Author Topic: new benchmark rules needed?  (Read 1709 times)

0 Members and 1 Guest are viewing this topic.

Offline mafrieger

  • Jr. Member
  • **
  • Posts: 79
new benchmark rules needed?
« on: October 08, 2018, 08:06:19 am »
Since writing of keyshot benchmarking guide back in 2010

https://www.keyshot.com/forum/index.php?topic=859.0

possibilities and features of hardware and software and especially their "interaction"
becomes a little more complex.

As one example:
Just think of CPU turbo modes taking current load details, temperature and energy consumption into account -and this in a hot summer day of 30°C or in winter at 20°C room temperature.
Or notebooks with same CPU but different cases with different dimensions and different cooling solutions....

-> same CPU may give speed results of + or -20%  :-O Ouch!

So producing comparable benchmark results, one should
do 2 things
1. first of all: benchmark within a steady temperature state: That's pretty easy: just let the rendering run for at least 15 minutes before noting benchmark numbers
2. second important point: provide some more details about system of Software, Hardware and Conditions.

For a good start the following details may help:

Software:
- Keyshot version
- OS Type
- OS Version
- OS Energy saving mode

Hardware:
- Sytem-name/number
- Manufacture

or/and

Hardware Details:
- CPU Type
- RAM total
- RAM number of modules (channels used)
- Motherboard
- System over-clocked yes/no
- CPU- Cooler Type
- CPU-Cooler rpm (if not dynamic)

External conditions:
- Room-Temperature


=> what do you think?
« Last Edit: October 08, 2018, 08:28:24 am by mafrieger »