busybee

1. Thou shalt prioritize belief in that which can be demonstrated empirically
2. Thou shalt not worship anyone, regardless of their social influence
3. Thou shalt strive to improve the situation for everyone, including yourself
4. Thou shalt allow yourself rest and hydration
5. Thou shalt leave yourself open to possibilities
6. Thou shalt not bring direct harm to others, and choose options which minimize indirect harm
7. Thou shalt respect the consent of others
8. Thou shalt not intentionally mislead
9. Thou shalt not treat the success of others as thine own failure
10. Thou shalt question everything, including these commandments

This is a followup to my previous cupping; this time around I evaluated the Decaf Project samples for S&W Roasting.

Quite some time ago, James Hoffmann stated his interest in doing a broad decaf test where he would get one set of bulk single-origin beans, split it in four, and process it four different ways. It took a while for this to happen, but finally he was able to make it happen, in the form of The Decaf Project.

Being super excited about decaf in general, I ordered two tasting kits, one from Rose Park (who I’ve never tried) and one from S&W (who I’m quite familiar with and a big fan of).

Both of my kits arrived in time for the big tasting event, although being at 7 in the morning my time I opted to wait until the afternoon, and further decided I’d only do one at a time. So for today I’ve just cupped Rose Parks roast.

One of the most poorly-defined yet critical things in configuring a VR avatar is where exactly to place the viewpoint. It can have a lot of impact on your viewing experience, especially with how the viewpoint ends up relating to the rest of the body. This becomes especially critical in systems like VRChat and Resonite where the physics of the body are directly impacted by this placement, and especially now that VRChat has built-in functionality for allowing a first-person view of your facial features (which the critter avatar calls “first-person snooter”).

When building a humanoid avatar that has proportions that more or less match one’s physical body, you generally want the viewpoint to be situated such that it’s placed at the same relative depth from your real-life view origin compared to your shoulders, as this means that your virtual head will be tracked with the same relative offset from your virtual shoulders as they are from your real ones. Otherwise, your virtual arms will seem to be the wrong length when fully extended.

But this raises the obvious question: where’s your physical view origin?

After getting in an extended discussion about the supposed performance tradeoff between #pragma once and #ifndef guards vs. the argument of correctness or not (I was taking the side of #pragma once based on some relatively recent indoctrination to that end), I decided to finally test the theory that #pragma once is faster because the compiler doesn’t have to try to re-#include a file that had already been included.

For the test, I automatically generated 500 header files with complex interdependencies, and had a .c file that #includes them all. I ran the test three ways, once with just #ifndef, once with just #pragma once, and once with both. I performed the test on a fairly modern system (a 2014 MacBook Pro running OSX, using XCode’s bundled Clang, with the internal SSD).