No, PCI 4.x is not dead. It's still in use and being developed for future standards.
No, PCI 4.x is not dead. It's still in use and being developed for future standards.
My watercooling suggestions suggest something about my Samsung SSD 960 PRO possibly not working. Using the Samsung Magician Software, temperatures reach 70-80 during heavy use, which is higher than usual (normal range 80-100). The 980 PRO with its heatsink stays around 60... I’m planning to get a new heatsink and recheck the 960 PRO.
Another issue is achieving 4 and 5 over a range, which explains why only the highest slots on AMD motherboards are 4.0 now. I think the main point is that the chipset or MUX needs more bandwidth to switch to 3.0 slots. It would be nice to see boards with more x4 slots rather than just x1, though that might not be possible since it could reduce space for M.2 drives. Personally, I prefer a board packed with many x4 slots and using M.2 adapters, as it gives more flexibility for expansion.
PCIE3x4 already supports 4GB/s, so boosting to 8Gbps isn't about saturation—it's more about performance expectations. Even for sequential tasks, it's unlikely to be sustained. "Oh my, the 4K UHD file took 5 seconds instead of 10": If you're compensated for video editing...well, that might apply to some.
It's beneficial when the duration of different technologies is decreasing once more, similar to past decades. This creates a positive outcome for all parties involved. In the long and short term, everyone stands to gain, with most benefiting soon after some adjustment period.
PCIe backward compatibility is what matters, not the speed of each generation. I recently purchased a PCIe 3.0 NVMe storage device for my system a couple of months back. My machine supports Gen 4, yet I opted for a Gen 3 model. I didn’t feel the need to pay an extra 50% for the performance increase, since it wouldn’t make much difference. What exactly does a regular user gain from a Gen 4 drive? You might think about large file transfers, but you’re missing the point of this discussion. Here’s a quick example: I recently moved a significant amount of data—over 600GB—from one NVMe drive to another. I even reformatted one of the disks in an attempt to resolve an issue. Most transfers ran between 100-200MB/s, sometimes dropping to just a few MB/s or even KB/s, with the process taking well over an hour. That’s roughly 20 times slower than what either drive is actually capable of delivering. "Why so slow?" you might wonder. The overhead from file systems limits performance, especially for smaller files where speed isn’t critical. Even a basic SATA SSD would suffice for most users, providing the necessary random read/write speeds for smooth system operation. In a simple comparison, you’d likely never detect the difference between a slower Gen 3 NVMe and a faster Gen 5 drive.