Annoying software

What's some software that's been pissing you off recently? For me it's Photoshop. It's just such an energy hog, but my work uses it religiously, so I kinda have to put up with it. Tried using GIMP once, but one of my higher-up's got all pissy about it.

>>76339 I absolutely hate qemu. It's incredibly slow. Arch linux boots in 15-20 seconds in virtualbox, but in qemu it takes about 1 minute on the same machine. If it's the first boot, it may take up to 3 minutes, which is an unbelievable amount of time, honestly. (It's important to mention that kvm is not availible on host os, which is windows 10, and that my arch linux installation has no de installed.) Unfortunately I have to use qemu since i like the portablity out of the box, and a good cli. I know that there are things like greenvbox and portable-virtualbox, but I don't want to trust any unknown third-party programs.

Obviously anything to do with chromium, that soykaf is so slow and hogs so many resources. Bring back the simple web!

>>76339 Discord, all my friends are on it and I don't have an account on anything else that copies it because the communities are somehow even less interesting. I also find blu-ray software annoying as soykaf

>>76342 I guess it's not that much of a difference in performance with Firefox. I would say that performance is more of the websites' bloatness issue, not a browser one. Especially considering the fact the internet becomes unusable if you don't block half of it's unnecessary garbage (ads, paywalls, tracking, etc.). Anyway, about browsers. My take on chromium is that it's completely soulless and unflexible product. Firefox is better at customization and methods of achieving privacy with things like about:config, css themes and user.js support. Chromium browsers don't have anything of that, nor do any of it's forks.

>>76341 It shouldn't be that slow. Have you tried using KVM as the accelerator and enabled CPU IOMMU?

VT (virtual terminals) pissing me off, ssh etc. the unix ecosystem of remote admin basically. >get VIM configured properly for use with Dvorak. >finally, I think I can gut used to it >haz wireshork (my fork with ssh connection logger tie-in) being promiscuous. >ssh to megacorp server to do some admin werk. >vi opens & VT pukes TERMCAPS dovn the link, ugh! >All my KEYBINDINGS they do NOTHING. come the fnck on, if a VT is gonna waste data on TERMCAPS why don't we have a PROGCAPS that remotely configures the progs so my local config is useful... At least let the remote VI receive event stream from my local VI. >calm down, this is tard world, think smarter not tarder. >OK, make a vi.sh wrapper to ssh back to myself and scp the file-to-modify back and forth, invoking vi locally >because tard world blue team locked down the server vi config for no reason: >I can invoke cpp and run results, even priv escalate with gdb attached to a root proc, trust me or fire me, plz! Today's Thick client model iz t3h suxx0rz! Mainframe L-users don't put up with this.

Obsidian. what's with all the hype for all the bells and whistles when it's really just textfiles and the behaviour can be replicated with, say, perl scripts, if not just sed and awk?

>>76366 Another annoyed obsidian user here. I like it because it feels more "fluid" thab typing on a terminal for a long while, and because i can embed pdfs on markdown files. But it hasn't been working for like a month now for me. I had my whole "knowledge base" or however it's called in there :( I'm guessing it's an electron problem, what a surprise, right?

Wine, I get that its not windows but man I hate windows.

>>76381 I'll second wine. I do cross platform code, I cross compile on linux to generate win binaries and do some of the testing on wine. Wine's behavior of "wait for single object" is broken, it never interrupts the wait say like when SIGIO is delivered, makes my nonblocking IO code test harness fail -- probably because it uses Pthreads buggy/nonstandards compliant library internally to impl wait for object; POSIX threads condition variable always waits for the whole duration instead of interrupting on OS signal delivery, even if you properly configure sigaction and enable SysV (interrupting) syscalls. Yep condvar wait is a cancelation point, so (as man signal says) enabling sysV syscalls should cause interrupiton upon signal delivery (so you can build quick/performant code sans busywait). Pthreads devs have a history of overly aggressive application of signal masking then gasslighting bug reporters. Last time this class of problems was encountered it took 5 years and a literal act of congress (convention of Posix standards body, issuing pthreads specifically a clarification of standard) for the pthreads devs to even admit the bug was real. I don't have time for the bickering FLOSS fiefdoms to haggle over whether or not my pull requests are neccesary or whether I simply do or don't understand the "very nuanced" issue and accompanied ASM code. My own multiprocessing / multi threading lib uses GCC atomic builtins to implement condvar. (Am I the onlyone writing peformant linux code? Seems that way). Primarily a CPU (or its processes) do 2 things either run or wait for IO. The only real reason one needs condvar is to wake from IOWAIT on SIGIO, and pthreads masks off (ignores) the damn signal (wake event), making nonblocking/asynchronous IO a pain to implement on linux. The "best practices"(tm) of how to do this keeps changing so I'm just building my own (rathr than rely on the behavior of newish complex kernel operations seemingly made just to work around pthreads bugs). I guess wine devs are just another fiefdom, provided a unit test to recreate/demonstrate the issue & a patch to use my free (MIT licensed) condvar code, they cite that no one else seems troubled by my problem. Of course they're not -- their users use wine instead of a proper VM w/ gpu passthrough. Well wine isn't going to be useful much longer for me too.

I will counter the thread's theme with some positivity! Years ago I had attempted to use Krita and I hated it. It was janky for me. But recently, upon trying it again I found it to be quite lovely. It's not a photoshop replacement for image editing, but for drawing it is completely viable. If any Krita devs happen to ever read this: thank you–I really appreciate your efforts.

>>76339 Unity. Mostly because I'm still relatively new to it and have to finish a project on a tight schedule. The GUI+script integration and having to constantly dragging and dropping stuff everywhere feels clunky and it's extremely time-consuming. Random bugs (for example with recognizing input devices, etc) and things that should be easy being complicated (like cloud collaboration) also grind my gears.