←back to thread

Replication of Quantum Factorisation Records with an 8-bit Home Computer [pdf]

(eprint.iacr.org)
112 points sebgan | 9 comments | 12 Jul 25 02:05 UTC | HN request time: 1.058s | source | bottom
Show context
wasabi991011 ◴[12 Jul 25 05:32 UTC] No.44539574[source]▶
Yeah there's a reason that the quantum computing field has moved away from attempting factorisations. Not that there's not still hype and misleading claims being punished, but the hardware has improved a ton since 2001 and ever closer to actual useful quantum computation (such as large size quantum chemistry calculations).
replies(1): >>44541356 #
1. thrance ◴[12 Jul 25 11:44 UTC] No.44541356[source]▶
Are those useful computations in the room with us right now? No, but seriously, I feel like factorization is the one application that could justify those massive investments QC is receiving (even though it would probably make the world strictly worse...).

All those other applications, no matter how neat, I feel are quite niche. Like, "simulate pairs of electrons in the Ising model". Cool. Is that a multi-billion dollars industry though?

replies(3): >>44541671 #>>44542131 #>>44542282 #
2. rgbforge ◴[12 Jul 25 12:45 UTC] No.44541671[source]▶
If results from methods with higher electronic structure accuracy than DFT (MP2, couple cluster) can be made cheap enough, it would hugely disrupt industrial chemistry, medical experimentation, pharmaceuticals, the energy sector, etc.
3. wasabi991011 ◴[12 Jul 25 14:00 UTC] No.44542131[source]▶
Ground state and activation energy estimation for chemistry would be really useful. I know chemists are looking specifically at nitrogen fixation as one useful example.

Or as another example, I'm currently at a conference listening to a PhD student's research on biomolecular structure prediction (for protein design).

replies(2): >>44542236 #>>44543377 #
4. DoctorOetker ◴[12 Jul 25 14:15 UTC] No.44542236[source]▶
Energy levels and activation energies can be acquired much more simply from Fourier Transform - Ion Cyclotron Resonance - Mass Spectroscopy...

Its a device that makes and analyzes at the same time, check out this primer:

https://warwick.ac.uk/fac/sci/chemistry/research/oconnor/oco...

replies(1): >>44543505 #
5. upofadown ◴[12 Jul 25 14:21 UTC] No.44542282[source]▶
Factorization could have number theory implications I suppose. Using quantum effects to break cryptography wouldn't have any real long term advantages unless you aspired to be some sort of a supervillain.
replies(2): >>44542328 #>>44543378 #
6. LeftHandPath ◴[12 Jul 25 14:29 UTC] No.44542328[source]▶
> Using quantum effects to break cryptography wouldn't have any real long term advantages unless you aspired to be some sort of a supervillain.

It's of interest to governments, for national security reasons. Quantum computing is an arms race.

7. dreamcompiler ◴[12 Jul 25 17:07 UTC] No.44543377[source]▶
One of the few genuinely useful accomplishments of modern "AI" has been protein structure prediction. I wonder if we still even need QC for this.
8. pclmulqdq ◴[12 Jul 25 17:07 UTC] No.44543378[source]▶
If you want O($10 billion per year) of funding, those numbers can only come from having $10 billion a year of impact balanced against your chance of success. The only application of QC worth $100+ billion is breaking cryptography.

PQC is as much a tool to reduce funding for QC as it is a tool against an actual eventual quantum computer.

9. wasabi991011 ◴[12 Jul 25 17:26 UTC] No.44543505{3}[source]▶
Cool stuff and thanks for the link, I'll have to learn about it when I have a bit more time.

I've always heard Qalgs for chemistry compared to classical methods though. Why do you think chemists are using CCSD and similar methods rather than the FT-ICR mass spectroscopy?