It's interesting that at the bottome of this article there is a recommened article about PCIE 7.0 published in June 2025. And it says "While PCIe 8.0 is still years away, PCIe 7.0 is a lot closer." Never know how these committees work.
Nevertheless, the perf increases of IO devices these days are in insane. I am wondering whether and when these perf promises will materialize. We are only on PCIE 5 this year and it's not that common yet. I am wondering how fast adoptions would be, which pushes manufacturers to iterate. The thing is that at the current level of PCIE 5, a lot of softwares already need to be rewritten to take full advantages of new devices. But rewriting softwares takes time. If software iterations are slow, it's questionable if consumers will continue to pay for new generations of devices.
It is. The biggest benefit for consumer devices is less lanes needed for same bandwith, but that is only really relevant to GPU's. SSD's are already constrained by thermals, not bandwith.
The downside - cost. Making these lanes run reliably at these speeds is not cheap.
I wouls say benefits for consumer devices peaked at pcie 4, anything above is just cost. Maybe gen 5 to allow gpus to run at x8 without bottlenecks, but given how massive those gpus are, you are not saving any space, just lanes.
It's interesting that at the bottome of this article there is a recommened article about PCIE 7.0 published in June 2025. And it says "While PCIe 8.0 is still years away, PCIe 7.0 is a lot closer." Never know how these committees work.
Nevertheless, the perf increases of IO devices these days are in insane. I am wondering whether and when these perf promises will materialize. We are only on PCIE 5 this year and it's not that common yet. I am wondering how fast adoptions would be, which pushes manufacturers to iterate. The thing is that at the current level of PCIE 5, a lot of softwares already need to be rewritten to take full advantages of new devices. But rewriting softwares takes time. If software iterations are slow, it's questionable if consumers will continue to pay for new generations of devices.
LLMs to the rescue :)
How fast is too fast? I wonder if this would only be usable in the largest of systems with multiple sockets and high core count processors.
It is. The biggest benefit for consumer devices is less lanes needed for same bandwith, but that is only really relevant to GPU's. SSD's are already constrained by thermals, not bandwith.
The downside - cost. Making these lanes run reliably at these speeds is not cheap.
I wouls say benefits for consumer devices peaked at pcie 4, anything above is just cost. Maybe gen 5 to allow gpus to run at x8 without bottlenecks, but given how massive those gpus are, you are not saving any space, just lanes.
Same old question that has been asked for 40-50 years now.
The answer is nothing is too fast.