What Quantum Computers REALLY Do



Quantum computers are crazy and have the potential to change how we understand the world around us. I got to go see one with Marques Brownlee recently at IBM. It was wild! But quantum computers are not just better computers. Here’s the analogy that helped me…

If you like fun tech videos, you’ll love the full adventure to see a quantum computer with MKBHD!

And if you want to support optimistic tech content, the best thing you can do is subscribe 🙂

#shorts #technology #quantum #stem

source

45 Comments

  1. Please check my comment on quantum computer
    Yes. Below is a merged conceptual design that combines:

    Quantum Harmonic Oscillator (QHO)

    Krotov optimal control (simplified framework)

    A transistor-inspired control layer (modeled as a switching element inspired by early Fairchild semiconductor logic, not a physical quantum transistor)

    A single Python program suitable for running in a standard Python interpreter or PyCrust with NumPy.

    import numpy as np

    # =====================================================
    # Quantum Harmonic Oscillator + Krotov Control
    # + Semiconductor Memory Inspired Control Layer
    # =====================================================

    HBAR = 1.0
    MASS = 1.0
    OMEGA = 1.0

    GRID = 128
    XMAX = 8.0

    x = np.linspace(-XMAX, XMAX, GRID)
    dx = x[1] – x[0]

    # Second derivative
    D2 = np.zeros((GRID, GRID))

    for i in range(1, GRID-1):
    D2[i,i] = -2
    D2[i,i-1] = 1
    D2[i,i+1] = 1

    D2 /= dx**2

    # Hamiltonian
    T = -(HBAR**2/(2*MASS))*D2
    V = np.diag(0.5*MASS*OMEGA**2*x**2)

    H = T + V

    energy, state = np.linalg.eigh(H)

    print("Ground Energy =", energy[0])

    psi = state[:,0]

    # —————————————————–
    # Semiconductor Memory Cell
    # —————————————————–

    class MemoryCell:

    def __init__(self):
    self.bit = 0

    def write(self,value):
    self.bit = value

    def read(self):
    return self.bit

    memory = MemoryCell()

    # —————————————————–
    # Transistor Controller
    # —————————————————–

    class FairchildController:

    def __init__(self):
    self.gain = 1.0

    def pulse(self,time):

    if memory.read()==1:
    return self.gain*np.sin(2*np.pi*time)

    return 0.0

    controller = FairchildController()

    memory.write(1)

    # —————————————————–
    # Simplified Krotov Update
    # —————————————————–

    def krotov_update(control,target):

    alpha = 0.05

    error = target-control

    control = control + alpha*error

    return control

    pulse = np.zeros(200)

    target = np.cos(np.linspace(0,np.pi,200))

    for iteration in range(100):

    pulse = krotov_update(pulse,target)

    # —————————————————–
    # Time Evolution
    # —————————————————–

    dt = 0.01

    history = []

    for step in range(500):

    t = step*dt

    field = controller.pulse(t)

    Hc = H + np.diag(field*x)

    U = np.eye(GRID) – 1j*Hc*dt

    psi = U @ psi

    psi /= np.linalg.norm(psi)

    history.append(np.real(np.vdot(psi,H@psi)))

    print("Simulation Complete")

    print("Final Energy =", history[-1])

    print("Stored Memory =", memory.read())

    This merged design contains four layers:

    1. Quantum layer: Builds and diagonalizes the quantum harmonic oscillator Hamiltonian.

    2. Control layer: Uses a simplified Krotov-style iterative optimization to improve a control pulse.

    3. Semiconductor-inspired layer: Models a memory bit and transistor-like gate that enables or disables the control pulse. This is an engineering abstraction inspired by early Fairchild memory logic, not a model of an actual quantum transistor.

    4. Simulation layer: Evolves the quantum state under the controlled Hamiltonian and reports the final energy.

    For a research-grade implementation, you would replace the simplified optimizer with the full Krotov algorithm (using adjoint-state propagation and forward/backward evolution), use unitary propagators such as scipy.linalg.expm, and optionally integrate QuTiP or the dedicated Krotov Python package for accurate quantum-control simulations.

  2. Traveling down a street-will be of no use with a boat. So unless you live on water, I suspect (based on this)-they will be way cool, and way limited.

  3. I think we need to be able to describe these things more accurately. I get what you’re trying to do but classical computers can 100% solve any problem a quantum computer can, just for very specific types of problem, it could take longer than the universe has existed to compute. It’s not really different terrain, it’s a subset of very specific problems in math – a lot of which we’ve made absolutely no progress on the software for.

  4. During your tour at IBM did you asked how many bits it takes to teleport someone without taking a plane nor train? I recently saw a video about Japan using quantum computers to teleport people. They're close, but more research need to be done.

  5. We will soon be in the last 7 yrs. the great tribulation.the mark of the beast. Satan will contol the world…..there will be no place to escape….your only hope is to repent of sin. Put your trust in Jesus as Lord God and Savior. For the forgiveness of sin…..and the free gift of eternal life in heaven… all Ai will do is usher in the satanic control grid .see revelation ch 13

  6. This isn’t a good example/explanation of quantum computing. Look at GI Quantum Computing. To me, it reminds me of an atom, molecules, instead of being limited to binary now. Binary uses 0 1. Quantum uses an infinite amount all at once.

  7. That is not a Quantum computer. They are pipes and plates. You got scammed and they are laughing at you. The actual chips are floating in space. 😂
    Imagine revisiting this video 10 years from now and listen to yourself.

Leave a Reply

Your email address will not be published. Required fields are marked *

You might like

© 2026 Cantinho do Vídeo - WordPress Video Theme by WPEnjoy