The question I'd ask is why do you want to know? Is it your plan to simply not use an effect that increases render time? Would you allow the needs of the program to go unfulfilled in order to get a quicker export?
As a general rule, it takes as long as it takes and short of spending money on better hardware there's often little you can do about it but get a cup of coffee, read a book, watch some TV, etc.
Codec in the time-line, codec used on export, bitrate setting, frame rate, scaling, blending, effects, transitions, system in use, version... and on and on.
Your question is like: I drive a Porsche 911, but when I stand in a traffic jam, how come that a FIAT 500 overtakes me on the right-hand lane?
Care to explain that?
I am essentially just trying to optimize workflow, as we are having quite high project throughput. So yes I would not use an effect, or think of a more efficient method to achieve said effect, if it was going to significantly bottleneck the workflow by several orders of magnitude, and yet only added negligible subjective aesthetic improvement.
It comes down to classic "Cost Benefit Analysis", thats why I was asking for technical documentation, to see if there is anything we are doing which is having a big hit on performance, which may have a more efficient alternative or may not improve the quality to a level which warrants to slowdown.
Provided you have a CUDA/MPE video card, rendering will be faster if you limit yourself to accelerated effects.
For the rest, provide details.
> Provided you have a CUDA/MPE video card, rendering will be faster if you limit yourself to accelerated effects.
For information on which effects are CUDA-accelerated, see this page, including the comments.