Articles: CPU

Bookmark and Share

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 ]

Application Tests

We benchmark 3D rendering performance with Cinebench R15. Maxon has recently updated it, so it can now measure the speed of computing platforms in the latest versions of the Cinema 4D animation suite.

Final rendering is a computations-heavy application which can make full use of the advantages offered by multicore CPUs. That’s why the Core i3 series, even the Haswell-based models, fall behind Intel’s quad-core and AMD’s 6- and 8-core CPUs. The Haswell microarchitecture makes the new Core i3 models faster than their predecessors by about 17%, though.

Using dBpoweramp Music Converter R14.4, we benchmark the speed of converting FLAC files into MP3 format with maximum compression quality.

Despite its single-threaded Lame encoder, dBpoweramp Music Converter can transcode several audio files concurrently, making full use of modern multicore CPUs. That’s why we have the same standings as in the final rendering test. The conversion speed is much lower on Intel’s dual-core (and AMD's quad-core) CPUs than on Intel's quad-core processors. For example, the gap between the senior Core i3 and the junior Core i5, both having the same microarchitecture, is up to 28%. By the way, the Haswell design is less advantageous in this MP3 encoding test than in the previous benchmarks.

The test scenario for Adobe Photoshop Lightroom 5.2 includes post-processing and exporting into JPEG format of two hundred 12-megapixel RAW images captured with a Nikon D300 camera.

Once again we see that it takes at least a quad-core CPU to efficiently run professional applications. The Haswell microarchitecture works faster in Lightroom than the Ivy Bridge, yet the 10% performance boost is insufficient to make the new Core i3 CPUs comparable even to the older Core i5 of the Ivy Bridge generation. Interestingly, AMD’s quad-core CPUs are also faster than the dual-core Haswell here.

We benchmark CPUs in Adobe Photoshop CC using our custom test that is based on the Retouch Artists Photoshop Speed Test and consists of typical processing of four 24-megapixel images captured with a digital camera.

Photoshop has always been well-optimized for Intel CPUs. The new Core i3 models beat the 8-core AMD FX-8350 but lag behind the quad-core i5-3330 of the Ivy Bridge generation. The i3-4340 is 11% faster than its predecessor i3-3250. The Core i3 4000 models have similar results, indicating that the amount of L3 cache isn’t a significant factor. The improvements brought about by the Haswell microarchitecture are far more important.

The processors’ performance in cryptographic tasks is measured with the built-in benchmark of the popular TrueCrypt utility that uses AES-Twofish-Serpent encryption. Besides optimizations for multi-core CPUs, it supports the AES-NI instructions.

One of the most notable innovations in the new Core i3 CPUs is their support for the AES-NI instructions which ensure substantial performance benefits for encryption algorithms. That’s why the i3-4340 is up to 35% faster than the old i3-3250 here. Still, the quad-core CPUs from both Intel and AMD remain a better choice for this type of applications.

To test the processors’ performance at data archiving we use WinRAR 5.0. Using maximum compression rate, we archive a 1.7GB folder with multiple files.

The speed of data compression depends on two factors: computing power (i.e. the number of CPU cores) and system memory performance. So we had expected to see the benefits of the 4MB L3 cache of the senior Core i3 4000 series models here. However, the i3-4340 isn't much better than the i3-4130. The enlarged L3 cache seems to be a marketing trick rather than a real performance-improving innovation but the new microarchitecture brings improvements of its own. The i3-4340 is 6% ahead of the i3-3250, for example. Of course, the new dual-core CPUs cannot compete with the quad-core i5 models of the Ivy Bridge generation, but perform very well for their price range. AMD's senior Socket FM2 model A10-6800K is substantially slower even than the junior i3-4140.

In order to measure how fast the tested CPUs can transcode video into H.264 format we used x264 FHD Benchmark 1.0.1 (64 bit). It measures the time it takes the x264 encoder to convert an MPEG-4/AVC video recorded in 1920x1080@50fps format with 30 Mbps bitrate. The results have high practical value, because the x264 codec is part of popular transcoding utilities, such as HandBrake, MeGUI, VirtualDub, etc. We regularly update the encoder used in this performance test. This time around, we use version r2358, which supports all contemporary instruction sets including AVX2.

The AVX2 support and the new Haswell microarchitecture raise the performance bar of the Core i3 series by about 22% in this test. The new CPUs can’t match the quad-core i5-3330 and fall behind the quad-core Haswell by over 40%, but now they at least have become as fast as the quad-core CPUs from AMD.

Encoding video with a bare encoder is hardly a real-life application, so we want to check out the speed of video transcoding with the popular free tool Freemake Video Converter 4.0.4. It uses the FFmpeg library and is based on the x264 coder too, but features certain optimizations. We disable CUDA and DXVA for this test to create maximum load on the CPUs’ computing cores.

As expected, the speed of encoding in Freemake Video Converter is comparable to the speed in the x264 test above. The new Core i3 4000 series CPUs have similar results. They are faster than their Ivy Bridge predecessors and compete with AMD’s quad-core products. Intel’s quad-core and AMD’s 6- and 8-core CPUs are much faster, though. So again, the Core i3 series is only good for the internet and office applications. It is better to use a CPU with more x86 cores for computers employed for content processing and authoring. And a junior CPU model of that kind wouldn’t be much more expensive than a senior Core i3.

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 ]


Comments currently: 8
Discussion started: 12/20/13 09:11:31 AM
Latest comment: 03/07/14 06:53:09 AM

View comments

Add your Comment