One of the main problems with Quantum Computing is a high error rate, especially if you want it to be done on more available hardware and without cost running wild.
IBM has seemingly managed to get the highly advanced error-correction used to reduced qubit errors to run on standard processors, according to a paper published on arXiv by IBM researchers Thilo Maurer, Markus Bühler, Michael Kröner, Frank Haverkamp, Tristan Müller, Drew Vandeth, and Blake R. Johnson titled "Real-time decoding of the gross code memory with FPGAs".
The team managed to get the error correction to run on standard AMD chips, and at 10x the speed needed to keep up with the quantum computer. This is significant, as error correction is a substantial hurdle for IBM and their goal scaling quantum computing beyond its current use and rather small scale.
IBM is also using a new decoding type called Relay-BP that is in itself significant faster than previous methods, important as the error correction until now has often been the limiting factor. This also offloads the error correction from the quantum computer to normal hardware, thus freeing more processing power.
While there is no guarantee that this will bring us closer to actual viable quantum computers within the next few years as experts do not agree on how much this will accelerate the theoretical timeline for such things, it will most likely speed up the process, and make it possible to reduce the cost of the final product.