CPU L1/L2 cache size over the years

I'm trying to find a graph with information on the CPU L1/L2 cache sizes over the years.

I have only succeeded to find an old chart from 2008 (here).

I know that the cache sizes have stayed roughly the same over the past twenty years, I just want a graphical representation of this info.

Does anyone know where to find one?

Thanks!

Solution

There is apparently an open database of historical CPU specifications at http://cpudb.stanford.edu/. I don't see the specific visualization you are asking for, but the data contains cache sizes for processors spanning from 1970s to 2016, so it might be possible to create your own plot from it.

What happens with the store "that lost race" to shared memory in x86 TSO memory model?
Pipeline on Registers calculation
If I have an 8-bit value, is there any advantage to using an 8-bit register instead of say, 16, 32, or 64-bit?
what is the purpose of using index caches in rigtorp's SPSCQueue
What does it mean by word size in computer?
How is cache coherency maintained on ARMv8 big.LITTLE system?
How does ARM's MTE prevent off-by-one memory errors?
FLOPs per cycle for Sandy Bridge and Haswell and others SSE2 / AVX / AVX2 / AVX-512
How can I implement the overflow flag in Logisim without having access to the second last carry?
Difference between x86, x32, and x64 architectures?
The most efficient way to test if a positive integer is 2^n (i.e. 1, 2, 4, 8, etc.) in C++20?
What is the cost of atomic operations?
How can the control hardware know which registers to read before it completes decoding the instruction?
Microarchitectural zeroing of a register via the register renamer: performance versus a mov?
Why do I get 'Error executing binary file' despite compiling for the correct architecture?
Hardware interrupt when Power Button pressed?
Program Counter incrementation
Why is processing a sorted array faster than processing an unsorted array?
Is lock acquiring function atomic?
Understanding micro-architectural causes for longer code to execute 4x faster (AMD Zen 2 architecture)
when we say a memory controller does it mean the component that manages communication between the RAM and the processor only?
In shift left instruction, why is rt used as source register instead of rs?
Intel vs AMD gather AVX performance
In a RISC-V architecture, do jump instructions (conditional or JAL/JALR) increase the PC by 4 as the rest of the instructions?
Why is memory aliasing needed?
Understanding LW in MIPS
How can x86 bsr/bsf have fixed latency, not data dependent? Doesn't it loop over bits like the pseudocode shows?
Getting docker-entrypoint.sh: exec format error while the architecture matches
Implementing closures and performance impacts of, for instance, relative vs absolute jump
Are data memory and instruction memory in data path L1 caches?