The OCUS Benchmark version 4
FAQ

- Last Modified: -

Why do I get a trailfile out of sequence error?

There can be a number of reasons why running the benchmark crashes Pro/E or gets a trailfile out of sequence error:
  1. You ran the wrong trailfile version.
    Make sure to select the trailfile appropiate to your Pro/E version.
  2. You did not start Pro/E from the Benchmark directory.
    Follow the instructions exactly. Open a command-shell (Run: cmd) and change dir to the directory where the benchmark is located (with cd). Then in the command shell type your command to start Pro/E. If you don't know your startcommand try proewildfire (for Wildfire) or proe1 (for Wildfire 2.0).
  3. You did not remove your own Pro/E configuration files.
    Remove the config.pro, config.sup, config.win, menu_def.pro and prodev.dat. files from the text directory in the loadpoint of Pro/E (e.g. c:\Progam Files\proewildfire\text) and from your HOME directory. Windows users should also verify there are no configuration files in the root directory c:\.
    In the Help -> Technical Support Info you can check which config files are read upon startup.
  4. The v4.0 trailfiles did not account for the license dialogbox which you can get at startup of Pro/E Wildfire. This makes the trailfile crash. The trailfiles in the v4.2 zipfile will skip this dialogue and continue.

Why are my overall results much slower than a comparable system?

There can be a number of reasons why your results are too slow.
  1. The most common reason is a configuration issue.
    First start with the solutions entered above (    Why do I get a trailfile out of sequence error?)
    Other reasons might be.
  2. You could be using a different OS version with different patches or service packs or graphics driver. Try the latest driver of your vendor.
  3. You might be using a license from a license server over a slow connection.
  4. You might be running the benchmark on a networked drive.

Why doesn't my high-end graphics card perform all that better?

  1. Check the version of your graphics driver. Is it too old or too new perhaps?
  2. This benchmark has been designed to perform on high-end systems as well as low-end systems. In order to make the test runable on the low-end systems the benchmark had to be carefull not to use any graphcis task which was too graphics intensive. Otherwise the low-end systems got exceptional slow results or couldn't run the benchmark at all.
  3. I also want this benchmark to run in it's fastest mode as possible.
    Some graphics features of high-end graphics cards only gets visible when you use advanced graphics options of Pro/E. I tend to leave those options off as much as possible.
    E.G. Users with high-end graphics cards can turn on shaded edges if they like and won't notice any real performance drop, but at some point they will, like when they open large assemblies or very large parts.
    So take my advice and leave those edges off ...

Why not expand the test to more Pro/E modules?

I want this test to run on any Pro/E system. So it should therefor only use modules in the Foundation package of Pro/E.

Why do you let vendors enter results as well?

Some of you are doubting the reliabilaty of result entered by vendors.
My opinion in this is, that if a vendor performs special tricks to get abnormal fast results he will soon be uncovered by users who have such a machine as well. If you consider buying a machine based on these results why not perform this benchmark on a demonstration model by that vendor and see if the fast results can be reproduced under your own working conditions?
I also compare the result with other results of comparable systems. If the results differ too much I get in contact with the persons who enters the result.

Where are your previous versions of the benchmark?

The previous versions of this benchmark are still online through the current result page. There you can find these links as well:

Why isn't there a version which can be run on 2001 as well as Wildfire?

The differences in the User Interface between 2001 and Wildfire are too drastic. It's very hard to create an OCUS benchmark which can run on both versions. Together with PTC I was able to compare most of the tests reliably (but not all of them). The test had to be run in a few steps and was too complex to release for general use.
Those results can be viewed here: a comparison of R2000i with R2000i2, R2001 and Wildfire