourne Australia to add one decoupling capacitor.

China reveals audit of 320,000 local apps, with 34 booted from app stores and hundreds of devs warned they could suffer same fate

JohnLH

"690,000 5G base stations"

All spreading COVID-19 eh?

Or maybe not...

ALGOL 60 at 60: The greatest computer language you've never used and grandaddy of the programming family tree

JohnLH

Re: It started my career move...

I also learned ALGOL programming on that 803 as a 5th form school student in Rugby. That would have been in 1965/6 probably. Apart from tunes on the console speaker I remember being told that the shortest valid Algol program was a single semicolon, but it compiled to many yards of paper tape so we were warned against it.

Surely the teletype tape reader on Colossus was using optical reading in WW2? After all, optical sound tracks were ubiquitous on talkie films before the war so the technology was well known.

I also remember looking at the Algol 68 User Guide edited by the great Philip Woodward and published I think by the Royal Radar Establishment. IIRC it starts "Begin comment Chapter 1...", and is presented in the form of comments to a program that calculates the date of Easter for any year.

Why is the printer spouting nonsense... and who on earth tried to wire this plug?

JohnLH

Re: DIY Electricians

Some appliances, for example storage heaters, nowadays have two supplies. One is from a time-switched off peak supply to charge the heater, the other is a standard non-switched mains supply to run a fan and the controller. Sometimes they have the ability to boost via one of the heating elements if the stored heat runs out. The switching is static (presumably triacs or something). It's important to make sure that both supplies are the right way round, as one having L and N interchanged could create magic smoke from the devices when both supplies are running. Fortunately the controller in the one I fitted seemed to have a lockout to inhibit the thing from charging if the supplies were crossed - but working out why the thing wasn't working led me to discover that the L and N wires were crossed on the high current input. All the wiring problems we have had in our house have been due to "professional" electricians, like the one who told us there was an earth fault on the immersion heater when he'd wired it to the lighting circuit, not to mention hiding a cheap and nasty LED transformer in the roof insulation so it melted and nearly burned the house down.

5G is 'ready' once you redefine 'ready'... and then redefine 'reality'

JohnLH

Subbing...

"But for the next 18 months, it's largely going to be a debugging exercise. Two coruscating expert summaries underlined this."

Sigh - the word you need is "excoriating" - look it up.

Sad Nav: How a cheap GPS spoofer gizmo can tell drivers to get lost

JohnLH

What's supposed to be new about this?

Grab your lamp, you've pulled: Brits punt life-saving gravity-powered light

JohnLH

So that's 12 Kg x 2 metres x 10 (approx value for g) = 240 joules = 4 watt minutes. Not a lot of power for very long when you've just lifted 12 bags of sugar. No wonder they need a solar panel and battery to make this sensible. This is just the kind of patronising "innovation" we don't need.

Half-ton handbuilt CPU heads to Centre for Computing History

JohnLH

Centre for Computing History is, according to its website, in Coldhams Lane, Cambridge.

Better mobe antennas a stretch goal for radiocomm boffins

JohnLH

This is warmed over news. What the authors describe is a way to implement in effect the combination of a circulator and an antenna. The previous paper (referenced in the Reg article) described just a circulator.

Circulators are non-reciprocal and usually use lumps of magnetised ferrite. They were used in early mobile phones (and in a host of other systems to this day) to allow a single antenna to be used for both Tx and Rx, but methods were developed of making duplexers that didn't need circulators (and indeed worked better). GSM didn't need a duplexer anyway as it is in effect TDD (like Wi-Fi), but 3G and 4G phones do. (Actually there is a new generation of sub-miniature ferrite circulators now available that are finding their way back into phones to help solve the multi-band nightmare.)

Another way to make a circulator is to use an active device such as an amplifier (which amplify one way but normally not the other). Another type of active device is a parametric amplifier which works by "pumping" a variable capacitor at a different frequency from the one you're trying to amplify - it needs input RF power just as a conventional amplifier needs input DC power. Since a major design criterion for a duplexer is power loss (attenuation), having to put additional power *into* a device to make it work is not very attractive.

Parametric amplifiers were quite popular in the 60s when RF transistors were poor but you could use things like klystrons to generate RF power as a pump, but generally are now an engineering footnote. I remember doing a whole exam course on "non-linear and parametric circuits" in my degree, just in timefor them to go out of fashion.

As power consumption is a major issue for mobile devices, I think this technology will remain a curiosity at least for mainstream systems.

Coding with dad on the Dragon 32

JohnLH

Made in Wales, designed in Cambridge, does yours have a strange ghostly colour effect on text? Reason for poor graphics - supposedly clever but fundamentally flawed usage of a standard Motorola graphics chip to save a few pence on a crystal.

New radio tech could HALVE mobe operators' bandwidth needs

JohnLH

Er, actually in an LTE terminal the maximum Tx power is +23 dBm and receive sensitivity ~ -100 dBm, according to 3GPP specs. So chop 20 dB off the problem. Maybe it wasn't clear from the original report but the technique has already been demonstrated working real time from relatively small hardware with ~15 dB isolation from a passive circulator and over 100 dB isolation from a combination of analogue and digital cancellation.