> our worst case projections did not come to pass. These loading time projections were based on industry data - comparing the loading times between SSD and HDD users where data duplication was and was not used. In the worst cases, a 5x difference was reported between instances that used duplication and those that did not. We were being very conservative and doubled that projection again to account for unknown unknowns.
> The pop-culture cargo cult description, however, takes features of some cargo cults (the occasional runway) and combines this with movie scenes to yield an inaccurate and fictionalized dscription. It may be hard to believe that the description of cargo cults that you see on the internet is mostly wrong, but in the remainder of this article, I will explain this in detail.
FWIW, I meant it strictly in the generic vernacular sense in which I've encountered it: doing something because it has the outward form of something useful or meaningful, without understanding whether or how it works.
Given the problematic history you shared, it seems a new term is needed for this... maybe "Chesterson's Folly"? It's related to Chesterson's Fence (the principle that it's unwise to remove a fence if you don't know why it was erected). If you leave in place all "fences" you don't understand, and never take the time to determine their purpose, fences which serve no good purpose will accumulate.
On the flip side I don't remember who did it, but basically extracting textures on disk fixed all the performance issues UE5 has on some benchmarks(sorry for being vague, but I can't find the source material right now). But their assumption is in fact a sound one.
Yes. Its quite common for games to have mods that repack textures or significantly tweak the UE5 config at the moment - and its very common to see users using it when it doesn't actually affect their use cases.
As an aside, I do enjoy the modding community naming over multiple iterations of mods - "better loading" -> "better better loading" -> "best loading" -> "simplified loading" -> "x's simplified loading" -> "y's simplified loading" -> "z's better simplified loading". Where 'better' is often some undisclosed metric based on some untested assumptions.
Non-made up numbers from Vermintide 2 (same engine): On PS4 when an optimized build took around 1.5 minutes to boot to main menu, the unoptimized version would take 12-15 minutes [1]. A different benchmark than SSD vs HDD, but shows that the optimization was certainly needed at the time.
Though the PS4 was partially to blame as well, with it's meagre 5400 RPM spinny drive.
For their newer instalment, Fatshark went with a large rework of the engine's bundle system, and players on HDDs are complaining about long loading times expectedly. That game is still large at ~80GB, but not from duplication.
They basically just made the numbers up. Wild.