As a general rule, the effectiveness of cache depends on the repeatability of the acquired data. If you hit a lot of random info at random locations, then a large cache will slow you down. If you have a collection of repeated accesses and / or same data accesses, then a cache will {usually} provide faster access to it.
Inflating a cache beyond the size of the most-often-accessed or repeated access data set size slows thing down because the system will go to the cache every time ... if it takes x amount of time, and you get no cache hits, it's that much time wasted, every time.
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)
Discussion Intel current and future Lakes & Rapids thread
