Welcome the latest DF Direct Weekly – the 29th, in fact! As you are likely aware by now, this is our regular show where DF team members down tools for a couple of hours to discuss the latest in tech and gaming news, kicking off this week with a little behind-the-scenes information on our collaboration with Gamers Nexus in getting to the bottom of the PlayStation 5 CFI-1100 – but also in sharing some more on data gathered during production of that project, where we discovered that the new PS5 firmware seems to have a small, but measurable boost to performance in some scenarios.
Today’s latest PS5 system software is yesterday’s firmware beta 3.1, where I discovered that in very certain circumstances, the software upgrade results in higher frame-rates on my launch model PS5. This was somewhat problematic during the CFI-1100 review process, where it seemed that my old PS5 was running faster than the new machine – something that shouldn’t happen. I got to the bottom of the issue by benchmarking all games three times in order to establish which games varied performance on a per-run basis. And then I repeated the process on a PS5 Digital Edition running the same firmware as the CFI-1100. I discovered that if I threw out the beta firmware data, the new PS5 ran exactly the same as the old PS5. And when I upgraded all of my PS5s to the new official firmware, all of them ran identically.
I benchmarked Control in graphics mode with its photo mode fps unlock, Devil May Cry 5 Special Edition in RT mode, Godfall and Resident Evil Village. Control and DMC5 do not vary by more than one percent on a run-by-run basis and yet the new firmware runs those titles two to three percent faster than before – a curious state of affairs. Resi has dynamic elements that adjust performance slightly each time, while I saw no change at all in Godfall. I’m wondering whether it’s perhaps a system level ray tracing optimisation? A couple of percentage points in select scenarios is hardly game-changing but intriguing nonetheless.
About Author
