Marrying Vega and Zen: The AMD Ryzen 5 2400G Review
by Ian Cutress on February 12, 2018 9:00 AM ESTBenchmarking Performance: CPU Rendering Tests
Rendering tests are a long-time favorite of reviewers and benchmarkers, as the code used by rendering packages is usually highly optimized to squeeze every little bit of performance out. Sometimes rendering programs end up being heavily memory dependent as well - when you have that many threads flying about with a ton of data, having low latency memory can be key to everything. Here we take a few of the usual rendering packages under Windows 10, as well as a few new interesting benchmarks.
All of our benchmark results can also be found in our benchmark engine, Bench.
Corona 1.3: link
Corona is a standalone package designed to assist software like 3ds Max and Maya with photorealism via ray tracing. It's simple - shoot rays, get pixels. OK, it's more complicated than that, but the benchmark renders a fixed scene six times and offers results in terms of time and rays per second. The official benchmark tables list user submitted results in terms of time, however I feel rays per second is a better metric (in general, scores where higher is better seem to be easier to explain anyway). Corona likes to pile on the threads, so the results end up being very staggered based on thread count.
Blender 2.78: link
For a render that has been around for what seems like ages, Blender is still a highly popular tool. We managed to wrap up a standard workload into the February 5 nightly build of Blender and measure the time it takes to render the first frame of the scene. Being one of the bigger open source tools out there, it means both AMD and Intel work actively to help improve the codebase, for better or for worse on their own/each other's microarchitecture.
LuxMark v3.1: Link
As a synthetic, LuxMark might come across as somewhat arbitrary as a renderer, given that it's mainly used to test GPUs, but it does offer both an OpenCL and a standard C++ mode. In this instance, aside from seeing the comparison in each coding mode for cores and IPC, we also get to see the difference in performance moving from a C++ based code-stack to an OpenCL one with a CPU as the main host.
POV-Ray 3.7.1b4: link
Another regular benchmark in most suites, POV-Ray is another ray-tracer but has been around for many years. It just so happens that during the run up to AMD's Ryzen launch, the code base started to get active again with developers making changes to the code and pushing out updates. Our version and benchmarking started just before that was happening, but given time we will see where the POV-Ray code ends up and adjust in due course.
Cinebench R15: link
The latest version of CineBench has also become one of those 'used everywhere' benchmarks, particularly as an indicator of single thread performance. High IPC and high frequency gives performance in ST, whereas having good scaling and many cores is where the MT test wins out.
Conclusions on Rendering: It is clear from these graphs that most rendering tools require full cores, rather than multiple threads, to get best performance. The exception is Cinebench.
177 Comments
View All Comments
Hurr Durr - Monday, February 12, 2018 - link
I don`t care about these. I want to see how AMD is holding up in notebooks, 15W range specifically.Hul8 - Monday, February 12, 2018 - link
For a low-end graphics part like this, it would be really interesting to have a section in the review exploring the "comfortable" settings in various games.It could be really useful information for potential buyers to know what kind of settings they'd need to run in a game to reach their preferred performance level (99th percentile), whether it's 30, 45 or 60 fps, and also to know if a product simply can't reach certain performance no matter how low you turn the settings.
DrizztVD - Monday, February 12, 2018 - link
Why do you only report total power consumption? I'd like to see power efficiency!!! Since I don't know what the performance per CPU is, these power measurements mean almost nothing. Also, the efficiency will change with the workload, so Prime95 is a very one-dimensional test of efficiency. Look at your power measurement graphs: they tell you what we already know - single core speeds are lower for Ryzen, and lower TDP CPUs use less power. That's kinda duh...JHBoricua - Monday, February 12, 2018 - link
I'm confused. The AMD vs. AMD section led me to believe there was going to be a comparison of Raven Ridge against Bristol Ridge APUs, which makes sense as it would have allow the use of the same motherboard for both APUs, even if the Bristol Ridge DDR4 memory was clocked slower. But then actual benchmarks is showing Kaveri parts?prtskg - Tuesday, February 13, 2018 - link
Kaveri was better at gaming/performance than Bristol. The latter had the advantage of efficiency.nwarawa - Monday, February 12, 2018 - link
Comparing with a competitive Intel platform with dGPU is kinda tricky right now. It's not just the dGPUs that are ridiculously priced right now. RAM is too. And to maximize performance on the R7 2400G, you WILL need to spend more than the basic $90 8GB 2400 kit. The cheapest 16GB Samsung b-die 3200 kit I found was $220. And you will want to go with a 16GB kit, because already some newer games use more than 8GB, and they use MORE when using graphics cards with less than 4GB. The iGPU takes some of that 8GB for itself...if runs out of system RAM, it has to use your system disk... enjoy the single digit frame rates...Here is what I found on newegg:
INTEL
$130 - i3-8100
$90 - 8GB 2400 RAM (or 170 for 16GB)
$120 - Z370 motherboard (no mainsteam chipset YET)
===
$340
AMD
$170 - R5 2400g
$220 - 16GB 3200C14 RAM
$80 - Motherboard(cheapest decently reviewed AM4)
===
$470
The intel system is a full $130 cheaper (or $50 if you spring for the 16GB), and that gap will only increase with the upcoming cheaper chipsets and/or upcoming coffee lake models. Now, I haven't included the dGPU yet, but the GTX1050 2GB currently goes for $150 - making the Intel system total only $20 more than than AMD system, and running rings around it in most games (although neither would be ideal for the latest games... the 2GB-GPU/8GB-SYSRAM Intel system would run out of memory and the Vega 11 just doesn't have the horsepower).
What would put things in favor of AMD would be if they made clear that the iGPU would still be of use when using a dGPU (such as with the new "ThueAudioNext") in the future.
What I would REALLY like to see, though, is AMD use the beefier Vega iGPUs Intel is using with their own 12nm Zen+ chips and slap on some HBM memory. THAT I could go for.
oleyska - Monday, February 12, 2018 - link
Jeez,1.\ HBM on budget chips pushes them into 250$ range by just adding HBM.
2.\ IGP solutions are not a GTX1060 replacement, it's not magic.
System comparison:
I3 8100 is inferior in cpu tasks.
it has half the memory (intel IGP still uses memory you know?, it's configurable on both systems.
The benchmarked system runs at 2933, not 3200 so o.0
it has inferior gpu 3X~
even at same memory speeds it would still be 2X as slow, slower cpu.
So what is the point of the argument ?
It not needed to have the extra memory frequency but if you want to replace a 80$ dedicated gpu you need to and definitely add 8gb extra memory and that's where the cost comes into place as a valid comparison if you subtract 20$ from AMD for 2933 ~
add 80-100$ for GT1030 you still end up with an intel rig with higher system power consumption, equal gaming performance, inferior cpu and you will have to buy a G-sync monitor if you want tearing free monitor while freesync is thrown at ya at any price range as an added bonus.
systems are comparable and Intel's I3 line is destroyed along amd's old R3 line too.
I5,I7 and R5 stands tall still and the R7 has it's place at times too.
nwarawa - Monday, February 12, 2018 - link
Did you not even read my post? Or the review for that matter? Did you think AT ALL about the real life application of anything said before posting? "I3 8100 is inferior in cpu tasks" WHAT tasks? I'll answer that for you. Rendering. If you are trying to get the cheapest CPU possible for rendering with as little RAM as possible to shave as much money off as possible... you are doing it wrong. Since we (or at least I am... not sure what you are going on about) are talking about GPUs, you can safely assume we are concerned about GAMES."it has half the memory" no s--t, sherlock, read the post again (or for the first time, apparently)
I could go on, but apparently you were just scouring the posts for someone to disagree with with a pre-defined little rant, so I won't bother.
Fritzkier - Monday, February 12, 2018 - link
Why you need 16GB? if you bought Ryzen APU, you probably only plays e-sport title anyway and some older games... E-sport titles doesn't need a huge RAM. And it already crush the intel counterparts, both in performance and price.You guys from first world countries are always complaining. Jeez. Try to live at poorer countries like South East Asia.
serendip - Tuesday, February 13, 2018 - link
An APU at $100 would still be expensive, especially when people in the developing world build machines with Pentium G chips. The speedy APU graphics would negate the need for a low-end discrete GPU though.