I usually compile with --quiet-build=y, it doesn't have to be configures and makefiles blasting into a shell window the whole time. On the rare occasions where a build fails there's still the log in /var/tmp/portage/....
Before I had a proper internet connection (had to ask permission to borrow a dial up account) I bought a magazine that had a picture of a cow on it saying that Larry the cow was different. It was a DVD image of the stage one mirror of this new fangled Gentoo thing.
Learnt from the magazine how to install a bootloader and so on and then "bravely" typed emerge world into the terminal after configuring the list of all the packages I wanted. Including a full desktop (KDE I think but may have been Gnome). And Firefox. And Open Office. And some multimedia stuff I don't remember.
I am rolling a few Gentoo VM’s these days and it’s really not that bad to compile things these days and I am on an old ass (10 year) dual Xeon setup.
I remember X taking a few days to a week to compile back in the 2000’s
Relatable. Me: wants musl libc and to build stuff with clang (so that it's not gnu/gentoo). Firefox: doesn't want neither muls, nor clang due to some god knows how old bug.
Even under FreeBSD and OpenBSD they use GCC for things requiring it, which kinda highlights Gentoo philosophy's problem in this regard. Setting USE flags mostly globally seems like a cool idea, but when for customization it gets down to setting them for every package - one could as well use FreeBSD ports.
I mean yes if time is an issue, but compiled code on your own hardware is specifically tuned to your machine, some people want that tiny tweak of performance and stability.
I use both for different purposes. Gentoo’s feature flags are the reason I wait for compiles, but only for computers a touch the keyboard with. Everything else gets Arch.
would you mind elaborating on the benefits? like what does one actually gain in a real-world scenario by having the software tuned to a specific machine?
disk space aside, given the sheer amount of packages that come with a distro, are we talking about 30% less CPU and RAM usage (give or take), or is it more like squeezing out the last 5% of possible optimization?
Closer to thr 5% . Between the intermediate code and final code writing there is an optimization stage. The compiler can reduce redundant code and adjust based on machine. i.e. my understanding is an old 4700 can have different instruction sets available than the latest intel.gen chip features. Rather than compile for generic x86 the optimization phase can tailor to the machine's hardware. The benefits are like car tuning, at some point you only get marginal gains. But if squeezing out every drop of performance and reducing bytes is your thing then the wasted compiling time may not been seen as waste.
