Ryzen APU Overclocking: A Focus on Memory Support

Also part of the demonstration about the new APUs was overclocking. For a number of years, AMD has hired Sami Makkinen, a well-known former world-record holder in various overclocking categories, to help tune the overclocking on some of its processors. So despite this being an integrated graphics platform, Sami was on hand to show the day-to-day overclocking potential. I was told that liquid nitrogen overclocking might be held at a later date…

Nonetheless, we were shown the system that would be distributed to reviewers for launch day: a mini-ITX motherboard, a Ryzen 5 2400G processor, 2x8 GB of DDR4 memory, and an AMD Wraith cooler. Using the popular benchmarking tool 3DMark Fire Strike, Sami showed a 39% increase over stock performance by overclocking only the memory and the integrated graphics:

  • DDR4-2400 + 1250 MHz GPU: 2911 Points - 'stock'
  • DDR4-3200 + 1250 MHz GPU: 3322 Points (+14%)
  • DDR4-3200 + 1550 MHz GPU: 3596 Points (+24%)
  • DDR4-3600 + 1675 MHz GPU: 4048 Points (+39%)

This shows a couple of interesting data points. First, that the DRAM speed directly influences the results a lot: the 14% gain moving to DDR4-3200 and a similar gain again up to DDR-3600 means that we are still dealing with a graphics system that is memory bandwidth limited, even with only 11 compute units. The Infinity Fabric communication pathways are tied to the DRAM speed, so these would also get a speed up – when asked if it would be possible to discern how much of the speedup is due to a pure DRAM improvement and how much from the Infinity Fabric boost, I was told that AMD could do it in house, but it would be impossible for consumers to do.

However, the DRAM speed gains match up with what we have seen with previous generations of AMD APUs, but also it means that Intel’s decision to combine a Vega GPU with high-bandwidth memory is going to be the way forward in this market.

The second point that this data brings is about memory support. One of the major criticisms of AMD’s initial launch of Ryzen was the lack of high-performance memory support, due to a young firmware. AMD has stated that the Ryzen with Vega graphics line of processors have better memory support with newer firmware revisions, capable of driving higher memory speeds. This is, of course, important to APUs. Additional updates will be coming with the new X470 motherboards in Q2.

Zen Cores and Vega: Ryzen APUs for AM4 AMD Ryzen Price Drops, New Wraith Prism
POST A COMMENT

136 Comments

View All Comments

  • jjj - Monday, January 8, 2018 - link

    Nor quite. Vega on 7nm is a SKU aimed at machine learning that samples in late 2018. That means volume in 2019 and other SKUs later , if there are other SKUs- I am unclear if AMD stated that there are other SKUs or that's an AT assumption.
    There is a possibility that Navi hits high end in 2019 and Vega lives bellow but that's kinda silly as AMD needs to be in a huge rush to replace Vega with something that is more competitive.
    Reply
  • haukionkannel - Monday, January 8, 2018 - link

    True. The machine learning chips Are the top priority I was just quessing that mobile part would Also get upgrade. My bad...
    So next Vega goes against machine learning Volta. Maybe Nvidia neither will release normal Volta, only those calculate monsters to AI projects. That would be interesting.
    Reply
  • jjj - Wednesday, January 10, 2018 - link

    Vega 7nm is 2019 for volumes it seem, Volta V100 will be replaced by something new likely this year and again in 2019- at the very least there will be one update.
    What Nvidia will do in consumer is unclear as there is no pressure on them to spend on releasing new things. The Intel+AMD part is problematic in laptop so likely they'll address that segment soon with something new.
    Reply
  • Pinn - Monday, January 8, 2018 - link

    Lisa is my waifu. I'm fine with my wife and daughter reading this. Reply
  • StevoLincolnite - Monday, January 8, 2018 - link

    Ugh. No Navi until 2019. So AMD's graphics are guaranteed to be terrible for another year. Reply
  • Pinn - Monday, January 8, 2018 - link

    and nvidia is guaranteed not to release volta for consumers.. fun. Reply
  • Xajel - Monday, January 8, 2018 - link

    I was really hoping for a low power 8C/16T Ryzen CPU's, these can be used in SFF & laptops with dGPU already... these will be for those who want more CPU performance than GPU, or they need both CPU & GPU power. Reply
  • A5 - Monday, January 8, 2018 - link

    You'll probably have to wait for 7nm for that. Current thermal budgets in laptop designs just aren't going to get you usable 16T performance from 35-45W. Reply
  • neblogai - Monday, January 8, 2018 - link

    Table of Ryzen Mobile CPUs has all Raven Ridge APUs with '1MB per core'. But this is not true for R3 2200U, which still has the same 4MB of L3: http://www.amd.com/en-us/press-releases/Pages/ces-...
    Also- I still hope that is a Banded Kestrel die. On that topic, Joe Macri, when asked about RR use for fanless designs back in October, said it would not work too well, and would have to be clocked real low; when asked if they are building a different chip for that fanless market- answered only broadly- 'we love PCs,.. we want you to find AMD everywhere where PC is'. So I'm not sure- does that still mean only Stoney Ridge (if it qualifies for fanless), or a new die. Just looking at the AMD's below average power efficiency in RR laptops at browsing and playing video- maybe AMD knew they are simply not ready for fanless, and need to put in more work. But I still wish they released Zen 2C/4T 3CU - if not low power- then at least for even cheaper normal laptops. After all- Raven Ridge is not very small die at 210mm2, and a very cheap ~110mm2 die with 2c/4t Zen could also be successful.
    Reply
  • T1beriu - Monday, January 8, 2018 - link

    ^This. Reply

Log in

Don't have an account? Sign up now