Interesting findings on the internet

[BU508A]

Not sure, if this is good (haven't watched it yet) but Marco Reps approves:

The complete YT channel:

https://www.youtube.com/@ericksonengineering7011/videos

[tggzzz]

And non-yootoob is http://www.djerickson.com/diy_smu/

[Cubdriver]

Someone replaced the VFD of a Keithley 2000 with an OLED display.

Technical details can be found here:
https://github.com/RM-Engineering/Keith ... inch-OLED-

After seeing this and also stumbling on a Zenwizard video on replacing the piss poor display in the HP 8904A Synthesizer this past weekend, I ordered a pair of replacement 40 x 2 LCDs to do mine. Need to see them in person to decide if I'm going with white characters on a blue background, or black on greenish-yellow. Either has got to be better than the barely readable one that's in it now!

-Pat

[Zenith]

The original display may not have been great to begin with and modern ones are better, and it may have degraded with age. However, some of those LCD character displays have a pot to adjust the contrast, and you may only need to tweak that. I was playing with an unused one a few days back and with everything wired correctly and a character string sent to it, it remained dark. Then I remembered the contrast pot and tweaked it, and it came to life.

[AVGresponding]

Someone replaced the VFD of a Keithley 2000 with an OLED display.

Technical details can be found here:
https://github.com/RM-Engineering/Keith ... inch-OLED-

After seeing this and also stumbling on a Zenwizard video on replacing the piss poor display in the HP 8904A Synthesizer this past weekend, I ordered a pair of replacement 40 x 2 LCDs to do mine. Need to see them in person to decide if I'm going with white characters on a blue background, or black on greenish-yellow. Either has got to be better than the barely readable one that's in it now!

-Pat

Dual channel, 600Ω, perfection!

[Cubdriver]

Yeah, took some time for one with those features to appear on eBuy back a few years ago when I got it.

-Pat

[tggzzz]

Only 37 songs, I believe - but still remembered over half a century later...

https://www.theguardian.com/music/2025/ ... l-satirist

This is important, not political correctness. I tell people, "I'll call you women instead of girls, just so long as I get paid more than you do." That's the issue, not all that PC stuff.

You can't be satirical and not be offensive to somebody.

https://www.avclub.com/tom-lehrer-1798208112

[Zenith]

A bit sad, but he was 97, which is a good run.

I recall many of his songs with fondness:

Poisoning Pigeons in The Park.
The Old Dope Peddler.
Lobachevsky.
The Vatican Rag.

and many others.

[Zenith]

Sale day at The National Museum of Computing.

https://www.youtube.com/watch?v=k6XxyaT ... unFixingIt

Apparently it took place on 19th July. Bargains galore (for those into 80s home computers, PETs, etc).

It drags on 43 minutes, but it's worth skimming through.

I though it worth mentioning but didn't know where to put it. Events - I didn't go. For Sale and Wanted - it's been and gone. Computers - I suppose so, but not really. So I dumped it here.

[tggzzz]

What's the Allen Deviation of your pulse (and other clocks)? See the PDF on http://www.leapsecond.com/ten/

Too much other timenuts stuff on that site

[Zenith]

That's a very good website for those even slightly drawn to time nuttery. I find for now that my ebay Chinese GPSDO is bags.

If you look through it, some of the links to other sites are also interesting. eg:

For those of us who think we have Test Equipment Acquirement Syndrome, we really don't.

https://www.kennethkuhn.com/hpmuseum/

And for calculator enthusiasts.

https://www.hpmuseum.org/

https://www.oldcalculatormuseum.com/#gsc.tab=0

[Cubdriver]

That's a very good website for those even slightly drawn to time nuttery. I find for now that my ebay Chinese GPSDO is bags.

If you look through it, some of the links to other sites are also interesting. eg:

For those of us who think we have Test Equipment Acquirement Syndrome, we really don't.

https://www.kennethkuhn.com/hpmuseum/

And for calculator enthusiasts.

https://www.hpmuseum.org/

https://www.oldcalculatormuseum.com/#gsc.tab=0

Inspiration!

-Pat

[BU508A]

When the earth was once a snowball:

https://agupubs.onlinelibrary.wiley.com ... 24JE008701

[tggzzz]

New paper about quantum computers is available...

"Similarly, quantum factorisation is performed using sleight-of-hand numbers that have been selected to make them very easy to factorise using a physics experiment and, by extension, a VIC-20, an abacus, and a dog."
https://www.schneier.com/blog/archives/ ... marks.html

and the original paper
https://eprint.iacr.org/2025/1237.pdf

If extended to include cats, I'm unclear where they would be in that paper's ranking
"In terms of comparative demonstrated factorisation power, we rank a VIC-20 above an abacus, an abacus
above a dog, and a dog above a quantum factorisation physics experiment. "

[Zenith]

I'm inclined to think quantum computing is the technology of the future, and always will be. Rather like fusion reactors, which are 50 years away from being practical, and have been since the 1950s. Occasionally a "breakthrough" is announced, but it never seems to amount to much.

Now I'm not saying these things don't have potential and it's a waste of time researching them, but it's easy to be sceptical.

[tggzzz]

"Incremental" would be OK if the increment was in a realistic direction without caveats.

I don't see quantum computing being general purpose, but then crypto doesn't have to be.

[bd139]

That's about it. Quantum computing is one of those things. There's progress on the left hand side (physics) and progress on the right hand side (mathematics). But for commercial viability there needs to be more than just a meet in the middle, there has to be a huge overlap and it has to result in viable applications. Alas at the moment there is a humungous gap and no real-world applications that can't be covered by classical computing. On top of that due to the physical properties of the equipment it's likely to be very difficult to commercialise on any scale anyway.

It's another graphene: even if we succeed, it doesn't mean it's going to change the world much once real-world practicalities are evident.

[Zenith]

On top of that due to the physical properties of the equipment it's likely to be very difficult to commercialise on any scale anyway.

It's another graphene: even if we succeed, it doesn't mean it's going to change the world much once real-world practicalities are evident.

AFAIK, quantum computing devices have to be run at very near absolute zero and that's not their only problem. Fusion reactors have in some way the opposite problem. They have to contain an unruly plasma at incredibly high temperatures. Both of these have extremely hard fundamental problems.

For graphene, see the Anastasi in Tech Youtube channel.

https://www.youtube.com/watch?v=yJSrX1u ... tasiInTech

It's also good for more likely integrated circuit developments. She knows what she's talking about and she's easy on the eye.

Over the years there have been a few developments which were hyped, such as room temperature super conductors, which were going to lead to all sorts of wonders, but really didn't, and others that were leapt on with enthusiasm and found practical applications, but were superceded by developments of more conventional technology. Around 1960 the fast switching properties of tunnel diodes were expected to lead to faster computers, and I've seen a few designs for gates using them. I'm not sure that any commercial computer used tunnel diodes.

[bd139]

On top of that due to the physical properties of the equipment it's likely to be very difficult to commercialise on any scale anyway.

It's another graphene: even if we succeed, it doesn't mean it's going to change the world much once real-world practicalities are evident.

AFAIK, quantum computing devices have to be run at very near absolute zero and that's not their only problem. Fusion reactors have in some way the opposite problem. They have to contain an unruly plasma at incredibly high temperatures. Both of these have extremely hard fundamental problems.

It's not just that. Stray neutrinos, EM fields, even physical motion can cause decoherence. The universe is fundamentally hostile to the notion of the qubit. Considerably less so with traditional bits of rock.

For graphene, see the Anastasi in Tech Youtube channel.

https://www.youtube.com/watch?v=yJSrX1u ... tasiInTech

It's also good for more likely integrated circuit developments. She knows what she's talking about and she's easy on the eye.

My professional qualification is actually microelectronics, albeit barely used in practice annoyingly so this my domain to some degree. The current idea of the graphene integrated circuit involves selectively introducing flaws into a perfect single atom thick graphene substrate. This is merely hype. There is little to no room for error on this so getting an actual production integrated circuit which is even remotely capable of something more than a few logic gates on a very low yield is a monumental stack of probabilities against you. I don't think we're anywhere near anything useful there or will be in the next 10 years. To put this into perspective your current mid range Ryzen has 8,315,000,000 transistors that need to work...

Over the years there have been a few developments which were hyped, such as room temperature super conductors, which were going to lead to all sorts of wonders, but really didn't, and others that were leapt on with enthusiasm and found practical applications, but were superceded by developments of more conventional technology. Around 1960 the fast switching properties of tunnel diodes were expected to lead to faster computers, and I've seen a few designs for gates using them. I'm not sure that any commercial computer used tunnel diodes.

Yes. Tunnel diodes are rather cool from a physics perspective but not that good really compared to other stuff. Incredibly difficult to bias reliably across any reasonably priced device spread and can't handle a lot of real power. Absolutely no good for building thousands of logic gates out of. Your normal BJT however, is a lot easier to bias reliably and consistently hence why TTL and ECL etc won that one.

Anyway as noted, there's a pattern here: worked on paper, doesn't scale for production.

[tggzzz]

Tunnel diodes were interestingly fast compared with contemporary Ge transistors

[Zenith]

Tunnel diodes were interestingly fast compared with contemporary Ge transistors

They certainly were and I've read there were some high current ones. But silicon planar transistors came in a little later and improved rapidly. By around 1970 there was ECL that could operate at 3GHz and TTL and CMOS had improved significantly. There were Tunnel Diode Amplifiers, mainly used by the military. Their main application was triggering and that continued until the early 80s. I've mainly seen them in scopes, although I do have an analogue function generator that uses a couple, I assume for triggering. I can't find a manual for it. Another application was low rise time pulse generators, although there wouldn't have been much interest in them outside some sections of the electronics industry.

They are often covered in text books, because as bd139 pointed out, they are interesting from the point of view of physics and as an example of a negative resistance device (along with tetrode valves, Lamda "diodes" and a few other things).

They were never common devices. I only saw them for sale once in the sources I used. I only saw one published design using them, and that was The Wireless World Oscilloscope, I recall from around 1970. I doubt many people made it. They are very easily damaged by over current and soldering. They are supposed to be static sensitive as well.

[tggzzz]

TD pulse generators were used for cable fault location, e.g. Tek 1502. Operates in driving rain, has a unique shorting BNC connector to discharge static on cables, used in Patriot missiles batteries.

My 1971 Philips 1.7GHz sampling scope has a 2GHz FET in the trigger circuits, BC107s elsewhere.

[vk6zgo]

TD pulse generators were used for cable fault location, e.g. Tek 1502. Operates in driving rain, has a unique shorting BNC connector to discharge static on cables, used in Patriot missiles batteries.

My 1971 Philips 1.7GHz sampling scope has a 2GHz FET in the trigger circuits, BC107s elsewhere.

We had a TU-5 pulser as an accessory for the 545B at my old work, which used a tunnel diode.
545s had a "calibrate" output that went up to 100v, which was used to drive the TU5.
A lot of them are probably still "sculling about" in various junk boxes, as later 'scopes didn't have the 100 v cal function.

Later, I had a TU-5 ex some random junk box & toyed with the idea of making up an astable using a 12AT7 to drive it, but my mate had a 545B so I ended up giving it to him.

[BU508A]

Here's a little quiz about valid / invalid email adresses:

https://e-mail.wtf/

[tggzzz]

Here's a little quiz about valid / invalid email adresses:

https://e-mail.wtf/

Lovely!

I especially like the valid fork bomb and zaldo text addresses.

Tried creating zaldo addresses on gmx and gmail. Gmx doesn't like the names, period. Gmail allows low complexity names such as "H̸i̵m̷ ", but refuses to have anything other than a-z,0-9,. in the address. So tough luck if your name is Þor Hornafjörður; you end up as thorhornafjordur@gmail.com. Typical US egocentricity!