vfxpedia

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Performance Comparison (GPU) ======

The following chart shows a list of the [[https://www.maxon.net/en/cinebench|Cinebench 2026]] results from Maxon on the three systems, all of which differ in terms of their graphics cards, though two of the systems are otherwise identical.

{{ :hardware:5090benchmark.png?|Benchmark 3 GPUs}}

In addition to the raw measurements, we also measured the actual power consumption of the systems during the test runs. These measurements are sufficiently accurate and robust, but could technically be made even more precise, since rendering processes on a graphics card also include the time required to upload data to the graphics memory, and full load does not persist continuously.

The table above lists all measurements by the three different graphics cards. The lower section shows the extent to which the different devices differ in terms of speed.

The values are neutral in terms of power consumption. This means that the RTX 3080 is 2.7 times more efficient at the same cost in comparison to the GTX 1070. **It therefore delivers 2.7 times more performance at the same cost and in the same amount of time.**

The RTX 5090 is 6.71 times more efficient than the GTX 1070. The comparison between the 3080 and the 5090 is 2.5 times better in favor of the newer card.

In terms of the CPU, the difference is smaller at 1.56.

 --- //[[r.engel@mediadesign.de|Rainer M. Engel]] 2026/04/24 12:04//