Many modern Particle simulation engines rely on user input for their maximum particle count, despite the maximum particle count being something that can be calculated ahead of time. All you need is the highest possible spawn rate (as particles per second), the minimum decay rate (as decay value per second), and the maximum life time (in seconds). Once you have these variables the following formulae will result in the maximum particle count:

This is a technique dating back to the days of DirectX 7 and was implemented differently from engine to engine - it was a wild west of ideas after all. The technique works by using the Alpha channel as a Multiplier, often a logarithmic or exponential multiplier. This allowed representing values above 255 somewhat safely, but using it reduced the precision significantly.

Thanks to advanced in computing performance we no longer have to rely on this technique as a 10/10/10/2 RGBA texture can safely represent values up to 1023 with no issue. It is still possible to use the Alpha channel as a multiplier, but with only 4 values for the multiplier it is significantly more limited. Both techniques, old and new, are still in use today.

No, but it's basically visually lossless. Usually it does not matter for final encoding, but if you do this for some effect, consider staying in a higher precision format for the converted values. Alternatively you can also use temporal dithering to hide the imperfections, which is what some streaming and recording software does to improve their output quality.

Dithering - a technique that is somewhat lost to time, thanks to advances in computing power. Dithering used to be everywhere in older games, as the hardware as incapable of rendering what people wanted to do. It works by abusing a significant flaw of Human vision, namely that we aren't that good at identifying colors surrounded by other colors, and will often take a shortcut instead. Thanks to advances in computing power, we can now also do temporal dithering - a technique that monitors already use to make HDR more affordable (see "Frame Rate Control").

You can also make use of this in your own games, either as a stylistic choice, or to enhance the visual fidelity of the output for any player. All you need is a higher precision input than output and a dithering function of your choice. My personal choice for spatial dithering is Bayer, seeded with some per-frame per-pixel random number to make it temporal. Take a look at this ShaderToy example, this image comparison, or this image comparison to see what Dithering can do for you.

The Record Size of a storage dataset in TrueNAS and FreeNAS matters a lot. It can be the difference between being able to transfer 2 GiB in a second, and taking multiple minutes to do so. While documentation is a bit sparse and fractured, through testing I was able to find out that the best record size is the default it has already selected. It offers the best mixture of performance for large and small writes, especially on faster storage.

My tests went through many record sizes, halving and doubling from 128K outwards. Every halving resulted in large read/writes getting significantly slower, but small read/writes getting slightly faster. Every doubling resulting in large read/write getting slightly faster, but small read/writes getting significantly slower. The best value ended up being 128K, which managed to keep most of the NVMe performance as is (~80MiB/s in RND4K, ~2.5GiB/s in RND1M).