![]() ![]() Alacritty github throws that claim in the front, but doesn't give any benchmarks next to this outrageous, thus they are assumed to be lying|misleading.īenchmarks so far have just involved running find /usr on my Linux system with Alacritty, st, and urxvt, and on macOS against Terminal.app and iTerm2. With a terminal emulator, interactivity is the most important thing. It's easy to claim the title of "fastest" when you don't elaborate at all, and don't say what you mean by fastest. ![]() The claim of "fastest" should have come with a massive asterisk. Noticeably faster.Ī final note is that kitty on MacOS sure looks competitive with iterm2 in terms of features. I can say that kitty is noticeably faster in general use than the other terminals I've tried (and I've been a linux user for almost 30 years). I can't speak to it being faster than kitty or other terminals. Kitty properly erases one preview and shows the next as you select them in ranger, where the others simply drew the next preview on top of the first - when the previews not the same size, you saw a mess of the previous ones with the new one on top. I was unable to get most other terminals (maybe urxvt worked) to do this or do this properly. Not only that, but if you're using zsh and oh-my-zsh (or similar), any things to help speed up calculating all the things printed in your prompt are useful.Īnother point is ranger running in kitty can display image previews in the terminal window, which is very neat. I use kitty, which is also GPU-accelerated.įaster is always better, especially if your program is printing a lot of lines and scrolling in the terminal. Bounce is to remove dupes, but shouldn't matter on the first press, so it's something else. I tapped it a ton of times and checked a few different presses to be sure.Įdit: Turns out Windows has a 100ms bounce delay to remove duplicate keypresses on shoddy hardware.Įdit2: Actually I'm talking rubbish. ![]() Wanna try it yourself? I filmed on my phone in Open Camera set to 120hz, transferred using WiFi File Transfer, and used Avidemux to count the time between my finger hitting the button and a char appearing on the screen. Either way though, it looks like terminals promising speed are a waste of my time. I'll try in the BIOS next time I reboot and figure out whether it's the laptop itself. Could be Windows's event loop, compositing, the laptop's drivers or hardware, but that seems pretty grim. Well it turns out on my Dell XPS 15 running at 60hz, Git Bash for Windows locally and remotely both have an input-to-render lag of around 100ms, as does Notepad. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |