Usually SSD storage is accessed by the CPU, this is a massive bottleneck. Not only do you not actually get the advertised transfer speeds of the SSD (as you need to jump through several hoops to actually get it into ram where the GPU can use it), it also hits CPU performance quite hard, meaning you need to design your game to take this into consideration (i.e. you can't start loading gigabytes of content in the middle of a gun-fight as it would just be too much of a hit to the CPU, so you end up in a corridor or lift or subway while this happens).

To overcome this, Sony built custom hardware into the PS5 that reads, decompresses, processes and places data into ram. So you can load data at 8-9gb/s (compressed) straight from the SSD->RAM without hitting the CPU at all and that data is ready to be used by the game straight away.

What this means is that you can have gigabytes of scene data displayed in front of you, turn around, and in that time as you're turning, stream the game content you actually need into ram. In short, game worlds can be insanely detailed.

It's confirmed. The way Microsoft promoted their RT performance is by saying XSX can do 380 billion intersections per second. This number is amount of TMUs times clockspeed. (TMU is Texture Mapping Unit, of which there are 4 per CU.) This gives an exact figure of 379.6; doing the same calculation for PS5 gives a value of 321.1. That's 15% lower, the exact same gap as general compute...which makes sense, since it's dependent on the same two things, amount of CUs and clock.

In other words, if a PS5 game is 15% lower resolution, then it should have the same quality of RT. Or, XSX could have 15% better RT, but then the two games would run at the same resolution. (In general, logical terms only, of course; real results will differ slightly from game to game.)