3D Graphics Performance Comparison – Part 6 (Alibre Design Revisited)

<< Part 5

The people at Alibre were pretty confused at my last post regarding the graphics performance in the new version of Alibre Design. Max Freeman of Alibre had previously shown me a video of the engine navigating quite smoothly on his computer. But on my computer, the improvement wasn’t very noticeable when compared to V12. So the graphics expert at Alibre was given the task of figuring out what exactly was going wrong. As it turns out, there was a bug in the Beta which treated files created in earlier versions incorrectly. As a workaround Max told me to change the minimum circular facets in file properties from 24 to 23 and hide and show the entire assembly. And this is what I got.

I think the point of changing the minimum circular facets was just to make the software recalculate the render mesh or something like that. I guess I could have increased it to 25. From the screen shot below you can see that decreasing it to 23 really didn’t have much of an impact on display quality.

Click image for larger view

So as you can see the graphics performance in the next version of Alibre Design will be a lot better than V12 and pretty close to the best in the industry so far. I am almost tempted to put Alibre Design in the same category as SpaceClaim and KeyCreator, but I don’t think I will. There is still some difference. So I will create a new category called “Awesome” and put Allibre Design in it. I think I am beginning to enjoy this comparison stuff. 😉

A few things to note about this comparison. My computer is definitely not the best hardware to run a CAD system. There are far better rigs out there. However, I must point out that it is a pretty common workstation (Dell Precision M6400) with a pretty decent graphics card (NVIDIA Quadro FX 2700M with half a gig of memory on it). Also I am running Windows 7 32 bit. I suspect the graphics performance on a 64 bit machine would be much better. But then all CAD systems would probably run equally fast on a top of the line computer and we would really not know the difference between then, as we are able to see now.

Another thing. This method of graphics comparison is most definitely not scientific. I could do a scientific one by using a ton of benchmarking software and posting all kinds of graphs and charts containing all kinds of metrics. The thing is those numbers, graphics and charts would make sense to mainly graphics experts who understand what those metrics mean. Even I wouldn’t be able to understand them completely. I don’t think I’m even qualified to do such a scientific comparison. This is a comparison targeted to a normal user who simply wants to navigate around his model in the fastest possible way and if possible learn how to set up his CAD system to do just that. That’s why I am posting videos and not cryptic numbers, charts and graphs which can be easily twisted around to point to any direction I want.

A video tells the real story. This is how I do it. I shut down all applications, let the machine idle for a while till all hard disk activity stops, fire up the CAD system, load the model, fire up the video recording software and start recording. As simple as that.

Part 7 >>

  • So it worked better, but it really doesn't matter since if you have to do this everytime… I guess it is a good wake-up call for the devs to fix this?

  • I bet they have already fixed this problem in the Beta already. Needless to say, the release version will not have this issue.

  • Max Freeman

    Correct – this is simply a bug in the beta version and is not necessarily to be unexpected since we did some replumbing. We'll be fixing it before it officially releases.