Retesting AMD Ryzen Threadripper’s Game Mode: Halving Cores for More Performance
by Ian Cutress on August 17, 2017 12:01 PM ESTCPU Web Tests
One of the issues when running web-based tests is the nature of modern browsers to automatically install updates. This means any sustained period of benchmarking will invariably fall foul of the 'it's updated beyond the state of comparison' rule, especially when browsers will update if you give them half a second to think about it. Despite this, we were able to find a series of commands to create an un-updatable version of Chrome 56 for our 2017 test suite. While this means we might not be on the bleeding edge of the latest browser, it makes the scores between CPUs comparable.
All of our benchmark results can also be found in our benchmark engine, Bench.
SunSpider 1.0.2: link
The oldest web-based benchmark in this portion of our test is SunSpider. This is a very basic javascript algorithm tool, and ends up being more a measure of IPC and latency than anything else, with most high-performance CPUs scoring around about the same. The basic test is looped 10 times and the average taken. We run the basic test 4 times.
Mozilla Kraken 1.1: link
Kraken is another Javascript based benchmark, using the same test harness as SunSpider, but focusing on more stringent real-world use cases and libraries, such as audio processing and image filters. Again, the basic test is looped ten times, and we run the basic test four times.
Google Octane 2.0: link
Along with Mozilla, as Google is a major browser developer, having peak JS performance is typically a critical asset when comparing against the other OS developers. In the same way that SunSpider is a very early JS benchmark, and Kraken is a bit newer, Octane aims to be more relevant to real workloads, especially in power constrained devices such as smartphones and tablets.
WebXPRT 2015: link
While the previous three benchmarks do calculations in the background and represent a score, WebXPRT is designed to be a better interpretation of visual workloads that a professional user might have, such as browser based applications, graphing, image editing, sort/analysis, scientific analysis and financial tools.
Overall, all of our web benchmarks show a similar trend. Very few web frameworks offer multi-threading – the browsers themselves are barely multi-threaded at times – so Threadripper's vast thread count is underutilized. What wins the day on the web are a handful of fast cores with high single-threaded performance, and it becomes a balance between cores and cross-core communication.
104 Comments
View All Comments
peevee - Friday, August 18, 2017 - link
Of course. Work CPUs must be tested at work. Kiddies are fine with i3s.Ian Cutress - Sunday, August 20, 2017 - link
https://myhacker.net hacking news hacking tutorials hacking ebooksIGTrading - Thursday, August 17, 2017 - link
It would be nice and very useful to post some power consumption results at the platform level, if we're doing "extra" additional testing.It is very important since we're paying for the motherboard just as much as we pay for a Ryzen 5 or even Ryzen 7 processor.
And it will correctly compare the TCO of the X399 platform with the TCO of X299.
jordanclock - Thursday, August 17, 2017 - link
So it looks like AMD should have gone with just disabling SMT for Game Mode. There are way more benefits and it is easier to understand the implications. I haven't seen similar comparisons for Intel in a while, perhaps that can be exploration for Skylake-X as well?HStewart - Thursday, August 17, 2017 - link
I would think disable SMT would be better, but the reason maybe in designed of link between the two 8 Core dies on chip.GruenSein - Thursday, August 17, 2017 - link
I'd really love to see a frame time probability distribution (Frame time on x-axis, rate of occurrence on y-axis). Especially in cases with very unlikely frames below a 60Hz rate, the difference between TR and TR-GM/1800X seem most apparent. Without the distribution, we will never know if we are seeing the same distribution but slightly shifted towards lower frame rates as the slopes of the distribution might be steep. However, those frames with frame times above a 60Hz rate might be real stutters down to a 30Hz rate but they might just as well be frames at a 59,7Hz rate. I realize why this threshold was selected but every threshold is quite arbitrary.MrSpadge - Thursday, August 17, 2017 - link
Does AMD comment on the update? What's their reason for choosing 8C/16T over 16C/16T?> One could postulate that Windows could do something similar with the equivalent of hyperthreads.
They're actually already doing that. Loading 50% of all threads on an SMT machine will result in ~50% average load on every logical core, i.e. all physical cores are only working on 1 thread at a time.
I know mathematically other schedulings are possible, leading to the same result - but by now I think it's common knowledge that the default Win scheduler works like that. Hence most lightly threaded software is indifferent to SMT. Except games.
NetMage - Sunday, August 20, 2017 - link
Then why did SMT mode show differences from Creator mode in the original review?Dribble - Thursday, August 17, 2017 - link
No one is ever going to run game mode - why buy a really expensive chip and then disable half of it, especially as you have to reboot to do it? It's only use is to make threadripper look slightly better in reviews. Imo it would be more honest as a reviewer to just run it in creator mode all the time.jordanclock - Thursday, August 17, 2017 - link
The point is compatibility, as mentioned in the article multiple times. AMD is offering this as an option for applications (mainly games) that do not run correctly, if at all, on >16 core CPUs.