/tech/ - Technology

>>10860
if I'm not mistaken your picture shows mosfets, transistors and thyristors.
From an electronics point of view these are all active components.

Electronics has what is called fundamental passive components those are resistor, capacitor and inductor but according to Leon Chua there is a missing fundamental component called a memristor.

I'm guessing this is what's going to be the next big thing in electronics, it can do persistent high speed random access memory for computers, but also act as a component that can be used for storing data and performing calculations at the same time. So we might get computers that will just pause operation when electricity is lost and resume operation when electricity comes back without loosing any data. This will make a lot of systems that use computers cheaper simpler and more reliable. It will also enable shrinking the circuitry needed for machine learning algorithms and it will be possible for to you have a machine learning accelerator chip in your computer that is decently capable at a low cost. Image and audio manipulation will get easier, video-games can have more clever non-player-characters, programs that need a lot of configuration will need fewer parameters to be set by the user. And if we are lucky widely available ML-chips will reach people that can figure out how to make program compilers more intelligent so we get faster and less buggy programs

>replacements
So you are thinking along the lines how transistors replaced vacuum tubes, which replaced mechanical logic-relays.

Yes there might be one more step, there is speculative physics about doing subatomic computers, sometimes this is called programmable matter or dimensional computers. But the machine for sub-molecular fabrication is probably going to be so big that it can't fit on earth.

Less futuristic is making semiconductors from carbon instead of silicon because those run at higher temperatures and clock-speeds which shrinks the "support components" for electronics like power-supplies or cooling, and more clock speed means faster computing at least as long as energy supply is not an issue.

>>10860
these are all useless questions.
Who makes them?

>>10863
These are not useless questions, it's just incredibly hard to predict technology fundamentals.
Not many people foresaw semiconductors taking off.

>>10863
op just wanted to say trans because they live rent free in his head.

>>10866
If you have an interest in electronics or computers, than yeah transistors will cross your mind. How is that relevant ?
It's not like there are people fighting over vacuum-tubes vs transistors, all the electronic components made peace, so we can have nice things.

>>10867
xd

>>10866
I wanted to make a thread about these things after I saw a watchmojo video of "Top 10 techonologies that changed the course of history" or something

The trans thing is a fun pun for the subject title, man

Don't be so jaded my dude

>>10861
Thanks for the effort post anon, I understand what you're saying and that sounds awesome.
Efficiency in technology even as an outsider is very much appreciated

These things draw so much power and get so hot, I hope it trickles down fast

> So we might get computers that will just pause operation when electricity is lost and resume operation when electricity comes back without loosing any data
WOOO, no more needing UPS for PCs

>available ML-chips will reach people
poggers
I want my next PC to have the chip fueled by the immortal science of marxist leninism

The only thing that worries me is the image manipulation getting easier, It's already almost here and I don't know how the powers that be are going to fumble this shit

Reality feels like it will be a subjective thing in the future

>>10871
>Machine learning will make reality subjective because of video manipulation
I doubt this logic, we already can make images resistant to manipulation. Camera sensors embed noise into the picture that can be used to create a cryptographic signature that breaks any attempt at making undetectable modifications. So if you just go by technical potential fake pictures should no longer fool anybody.

Machine learning will however change how videos are a encoded, because they can recognize objects, it will now be possible to have a library of things that are in the video and a instruction table that describes how the objects move through the video-frame. That means much better compression. A video with a talking head in front of a static background will be a few megs even if it's hours long. Videos will also encode a considerable amount of depth and that will again make it harder to produce convincing fakes.

If machine learning chips become cheaper it will also make it much easier to cross reference scientific data, and screwing with the reality perception of scientists will get much harder. So reality will become much more objective because the ability to detect inconsistencies will increase dramatically.

The enlightenment hasn't really caught up with the digital age, it takes longer to learn how to use new tools for creating accurate models of reality than to make up imaginary stuff. Consider how long it took for the scientific revolution to get going after all the tools it needed were already in existence. Consider how hard it hit when it got going, how much the bullshitters had to give, and then project this into the future when all the new technical tools have filtered down into scientific praxis.

>>10872
Huh didn't think of it that way
I was thinking more in terms of misinformation being more high-quality nowadays but still it's good to know that the science guys got ways to find what has been modified with these signatures and shit

now the future burden falls on the regular person on waiting and finding what has been debunked then, harder to train everyone to do that but still possible and not as bleak as i had thought

>>10865
>These are not useless questions
They are, I want to hear about transistors from a marxist pov.

>>10875
Tell us about transistors from a marxist pov

>>10876
I'm asking you. I don't know shit about them.

>>10878
Well, start!

>>10875
The basic signficiance of the transistor is that it provides an efficient way of using analog signals to do digital logic. I would think that the replacement for the transistor would basically be a more efficient digital logic component. Now I will speak out of my ass: Cockshott has remarked that physical matter itself is discrete, which is to say that "continuous" electro-magnetic signals are actually quantized. So I imagine that quantum computing makes use of these discrete signals to do its logic operations, but I actually have no idea. A quick search brings up some articles on quantum transistors.

>>10893
I asked for marxism, not for cockshott

>>10893
>hysical matter itself is discrete, which is to say that "continuous" electro-magnetic signals are actually quantized. So I imagine that quantum computing makes use of these discrete signals to do its logic operations
very Interesting thesis: Towards a theory of quantum computers that we might actually understand.

>>10894
nobody asked for sectarianism, so …

>>10894
I'm not sure what a marxist understanding of a transistor would be besides materialist