>>>I think VFPA is permanently under development, so you can simply talk to the guy and explain to him how wrong is it to have it faster without mdot. I'm convinced that the power of your proof will make him see the error in his ways.
>>
>>Ah. I see. Not a friend as I thought. Thanks for the revelation. Good riddance.
>
>No you don't see anything. You claim your passion is science, and evangelize it to the point where you forget how relevant is this. Just calculate the total time the twenty-some of us have spent reading and writing in your crappy coder threads, and compare that with the cumulative time wasted on not using mdot where it would shaved off some milliseconds.

It is not about saving milliseconds alone. It is about adhering to Knuth's point that a 12% gain for free is what should be acceptable to any group of engineers, but then I am mistaken to think any here are more than hacks.

You make a better point than you know. Too many self-proclaimed geniuses here. A lot of pathetic losers too. All draining the life from people who used to give a crap about this community. I forget Christ advised not to cast pearls before swine. I also forgot where ignorance is bliss it is folly to be wise.

>
>It's not that I'm saying the mdotted code is slower - it isn't (except perhaps in VFPA). It's that the speed gained is minuscule compared with the time we have wasted on these threads. And that in most scenarios these speed gains, even with possible confusion with field names (which should and can be avoided by naming schemes, rigorously using different ones for variables and fields), are small compared to the time wasted on discussing the subject.
>
>Sure, 0,00000034 seconds is twice 0,00000017 seconds. They're both laughably small compared to whole seconds, not to mention something like six or twelve hours a week (my rough estimate to the time we have wasted on these threads).
>
>But if belief in the supreme importance of mdot over everything else is required, then okay, consider me under anathema... I was never a believer anyway.