The AMD Ryzen Threadripper 1950X and 1920X Review: CPUs on Steroids
by Ian Cutress on August 10, 2017 9:00 AM ESTCPU 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.
Corona loves threads.
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.
Blender loves threads and memory bandwidth.
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.
Like Blender, LuxMark is all about the thread count. Ray tracing is very nearly a textbook case for easy multi-threaded scaling. Though it's interesting just how close the 10-core Core i9-7900X gets in the CPU (C++) test despite a significant core count disadvantage, likely due to a combination of higher IPC and clockspeeds.
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.
Similar to LuxMark, POV-Ray also wins on account of threads.
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.
Intel recently announced that its new 18-core chip scores 3200 on Cinebench R15. That would be an extra 6.7% performance over the Threadripper 1950X for 2x the cost.
Facebook
Twitter
RSS
347 Comments
View All Comments
blublub - Sunday, August 13, 2017 - link
From what I have read is that all TR do 3.9hhz and some even 4-4.1ghz on all cores .What are your temp when running all 10c @4.6ghz prime for 1-2hrs
Zingam - Sunday, August 13, 2017 - link
Ian, how about testing mobile CPUs - for games and for office work. Aren't mobile CPUs selling much larger numbers thatn desktop ones these days?I can't find a single benchmark comparing i5-7300hq vs i7-7700hq vs i7-7700K showing the difference in productivity workloads and not just for rendering pretty pictures but also for more specific tasks as compiling software etc.
I also would like to see some sort of comparison of new generation to all generations upto 10 years back in time. I'd like to know how much did performance increase since the age of Nehelem. At least from now on there should be a single test to display the relative performance increase over the last few generations. The average user doesn't upgrade their PC every year. The average user maybe upgrades every 5 years and it is really difficult to find out how much peformance increase would one get with an upgrade.
SanX - Sunday, August 13, 2017 - link
I agree, there must be 5-7 years old processors in the chartsSanX - Sunday, August 13, 2017 - link
Why one core of Apple A10 costs $10 but one core of Intel 7900x costs 10x more?oranos - Sunday, August 13, 2017 - link
so its complete dogsh*t for the segment which is driving the PC market right now: gaming. got it.ballsystemlord - Sunday, August 13, 2017 - link
Hey Ian, you've been talking about anandtech's great database where we can see all the cool info. Well, according to your database the Phenom II 6 core 1090T is equally powerful when compared to the 16 core threadripper!!!!!!! http://www.anandtech.com/bench/product/1932?vs=146With those sorts of numbers why would anyone plan an upgrade?
(And there is also only one metric displayed, strange!)
Not to play the Intel card on you as others do, but this is a serious problem for at least the AMD lineup of processors.
jmelgaard - Monday, August 14, 2017 - link
o.O... I don't know how you derived that conclusion? you need a guide on how to read the database?...BurntMyBacon - Monday, August 14, 2017 - link
For anyone looking for an overall fps for two pass encoding here is your equation (hope my math is correct):FPS = 2*FPS1*FPS2/(FPS2+FPS1)
No, you can't just average the FPS scores from each pass as the processor will spend more time in the slower pass.
For the x264 encoding test, for example, a few relevant FPS scores end up being:
i9-7900X: 122.56
i7-7820X: 114.37
i7-6900K: 95.26
i7-7740X: 82.74
TR-1950X: 118.13
TR-1950X(g): 117.00
TR-1920X: 111.74
R7-1800X: 100.19
Since two pass encoding requires both passes to be usable, getting an overall FPS score seems somewhat relevant. Alternately, using time to completion is would present the same information in a different manner. Though, it would be difficult to extrapolate performance results to estimate performance in other encodes without also posting the number of frames encoded.
goldgrenade - Thursday, January 4, 2018 - link
Take all those Intel FPS performance counters and multiply them by .7 and you have what their chips actually run at without a major security flaw in them.Let's see that would be...
i9-7900X: 85.792
i7-7820X: 80.059
i7-6900K: 66.682
i7-7740X: 57.918
And that's at best. It can be up to 50% degradation when rendering or having to do many small file accesses or repeated operations with KAISER.
Gastec - Tuesday, August 15, 2017 - link
I've having a hard time trying to swallow "Threadripper is a consumer focused product" line considering the prices to "consume" it: $550 for the MB, $550 for the TR1900X ($800 or $1000 for the others is just dreaming) then the RAM. The MB(at least the Asus one) should be $200 less, but I get it, they are trying to squeeze as much as possible from the...consumers. Now don't get me wrong and I mean no offence for the rich ones among you, but those CPU are for Workstations. WORK, not gamestations. Meaning you would need them to help you make your money, faster.