| Quote: | “even if your RNG is statistically perfect, people will still complain  “ |
| Another Quote: | “On modern x86 CPUs, single core can generate 150M+ random bytes/second this way (and this is a Damn Lot).” |
| [→] |
Development & Deployment of Multiplayer Online Games, page 11:
Implementing Queues for Event-Driven Programs
| Quote: | “full queues SHOULD NOT happen during normal operation” |
| Another Quote: | “With queues-implemented-over-mutexes like the ones we’ve written above, the most annoying thing performance-wise is that there is a chance that the OS’s scheduler can force the preemptive context switch right when the thread-being-preempted-is-owning-our-mutex.” |
| [→] |
C++ Guidelines – Made-to-Measure vs One-Size-Fits-All
| Abstract: | “After bashing all those Big Name guys for making those over-arching sets of guidelines, it is perfectly logical for me myself to do the same  ” |
| Quote: | “DON’T use C-style cast, EVER. If you DO need a cast – DO spend time on figuring out which of C++ *_cast<>s you really mean (and maybe, you’ll find a way to avoid that cast at all)” |
| [→] |
C++ Performance: Common Wisdoms and Common “Wisdoms”
| Quote: | “over(ab)using C++ features is a different story, we’ll discuss these features below on case-by-case basis” |
| Another Quote: | “Compiler will use all its Next-to-Divine Wisdom to show you that it is smarter than you are, and to ignore most of those inline specifications you carefully wrote.” |
| [→] |
C++ for Games: Performance. Allocations and Data Locality
| Abstract: | “Allocations and related lack of spatial locality can be DAMN EXPENSIVE” |
| Quote: | “During the times of Ancient Gamers (also known as times of Atari, ZX Spectrum, and Commodore64), CPUs were running at clock speed of a single-MHz (and one operation usually took several clocks). At the same time, memory run at about the speed comparable to that of CPUs.” |
| [→] |
Marshalling and Encodings
| Quote: | “Encoded data size is especially important for Client-Server communications, and less important (within reason) for Server-to-Server ones and for locally-stored format.” |
| Another Quote: | “On the other hand, due to being optimized for CPU operation, Flatbuffers are NOT optimized space-wise; even compared to not-so-optimal-space-wise Google Protocol Buffers, Flatbuffers can lose additional 1.5x in size” |
| [→] |
Network Programming: Socket Peculiarities, Threads, and Testing
| Quote: | “I am not saying that this architecture is the only viable one, but it does work for TCP for sure (and performs reasonably well too)” |
| Another Quote: | “The whole task of optimizing performance beyond, say, 20-50K packets/second per box tends to be Quite Elaborated, and involves quite a few things which are platform- and hardware-dependent.” |
| [→] |