Conclusions

In this mini-test, we compared AMD’s Game Mode as originally envisioned by AMD. Game Mode sits as an extra option in the AMD Ryzen Master software, compared to Creator Mode which is enabled by default. Game Mode does two things: firstly, it adjusts the memory configuration. Rather than seeing the DRAM as one uniform block of memory with an ‘average’ latency, the system splits the memory into near memory closest to the active CPU, and far memory for DRAM connected via the other silicon die. The second thing that Game Mode does is disable the cores on one of the silicon dies, but retains the PCIe lanes, IO, and DRAM support. This disables cross-die thread migration, offers faster memory for applications that need it, and aims to lower the latency of the cores used for gaming by simplifying the layout. The downside of Game Mode is raw performance when peak CPU is needed: by disabling half the cores, any throughput limited task is going to be cut by losing half of the throughput resources. The argument here is that Game mode is designed for games, which rarely use above 8 cores, while optimizing the memory latency and PCIe connectivity.

A simpler way to imagine Game Mode is this: enabling Game Mode brings the top tier Threadripper 1950X down to the level of a Ryzen 7 processor for core count at around the same frequency, but still gets the benefits of quad channel memory and all 60 PCIe lanes for add-in cards. In this mode, the CPU will preferentially use the lower latency memory available first, attempting to ensure a better immediate experience. You end up with an uber-Ryzen 7 for connectivity.


AMD states that a Threadripper in Game Mode will have lower latency than a Ryzen 7, as well as a higher boost and larger boost window (up to 4 cores rather than 2)

In our testing, we did the full gamut of CPU and CPU Gaming tests, at 1080p and 4K with Game Mode enabled.

On the CPU results, they were perhaps to be expected: single threaded tests with Game Mode enabled performed similar to Ryzen 7 and the 1950X, but multithreaded tests were almost halved to the 1950X, and slightly slower than the Ryzen 7 1800X due to the lower all-core turbo.

The CPU gaming tests were instead a mixed bunch. Any performance difference from Game Mode over Creator Mode was highly dependant on the game, on the graphics card, and on the resolution. Overall, the results could be placed into buckets:

  • Noted minor losses in Civilization 6, Ashes of the Singularity and Shadow of Mordor
  • Minor loss to minor gain on GTX 1080 and GTX 1060 overall in all games
  • Minor gain for AMD cards on Average Frame Rates, particularly RoTR and GTA
  • Sizeable (10-25%) gain for AMD cards on 99th Percentile Frame Rates, particularly RoTR and GTA
  • Gains are more noticable for 1080p gaming than 4K gaming
  • Most gains across the board are on 99th Percentile data

Which leads to the following conclusions

  • No real benefit on GTX 1080 or GTX 1060, stay in Creator Mode
  • Benefits for Rise of the Tomb Raider, Rocket League and GTA
  • Benefit more at 1080p, but still gains at 4K

The pros and cons of enabling Game Mode are meant to be along the lines of faster and lower latency gaming, at the expense of raw compute power. The fact that it requires a reboot to switch between Creator Mode and Game Mode is a main detractor - if it were a simple in-OS switch, then it could be enabled for specific titles on specific graphics cards just before the game is launched. This will not ever be possible, due to how PCs decide what resources are available when. That being said however, perhaps AMD has missed a trick here.

Could AMD have Implemented Game Mode Differently?

By virtue of misinterpreting AMD's slide deck, and testing a bunch of data with SMT disabled instead, we have an interesting avenue as to how users might do something akin to Game Mode but not specifically AMD's game mode. This also leads to the question if AMD implemented and labeled the Game Mode environment in the right way.

By enabling NUMA and disabling SMT, the 16C/32T chip moves down to 16C/16T. It still has 16 full cores, but has to deal with communication across the two eight-core silicon dies. Nonetheless it still satisfies the need for cores to access the lowest latency memory near to that specific core, as well as enabling certain games that need fewer total threads to actually work. It should, by the description alone, enable the 'legacy' part of legacy gaming.

The underlying performance analysis between the two modes becomes this:

When in 16C/16T mode, performance in CPU benchmarks was higher than in 8C/16T mode.
When in 16C/16T mode, performance in CPU gaming benchmarks was higher than in 8C/16T mode. 

Some of the suggestions from comparing AMD's defined 8C/16T Game Mode for CPU gaming actually change when in 16C/16T mode: games that saw slight regressions with 8 cores became neutral at 16C or even had slight improvements, particularly at 1080p.

One of the main detractors to the 8C/16T mode is that it requires a full restart in order to enable it. Disabling SMT could theoretically be done at the OS level before certain games come in to play. If the OS is able to determine which core IDs are associated to standard threads and which ones would be hyperthreads, it is perhaps possible for the OS just to refuse to dispatch threads in flight to the hyperthreads, allowing only one thread per core. (There's a small matter of statically shared resources to deal with as well.) The mobile world deals with thread migration between fast cores and slow cores every day, and some cores can be hotplug disabled on the fly. One could postulate that Windows could do something similar with the equivalent of hyperthreads.

Would this issue need to be solved by Windows, or by AMD? I suspect a combination of both, really. 

Update:

Robert Hallock on AMD's Threadripper webcast has stated that Windows Scheduler is not capable of specifically zeroing out a full die to have the same effect. The UMA/NUMA implementation can be managed with Windows Scheduler to assign threads to where the data is (or assign data to where the threads are), but as far as fully disabling a die in the OS requires a restart.

 

Related Reading

Analyzing Creator Mode and Game Mode
POST A COMMENT

104 Comments

View All Comments

  • ddriver - Friday, August 18, 2017 - link

    Why not? We've had 16 core CPUs long before W10 was launched, and it has allegedly been heavily updated since then.

    But it is NOT the "coder"'s responsibility. Programmers don't get any say, they are paid workers, paid to do as they are told. Not that I don't have the impression that a lot of the code that's being written is below the standard, but the actual decision making is not a product of software programmers but that of software architects, and the latter are even more atrocious than the actual programmers.
    Reply
  • HollyDOL - Friday, August 18, 2017 - link

    Sadly, the reality is much worse... those architects are ordered by managers, economic persons etc. who, sadly often, don't know more about computer than where's power button. And they want products with minimal cost and 'yesterday was late'. Reply
  • ddriver - Friday, August 18, 2017 - link

    Well, yeah, the higher you go up the latter the grosser the incompetence level. Reply
  • BrokenCrayons - Thursday, August 17, 2017 - link

    Interesting test results. I think they demonstrate pretty clearly why Threadripper isn't really a very good option for pure gaming workloads. The big takeaway is that there are more affordable processors with lower TDPs offer comparable or better performance without adding additional settings that few people will realize exist and even fewer people will fiddle with enough to determine which settings actually improve performance in their particular software library. The Ryzen 7 series is probably a much better overall choice than TR right now if you don't have specific tasks that require all those cores and threads. Reply
  • Gothmoth - Thursday, August 17, 2017 - link

    "I think they demonstrate pretty clearly why Threadripper isn't really a very good option for pure gaming workloads."

    wow.... what a surprise.
    thanks for pointing that out mr. obvious. :-)
    Reply
  • Gigaplex - Thursday, August 17, 2017 - link

    These are single GPU tests. Threadripper has enough PCIe lanes to do large multi GPU systems. More GPU usually trumps better CPU in the high end gaming scene, especially with 4k resolution. Reply
  • BrokenCrayons - Friday, August 18, 2017 - link

    Yes, but multi-GPU setups are generally not used for gaming-centric operations. There's been tacit acknowledgement of this as the state of things by NV since the release of the 10x0 series. Features like Crossfire and SLI support are barely a bullet point in marketing materials these days. With good reason since game support is waning as well and DX12 is positioned to pretty nail the multi-GPU coffin shut entirely except in corner cases where it MIGHT be possible to leverage an iGPU alongside a dGPU if a game engine developer bothers to invest time into banging out code to support it. That places TR's generous PCIe lane count and the potential multi-GPU usage in the domain of professional workloads that need GPU compute power. Reply
  • Bullwinkle J Moose - Thursday, August 17, 2017 - link

    I agree with ddriver

    We should not have to fiddle with the settings and reboot to game mode on these things

    Windows should handle the hardware seamlessly in the background for whatever end use we put these systems to

    The problem is getting Microsoft to let the end users use the full potential of our hardware

    If the framework for the hardware is not fully implemented in the O.S., every "FIX" looks a bit like the one AMD is using here

    I think gaming on anything over 4 cores might require a "proper" update from Microsoft working with the hardware manufacturers

    Sometimes it might be nice to use the full potential of the systems we have instead of Microsoft deciding that all of our problems can be fixed with another cloud service
    Reply
  • Gothmoth - Thursday, August 17, 2017 - link

    but but.. what about linux.

    i mean linux is the savior, not?
    it has not won a 2.2% marketshare on teh desktop for nothing.

    sarcasm off....
    Reply
  • HomeworldFound - Thursday, August 17, 2017 - link

    What can we expect Microsoft to do prior to a product like this launching. If a processor operates in a manner that requires the operating system to be adjusted, the company selling it needs to approach Microsoft and provide an implementation, and it should be ready for launch. If that isn't possible then why manufacture something that doesn't work correctly and requires hacky fixes to run. Reply

Log in

Don't have an account? Sign up now