Why is nobody talking about optical computing and its potential? Is the technology being suppressed?>environmentally friendly>far more computing power than traditional computing>needs less energy>doesn't need as much cooling>cheaper in resources
>>536244054because this shit is a pipedream and the jew vision of computers is that you rent your desktop space in a datacenter.also you wont be able to afford these things anyway.
>>536244054because it's even less real than quantum computing
>computingThe proper term is "compute"
>>536244054That "light speed" part is probably false advertising here but I was under the impression that such an approach might experience less heating issues, which could be quite an advantage ...
>>536245226Would be a great advantage. Apparently the issue replacing the classical transistor and data storage, but then they also argue the possibility of some kind of hybrid approach. Also using stuff like accoustics maybe.
>>536244054>muh optics
>>536245226computing already happens at relativistic speeds because electrical signals in chips travel at like 0.5c (though I guess maybe capacitance slows it down?)
>>536244054because it's still far away from reaching the market
>>536245713Is it? Wasn't there some progress with some kind of chip?
>>536244054One simple question breaks whole thing apart. Make me a light transistor. Basically you have a glass fiber that lgiht passes through. You need to somehow be able to control with another light the light conductivity of this fiber
>>536245460kek
>>536245954tbf i don't know but I doubt people will be jumping to go all in on it, especially with how tense people are about not missing the AI boat. Hell traditional ASICs are getting neglected. Also the optical chips need to find a niche or somehow compete with newer chips to even attract customersIf you could make lower latency, higher bandwidth memory with optical chips then that would be great. I feel optical interconnects and networking is more promising than optical computing
>>536244642NNNNNNNNNNNGGHHH I'M COMPOOOOOOOTIIIINNG
itd be possible if you can yknow control gravity. otherwise best case scenario is blasing light into bubble ram. let me know how it goes.
>>536246393I dunno, couldn't you etch pathways into two plates of silica glass and just sandwich them while painting the outside black?
>>536244054https://www.youtube.com/watch?v=4Jfhq0rLE_Yrelated
>>536245433Would inevitably be a hybrid, somewhere you will translate your signal into an electrical one.>>536245686Well, what really gives you a speed limit here is from my understanding still the damn heat issue. If the atoms in your chip microstructure start wobbling too hard ...
threads like this are why i come to /pol/nice job OP
>>536244054because so far the breakthrough tech is for data center only not for consumers so no can use it. Its mostly used for communication links not for processing which is its real strength. Magnonics is in a less advanced stage but offers faster way to accelerators.The whole quantum computing field by big tech is a scam. Only the nips with the qfp tech are the real players.
why do people need increased rate of communication? to gain an advantage in war?we already do everything quickly enoughit isn't like people are curing cancer kekeveryone is in a rush to gain convenience or an advantage over others or to become even more lazy and unskilled than they already are, which is a recipe for failure for a societypic related is a test for an EIGHTH GRADER in ~1880s, which 90% of adults globally couldn't pass todayunfortunately, tech is only making things less secure, more complicated and more expensive kek
Photonic
>>536250663>~1890s*
>>536245982its not fiber its waveguides made up of silicon or some other semiconductor material. Their issue is minimizing the laser sources they are too big for ICs there was one thats gotten closer but its hasnt been integrated yet.
>>536250663>unfortunately, tech is only making things less secure, less private, more complicated, less sustainable, less correct,* and more expensive kek
>>536247710the way we reduce power consumption is by reducing the clock frequency, not the speed of the signals. We can't adjust the speed of the signals.
>>536244132>because this shit is a pipedreamIts not. There are literally light based chips running right now in germany as prototypes. You should be proud you kraut autist.As i understand the computation is light based and EXTREMELY fast and effective, but the bottleneck is storage/memory/cache. Basically storing any kinda computation result as light is is not a thing atm and maybe never(mybe crystals?)But i really like where this is going and the potential upside is huge - lightspeed latency, no waste heat, insane perfomance - cool shit.
>>536244054any us traded companies you recommend
>>536244054It is being researched and developed but it cannot work on its own.You need classic electronics to somehow deliver power to diodes that emit light.So it would have to be a hybrid contraption just to barely be viable.The potential is mostly in getting higher clock speeds and not being concerned with the limitation of transistor size and density.But I think their main problem is how to make these signals travel and get received in a reliable way, due to light dispersion factors. The signal would have to be very strong and travel a short distance to have a more reliable reading.
>>536251789>crystalsEvery day Stargate feels more and more like soft disclosure
