• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Just 19-20 years ago, extra L3 cache or any L3 cache used to help in almost all workloads

Cyber-Mav said:
that data would prove if more l3 cache hinders performance as the op has stated. my suspicions are that the cpu with more cache would out perform the one with less cache in everything tested, clock for clock at same power targets.
Click to expand...

The OP is talking about a single core 32bit chip running 20+ year old code.

I think if everything was even, the performance would be margin of error because AMD seem to have sized its caches pretty well for most tasks. Once the cache is full the difference would be the round trip journey to RAM. So massive.

What is needed is a CPU with 32gb of L1 running at the Thz range.
 
Amds architecture does suffer from memory latency a lot unlike intels hence going from zen1 to zen2 the doubled cache was mostly the main reason for the performance uplift especially in gaming.
Only time more cache can have a performance loss is if its got more latency than the smaller cache, again that would be application depending, if app needs faster cache access or just lots more cache due to size and randomness of data.
Im wondering if insted of going with vcache was it not possible to just add more regular cache e.g go from 32mb per ccx to 48mb?
 
You must log in or register to reply here.
Back
Top Bottom