CPU Benchmarks

The dynamics of CPU Turbo modes, with 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 the CPU manufacturer 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.

HandBrake v0.9.9: link

For HandBrake, we take two videos (a 2h20 640x266 DVD rip and a 10min double UHD 3840x4320 animation short) and convert them to x264 format in an MP4 container. Results are given in terms of the frames per second processed, and HandBrake uses as many threads as possible.

HandBrake v0.9.9 LQ Film

HandBrake v0.9.9 2x4K

The variable turbo speeds of the CPUs results in a small difference in low quality conversion, and the high single core frequency of the 4790K wins there. For 4K conversion the problem becomes more parallel and the extra cores of the 5960X push it ahead of the pack. The 5930K and 5820K are both behind the 4960X however.

Agisoft Photoscan – 2D to 3D Image Manipulation: link

Agisoft Photoscan creates 3D models from 2D images, a process which is very computationally expensive. The algorithm is split into four distinct phases, and different phases of the model reconstruction require either fast memory, fast IPC, more cores, or even OpenCL compute devices to hand. Agisoft supplied us with a special version of the software to script the process, where we take 50 images of a stately home and convert it into a medium quality model. This benchmark typically takes around 15-20 minutes on a high end PC on the CPU alone, with GPUs reducing the time.

Agisoft PhotoScan Benchmark - Total Time

Photoscan's four separate components rely on different amounts of high frequency vs. many cores: check our Bench database for more detailed results but overall the 5960X comes out on top. That being said, the 5820K is less than 40% of the price and is only 1.2 minutes behind.

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.

Dolphin Emulation Benchmark

Dolphon loves single core speed and efficiency, meaning the 4790K wins out again. Interestingly the large L3 cache of the 5960X also helps here against the 5820K, despite the 5820K having a higher single thread frequency.

WinRAR 5.0.1: link

WinRAR 5.01, 2867 files, 1.52 GB

WinRAR is a variable thread workload, but more cores still wins out. Interestingly the xx60X CPUs are ahead of the xx30K CPUs followed by the xx20K. After this comes the 4790K, and then the 990X on par, showing how far three generations of Intel CPU have developed.

PCMark8 v2 OpenCL

A new addition to our CPU testing suite is PCMark8 v2, where we test the Work 2.0 and Creative 3.0 suites in OpenCL mode.

PCMark8 v2 Work 2.0 OpenCL with R7 240 DDR3

PCMark8 v2 Creative 3.0 OpenCL with R7 240 DDR3

PCMark v8 relies on a number of factors, and it would seem that frequency is preferred over cache and memory. Interestingly the 4930K beat the 4960X in the Creative Suite with no obvious explanation.

Hybrid x265: link

Hybrid is a new benchmark, where we take a 4K 1500 frame video and convert it into an x265 format without audio. Results are given in frames per second.

Hybrid x265, 4K Video

Converting 4K video gets another step in the preference for more cores in Hybrid x265. The 5820K matches the 3960X, showing the progression of CPU generational development.

Cinebench R15

Cinebench R15 - Single Threaded

Cinebench R15 - Multi-Threaded

3D Particle Movement

3DPM is a self-penned benchmark, taking basic 3D movement algorithms used in Brownian Motion simulations and testing them for speed. High floating point performance, MHz and IPC wins in the single thread version, whereas the multithread version has to handle the threads and loves more cores.

3D Particle Movement: Single Threaded

3D Particle Movement: MultiThreaded

FastStone Image Viewer 4.9

FastStone is the program I use to perform quick or bulk actions on images, such as resizing, adjusting for color and cropping. In our test we take a series of 170 images in various sizes and formats and convert them all into 640x480 .gif files, maintaining the aspect ratio. FastStone does not use multithreading for this test, and results are given in seconds.

FastStone Image Viewer 4.9

FastStone is a purely single threaded exercise, showing here how the lower core CPUs with high turbo perfom best, and by quite a margin.

Power Consumption, Test Setup Gaming Benchmarks
POST A COMMENT

203 Comments

View All Comments

  • swing848 - Thursday, September 04, 2014 - link

    Regarding your statement about game benchmarks, "It makes sense that we should test this with 4K in the future".

    You did not say how far into the future, in the near future it makes NO since. The video card you used was an 770; there is no way that GPU can handle 4K at high game settings, even medium settings will bring it to it's knees. First of all it is a mid-range GPU, secondly, it does not have enough local memory.

    Lower resolutions more indicate what the CPU can do because the GPU is not overtaxed therefore not becoming a factor.

    If you change anything, pick a higher end video card, to make sure the GPU is not bottlenecking tests.
    Reply
  • M.Q.Leo - Saturday, September 06, 2014 - link

    This generation has not much improvement I think. Especially 5820K, even less PCI-E Lanes there is! :( Reply
  • djemir - Monday, September 08, 2014 - link

    I owned a first generation 2008 I7 running at a measly 2.7Ghz and I can tell you the new motherboard alone having 6GB/s SATA and 10GB/sec SATA interface made up for speed losses not to mention the 4790K running at 4.5GHz on a stable overclock. For photoshop and video the new CPU and motherboard have made a world of difference. Save time and open / read time has been reduced from 3-15 minutes per file down to "wait what I can count it on fingers???!" 7 seconds. That means no more "oh well it's saving let me go to the bathroom or something while I wait for this slow a$$ computer. I'm kind of wanting to kick myself for buying a z97 on the very day that the new x99's came out but when I did a price check it just wasn't worth it. This will hold me over til those crazy prices drop. I looked at bench marks for the 4790k vs the lower spectrum of the newer x99's and it looks like the 4790K does better in Photoshop than the X99 due to higher frequency. I even had to drop 8GB of Ram because my old motherboard had 6 slots and was holding 24GB of Ram in 4GB sticks now my new motherboard only has 4 slots. That made me sad. But even with less RAM the motherboard and processor are much more efficient and they can actually use the higher speeds of my SSDs and my 3GB/s and 6GB/s internal and external hard drives as well. Everyone arguing that their old processors are amazing need to open their eyes. On paper it all sounds like they are very much equal. But people forget that motherboards have been improving as well. I was having consistent blue screen crashes on my old system eve after refreshing the system a few times. This new configuration (Asus with i7-4790k) runs like a champ no blue screens at all, nothing but blue skies. I think Intel needs to just drop their prices a little, I would have prefered to get the 8 core i7 or 12 core xeon Yes that sucker is out there as well but at $1000-2500 for these units it's just not worth the small amounts of improvements vs the price. Whereas the huge increase i felt in performance at a much lower price point was worth the upgrade. Reply
  • untoreh - Friday, September 12, 2014 - link

    What about benching some games that have decent multithreading? Like games on the bitsquid engine like War of the roses/war of the vikings? Or Natural Selection 2 with its immensive poligon count? No. Lets just benchmark GPU Heavy AAA titles that generally push the GPU Market more than the gaming market. If you wanna benchmark a 8 core CPU with games, you should AT LEAST let them be half with decent MT support. Reply
  • IUU - Thursday, September 18, 2014 - link

    Well, except for gaming..
    Mostly true but not completely true.
    If you play chess, the 8 core version will smoke any 4 core version just for fun.
    There are other games that are not in the front scene right now that are mainly cpu demanding.
    Any fps game currently in the market, is heavily based on showing realism, so it requires graphics processing power,and demanding from a powerful cpu to run such a game faster than the fractions of a second a slower cpu would, is well.. pointless.
    But gaming is vast , bigger than your encoding software, your bitcoin mining, and your much advertised enterprise software. When developed further, it will require what the most demanding scientific applications require and probably more. See that it is already a main driving force(if not the main driving force) of modern supercomputer improvement. And it will be for the future.
    Think of ingame ais, multiple ais, that will interact with the game world like a human would. Think of voice and pattern recognition, of tracking thousands or millions of objects, etc etc.
    If your only aspiration in buying such a cpu is how good it will run current gen games, you wouldn't ever belong in the category of appropriate customers for this cpu. You would rather be excited by the mobile parts, which while anemic compared to the 8-core haswell, are fancy and fashionable and satisfy your vanity.
    Of course it still remains as a problem, because of this sad market turn, the rise in the pricing of the "extreme" parts which puts the most of us off.
    Reply
  • kelendar - Tuesday, September 23, 2014 - link

    The one thing I like about the X99 chipset over my ASUS P8P67 Deluxe are the plethora of SATAIII ports and vastly improved onboard sound. The problem is the SB is a dead platform; there is no upgrade path with it and all boards ever made have few ports. I actually had to buy a separate add-in card but it suffers from being on the PCI-8x port - connect more than 1 drive, and they share the throughput.

    That's why I'm looking at this build. I figure the ASUS Rampage IV + 5820 + 16 GB RAM should set me back about $1,100 but it gives me a bit of future-proofing.
    Reply
  • Spartan 363 - Wednesday, October 01, 2014 - link

    I just purchased the i7 5930K along with an EVGA X99 Classified mobo with 16 GB of Corsair Dominator Platinum DDR4 3000 RAM for now along with 2 EVGA GTX 980 Classified for SLI. I purchased both the Classified versions of the mobo and GTX 980's because I love tinkering and overclocking to see what is the best stable clocks I can achieve without heating up my bedroom like my PC were a space heater which would be handy in a month or so as the cold weather returns to Northern VA where I live. My current PC and even my older PC that this build replaced are air cooled for now. I will buy WC blocked and such for my GTX 980's since I have fantastic components for overclocking with the Classified things from EVGA and a CPU that OC's well too on air.

    Sure,this was a very expensive upgrade, but at least the CPU, RAM and the MOBO will be good for the next 4-5 years just like my old aging i7 920 D0/MSI X58 Pro-E mobo w/12 GB of Corsair XMS3 DDR3 1600 Triple Channel and that PC had aging SLI GTX 680's that have been great so far, but it was time to upgrade that system, it was a great PC for all these years since I purchased it when the X58 and i7 series first launched back then.

    This upgrade over my aging X58 build is a massive jump as I was running into CPU bottlenecks with my SLI GTX 680 even with the CPU overclocked to a stable 3.8GHZ on air cooling with good temps in mind. I tested my new system with both my old EVGA GTX 680 FTW+ 4GB cards in SLI and a single GTX 980 beats it in games that don;t have as great SLI/Crossfire support and when I placed my second GTX 980, it was overkill compared to the 680's but also using less power at the reference clocks at first. I may get a third GTX 980 but I'm holding out on the possible rumored GTX 980 Ti that may come around Spring 2015 and I hope to get 2 at least for SLI if they are a comparble upgrade like the GTX 780 Ti was over the 780 and I hope to have 6 to 8GB of VRAM on the cards too as I am gaming at 4K and there are times where the VRAM is topped out and more VRAM is needed. I'm just glad the that GTX 980 was such a great buy this time around and made a big difference for those of use on aging GTX 680's and earlier cards. I usually upgrade every 2 years for my GPU anyways. I am currently reading up and also overclocking my new build to to tutorials and such on these components that I have since it's been years since I really used to tinker with voltages, RAM memory timings and of course serious voltage tweaks to the GPUS for better overclocking. I actually miss the old days of my tinkeirng my PC for hours a day when I was younger, but at least I have the weekends off to tinker with my new gaming beast that should last me for a few more years. For now, I'm happy at the performance that I have with a mild overclock and my games run and look fantastic on my new 55 inch LG 55UB8500 4K TV that has HDMI 2.0 so that I can play at 60 FPS, but with my current setup,I'm averaging 40+ FPS in most games on ultra settings with AA set to 2 or non at all since in my opinion, a 55 inch 4K TV is the perfect size foe the resolution and turning off AA in most games makes no major difference in appearance on the screen due to the high resolution, so that's a major plus for 4K gaming and it's easier on the hardware too. That TV only cost me about $1500 in Best Buy during the Labor Day weekend sales. It was a steal for a TV that has great features and most importantly, HDMI 2.0 to take advantage of my GTX 980's in 4K instead of having to use Display Port to HDMI adapters I was using with my GTX 680 SLI to achieve 30+ FPS in a few games at 4K with no AA and such.

    For those that want great performance at 4K, SLI GTX 980's are great and are great for systems that have PSUs typically that can comfortably handle both cards at load with a quality 750W PSU vs needing a 1000W unit for older cards.
    Reply
  • Spartan 363 - Wednesday, October 01, 2014 - link

    I forgot to mention that I am currently using 2 EVGA GTX 980 SC, but I will be sending them back next week to exchange them for the Classified cards that will be the aire cooled ones, but I am currently running clocks on both my current cards at a mild 1300 mhz OC on the core while the memory is at the factory clocks for now, not bad for air cooling, I know I can push for 1400 to 1500 but I will wait for the Classified cards for that type of OC as I will be getting EK WC blocks for both cards this time. Reply
  • sandwich_hlp - Tuesday, October 07, 2014 - link

    Been trying to decipher Intel roadmaps and the like, to no avail, so... can anyone tell me approximately when the DDR4-supporting Core i5 line is expected to launch? I'm needing to upgrade my aged Core 2 Duo (DDR2!), but I don't wanna hop on the DDR3 bandwagon just as it's being superseded by DDR4... :-/ Reply
  • GGuess - Saturday, December 06, 2014 - link

    There should be a line of discussion of why haven't CPU speeds increased in the past 5years in a significant way. My 5yr old Intel I7 is a 4 core at 3GHz. The ones discussed here are only 6 or 8 cores and run stock in the mid-3 to 4GHz range. So over 5yrs, the CPU capability has not grown 2 to 3times faster, and that only applies to applications that can use the extra cores and hyper-threading. The usual rule we work to is that people won't even notice a 50% speed increase. It has to be 2 to 3 times before it is noticed. Previously, a 3year refresh of a computer resulted in a 5 to 10x computer speed increase.

    With the current barely noticeable 2x, why bother with the trouble of an upgrade? No wonder Intel's and AMD's sales figures are failing to grow.
    Reply

Log in

Don't have an account? Sign up now