CPU Benchmarks

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.

HandBrake, SD Filmlink

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

For low quality encoding, single threaded performance wins out over threads despite the extra multi-threading of i3 processors.

HandBrake, 4K60 Animationlink

HandBrake v0.9.9 2x4K

For larger frame sizes, the extra MHz of the overclock brings it on par with the i3-4360, although one might suggest spending the extra $40 for the i3 to ensure full stability.

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

Dolphin Benchmarklink

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

Dolphin is another example where Haswell combined with strong single threaded performance wins.

WinRAR 5.0.1link

This test compresses 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 5.01, 2867 files, 1.52 GB

PCMark8 v2 OpenCL on IGP

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.  As this test is new, we have not run it on many AMD systems yet and will do so as soon as we can.

PCMark8 v2 Work 2.0 OpenCL IGP

PCMark8 v2 Creative 3.0 OpenCL IGP

In both of our PCMark8 v2 results, overclocking the CPU gave a significant jump in performance.  This would be down to the single threaded nature of parts of the benchmark, allowing web browsing and the snappyness of the system to be sped up.

Hybrid x265

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

Almost an extra 50% performance for Hybrid x265 encoding.

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

Web Benchmarks

On the lower end processors, general usability is a big factor of experience, especially as we move into the HTML5 era of web browsing.  For our web benchmarks, we take four well known tests with Chrome 35 as a consistent browser.

Sunspider 1.0.2

Sunspider 1.0.2

Mozilla Kraken 1.1

Kraken 1.1

WebXPRT

WebXPRT

Google Octane v2

Google Octane v2

The usage model for an overclocked G3258 comes in highly single threaded environments. 4.7 GHz is nothing to be sniffed at, especially when it comes to web browsing or simple photo editing where it beats out more expensive processors. With multithreaded scenarios, it battles with the i3 depending on if the software can use hyperthreading to its full advantage.  If the software can do that, then the i3 still wins out.

Overclockable Pentium Anniversary Edition Review: The Intel Pentium G3258 Gaming and Synthetics on Processor Graphics
Comments Locked

96 Comments

View All Comments

  • Computer Bottleneck - Monday, July 14, 2014 - link

    Regarding cooling at slightly slower OC Pentium G3258 speeds, please read this Anandtech thread:

    http://forums.anandtech.com/showthread.php?t=23910...

    Also for Non-Z overclocking Pentium G3258, please see this thread:

    http://forums.anandtech.com/showthread.php?t=23899...
  • xrror - Monday, July 14, 2014 - link

    You will never see an i3-K from Intel because that would actually be popular/useful enough for the "average buyer" to seriously consider one.

    OR IF Intel did release one, it would be (over)priced to be so close to an i5 that you wouldn't bother with the "risk" of clocking one.

    I would absolutely love to be proven wrong mind you, but yea. Pentium-AE is so crippled it's painfully cynical.
  • boozed - Monday, July 14, 2014 - link

    "Aside from the awkward/inaccurate scaling on the slide shown"

    I call it either deliberately misleading or absolutely ridiculous. They should know better.
  • Spoelie - Tuesday, July 15, 2014 - link

    So all benchmarks were run at 1333 memory speed? Was it not possible to reach a higher speed?

    It seems that at least for some benchmarks, it would've made quite the difference. Anyone bothering to overclock this one will want to overclock the memory as well.
  • Torpe - Tuesday, July 15, 2014 - link

    Hello Ian. Do you have any Quick Sync data on this CPU? I'm looking at upgrading from my E8400@3.6, which does everything I need well except compress Blu-Ray in a timely manner. Thank you.
  • jamescox - Tuesday, July 15, 2014 - link

    I don't think I would recommend one of these to anyone for a low-end gaming machine. With DX12 being able to take better advantage of multiple threads, the low-end for gaming probably should be 4 threads. With xbox one and ps4 both using a low single thread performance, 8 thread cpu, I would expect to see a lot more optimization aimed at 4 to 8 threads. I doubt these low-end chips have sufficient caches and other resources to effectively support multi-threading. I wonder if turning on vector extensions would use too much cache also. The vector extensions should be getting less use anyway, since almost anything that could make good use of the vector extensions would run faster on a gpu.
  • Laststop311 - Wednesday, July 16, 2014 - link

    too many bottlenecks, the i5-4670k is still the gold standard for gaming performance and value.
  • Casecutter - Wednesday, July 16, 2014 - link

    Oh very nice thanks!
  • Mayuyu - Thursday, July 17, 2014 - link

    I must be really unlucky. My G3258 isn't even stable at 3.7 Ghz at 1.2V. The best I can do is 3.6Ghz at 1.1V.
  • C.C. - Sunday, July 27, 2014 - link

    You are doing something wrong, or that is a defective chip. I would RMA it (if that doesn't bother your moral code like it does some people).

Log in

Don't have an account? Sign up now