h a l f b a k e r yNot so much a thought experiment as a single neuron misfire.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
I have received a message from myself in 1978, who has
asked me to post this on here. I thought I already had but
considerable perusal of my page suggests that I haven't, so
here it is:
Consider the computer, with its tape drives, terminal, disc
packs and blinking lights, oh, and its wiring,
along which
signals move rather slower than necessary. Or, better,
consider the computer as we know it, with its 6502 CPU
and PCB covered in TTL integrated circuits and stuff.
Replace all the external wiring to the chips and integrated
circuits with fibre optics and attach semiconductor lasers
and photocells to the very last bits of the circuits which
absolutely must use electricity. This will increase the
velocity of the signals to the speed of light in the
appropriate medium. Where possible, replace a wire with
a laser beam moving through a vacuum.
There's some stuff here about one-way glass being
analogous to semiconductors but at the time I was only
eleven so that's forgivable. Having said that, is there a
way of using fluorescence to implement logic gates?
Achieve a display using the actual lasers themselves
projected onto a wall or other screen device, thereby
eliminating the need for any electronic display at all.
Now the questions: Have I in fact posted this before? I did
post something about replacing an electricity supply with
light.
Would this really make a lot of difference to processing
speed? I suspect it wouldn't, and also that many of the
components which are currently smaller than a wavelength
of red light would have to be larger. It also sounds like it
would use a lot of power. On the other hand, fibre optics
does seem to speed up the internet, so maybe.
Two-photon absorption
https://en.wikipedi...o-photon_absorption
A nonlinear optical phenomenon that could be useful for // Having said that, is there a way of using fluorescence to implement logic gates? // [notexactly, Feb 26 2017]
Saturable absorption
https://en.wikipedi...aturable_absorption
Another NLO phenomenon that could be useful for the same purpose. Maybe you could even make a saturable absorber function as a NOT gate, possibly by adding two-photon sensitivity somehow? [notexactly, Feb 26 2017]
The Machine Stops
https://en.wikipedi...i/The_Machine_Stops
Short story that predicted something like the Internet, complete with video chat. Link to full text in the external links section. At my old university I attended a production of "The Machine Starts", which was an interactive play written by the students as a prequel to this story. [notexactly, Feb 26 2017]
Memex
https://en.wikipedia.org/wiki/Memex
Hypothetical proto-hypertext system described by Vannevar Bush in 1945. An alternative imagining of what might have been instead of the Internet. [notexactly, Feb 26 2017]
Data written on paper, read by fire. Also the idea of burning-paper logic.
https://www.youtube...watch?v=eafEPGx-9rE
// I wonder why we didn't invent paper computers first? // & // It occurs to me, in fact, that even making the right pencil marks on paper ought to be able to do some data processing // [notexactly, Feb 26 2017]
We did that.
https://en.wikipedia.org/wiki/Dekatron
// I would've thought that the earliest attempts at electronic computers would either involve attempts to replicate mechanical calculators using circuits (e.g. something which stores digits in the electronic equivalent of a dial which then trips the next dial when it goes round the clock like an odometer) […] // [notexactly, Feb 26 2017]
We did that too.
http://www.ee.ryers...s/relay-counter.pdf
// […] or relays. // [notexactly, Feb 26 2017]
Technology for a better fire computer
http://www.sparkban...me-amp/flameamp.htm
I should have remembered this the other week. See next link for where I saw this again. [notexactly, Mar 11 2017]
Electrolytic diodes made with baking soda
https://hackaday.co...-7-segment-display/
A comment from [Hales] (the maker of the baking soda diodes) gave the link immediately above. [notexactly, Mar 11 2017]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Destination URL.
E.g., https://www.coffee.com/
Description (displayed with the short name and URL.)
|
| |
//I have received a message from myself in 1978// -
Does the medium you received this message through
allow two-way conversations? If so, you could make a
lot of money. |
|
| |
//if so, you could make a lot of money// |
|
| |
I've had a debate with myself about that but haven't heard
anything back yet. I'll let you know when I do. |
|
| |
You can actually go quite a bit further than just moving
signal around optically, you can actually do some
computing. There's a few ways of doing it too. You can
fire a light beam at a 45 degree half silvered mirror,
half your light passes straight through and half is
reflected. Put a detector at both end points, now you
have an AND gate. By meddling around with
fluorescence, you can for example, illuminate a red
fluorescent quantum dot with either a UV, blue, green,
or orange light and get red output from all of them.
That's a single molecule triple AND gate. If you have
something really dull to do, you can spend a whole
afternoon avoiding it by designing all of the logic gates
with single/dual molecule sizes, all operating at the
speed of light. Then you realize you've invented optical
computing. Unfortunately, that's already been
invented. |
|
| |
While the 6502 microprocessor was only 3 years old in
1978, the HalfBakery didn't exist back then. How could
you
have known of it to tell yourself to post a message to it?
Perhaps you did receive a message from the future? |
|
| |
Oh, and as for this Idea, you are not accounting for the
time it takes to convert signals from one medium to
another and back again (electric signals to photon
signals to electric signals). To achieve maximum speed,
all the circuitry
needs to be photonic, not just the interconnects. |
|
| |
In fact, [Vernon], that was very much in my mind and it's
a question I'm kind of asking by posting this. How much
difference would it make? |
|
| |
[Ian], right, okay, I think I was thinking of an Apple ][ PCB
and not noticing what the chips were but assuming they
were 74LS doobries. Not having an Apple ][ handy at the
time, I had to guess. |
|
| |
[bs0u0155] thanks, I've long wondered about it. I imagine
it might be a bit similar to fluidic logic in that there
would be certain things which are undoable. For
instance, you mention AND gates but not inverters and
that's not enough for functional completeness although I
imagine some of it is. |
|
| |
[Vernon] again: Something like the internet is
predictable from way off. Tomorrow's World and Profiles
Of The Future both mention something similar and
Asimov's Multivac has similar qualities. The internet
seems to be one of the most predictable inventions ever.
The question is though, how much detail in the internet is
predictable? Could I have imagined an ideas bank? Not
sure. I did write a lot of stuff down though. |
|
| |
Multifrequencies or a holistic component might be easier to implement in light. |
|
| |
It occurs to me, in fact, that even making the right pencil
marks on paper ought to be able to do some data
processing because graphite can to some extent conduct
electricity and you can adjust a CPU's clock speed with a
pencil mark because of that. However, I would've thought
that the earliest attempts at electronic computers would
either involve attempts to replicate mechanical
calculators
using circuits (e.g. something which stores digits in the
electronic equivalent of a dial which then trips the next
dial
when it goes round the clock like an odometer) or relays. |
|
| |
Also, I wonder about how feasible making a computer out
of silver on a glass plate photographically would be, as
far back as the nineteenth century. |
|
| |
Wow! That's true. Wouldn't it be amazing if there was a
city somewhere which turned out to be a CPU if
photographed from the air? |
|
| |
I have so many [links] to add to this! |
|
| |
Thanks, [notexactly], a lot to get my teeth into there. |
|
| |