Zen 3 is rumored to be flaunting monumental IPC gains in early testing

4700x here I come and that’ll be for my mainstream PC next year. My high end PC will wait and get an upgrade to Zen 4 or 5.
 
Lol the only way that can manage 50% more performance in scientific workloads is avx512 support, so this leak is complete bs and no, you don't give away pre es samples to your partners to test them and how they perform, another bs. Motherboards aren't ready, bioses aren't ready, microcode isn't ready, who is going to test them and how?
Believe it or not, AMD actually makes boards for testing, so do OEMs. Are you confusing early production boards, meant for testing, with retail ready boards? And you are forgetting that server hardware and software is done many months earlier than consumer hardware. (after which it gets ported/modified for the mainstream market)

IPC can be checked even when things are not yet 100% ready since the chips don't need to maintain high clock speeds and are also running on weird voltages to keep them stable. At this point though, AMD should have close to retail versions of the chips.

As for AVX 512 they could potentially add full support for it since the higher node density could allow AMD to fit it in without increasing the die size (compared to Zen2). Even if they don't add full AVX 512 support, they can still do a lot of optimizations to improve performance significantly. For example they could have solved some sort of instruction fetch and decoding bottleneck (which is actually something that happens often) by reworking the cache. There are many things that AMD should have learned from their first proper AVX 2 implementation.
 
Last edited:
If I lose support for my current CPU and forced to purchase a current model just so the board I have can support Zen 3, then so be it. I was planning on two CPUs, the introductory, and then the best the board can handle. AMD has had great upgrade support and if my board won't support Zen 3, I really can't be too upset. I can drop a 3950 in it so it's all good. Zen 3 would certainly be a nice bonus. If this information holds up, then Zen has finally matured in nearly every aspect.
 
If I lose support for my current CPU and forced to purchase a current model just so the board I have can support Zen 3, then so be it. I was planning on two CPUs, the introductory, and then the best the board can handle. AMD has had great upgrade support and if my board won't support Zen 3, I really can't be too upset. I can drop a 3950 in it so it's all good. Zen 3 would certainly be a nice bonus. If this information holds up, then Zen has finally matured in nearly every aspect.

The fact that we have been presented blessed with a possible 4 CPU generational fully (with a couple small exceptions) backwards / forwards platform, is mind blowing!

Even to present the idea we might only get three, is complete, unnaceptable BS!!

Of course, I am being facetious, :)

<Insert slow clap meme gif here>
 
Last edited:
The latency that hinders Ryzen against Intel is addressed here, moving data across the core complexes (one CCX reading the other's cache) incurs a latency penalty. The changes being made for Zen 3 eliminates that, at least for anything up to 8 core parts.

I would say the advantage averages out to less than 10 percent at the high end. Techspot themselves did a recent article with tuned memory and showed that was the case. This was fairly impressive in the face of the high clocks of the 9900ks. I would expect that gap to reduce just by the passage of time alone, as new consoles employ 8 core Zen 2 CPUs.

In terms of emulation that's mostly a software thing. With most emulation software being built on and for Intel designs the past 15 years and AMD's inferior single threaded performance (up til at least Zen 3) it's not going to change too quickly.

If AMD manage to surpass the IPC of Intel then it'll be much better than older Zen parts at emulation. But I wouldn't expect it to be as good unless the market sustains a heavy shift in favour of AMD.
Latency is not only CCX to CCX but also CCX to memory. Did you forget that Intel desktop CPUs have about 45ns latency to memory while ZEN2 have 65ns? By the same token, IF link Ryzen has higher latency to everything - to memory, to PCIEx (GPUs and other), to NVMe (and that is why NVMe performance suffers compared to Intel). While above may not impact gaming a lot (except memory access) increasing IF frequency and decreasing latency would help a lot - but that has nothing to do with CCX 2 CCX latency.
 
Its only 4% if you tune your ram up a bit ... and its not just about 1080p -- its 4% @1080P with a 2080ti ... any less video card, and that 4% shrinks to ~0% at any resolution.

So for 99.99% of gamers that don't own a 2080ti, and the ones that don't use theirs to play at 1080p ... the difference is entirely "academic" and not experienced in the real world.

Who really cares about 4% in a hardware and setup scenario that less than 0.1% of all gamer's use where its less than 4% and most likely 0% in the setup you actually have?

What if AMD sneaks out the win by a tiny 4%? It'll be claimed as "insignificant" by a certain some, you can be assured, but currently a 4% advantage with a 2080ti at 1080p is ABSOLUTELY DESTROYING! AMD in gaming.

The differences are all already entirely "academic"; that's why an artificial bottleneck is induced with a 2080ti and low resolution for the reviews ... because the graphs would all be the same if they didn't.

Also keep in mind some of us use our CPUs for adult stuff, like work, video making, multitasking, design, etc. ... gaming is just one facet.
Gamers that want to play at 1080p on 240hz monitors care - while it is not a large group, they are "loud" and they are those that spend most money on their gaming needs.
 
So here are my observations about the current state of the CPU industry. We're almost exactly where we were 20 years ago, with AMD making serious advances and Intel relying on raw clock speed to complete. The industry has come full circle. Very interesting.
 
Back