What does this mean?

Well AMD makes chiplets (Core Compute Dies), and these each hold upto 4 CCX's (Core Compute Complex's).
Now when you spread out the cores, to be amoung differnt CCX's, theres a "lag" moveing data around between the CCX's, and cache's.
You could call this a Interconnect Latency. When all the CPU cores are in the same package, or same CCX theres a performance uplift.

This is shown in the AMD Ryzen 3 3300x and the Ryzen 3100.
(the 3300x is only a few mhz faster, but being in 1 ccx instead of 2, theres a massive performance differnce).


Why does this matter?
Well remember Sony Playstation 5 roadmap video?
Where they keep talking about latency and how they went out of their way to do away with it where ever they could?


Rumor has it that applies to the cpu too.
With Zen 3, AMD is going to make CCX's not have this issue, when your useing 8/16 core/threads.
Playstation 5, wont be zen3, but it could have some of things used in zen3, used in the Zen2 cores it has.
Ei. fixing this issue of interconnection latency in the cpu.

Hint 1:  (redgameingtech and his source's)

https://youtu.be/4BlDm-FD-ls?t=1254

timestamp : 21:19

"the zen 2 achitecture inside the playstation 5 has seen some modifications"
"I believe the playstation 5 might have some features that belong to the zen 3 and that feature is the unified CCX design"

and then theres this:

Let’s say you have a CPU whose cores are 110mhz faster, but with higher interconnect latencies, versus another CPU. That’s fine if you are hitting on-core L1/L2 cache almost all the time, otherwise you’re stalling. I’d personally favor the lower latencies vs marketing numbers.

— Matt Hargett (@syke) June 30, 2020

"Let’s say you have a CPU whose cores are 110mhz faster, but with higher interconnect latencies, versus another CPU. That’s fine if you are hitting on-core L1/L2 cache almost all the time, otherwise you’re stalling. I’d personally favor the lower latencies vs marketing numbers." - Matt Hargett


Thats the wink wink, hypothetically cpu inside the xbox series x, vs the playstation 5.

Hes saying that the modifications to the playstation 5 cpu, mean that it doing away with this "Interconnect Latency" is a bigger boost to the cpu, than it being 100mhz slower than the xbox series x's cpu.

TL DR :
Playstation 5 cpu, is perphaps slightly faster than xbox series x's cpu, even while running 100mhz slower.

Back in May redgameingtech (who admits he doesnt have great sources for xbox side of things), say's he could see Xbox haveing done the same.
However from the xbox series x reveal video, theres a render that supposedly shows the oppersite.

So its unlikely Xbox went the same route, with the cpu.

It would make things more interesting, however isn't the PS5's CPU throttled? Meaning it wont always be 100mgz slower, it could be even more depending on the game.. however this solution does seem like it might solve this problem. I don't think either console will be going that route. Not yet anyway, maybe in the mid gen upgrades.

From neogaf: BadBreathoftheWild (his explaination of this)

"At the speed a CPU runs at, having to go get some data from RAM is like walking to the next building, where having it in your cache is like having the data you need in your room.

When you only have 8ms per frame (120 FPS) this kind of latency matters.

Bandwidth is like how big your pockets are, and latency is like how fast you run to go and get that data to stuff into those pockets.

It doesn’t matter if you have massive pockets if the data turns up too late to get the frame out before 8ms has elapsed.

A system that needs to take less frequent trips to RAM because its caches are faster and don’t get entirely invalidated (Coherency Engine) should see less frame drops and have a higher average frame rate as frame times get smaller and smaller.

High frame rate is critical for something like VR, and Sony may have specifically included Coherency Engines, cache scrubbers and be running the GPU as fast as they are doing all to aid latency so PS5 can deliver better VR than PS4 could.

From the wording of what Matt was saying, and from RGT’s sources (no idea how reliable he is), it could be that PS5 CPU will spend less time waiting for a RAM fetch and be doing useful work more often as it either has a unified L3 cache between all cores like Zen3 architecture, or it manages the problems of cache coherency better.

The GPU will already be doing this, and Cerny did say that “one” of the jobs of the Coherency Engines was working with the GPU cache scrubbers, implying it also has a role to play on the CPU side, too.

It could be that XSX has the potential to push more pixels per frame, but PS5 has the potential to push more frames per second.

The relationship between both aren’t linear exactly because of cache/RAM latencies.

PlayStations IO patent also has just as much emphasis on latency as it does on bandwidth, which is why so much of the pipeline is offloaded to fixed function accelerators. Low latency is down as a key requirement, just as much as high bandwidth is."

Yes depending on the workload, the cpu could be even more than 100mhz slower on the playstation 5.
Depends on how much power the gpu workload is useing.

Typically, for all but the most depending games, I suspect the playstation 5 will run at its rated speeds though.

if that means the ps5 can also do 120fps engines and actually does create games with it, i'll be happy

Both the GPU and CPU are likely to be fairly vanilla AMD parts. There isn't much in standard PCs that can be left out, and it seems pointless for AMD to develop major functionality that they wouldn't also included in their PCs parts.

The days of heavily customised console hardware are over, now it's about smart ways to wire together standard PC parts to maximise value for money.

Sounds like it is more expensive then.

This isnt something "heavily customised", its like takeing part of Zen 3 (which should be releaseing this year), and useing it on a zen2 cpu, with slight modifications.
Its not something brand new developed just for sony.

AMD is constantly working on makeing their cpu's better, so when they make advancements, takeing something new and putting it into something old, often probably doesnt take too much work. The cost in R&D are near zero, since its something they already worked out long ago for a future product.

Honestly dont think so. Probably not a huge price attachted to this sort of thing.

Theres also this:

https://wccftech.com/keoken-on-next-gen-we-dont-see-many-differences-between-ps5-tempest-will-free-cpu-resources/

KeokeN founder (game director Koen Deetman) :
"For now, we don't see too many differences, they seem to be competing well against each other and both are pushing new boundaries."

"We Don’t See Many Differences Between PS5 & XSX (CPU); Tempest Will Free CPU Resources"

"Audio will always play a crucial role in our games and the next generation of consoles boasts a host of fantastic audio features. It looks like Tempest will free up CPU that we used to need for audio, so that leaves room for us to use for other aspects of the game. The HRTF functionality seems particularly interesting, too. We're also excited about Project Acoustic's wave-based technology. It's all very promising and we're looking into how it would tie into our existing workflow and technology."

One way of getting CPU to punch above its weight, is to off-load tasks from it.
Like if audio doesnt take up cpu usage anymore, that can be spent on other cpu tasks instead.


I honestly think CPU wise, the differnce between the Xbox Series X, and the Playstation 5 will be like near 0.
GPU wise, that 15% wont be that big a deal either.

Some games will be.

Valorant dev's have said that 4k 120fps is a achievable goal on next gen consoles.
That means on both the xbox series x, and playstation 5.

Polyphony Digital (Grand Turismo) have said that 4k is enough, and instead they ll chase 120fps or 240fps.
(ironic considering they showed it off running 8k, I believe?)