For your orientation we have tested a selection of hardware configurations.

The result may help you to find the optimized hardware configuration for your need. 

Download the performace sheet here

We recomment that you read this important information describing the test-procedure:

Tested software:

easyDCP Player+ 2.0.8 using the CPU for JPEG 2000 decompression. The used memory-, storage- and graphics card-configuration had NO impact on the decoding performances.

Test material:

We generated Test-DCPs with white noise. White noise images can be regarded as a worst-case scenario for JPEG 2000 as well as many other image/video compression formats. We generated white noise DCPs for 2K/2D, 2K/3D and 4K/2D. We also used different container-sizes known as Flat (2K: 1998x1080, 4K: 3996x2160), Scope (2K: 2048x858, 4K: 4096x1716) and Full-Container (2K: 2048x1080, 4K: 4096x2160). Even if easyDCP Player has to decode more pixels in Full-Container-mode compared to either Flat or Scope, we did not notice a huge difference in the decoding-performances. 

Another effect that has to be considered is a special encoder setting when generating DCPs: easyDCP Creator, for example, generates JPEG 2000 files using Variable Bit Rate (VBR). This means, that, regardless of the maximum bitrate, source images with little details result in compressed JPEG 2000 files comprising a small file-size. On the other hand, source images with plenty of details result in larger file-sizes. Only the maximum size is limited by the maximum bitrate that was specified when encoding the images. When required, the encoder stops to include more information in the compressed images so as not to violate the maximum bitrate. For our worst-case DCPs, the encoder has constantly hit the maximum bitrate, so the data-rate appears in fact to be constant. VBR-encoded DCPs, however, usually show a varying data-rate throughout the playback of the DCP. The data-rate has a very strong impact on easyDCP Player’s playback performance. If a source DCP was encoded with a very low data-rate it may be possible to playback the movie in real-time, even when using a slow hardware configuration. Since the data-rate in our test-DCPs does not vary at all, the measured results show a worst-case behavior. So it may be possible to playback a DCP without any issues in real-time, even if the table shows an orange bullet-point for the test case.

* Quality Slider (QS):

Since version 2.0, easyDCP Player is capable of trading decoding precision for a better performance. Especially, when viewing DCPs (which have 36 bits per pixel) on a computer monitor (24  bits per pixel), some information from the JPEG 2000 codestream can safely be ignored without a significant impact on the resulting image quality.

** Level Skipping:

Using easyDCP Player’s Level-Skipping feature , one or more resolution levels of the JPEG 2000 code-stream can be discarded. Of course, the reduction of resolution levels has direct impact on the playback performance on the one hand. On the other hand, skipping a resolution level will half the resolution in both dimensions. Discarding the highest resolution level is often sufficient to allow for real-time playback even on notebooks.