![]()
I'm not sure the data read is correct unfortunately. If you look here you can see low pci-e utilization and yet the game has clear performance penalty at x8 vs x16. That would mean it's a very narrow bottleneck. There's just something screwy with this game but without a proper gpu profiler I doubt the data from these tools (gpuz etc) as well.
One of the things with PCI-e and games - while you have lots of bandwidth that is measured over a second the actual amount of bandwidth per a game frame and any latency involved can be a more complicated story.
I keep seeing mention of PCI-e bandwidth constraints with HZD but every time I look into them it just comes back to one person's remarks in an article they did without actual graphs or a study of it? and I can't see anything about it in that video? (other than they chart bus controller utilisation - which does tend to give an indication but won't show spikes in bandwidth use, etc.).