IBM Says Conventional AMD Chips Can Run Quantum Computing Error Correction Algorithm (reuters.com) 23
IBM announced that its quantum error-correction algorithm can now run in real time on standard AMD field-programmable gate array (FPGA) chips -- a major step toward making quantum computing more practical and affordable. Reuters reports: In June, IBM said it had developed an algorithm to run alongside quantum chips that can address such errors. In a research paper seen by Reuters to be published on Monday, IBM will show it can run those algorithms in real time on a type of chip called a field programmable gate array manufactured by AMD.
Jay Gambetta, director of IBM research, said the work showed that IBM's algorithm not only works in the real world, but can operate on a readily available AMD chip that is not "ridiculously expensive." "Implementing it, and showing that the implementation is actually 10 times faster than what is needed, is a big deal," Gambetta said in an interview. IBM has a multi-year plan to build a quantum computer called Starling by 2029. Gambetta said the algorithm work disclosed Friday was completed a year ahead of schedule.
Jay Gambetta, director of IBM research, said the work showed that IBM's algorithm not only works in the real world, but can operate on a readily available AMD chip that is not "ridiculously expensive." "Implementing it, and showing that the implementation is actually 10 times faster than what is needed, is a big deal," Gambetta said in an interview. IBM has a multi-year plan to build a quantum computer called Starling by 2029. Gambetta said the algorithm work disclosed Friday was completed a year ahead of schedule.
How nice for AMD (Score:4, Interesting)
That Xilinx acquisition keeps paying off for them. It's surprising Intel didn't get more mileage out of Altera.
Re: (Score:3)
We'll not know until Monday, of course.
Tell me you're not technical without telling me (Score:4, Funny)
In this glorious age of AI, the punchlines can write themselves.
Re: Tell me you're not technical without telling m (Score:3)
What is this nonsense? (Score:4, Informative)
You need to run quantum error correction on the qbits, or it does nothing. Well, I guess they have stopped pretending QCs are ever going to be useful...
Re: (Score:2)
I expect to see a lot of implementations of CNOTs in silicon represented as quantum in the next few years, but not big enough that anyone could actually use them in anger and discover they don't quite work as advertised.
They have to salvage their investments somehow, and pivoting the branding while betting that nobody knows any better seems to be tactically what they do by default these days.
Re: (Score:2)
Indeed. That is basically the only thing that makes sense for this move. I guess they bet on nobody understands that faking it is not a step to making it here.
Re:What is this nonsense? (Score:4, Informative)
Apparently not. From https://thequantuminsider.com/2025/10/24/forthcoming-ibm-paper-expected-to-show-quantum-algorithm-running-on-inexpensive-amd-chips/ [thequantuminsider.com],
But qubits are fragile and they can lose coherence or produce calculation errors from the slightest environmental interference, making error correction one of the biggest challenges in building reliable quantum systems. IBM’s algorithm, first announced in June, is designed to detect and correct these errors dynamically while the quantum chip is running, allowing for continuous and stable operation.
According to IBM, the experiment shows that its algorithm can run effectively on AMD’s reprogrammable chips — hardware that is common in data centers and embedded systems — rather than on specialized quantum control units.
and
The research suggests a path to building hybrid quantum-classical systems that can operate with lower costs and faster development cycles.
It sounds like the error correction algorithm takes in data about the qubits and suggests some way to manipulate them to correct errors. So qubits still need to be monitored and manipulated, but the procedure for doing so can be computed on conventional hardware.
Re: (Score:2)
Re: (Score:2)
Naa, you just do not know how a QC is supposed to work and hence you do not see the elephant in the room. Well done for being proud of that!
Re: (Score:2)
Yes, theoretically, that is possible. But it requires measurements on the qbits and it requires manipulation on the qbits, both with high chances of breaking the entanglement. Hence this does not make any sense at all. The very point of quantum-error correction is that you do not need to touch the qbits outside of quantum operations to minimize that collapse risk while reducing noise.
Re: (Score:2)
They're publishing the actual paper on Monday; stay tuned to see what it says.
Re: (Score:2)
I have heard the empty promises of QCs for 35 years now and it was clear back then that this will likely never amount to anything. I don't care what they publish. Factor a 256 bit number on a QC and I may care again. Before that it is all just a scam. Oh, incidentally, the record (!) for actual, real, non-faced QC factorization is 35. Not 35 bits, 35. And it was done with a circuit specifically crafted to factor 35.
Re: (Score:2)
Those who say something cannot be done should not interrupt those who are doing it. If research into quantum computing continues, I have no doubt they will eventually figure out how to make a useful quantum computer. This stuff is really hard; it's not surprising that it takes a lot of time, funding, and expertise. Hopefully they create neat spinoff technologies along the way.
I have heard the empty promises of QCs for 35 years now
The amount of hype in news articles about scientific discoveries is unfortunate but not particularly surprising. Given your user
Re: (Score:2)
I have heard the empty promises of QCs for 35 years now
The amount of hype in news articles about scientific discoveries is unfortunate but not particularly surprising. Given your user ID, I would have expected you to be aware of this. Per https://en.wikipedia.org/wiki/Timeline_of_quantum_computing_and_communication [wikipedia.org], the first quantum logic gate was built in 1995 (30 years ago), the first quantum computer—a whopping 2 qubits—in 1998 (27 years ago). If anyone was talking about what quantum computers could do 35 years ago (1990), it was pure future-looking theoretical possibility and nothing more.
And fail. I am not talking about press reporting. I am talking about scientific publications. And you are quite wrong about the state of things back then. It was in no way theoretical only, people were intently researching QBits and entanglement for them.
As to your prediction, see any computers running Josephson circuits? Or one of the many other failed approaches? Because I do not.
Re: (Score:2)
Ok, sure. People hype the potential usefulness of their research too. It seems like you're expecting every research direction of a complex new technology like quantum computers to quickly pan out. That's unrealistic. I also don't get why this bothers you so much. Superconducting quantum computers [wikipedia.org] use Josephson junctions, and a few of those exist. But it doesn't matter: Even if that technology was a dead end, there are still other viable ways to build quantum computers, some of which might have more favo
Re: (Score:2)
It seems like you're expecting every research direction of a complex new technology like quantum computers to quickly pan out. That's unrealistic.
What gave you that idea? I said I have been following this for 35 years. In what universe is 35 years "quickly"?
Re: (Score:2)
Ahhhhhh, I think I get what's going on here. You're a conspiracy theorist about whether quantum computers will ever become powerful enough to crack RSA or whatever else. You know they never will. You know the truth. You have the secret knowledge.
And all these companies and scientists with PhDs and whoever else are just wasting their time and throwing money in a furnace. They're doing it to enrich themselves or to generate hype or for job security, but you know the truth that it will never become actual
Re: (Score:2)
Well, you have understood nothing, clearly.
But I have your number now: You are just an asshole with a gigantic ego and rather pathetic skills.
Re: (Score:2)
Well, you have understood nothing, clearly.
But I have your number now: You are just an asshole with a gigantic ego and rather pathetic skills.
100%
Re: (Score:2)
Re: (Score:2)
It looks that way. In fact, that seems to be the only reasonable explanation for this "breakthrough".