Core-to-Core, Cache Latency, Ramp

For some of our standard tests, we look at how the CPU performs in a series of synthetic workloads to example any microarchitectural changes or differences. This includes our core-to-core latency test, a cache latency sweep across the memory space, and a ramp test to see how quick a system runs from idle to load.

Core-to-Core

Inside the chip are eight cores connected through a bi-directional ring, each direction capable of transmitting 32 bytes per cycle. In this test we test how long it takes to probe an L3 cache line from a different core on the chip and return the result.

For two threads on the same core, we’re seeing a 7 nanosecond difference, whereas for two separate cores we’re seeing a latency from 15.5 nanoseconds up to 21.2 nanoseconds, which is a wide gap. Finding out exactly how much each jump takes is a bit tricky, as the overall time is reliant on the frequency of the core, of the cache, and of the fabric over the time of the test. It also doesn’t tell us if there is anything else on the ring aside from the cores, as there is also going to be some form of external connectivity to other elements of the SoC.

However, compared to the Zen3 numbers we saw on the Ryzen 9 5980HS, they are practically the same.

Cache Latency Ramp

This test showcases the access latency at all the points in the cache hierarchy for a single core. We start at 2 KiB, and probe the latency all the way through to 256 MB, which for most CPUs sits inside the DRAM.

Part of this test helps us understand the range of latencies for accessing a given level of cache, but also the transition between the cache levels gives insight into how different parts of the cache microarchitecture work, such as TLBs. As CPU microarchitects look at interesting and novel ways to design caches upon caches inside caches, this basic test proves to be very valuable.

The data here again mirrors exactly what we saw with the previous generation on Zen3.

Frequency Ramp

Both AMD and Intel over the past few years have introduced features to their processors that speed up the time from when a CPU moves from idle into a high-powered state. The effect of this means that users can get peak performance quicker, but the biggest knock-on effect for this is with battery life in mobile devices, especially if a system can turbo up quick and turbo down quick, ensuring that it stays in the lowest and most efficient power state for as long as possible.

Intel’s technology is called SpeedShift, although SpeedShift was not enabled until Skylake.

One of the issues though with this technology is that sometimes the adjustments in frequency can be so fast, software cannot detect them. If the frequency is changing on the order of microseconds, but your software is only probing frequency in milliseconds (or seconds), then quick changes will be missed. Not only that, as an observer probing the frequency, you could be affecting the actual turbo performance. When the CPU is changing frequency, it essentially has to pause all compute while it aligns the frequency rate of the whole core.

We wrote an extensive review analysis piece on this, called ‘Reaching for Turbo: Aligning Perception with AMD’s Frequency Metrics’, due to an issue where users were not observing the peak turbo speeds for AMD’s processors.

We got around the issue by making the frequency probing the workload causing the turbo. The software is able to detect frequency adjustments on a microsecond scale, so we can see how well a system can get to those boost frequencies. Our Frequency Ramp tool has already been in use in a number of reviews.

A ramp time of within one millisecond is as expected for modern AMD platforms, although we didn’t see the high 4.9 GHz that AMD has listed this processor as being able to obtain. We saw it hit that frequency in a number of tests, but not this one. AMD’s previous generation took a couple of milliseconds to hit around the 4.0 GHz mark, but then another 16 milliseconds to go full speed. We didn’t see it in this test, perhaps due to some of the new measurements AMD is doing on core workload and power. We will have to try this on a different AMD Ryzen 6000 Mobile system to see if we get the same result.

AMD's Ryzen 9 6900HS Rembrandt Benchmarked Power Consumption
POST A COMMENT

92 Comments

View All Comments

  • mode_13h - Thursday, March 24, 2022 - link

    > Mulholland Drive is perhaps my favourite film of all films.

    After your last post, I was already going to start (re-)watching Lynch's films, but maybe I'll start there.
    Reply
  • GeoffreyA - Monday, March 28, 2022 - link

    MD is the best. All his other films tend to be something of a mess or too obscure or too excessive. In particular, I despise Lost Highway. Straight Story is good for the whole family though, and Elephant Man. Reply
  • GeoffreyA - Monday, March 28, 2022 - link

    Another film I love is Kieslowski's "Trois couleurs rouge" (Three Colours: Red). Reply
  • mode_13h - Wednesday, March 30, 2022 - link

    I saw Red, long ago. I remember it made an impression on me, but not much else.

    I remember watching bits of Elephant Man with my parents, but they rented and watched it at home and I think I was too young to really sit and watch the whole thing.

    Never saw Lost Highway, but I think one of the first MP3s I got was a rip of the Trent Reznor song from it + remixes.

    Heh, I just watched the documentary: Jodorowsky's Dune. What a trippy guy! He literally cast Salvador Dali as the Emperor and seems to have been H.R. Geiger's route into the movie business. We can probably thank the fact that he cast Mick Jagger as Feyd-Rautha as the reason Sting ended up in Lynch's version.

    I think it's safe to say it would've been a *very* different movie. He seemed most fascinated by the aspects of the book that least impressed me (i.e. the mysticism and psychedelic stuff). He said he wanted the movie to seem like a long acid trip and even admitted to changing the ending. Unapologetically. He's *that* kind of film maker ...which is okay, but just not if you're a fan of the book.
    Reply
  • GeoffreyA - Friday, April 1, 2022 - link

    The first time I saw Red, I didn't think much of it, except that it was very polished stylistically; but it's the sort of film that gains a lot from repeat viewings, and possesses a beautiful symmetry and design. It seems to distill, in a simpler fashion, the "plot" we sometimes see in life.

    I've often heard about Jodorowsky's Dune and need to watch that documentary. Well, if Geiger's designs were supposed to be there, it would've been a striking movie, that's for sure! I get the feeling I might have actually enjoyed it. Anyhow, looks as if the faithful Dune adaptation is still to be made. Where is this mighty director?
    Reply
  • GeoffreyA - Friday, April 1, 2022 - link

    (Come to think of it, DeMille might have done a good job.) Reply
  • mode_13h - Saturday, April 2, 2022 - link

    > I've often heard about Jodorowsky's Dune and need to watch that documentary.

    You'd get a little more from the documentary if you'd read the book, first. Otherwise, you wouldn't have as much appreciation for the aspects of the book that attracted him. Plus, there's a bit of a spoiler, later in the documentary, where he talks about how he was going to change the ending.
    Reply
  • GeoffreyA - Sunday, April 3, 2022 - link

    I see. Thanks for that warning! Reply
  • Violet Giraffe - Friday, March 11, 2022 - link

    Are you mad? Apple is at least a year behind the real CPU manufacturers. And Zen 4 will leave it on the side of the road. Reply
  • Violet Giraffe - Friday, March 11, 2022 - link

    M1 is cute, nothing more. It's GPU is relatively powerful compared to other mobile chips, otherwise it sucks. I have an M1 Mac Mini so it's first-hand experience. Reply

Log in

Don't have an account? Sign up now