It's 2024, why are computers still binary? Wouldn't it make more sense to add more states so they run much faster?
"Ternary computers" do exist. And they provide no outstanding benefit. Just look up the term on google.Quaternary computers would be silly. It's just binary with an extra step.Quantum computers do have some usecase. But they will be "in developpement" until the DoD is done weaponizing them.
adding more voltage levels tends to make things slower, since then they have to be more accuratesee ssd's for example, SLC flash has two states, they are both way faster and longer lasting than multi-state flash, the more states you make (TLC = 3 state, QLC = 4 state, etc), the slower and less reliable they becometrinary (3 state) computer have existed in the past, but it's needless complexity when you can make a simpler binary computer go faster
>>100142269>And they provide no outstanding benefit.that's what (((they))) want you to think
>>100142308correction: TLC is 3 bit, and QLC is 4 bit, not state, 3 bits has 8 states, and 4 bits has 16 statesthere's also 2 bit MLC with 4 states
>>100142331meant for>>100142299
>>100142299Quantum computers just have no use case yet. They produce statistical garbage that a binary computer has to analyze to get a "confidence level" on the quantum computers output. Sounds like we are far away from fully understanding quantum mechanics and making any use of the tech.
>>100142358Y-Yeah, totally!The DoD approve of your opinion :thumbup:
>>100142371I'm over the "government is secretly super powerful" meme. They are bogged down military bureaucracy and do worse than the private sector. Even their contractors are shit.
>>100000000Too expensive
>>100142269>so they run much faster?Computers already run as fast as they need (and possibly even as the can).Any further improvement benefits no one other than governments and big corpos.
>>100142269>Wouldn't it make more sense to add more states so they run much faster?What a retarded question. Why would it? Do you have anything to support this ridiculous implied claim?
>>100142300>implying midwits with their "serious" discussions aren't the main problem with the board
>>100142269Why is /g/ so known for retards asking midwit questions?
>>100144507this is really fires up the ol' factory
>>100142269no because the complexity of the design is so much greater that that pathways end up eating up space faster than they displace part of the reason for this is that to make a binary computer all you need is a way to make the current turn on or off, no matter how much power is actually flowing throughthis can be done with as little at 1 transistor, or 2 mosfets in the cast of CMOS, with 4 mosfets being the most commonyou can create a cmos ALU bit that does all logical operations (not shifting), addition, and subtraction using only 80 mosfets, compare that to the billions of transistors on the chip and it's fucking nothingmeanwhile how would you have a third state? negative voltage? CMOS already uses negative voltage; you HAVE to connect the input from positive to negative rail to change the state, otherwise it just stays as is. Voltage levels? well now every single logic gate will need some way to detect the difference in voltage. even the simplest designs would increase the transistor count to the multiple hundreds, perhaps thousands to perform the same function for a single trit (trinary digit)yes it's possible, it even exists, but it's stupid and has limited uses, it's like the retards who think everyone is gonna get a quantum computer, those things have severely limited use that even if you could overcome the physical problems with them you wouldn't really want to anyway cause it's pointless and stupid
>>100142338But thats still fundamentally binary, and thats the issue. True ternary computers don't exist, and the closest weve come to that are quantum computers with the superposition of states being the third.In theory a ternary computer would make things a whole lot faster as it's've been shown so far, but two a make a true ternary system you'd need to have a way to express in the simplest way on hardware the signals 1, 0 and 2. All of them being clearly different to each other
>>100145793i bring up flash cells because they work similar to something like ternary values, in that they convey more than 1 bit of information with a single voltage/charge stateyes, logically they are split into multiple bits, but that's only a binary interpretation
