Reading the Digital Computer Electronics eBook (third edition):
Today we have heard about a new accident involving a teenager electrocuted by her mobile phone. Luckily, this time it was not a lethal case, but a quick search on the web has revealed that this is no joke: several teens have died in just a few years because they were either holding their phone with wet hands while the phone was being charged at the same time, or dropped their phone into the bath tub while the phone was plugged in, or because they were using wired headphones while charging their phone!
At first Simon and I didn’t believe this could be so dangerous, as he knew for sure that a mobile phone adaptor always has a voltage control built into it that reduces the voltage from 220V to something like 5 to 20V. But then we dove into it and found out that apparently, once a short circuit occurs, the adaptor’s voltage control unit also malfunctions and lets the 220V current through!
In October and early November, Simon was busy with another attempt to simulate SAP-1 (simple as possible processor, an 8-bit computer) in Circuitverse (something that he hadn’t managed to complete when he tried it last time). I’m not even sure if anyone uses Circuitverse for such large-scale projects.
On November 7, Simon finally managed to finish the RAM on his simulated 8-bit computer (a computer where every general-purpose register contains 8 bits and therefore can only process 8 bits of data)! Although he is far from the end of the project, he is convinced that the RAM is the hardest part, so “now everything is going to be okay!”
“RAM was the hardest mainly because I have been trying to build the subcircuit for the RAM myself, which is not going to do it for SAP-2”,(Simon’s next ambition, also an 8-bit computer but with 64K memory, 2K PROM + 62K RAM). “This time the RAM I needed was particularly small, so I built a mini-RAM myself”.
You can view and launch this (unfinished) project via this link: https://circuitverse.org/users/7241/projects/35775
All of Simon’s projects on his Circuitverse page: https://circuitverse.org/users/7241
Simon’s current plan is to record a series of videos based on the Digital Computer Electronics book he uses as a guide in his engineering projects.
These are some simpler circuits from late September, simulated on Tinkercad:
Simon’s way to celebrate Helloween: a little demo about how red marker reflects red LED light and becomes invisible. A nice trick in the dark!
We also had so much fun with the blue LED lamp a couple days ago when Simon discovered that it projects perfect conic sections on the wall! Depending on the angle at which he was holding the lamp, he got a circle, an ellipse, a hyperbola and a parabola! Originally just a spheric light source we grabbed after the power went out in the bathroom, in Simon’s hands the lamp has become an inspiring science demo tool.
This is the text of the mini-lecture on Supersymmetry that CERN Research Physicist, CMS supersymmetry group convener and Deputy LHC Programme Coordinator Filip Moortgat kindly gave us during our visit to CERN and the Large Hadron Collider last week.
Filip Moortgat: Supersymmetry stands out among all the other Beyond Standard Model theories (like extra dimensions and so on). It’s particularly interesting because it answers multiple things at the same time. I would say that most other extensions of the Standard Model solve one problem but not five like supersymmetry.
The first problem: because it connects the internal property of a particle to spacetime, it actually opens a way of gravity entering the Standard Model. As you know, the main problem with the Standard Model is that gravity is not in there. So one of the major forces that we know exists is not in there. Nobody has succeeded to make gravity part of it in a way that is consistent. People hope that supersymmetry can do it, although we’re not there yet.
The second problem is called the hierarchy problem. What that means is that you have a base mass for a particle and then you have corrections to it from all the other particles. What happens is that if you don’t have any other particles beyond the Standard Model particles you get corrections that become gigantic. What you need to do is tune the base mass and these corrections so that you get the mass that we measured for the Higgs Boson or for the w and z bosons. It’s like 10^31 minus 10^31 is a 100 type of tuning, and we find it unnatural. It’s ugly mathematics. In supersymmetry, you get automatic cancelation of these big corrections: You get a big one and then you get minus the big one (the same correction but with a minus in front of it), it cancels out and it’s pretty, it’s beautiful.
The third thing is dark matter, a big problem. 85 procent of the matter in the universe is dark matter (if you also include the energy in the universe, you get different numbers). And the lightest stable supersymmetry particle is actually a perfect candidate for dark matter, in the sense that it has all the properties and if you compute how much you expect it’s exactly what you observe in the universe. It works great. It doesn’t mean that it’s true, it would work great if you could find it.
And then there’re more technical arguments that make things connect together in nicer ways than before. Normally, the electric symmetry is broken in the way that everything becomes zero. All the masses would be zero, the universe would just be floating particles that wouldn’t connect to each other, it would be very boring. But that’s not what happened. To show what actually happened you need to drive one mass squared term negative, which is kind of weird but that is what supersymmetry does automatically! Because the top quark mass is so heavy. Heavier than all the other quarks. For me it’s the most beautiful extension of the Standard Model that gives you a lot of solutions to problems in one go.
The problem is that we haven’t seen anything, yet! We have been looking for it for a long time and we have absolutely zero evidence. We now have reasons to believe that it’s not as light as we have originally thought, that it’s a little bit heavier. Which is not a problem. The LHC has a certain mass range, for supersymmetry it’s typically up to a couple of TeV. But it could be 10 TeV and then we couldn’t get there, we can only get up to 2 or 3 TeV. It could be factor 10 heavier than we think!
This why we are starting to discuss the planning of the Future Collider that will be able to go up the spectre of 10 TeV in mass, for supersymmetry and other theories. There’re several proposals, some of them are linear colliders, but my favourite one is a 100 km circular collider which will connect to the LHC, so that we have one more ring. That ring will actually go under the lake and that would be quite challenging, but in my opinion – although we don’t have any guarantee – we will then have a very good shot, at least in terms of supersymmetry. At the LHC we also have a good shot but don’t have enough reach that we need to really explore the supersymmetry.
When we use conservation of energy and momentum at the collision point, what we do is we measure everybody, we sum it all up and what we need is we need to get the initial state. If something is lacking, then we know there’s something invisible going on. It could be neutrinos, or neutralinos, or it could be something else. So we have to look at the properties and the distributions to figure out exactly what we’re seeing. It’s not a direct detection but it’s a direct derivation if you want, from not seeing something, from lacking something, that we can still say it is consistent with neutralinos.
How do you know if it’s neutrinos or neutralinos?
Neutrinos we know well by now so we know what to expect with neutrinos. Otherwise it could be neutralios but it could be something else. And then to actually prove that it’s neutralinos we have a long program of work.
And is that mainly math?
No, it’s everything. It needs all the communities to work together, because we need to measure certain properties, distributions with the detector and we will need the theoretical ideas on how to connect these measurements to the properties of the particle. So we will need both the mathematical part and the experimental part. Translating the mathematics into the particle predictions, we will need all of that.
As some of you may know, Simon is working on building a real-life 8-bit computer from scratch, guided by Ben Eater’s tutorials (it’s a huge project that may takes months). He has also been enchanted by the idea to build the computer in a simulator as well, researching all virtual environments possible. The best simulator Simon has tried so far has been Circuitverse.org, although he did stumble upon a stack overflow error once, approximately half-way through (maybe the memory wasn’t big enough for such an elaborate circuit, Simon said). You can view Simon’s projects on Circuitverse here: https://circuitverse.org/users/7241
Link to the project that ended up having a stack overflow: https://circuitverse.org/users/7241/projects/21712
And here is a link to Simon’s new and more successful attempt to put together a SAP-1 (simple as possible) processor (work in progress), something he has been reading about in his new favourite book, the Digital Computer Electronics eBook (third edition): https://circuitverse.org/users/7241/projects/22541
Simon has also tried building an 8-bit computer in Simulator.io, but it was really difficult and time consuming:
The next hopeful candidate was the Virtual Breadboard desktop app for pc. Simon downloaded it about ten times from the Microsoft store but it somehow never arrived, most probably because our Windows version was slightly outdated but who knows.
And finally, Simon has also discovered Fritzing.org, an environment for creating your own pcbs with a real-life look. He may attempt actually making a hardcopy SAP-1 via Fritzing after he’s done with the Ben Eater project. Conclusion: sticking with Circuitverse for the time being.
Simon has been mesmerised by this book for a couple of days by now, the Digital Computer Electronics eBook (third edition). He has downloaded it online and has been reading about the so called “simple as possible” processors or the sap’s (he loves the name) one of which is like the 8-bit computer he is currently trying to build from scratch.
One more blog post with impressions from our vacation at the Cote d’Azur in France. Don’t even think of bringing Simon to the beach or the swimming pool without a sketchbook to do some math or computer science!
Parts 1 and 2 in Simon’s new series showing him attempting to build an 8-bit computer from scratch, using the materials from Ben Eater’s Complete 8-bit breadboard computer kit bundle.
Simon is learning this from Ben Eater’s playlist about how to build an 8-bit computer.
It’s all Ben Eater‘s fault! Simon is more of a software and pure math champion, but Ben Eater’s videos have sparked Simon’s interest in logic and electronics, anew. Back in mid July (yes, I know, I’m a little behind with the blog), while waiting for his Complete 8-bit breadboard computer kit bundle to arrive from the US, Simon was playing with virtual circuits that he built on two wonderful platforms: Circuitverse.org and Logic.ly. You can view Simon’s page on Circuitverse at https://circuitverse.org/users/7241
Simon’s favourite was building the Master-Slave JK Flip-Flop https://circuitverse.org/simulator/edit/20037
Simon gave me a whole lecture on the differences between Sequential and Combinational Logic: in the former, there’s a presence of a feedback loop (the output actually goes back to somewhere else in the circuit), and the latter has everything going in one direction (the inputs come in and the outputs go out).
It’s a little bit like the difference between a Feed Forward neural network where the output only depends on the input and a recurrent neural network where the output also depends on what the output was previously,
Here’s a problem with sequential logic circuits: they go crazy like this very often (confused NOR gate). That’s why most sequential logic circuits have a clock in them. A clock acts like a delay so that it won’t go crazy.
That’s the power of sequential logic: you can have the same input but a different output. This is useful for storing data: I release the input, but the data is stored. It can only be archived in sequential logic.
The delay comes in error detection (on the rising edge of the square wave).
The following circuits are buit in Logicly https://logic.ly/demo