Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I want to test on Windows 7 if my old ATI Radeon X1650 card can do better at rendering Direct3D 9 or OpenGL 2.

Does anyone know any software that can help with determining this, like a benchmark?

share|improve this question
Which OS are you operating ? – Mr-Right Jan 24 '10 at 13:35
Running Windows 7 – Fladur Jan 24 '10 at 17:50
up vote 0 down vote accepted

Here are 2 tools that allow you to test and compare both Direct 3D and OpenGL:

FogCity measures the Direct3D and OpenGL speed. (quite old, but it still works with Windows 7, free & portable)

alt text

Fraps is a universal Windows application that can be used with games using DirectX or OpenGL graphic technology. In its current form Fraps performs many tasks and can best be described as:

Benchmarking Software - Show how many Frames Per Second (FPS) you are getting in a corner of your screen. Perform custom benchmarks and measure the frame rate between any two points. Save the statistics out to disk and use them for your own reviews and applications.

share|improve this answer
Furmark looks kind of dangerous =) – Fladur Jan 24 '10 at 20:45
@Fladur - Looks more like some overclocking gone bad. but then, video card benchmarking is not for the faint at heart. it ain't solitaire, you know :) – Molly7244 Jan 24 '10 at 20:48
I know, just kidding xD. I damaged a GeForce FX5500 xD with 3D Mark xD – Fladur Jan 24 '10 at 20:50
tss ... nonsensical downvotes from some anonymous lackwits. – Molly7244 Jan 24 '10 at 21:42
Sorry to hear that u.u – Fladur Jan 24 '10 at 23:02

Games and most benchmark programs will not work, it is like comparing the acceleration of two different cars under two different sets of circumstances.

Games for DirectX will use one set of features and if they have a port do OpenGL, it will probably have a few different textures / feature and run differently - again, comparing apples and pears! (or whatever the saying is!).

What I would recommend instead is programs that are actually designed for both - such as 3dStudio Max (You can download a demo), and try rendeing a simple scene on either and time it.

share|improve this answer

You should be able to compare D3D and OpenGl performance on the same card.

Find a tool that allows you to choose the rendering engine - Unigine Heaven for example - run it in each mode and compare the results.

You can download it from here

share|improve this answer
Just had a very quick look and I do not know for sure, but what I tried to say in my answer is this is a very awkward thing - DirectX and OpenGL are quite different. It is possible that they take advantage of a feature in one API that doesn't exist in another, so one may produce higher FPS/benchmarks but the other may look a lot better. – William Hilsum Jan 24 '10 at 21:55
@Wil - A properly designed benchmark program should take care not to "cheat" in this manner, or at least tell you that it is. I worked in 3D graphics for a while in the 1990's to early 2000's so I know it's possible. – ChrisF Jan 24 '10 at 22:31
But, I thought most benchmarking software is designed to test the same software on different hardware... OpenGL v DirectX is another kettle of fish, even if it doesn't cheat, there are still things that one is better at than the other in certain experiments, so any benchmark will not really be the same as real world usage.... I don't mean to overstep the mark as I never worked with (or really know that much about) 3d graphics, but again, in all my experience, I have seen benchmaring software for various scenarios, but not DirectX v OpenGL. – William Hilsum Jan 24 '10 at 22:39
@Wil - well I'll admit my knowledge is 5+ years out of date, but the whole point of comparing D3D to OGL for a particular scenario is to show which is better so that you can choose what's right for your application. – ChrisF Jan 25 '10 at 8:58
Exactly, one performs better in certain scenarios, you can't really compare one against the other outright.... e.g. in benchmarks, for example, a Ferrari is faster than a standard car, but a standard car is probably a lot more comfortable... It doesn't mean a standard car is better just because it wins in one area. Something may take advantage of one technology and appear to have a lower benchmark score whilst actually producing a better result. Anyway, I just really do not know enough to say anything more and don't want to argue especially as 3D is not my area... but +1 to you! – William Hilsum Jan 25 '10 at 10:18

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .