The Skylake Core i3 (51W) CPU Review: i3-6320, i3-6300 and i3-6100 Testedby Ian Cutress on August 8, 2016 9:00 AM EST
Real World Performance
The dynamics of CPU Turbo modes, both Intel and AMD, can cause concern during environments with a variable threaded workload. There is also an added issue of the motherboard remaining consistent, depending on how the motherboard manufacturer wants to add in their own boosting technologies over the ones that Intel would prefer they used. In order to remain consistent, we implement an OS-level unique high performance mode on all the CPUs we test which should override any motherboard manufacturer performance mode.
All of our benchmark results can also be found in our benchmark engine, Bench.
Dolphin Benchmark: link
Many emulators are often bound by single thread CPU performance, and general reports tended to suggest that Haswell provided a significant boost to emulator performance. This benchmark runs a Wii program that raytraces a complex 3D scene inside the Dolphin Wii emulator. Performance on this benchmark is a good proxy of the speed of Dolphin CPU emulation, which is an intensive single core task using most aspects of a CPU. Results are given in minutes, where the Wii itself scores 17.53 minutes.
Skylake saw a good boost with our Dolphin benchmark, and it remains a single-thread driven event. The 100 MHz difference between the i3-6320 and i3-6300 amounts to just over a second difference, but one step back to the i3-6100 is another 42 seconds. This is one of the few instances where the 3 MB of L3 cache on the i3-6100 makes a significant difference. The i5-6600 sits in the middle of our i3 parts due to its lower CPU frequency but higher cache.
WinRAR 5.0.1: link
Our WinRAR test from 2013 is updated to the latest version of WinRAR at the start of 2014. We compress a set of 2867 files across 320 folders totaling 1.52 GB in size – 95% of these files are small typical website files, and the rest (90% of the size) are small 30 second 720p videos.
WinRAR is a variable thread test, so the CPUs with more threads all pull ahead of the i3 parts. Again, we see a crescendo from the i3-6320 to the i3-6300 and the i3-6100, with the i3-6100 being slightly further out due to its reduced L3.
3D Particle Movement v2
The second version of this benchmark is similar to the first, however it has been re-written in VS2012 with one major difference: the code has been written to address the issue of false sharing. If data required by multiple threads, say four, is in the same cache line, the software cannot read the cache line once and split the data to each thread - instead it will read four times in a serial fashion. The new software splits the data to new cache lines so reads can be parallelized and stalls minimized. As v2 is fairly new, we are still gathering data and results are currently limited.
The updated 3DPM benchmark likes good cache management and a high frequency with lots of threads. As from our Carrizo review, the AMD X4 845 does really well here, beating the i3-6100TE which has a much lower frequency. For the i3 parts, we see a regular staircase between the three, outpacing the much older FX-6350.
On the lower end processors, general usability is a big factor of experience, especially as we move into the HTML5 era of web browsing.
Kraken also enjoys Skylake CPUs, with a preference for high frqeuency parts as well. The regular staircase is in effect between the three, showing how L3 cache makes little difference here.
Google Octane v2
Octane has a similar performance to Kraken, with the Skylake CPUs out in top with the staircase results. The i5-6600 sits in the middle, despite having four physical cores, but restrained to four threads. Because of this and our regular staircase, we can see that the L3 cache plays little role here and the threads are very lightweight. The resources on the FX CPUs unfortunately do not do well here.
Post Your CommentPlease log in or sign up to comment.
View All Comments
tipoo - Monday, August 8, 2016 - linkLooks like even a Skylake i3 may be able to retire the venerable 2400/2500K, higher frame rates and better frame times at that. However a native quad does prevent larger dips.
Kevin G - Monday, August 8, 2016 - linkI have a feeling much that is due to the higher base clock on the SkyLake i3 vs. the i5 2500K. Skylake's IPC improvements also help boost performance here too.
The real challenge is if the i3 6320 can best the i5 2500k as the same 3.9 Ghz base clock speed. Sandy Bridge was a good overclocker so hitting those figures shouldn't be difficult at all.
tipoo - Monday, August 8, 2016 - linkThat's true, overclocked the difference would diminish. But you also get modernities like high clocked DDR4 in the switchover.
At any rate, funny that a dual core i3 can now fluidly run just about everything, it's two cores are probably faster than the 8 in the current consoles.
Lolimaster - Monday, August 8, 2016 - linkBenchrmarks don't tell you about the hiccups when playing with a dual core. Specially with things like Crysis 3 or even worse ROt Tomb Raider where you get like half the fps just by using a dual core bs a cheapo Athlon 860K.
gamerk2 - Monday, August 8, 2016 - linkThat's why Frame Times are also measured, which catches those hitches.
Samus - Tuesday, August 9, 2016 - linkI had a lot of issues with my Sandy Bridge i3-2125 in Battlefield 3 circa 2011 with lag and poor minimum frame rates.
After long discussions on the forums, it was determined disabling hyper threading actually improved frame rate consistency. So at least in the Sandy Bridge IPC, and probably dating back to Nehalem or even Prescott, Jackson Technology or whatever you want to call it, has a habit of stalling the pipeline if there are too many cache misses to complete the instruction. Obviously more cache resolves this, so the issue isn't as prominent on the i7's, and it would certainly explain why the 4MB i3's are more consistent performers than the 3MB variety.
Of course the only way to prove if hyper threading is causing performance inconsistency is to disable it. It'd be a damn unique investigation for Anandtech to do a IPC improvement impact on it's affect on hyper-threading performance over the years, perhaps even dating back to the P4.
AndrewJacksonZA - Wednesday, August 10, 2016 - linkHOW ON EARTH DID I MISS THIS?!?!
Thank you for introducing me to Intel's tech known as "Jackson!" This is now *SO* on my "To Buy" list!
Thank you Samus! :-D
bug77 - Monday, August 8, 2016 - linkNeah, I went i5-2500k -> i5-6600k and there's no noticeable difference. The best part of the upgrade was those new I/O ports on the new motherboard, but it's a sad day when you upgrade after 4 years and the most you have to show is you new M2 or USB 3.1 ports (and USB 3.1 is only added through a 3rd party chip).
Sure, if I bench it, the new i5 is faster, but since the old i5 wasn't exactly slow, I can't say that I see a significant improvement.
Now, if you mean that instead of getting an i5-2500k one can now look at a Skylake i3, I'm not going to argue with you there. Though (money permitting) the boost speed might be nice to have anyway.
Cellar Door - Monday, August 8, 2016 - linkThis is a poorly educated comment:
a) Your perceived speed might be limited by your storage
b) You don't utilize your cpu's multitasking abilities fully(all cores)
Duckeenie - Monday, August 8, 2016 - linkWhy did you continue to post your comment if you believed you were making poorly educated points?