A tiny universe in a bottle reveals clues to the origin of life

Asteroid and meteorites get quite hot (or at least their outsides do) when they go through the atmosphere on their way down to earth. There's not a lot of organic chemistry that would survive that.

The sun now emits quite a lot of hard ultra violet light and a certain amount higher energy radiation which doesn't make it through the earth's atmopshere.

The earth seems to have had a molten rock surface early in its career, and that wouldn't have been all that receptive either.

It is an interesting area of study, but getting the results down to earth may be tricky.

The mystery of life isn't ingredients, it's structure.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Jan Panteltje snipped-for-privacy@comet.invalid spake the secret code <113irfo$fa0j$ snipped-for-privacy@dont-email.me thusly:

And this has what, exactly, to do with electronics design?

Not "science." Not "interesting." Not "contains a glass tube." Not "somebody somewhere used an instrument that had electrons in it." What does copy/pasting a university press release about a hyped-up origin-of-life experiment from a PhD student have to do with designing electronics?

This is sci.electronics.design, not sci.random-press-release-pastebin. If there is an electronics angle, state it. What circuit? What instrumentation? What detector? What vacuum control? What measurement problem? What design tradeoff? What failure mode? What schematic question? Anything?

Last week it was Chinese rocket cheerleading with a pointless little "bye bye Elon" garnish, as if sci.electronics.design needed another geopolitical fortune cookie from the Ministry of Copy/Paste. Now it is cosmic dust in a bottle, because apparently anything with a ScienceDaily headline and enough shiny nouns qualifies as electronics design after it has passed through your newsreader.

You are not starting a technical discussion. You are flyposting headlines. That is not contribution, it is RSS with a pulse.

Try this next time: before pasting the press release, identify the actual electronics design issue you want discussed. If you cannot find one, that is the clue that the article belongs somewhere else, preferably in the same drawer as the rocket propaganda and the Elon reflex.

What are you designing now?

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Could you with your wisdom elaborate a bit on that?

But the ingredients have structure. DNA was labelled the golden helix because it has a spiral structure.

Surprisingly they do survive Earth entry. The outer fused heat damaged skin is generally not that thick. They are also not very hot when they reach the ground but may disintegrate due to air pressure when they reach the thicker atmosphere generating a shower of pebbles.

One such meteorite shower hit Paris just after their Academy of Sciences had declared in a report that "nothing" ever fell from the sky!

Obviously stony and pure iron ones usually stay together in big lumps.

Carbonaceous chondrites like the one that fell in the UK during Covid end up with an outer hard crust that is well toasty but the innermost parts are virtually pristine provided that nobody touches them. This wasn't the one I recall but it is in the news now and will do as an example:

It is still thought that during the early phases of the solar system and after the Earth had a mostly solid surface that the seas we now have were at least in part a result of cometary bombardment (dirty snowballs) and steam/CO2/SO2/N2 escaping from the core by vulcanism.

The balance between the two mechanisms is not clear. Both have certainly played a part. Extreme vulcanism can significantly affect climate, and has done in recorded history 1816 Tambora and 1883 Krakatoa.

john larkin snipped-for-privacy@glen--canyon.com spake the secret code snipped-for-privacy@4ax.com thusly:

I've got several projects in the works, some of which are "slow burn" and some are moving faster:

- LED blinky light employee name badge for a local Mom & Pop burger chain

- 6805 SBC project to play around with making an 80s 8-bit SBC from some parts that were donated:

- Tektronix 401x storage tube terminal "printer". The printer for this thing was really not what you'd expect. The printer essentially puts the terminal into a slave mode and drives the tube's electron beam with scanning ramps generated by the printer. The terminal would report back a voltage ("target signal") representing the charge stored on the tube at the position of the scanning ramps. The printer would take this analog signal and use it to print on silver halide paper. While the printer units can be found, the paper is unobtainium at this point and even if you did find some, it would be 40+ years old at this point and likely the active chemicals are no longer very responsive. So the idea was to make a circuit that pretends it's the printer: start the handshake with the terminal to put it in slave mode, generate the scan ramps, digitize the return target signal and generate a grayscale image. Many vintage compy folks still have these terminals and the idea of getting a "hardcopy" of the screen image would be nifty.

Other projects, too, but they're mostly of the repair, refurbish, or reverse-engineer variety for vintage computing gear.

You can't shake up a bucket of parts and get a radar. Or digitize thermal noise into bytes and get a word processor program.

The reproduction of a DNA-based cell requires the incredible complexity of a DNA-based cell.

That didn't evolve by incremental trial and error from a mud puddle with a few organic compounds mixed in. Nothing works until it all works.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

I've some multicolor "light fixtures" for which I have been developing an animation package so I can create "effects" without having to manually determine when each lamp illuminates and with which color. I.e., the

A 64180 would be far more capable -- and a '7180 far more USEFUL! (a single chip solution that can actually *do* something nontrivial)

If you enjoy self-abuse, a 99115 would be more exotic. Or, 43201/02,

If you really want to LEARN something, throw together a 2900 chipset and design your own instruction set! Application-specific processors were all the rage "way back then" (hunt for a copy of "Mick & Brick"). The design effort is more "what is of value to you in a CPU" than "how do I wire these things up".

If it is anything like the '010, you may regret the effort. (TI never really seemed to get the hang of "microprocessors"). The 9918 might be an interesting (though sorely constrained in its abilities) as a "display controller" -- with some other CPU as host.

Motogorilla's RGP would have been an interesting device but I don't think they ever released production silicon (and I have no one to "return" the NDA-released materials to, after all these years -- but "Do Not Destroy!" <frown>)

I refer you to New Scientist, 11th July 2026, page 11.

" Traces of earth's primordial magna ocean spotted in lava".

The text makes it clear that it wasn't actually primordial, but rather a transient consequence of the early collision with a Mars-sized object that left us with our remarkably large moon.

So the surface probably did get sterilised at least once.

Getting hit by a Mars-sized object produces effects that rather overshadow "extreme vulcanism".

On 7/20/26 18:15, john larkin wrote: [...]

We'd guesses as much, but there it is stated clearly at last.

Where did the creator come from?

Jeroen Belleman

Don Y snipped-for-privacy@foo.invalid spake the secret code <113lhgk$1agv7$ snipped-for-privacy@dont-email.me thusly:

In my case, I'm trying to keep the parts BOM as small as possible and the design as simple as possible to keep the price per badge down.

As I stated, the point is to use the parts I have, not make the "best" SBC one could make.

Already did that in college.

Many arcade machines and high end PC graphics cards used this chip. Being familiar with this chip got me a job offer with Evans & Sutherland in the 90s.

Maybe, but again, I'm not asking for different alternatives to what I'm already working with. I was asked what I'm designing and I posted a list.

Everything I've listed has value to me, which is all that matters.

Some other people in the vintage compy community would likely have interest in the 401x dongle. A couple other people have already expressed interest in the 6805 SBC.

One theory (mine) is that life did evolve incrementally somewhere in the universe billions of years ago, got to be very smart, and designed us carbon-based life as a practical joke or a high school science project. [1]

What I can't understand is that if the little green men are smart enough to have interstellar flight, why do their flying saucers crash so often?

A 64180 wouldn't be the "best"; it would just be considerably more "useful" (i.e., applicable to real-world problems). The '7180 lets you create real products in a single chip (of complexity comparable to a 6805-based design).

But, did you just create a same-ol', same-ol' instruction set? Or, did you explore different approaches to putting "utility" into the set of operations supported? E.g., I was heavily into graphics so had opcodes to draw line, a blitter, etc.

Yes, I used the '010 in the design of a video slot machine many decades ago.

Fair 'nuf. To me, value comes from creating something that doesn't already exist. Otherwise, you're just spinning your wheels making something that you could likely *buy*.

E.g., my current design emulates (in software) many *hardware* features of (really) old CPUs -- like the 5500. And, given how much faster devices are, today, they can almost approximate the equivalent performance (without having to resort to VHDL to reify the design)!

Whoop-de-sodding-doo to you!

Jan has every right to post whatever he wants whenever he wants.

Don Y snipped-for-privacy@foo.invalid spake the secret code <113lsl5$1ed5q$ snipped-for-privacy@dont-email.me thusly:

At the risk of repeating myself once more, the point is to make something with the pile of parts I was gifted and learn something about SBC design along the way. The 8K address space makes for some interesting ROM challenges in order to make a system that is not built for a specific microcontroller problem. The funky gaps in the memory map into which you can add your own peripherals made for some fun address decoder practice. See the repository for deets.

It was 40 years ago, TBH I don't remember the detailts, but it was a generic CPU. I'm pretty sure it was straight from Mick & Brick and didn't deviate from one of the designs presented there. Honestly, the UDel EE department at the time was pretty crappy at teaching any useful digital design beyond a simple state machine for a traffic light controller. When I went to UofU for CS as a grad student I discovered that undergrad CS students did more digital design in a one year course than I did during 4 years at UDel. Kurt Akeley was like one or two years ahead of me, but I never interacted with him at that time. I learned more about practical digital design by assisting John Kelly on his PhD graphics processor. I wire wrapped something like

The designer of the chip recently gave a talk on it:

There are two types of value I am getting out of these projects:

- actually building *something*

- having fun building them

I was subscribing to the AdaFruit AdaBoxes for a while but they aren't very interesting because you don't *build* anything. You connect prefabricated modules with cables and load firmware. The HackerBoxes still actually require you to solder stuff, even if you're not designing boards/circuits from scratch. Sure, they use a bunch of modules with microcontrollers and you load firmware into them, but they are less oriented as a complete thing than as a platform into which you can experiment. Plus HackerBoxes has managed to deliver a monthly project on time for multiple years, whereas AdaBox flaked out and took like a 6 yr break between the last one I got for my "subscription" and the next one that came out.

Pi Pico projects are fun that way and I've done a little fiddling around with them, but nothing serious yet. The 401x printer dongle thingy will like be a Pico with some analog ramp circuitry attached.

Suit yourself. If I'm going to invest time in a project, I want something that I can *use* from it. Otherwise, spend the time reading a book...

E.g., I am currently hoping the local maker house gets a waterjet or plasma CNC so I can use *it* to cut the copper sheeting I will be using in my "light fixtures". I have no desire to purchase one and paying someone else doesn't benefit anyone -- as it would the maker house (it would also likely drum up some interest due to its scale)

The advantage of tailoring a processor to a specific application comes from the performance benefits that accrue. E.g., having to write subroutines/functions to provide the same sort of capability makes it abundantly clear how inefficient the COTS solutions are for these tasks. That's why we have graphics accelerators, crypto coprocessors, etc. -- it's not that you *can't* perform those actions on a generic processor but that they are horribly inefficient when done thusly.

E.g., having a blitter in hardware means the processor doesn't waste its time converting linear addresses to rectangular (as well as all of the opcode fetches involved in doing so -- just move the data directly!)

I met Karl in the early 80's when he was pitching the 99K to us. He had a very different idea of how the technology should be applied (and, given that we were making lots of *product*, I had more faith in our understanding of the problem than his).

Exactly. Slap together things that someone else designed (and MADE!) and pat yourself on the back for your accomplishment.

There is (was) a TV commercial where two kids are spying on their mom in the kitchen: "She's BAKING!!!!!", excitedly.

No, she's taking a refrigerated, oversized cookie and HEATING IT UP.

It is not easy to come up with ideas for projects that others would LIKE to make -- let alone document them in enough detail for them to do so! My arrangement with my colleagues is that they can "steal" whatever I've done but don't expect me to redirect my attentions to make your life much easier (if you want a finished product, wait until it is FINISHED!)