/IonQ leader in Quantum Compute goes public via DMYI SPAC merger (via Qpute.com)
IonQ leader in Quantum Compute goes public via DMYI SPAC merger

IonQ leader in Quantum Compute goes public via DMYI SPAC merger (via Qpute.com)


IonQ, pending merger with DMY Technology Group (DMYI) is a leader in quantum computing. They are developing a general-purpose trapped ion quantum computer and software to generate, optimize, and execute quantum circuits.

IonQ leader in Quantum Compute goes public via DMYI SPAC merger

This is how you can envision a quantum circuit. What matters here and what is the number one thing quantum computing is trying to achieve, is getting a program through the circuit. The problem with QC is that those “qubits” (which comes from bits), have errors when doing computation on them.

Therefore if you give them 5 things to do (5 gates), they might end up somewhere you didn’t intend them to end up at. With every step they make, the error gets bigger. Getting this faultiness out of them is the end goal of quantum computing. Unfortunately, this is really hard.

Now let’s look closer at what a quantum computing circuit consists off: We have qubit number, 1 gate fidelity, 2 gate fidelity and connectivity. We have the number of qubits on the left. You can perform gates on those qubits. Those usually involve 1 or 2 qubits. If you want to know how “precise” a operation on 1 qubit is, you check the single gate fidelity.

If you want to know how precise a operation on 2 qubits is, you check 2 gate fidelity (they move together due to entanglement). The higher the fidelity, the more operations you can do on a qubit until the result becomes too faulty.

What isn’t visible in the picture is connectivity between qubits. If you have all to all connectivity, you can perform 2 qubit gates with no additional steps. If you don’t have all to all connectivity, you need to “move” the information of the qubits in order for them to be next to each other, which requires additional gates.

Quantum Volume — I will talk about quantum volume a lot, therefore a short explanation of this benchmark is necessary. For some time, the number one thing a quantum computing company touted as a success was the number of qubits. The number of qubits alone doesn’t really matter for a quantum algorithm, since they are faulty. This brings us back to the quantum circuit picture. What quantum volume tells you is, how many 2 qubit gate operations you can make until its becomes too noisy.

For example, you have 8 qubits that you randomly perform 2 qubit gates on. If you can do this 4 times in a circuit, you have a circuit depth of 4. Each depth means that you randomly paired up the qubits and made a 2 qubit gate. 8 times and you have 8 depth and so on (2).

In order to not let depth or qubit number become dominant, the smaller of those two is counted. E.g. if you have 10 qubits and a depth of 8, you pretend that you only have 8 qubits and vice versa. (3) If you get an 8 for example, the quantum volume is 2^8= 256.

If you get a 12 the quantum volume is 2^12 = 4069 This number you put after the “^ “ is what IonQ calls “algorithmic qubit”. You get the number by taking the log2(quantum volume) (4). In our example this would be an 8 for 2^8 and a 12 for 2^12. The number in quantum volume that IonQ will reach in the future becomes too big, therefore they made this adjustment.

The TAM of quantum computing is estimated at 65 Billion by 2030 and the gross profit margins of 90% are only rivalled by Software.

Concerning the TAM: There are two main areas of uses for quantum computers. One is simulating nature, the other is running algorithms. In nature, many processes are depended on quantum physics. Quantum physics is really hard to simulate on a normal computer. Faced with this problem, the initial idea behind quantum computers was to simulate quantum physics (1). It is pretty uncontroversial that quantum computers will create value here. Therefore the expected TAM coming for this is predictable, especially since modern supercomputers are already used for that.

What is noteworthy in this regard is that we starting to see a slow down in Moore’s Law. Therefore the reduction of costs for computation is slowly starting to decrease less, while the need for computational power is steadily increasing. Therefore the chip industry will start to see problems when it comes to meeting the ever increasing demand.

By 2023, 20% of organizations intend to budget for quantum-computing projects, compared to less than 1% in 2018 (3).

Concerning the sales strategy, IonQ is the only company that is available on all 3 major cloud providers. Once QC are sufficiently strong to be commercially interesting, the customers will come based on a cost/benefit calculation. Therefore what matters most here is creating the access.


This is a syndicated post. Read the original post at Source link .