The OCUS R19 Benchmark Results

- Last modified: Feb 26 1999 -

Here are the results of my OCUS R19 Benchmark.

In february 1998 I have sent out a trail file on the Pro/USER mailing list with a request to everyone to run it on their computer. The trail file was created with my OCUS pro_bench script (version 4.1; R18+) and performs several tasks. This mainly checks the speed of the CPU and the Graphics card.

Below are the results I received. You can view several lists:

• Total score
• CPU tests only score (shade calculations, regenerations, pattern x-section)
• Graphics test only score (wireframe redraws, shaded redraws, menu redraws, dialogue box open, mouse spins)
• Total score including sub-scores

You can also view the FULL RESULTS, download the results in Excel 7.0 and text format (30Kb) or view the results as a graph (39Kb; only results faster than 900 secs. in total).

This kind of benchmarking can also be done to compare Pro/E builds on one and the same machine: See how Pro/E has "grown" the last couple of builds

Remarks, conclusions and tips...

• Do not take these result for granted. You can make a lot of adjustments on every machine to make Pro/E run faster or slower. Furthermore, note that these benchmarks were performed at all kinds of sites and with a lot of different builds. This surely influences the results.
• Note that in release 9725 PTC implemented a new memory manager which had significant performance gains for NT machines. There is also a "NetPower switch" that one could set that turned on memory management for some NT boxes. However, this would only be an issue on NT 4.0 without SP3, since turning on that switch will crash you on SP3.
• This benchmark needs at least 128Mb of RAM to run without swapping. So results from machines with less are not accurate at the end of the test.
• There are some Dec Alpha NT users who cannot equal the result by Don R. Fye (total of 242 seconds). This is how he configured his machine:
  • Display Plus! setting "Show window contents while dragging" unchecked
  • Display Plus! setting "Smooth deges of screen fonts" unchecked
  • FX!32 disabled
  • Pro/E installed with FX!32 turned off
  • McAffee turned off
  • NetMeeting turned off
  • Install and configure NetBEUI as first network protocal (My machine had only TCP/IP, adding NetBEUI & putting it first helped)
  • I use DHCP, not a fixed IP address, error in std.out about TCP/IP configuration when pro starts.
  • Pro/E software is loaded locally, license is obtained from Unix server across town.
  • All config.pro's on the machine were renamed to config.old except the config.pro provided with benchmark.
• Here is some more information for DEC users:
  • For configuration information specific to running Pro/ENGINEER benchmarks on Digital Workstations, reference the following web site:
    http://www.workstation.digital.com/application/config_guide/config_guide.html
  • The 4D30T graphics card is a mid-range card. The 4D60T, having the best score in the benchmark results is a high-performance card. The disk used is an older, slower model than that currently available.
• If your machine is low only on mouse spinning, check your own config.pro (also in the loadpoint of Pro/E) and remove the options SPIN_WITH_SILHOUETTES and SPIN_WITH_PART_ENTITIES (or set them to NO).
• I did not add any prices. Too much uncertainty ... You'll have to calculate your own price/performance values.
• Another one pointed out that floating licenses could also make a difference, since Pro/E has to check the network for free licenses.
• Whatever the results are. If a machine runs the test really fast it will probably be a really fast machine (check for no abnormal sub-scores). If it is really slow on the other hand you will first have to check if this is not some kind of configuration issue before judging the computer.