Intel cpus = little cache

[KGGaming]

So recently I was comparing my cpu to a friend of mines he has a i9 9600kf and i have a r5 3600, why does mine have 35mb of cache and his has 9.5mb. Why does ryzen cpus in general just have so much more cache then intel?

[5x5]

Because AMD could and decided "why not?" - so they added more cache. Cache sizes are reduced in lower end models - sometimes due to design (removing a CCX will remove the cache on Ryzen) - or just to separate products (mostly Intel does this so as to make i7 and i9 models appear faster in certain cache-heavy workloads)

[Haro]

Because amd has more than one ccx and because they need to compensate for the latency somehow 

And one of the ways was to add more chache and obviously they have an extra ccx 

Meaning that each ccx has its own batch of cache.

[Hairless Monkey Boy]

Maybe it is needed to achieve good performance?

 

I know that Ryzen has slower RAM latency vs Intel. I wonder if this increase in latency carries forward into the cache as well?

 

Does the Ryzen cache have slower read/writes than Intel, so more is needed?

[Jurrunio]

Different architecture so not comparable. Just like asking why AMD graphics cards have greater memory bandwidth against Nvidia's competitor at the same performance range.

[porina]

The way you count it matters too. Don't add it all up, unless you know you can.

 

Mainstream Intel CPUs historically had an inclusive cache. Whatever is in L2 is also in L3. So you just look at the L3 as indicator. The 9600KF has 9MB, or 1.5MB/core. It is easier to look at them per-core as generally Intel gives you 1.5MB/core on i5, 2MB/core on i7 and above, and some historic Xeons have 2.5MB/core. There are exceptions so don't take that as a hard rule, but it is often the case. For example, Broadwell on desktop only had 1MB/core of L3, but it had a 128MB L4 cache, which was unusual.

 

With Zen 2, AMD went large at 16MB of L3/CCX. Here we don't look at it per-core so much as per-CCX, since that is the more logical unit. I don't think AMD changed from recent history, and they run an exclusive cache. That means data in L2 doesn't get duplicated in L3, thus you can count them both for best case effective total cache. In practice, the L3 is so much bigger than the total L2, the L2 isn't really significant and may be ignored for simplicity.

 

Note AMD only started going large with Zen 2. Older Ryzen was still closer to 2MB/core than 4MB/core (assuming a fully loaded CCX, since the effective cache/core goes up if they disable more cores).

 

Oh, one more factor which isn't obvious. Intel's CPU cache is unified. Any core can reach any part of the L3 cache with only a small difference in latency depending on how far around the ring it has to go. On Ryzen, each CCX is isolated from each other. If it isn't local, you take a big penalty to go outside. Rumours are Zen 3 will go to a bigger CCX which will help that a lot.