• 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.

Ryzen "2" ?

It's exactly what Ryzen needs. Biggest drawback with Ryzen at the moment is the much lower max frequency compared to Kaby/Coffee.

Depends on the clocks it hits.
Intel got a lot of flack for making meagre gains. If Ryzen 2/+ whatever it's called is the same, in the name of neutrality, I'd expect the same response.
 
This Ryzen refresh isn't exactly sounding good.
I'm going to expect it'll be given the same reception Kabylake was?
Zen+ is expected to bring some actual architecture improvements and tad more frequenzy. Its also highly nothworthy Intel sold PREMIUM 4 CORE parts in 2017, they could given you more cores if they wanted, but they only wanted to be a cashgrab. So as expected, there might be some pushback at times like those.
IPC at equal clocks there is absolutely no way to discern the Kaby Lake 7700K from the Skylake 6700K
 
Even if it is just was a frequency hike, which it isnt, it would be worth it as thats the big drawback to the first gen.
4.4 or 4.5 ghz with 5% ipc improvement would be great.
 
I would rather they worked on improving the inter-CCX communication and the latency of the current DDR4 controller,which seems a bit high.
That's where I expect most architectural improvements to be.
Doing more wide spread changes to CPU's internal design needs lots of R&D resources and it's not like AMD has excess of resources.
(Zen2 with new manufacturing node needs lots of R&D to keep it in schedule)
So for Zen+ IPC improvements are likely tweaking of some bottle necks and such.
 
Depends on the clocks it hits.
Intel got a lot of flack for making meagre gains. If Ryzen 2/+ whatever it's called is the same, in the name of neutrality, I'd expect the same response.

Intel got flack for making 3.5% increments to their CPU's, granted the step up from Devils Canyon to Skylake was a lot more than that and pretty good but Sandy > Ivy > Haswell > Devils Canyon | Skylake > KabyLake the flack was well deserved.
 
If you take "actual" figures out of the equation and look at what AMD have kind of hinted at previously (referencing low hanging fruit and 10%'s). AMD will be making another big wave in the CPU market purely for the fact that an upgrade will make more than a 5% increase. It should be raising a lot of questions of Intel as to why they dripped out the improvements an kept the core count so low for so long.
 
It isn't that great still relative to what they were saying though hey ;)

https://www.youtube.com/watch?v=1VXpTDOxL1c

No, it never is. :)

If you take "actual" figures out of the equation and look at what AMD have kind of hinted at previously (referencing low hanging fruit and 10%'s). AMD will be making another big wave in the CPU market purely for the fact that an upgrade will make more than a 5% increase. It should be raising a lot of questions of Intel as to why they dripped out the improvements an kept the core count so low for so long.

As a collective we don't really care, we never have, we just want AMD to be competitive so we can buy more Intel stuff.

And that's why Intel don't care.
 
Depends on the clocks it hits.
Intel got a lot of flack for making meagre gains. If Ryzen 2/+ whatever it's called is the same, in the name of neutrality, I'd expect the same response.

This is not me trying to bait you at all but what would you define as a meagre gain? What have the actual total gains been for Intel for their last 4 chips as a comparison?

I don't believe AMD have made a definitive announcement but from their graph and the foundry announcement they are suggesting a 8-12% improvement. Anything north of that is more or less hype train hope. I know we are hoping for a 5%+ IPC gain and a 10%+ clock improvement for a 15% uplift, but AMD have not said that.

I would be pointing some flack at them if there is less than a 7% improvement as they were clearly indicating that level or more.

I would love to have an internet Flakometer -
<6% WTF and you though Intel was bad?
7%-8% C+ It is a pass, but barely.
9%-11% Awesome - Intel looking hard to justify right now.
12%-15% Epic win, the hype delivered.
15%+ The internet explodes.
 
10% makes it hard to justify the 8700k over AMDs stuff, anything like 15% and its impossible to justify Intel stuff at all until they spurt out something else later in the year.
 
AMD could quite easily get a 10% improvement from clockspeed alone on the new process.

The average OC for a Ryzen is ~3.8GHz, so if the new chips can do 4.2GHz without much hassle then then that's your 10% uplift without even tweaking anything under the hood...
 
AMD are paper launching Ryzen 2 at GDC 2018, 19 to 23 March.

Likely this event.

Optimizing for the AMD Ryzen Family of CPU and APU Processors

Join AMD Game Engineering team members for an introduction to the AMD Ryzen™ family of CPU and APU processors followed by advanced optimization topics. Learn about the "Zen" family SoCs, upcoming next-generation Ryzen processors, and profiling tools. Gain insight into code optimization opportunities and lessons learned. Examples may include C/C++, assembly, and hardware performance-monitoring counters.

Other AMD events.

Engine Optimization Hot Lap
Taking the Red Pill: Using Radeon GPU Profiler to Look inside Your Game
Simulating and Rendering Physically-Realistic Curly Hair
The Art of Profiling: Radeon GPU Profiler & RenderDoc
Real-Time Ray-Tracing Techniques for Integration into Existing Renderers
Getting Explicit: How Hard is Vulkan Really?
Advanced Graphics Techniques Tutorial: Water Rendering in 'Far Cry 5'
HLSL in Vulkan: There and Back Again
Advanced Graphics Techniques Tutorial: "New Techniques for Accurate Real-Time Reflections" & "Memory Management in Vulkan and DX12"
 
Last edited:
Clockspeed is not the whole ballgame though.

No, but it is probably the easiest way of improving performance without huge R&D costs and is generally what is expected from a smaller fab process.

I don't expect there to be huge architectural differences between Ryzen and the refresh, just a few easy wins that tick up IPC a few points, with the bulk of the performance improvements coming from using a process node that is designed for higher clocks rather than low power.
 
My only issue with this is that, however good this might be (or not), are there not rumours saying that the 2019 reiteration, Zen 2, is going to be the game changer, which makes me want to wait to see what it could be regardless.

I know the logic behind not waiting for the latest thing, as you'll in theory always be waiting...but I don't mind waiting for a bigger change instead of the usual 10% or whatever.
 
Last edited:
Back
Top Bottom