Best BSD alternative(s) to FreeBSD?

FreeBSD is usually the first of the BSD systems to support new hardware, that's why I use it on my desktop. Video drivers are one of the more critical ones. Video processing is a big part of a desktop computer's usefulness for me.
 
I've read the number of lines of code in the host driver for recent accelerated 3D cards is exploding. Making it problematic in resources to port.
 
I've played a video on DOS once, but never tried any editing. I doubt there's any viable program for that. I actually never was a fan of applications and only use DOS with small utilities which I sometimes string together. That's why I found *nix comfortable eventually.

BTW, my comment about video was a semi-sarcastic one stemming from the difficulty of searching for information about video performance for non-game or video-editing usage. :)
 
I don't do any complex editing that requires a GUI program, but ffmpeg works quite well on FreeBSD. I do have to use the port which usually winds up, a few months later, causing conflicts with packages as I want a few non-standard things. I should add that I requested x265 be included in the default package and the maintainers and others who saw the RFE agreed. I repeat my stuff is all pretty simple, adding an audio track, perhaps combining videos, perhaps re-encoding so a Mac can easily play it, stuff like that.
 
These hardware vendors contribute to Linux kernel because Linux is the most widely used server OS, but also want/(implicit)allow other OSes to take and reuse/port the code. So these code mostly licensed with permissive licenses. But kernel modules (in-tree, mainlined) are a different story, they're forced to be GPLed.
I doubt their intentions are so altruistic. They are simply avoiding using copy left licences for their code.

Also the code in question is actually "in-tree". Refer to the link I posted and you'll observe the licences used in the Linux kernel itself and that the amdgpu driver is part of the source tree (as with the intel, radeon, nouveau GPU drivers).
 
I doubt their intentions are so altruistic. They are simply avoiding using copy left licences for their code.

Also the code in question is actually "in-tree". Refer to the link I posted and you'll observe the licences used in the Linux kernel itself and that the amdgpu driver is part of the source tree (as with the intel, radeon, nouveau GPU drivers).
I guess they would have more benefits to use FreeBSD, OpenBSD or NetBSD, because they would have increased stability of their products.
Licensing is a long story. Since Linux is being used for Servers, they go for it.

Looking the logs, it seems that sourceforge uses Nginx on ubuntu server ;)
 
I've played a video on DOS once, but never tried any editing. I doubt there's any viable program for that. I actually never was a fan of applications and only use DOS with small utilities which I sometimes string together. That's why I found *nix comfortable eventually.

Does Canada have patents? If not, there's a DOS version of ffmpeg that you could use for video editing.
 
I've read the number of lines of code in the host driver for recent accelerated 3D cards is exploding. Making it problematic in resources to port.

3D games have always been the motivation behind advances in 3D acceleration, but there are other uses for it such as 3D modeling. Though it seems those other applications don't get much attention. But yeah those drivers are surprisingly large. I wouldn't want to be the one to sift though all that code.
 
I think Linux can fill a gap where hardware support concerns. Distros like Alpine Linux is minimal enough to build your system. One can install Xserver/Wayland, window managers etc. However Alpine Linux uses musl which is lightweight glibc alternative. Therefore to use nonfree applications, one must (i.e. the author etc) need to build those against musl libc. However there is a libc6-compat mode where certain statically liked glibc apps claim to run. I am a minimalist guy and I like both FreeBSD and minimalist Linux systems (except those general purpose systems with tons of internal complexity, such as Ubuntu, and desktop oriented distros , mostly). FreeBSD chroot system is quite simple, due to availability of base system as tar.xz package. I think I can get some luck with stage 3 tarball :)

You can try experimental, minimalist, DIY distros like Alpine Linux, Void Linux, Sabotage Linux, Gentoo, GoboLinux, GuixSD, Slackware and CRUX.

I don't know, I just prefer a plain shell install over any other method. In this case, both {Free,Open}BSD are exemplary to the point where I don't even have to look up the wiki or copy / paste any boilerplate. I also find the Anaconda installer, featured in most RedHat sponsored distros to be the most horrible among existing GUI installers. Not only is it slow and bloated but also unlike most, it follows a non linear process which can be quite confusing to a new user.

"RedHat sponsored distros" - There you go. Advertisement leads to consumerism.
 
For sure! I don't think that video people and gamers will ever find out about that though. We're invisible to them.

I agree with this! I am a computer graphics person but I see a worrying trend these days, especially with Linux (and more specifically, what Linux users want) which will possibly pull through into FreeBSD unfortunately.

*Everything* is becoming about the desktop visual experience, high performance graphics and smooth playback. We have things like Wayland which is destroying crucial enterprise features such as multiple user sessions for VNC. Yes, X11 directly is a bit old hat for remote desktops across the internet but VNC relies on XDMCP to kickstart a remote session. Without this VNC will be like Windows where you can only stream the one session (the same as what appears on the monitor).

Likewise a lot of cli tools now in Linux are seemingly designed to be automated by GUI tools. They are far too verbose to be used directly from a terminal. For example ifconfig has been replaced by "ip" which requires a lot of instructions to achieve simple tasks. Another one is the firewall, getting that to kick in is not very elegant, having to write a bunch of post.ip.rules.d crud.

One day perhaps we will have a beautiful desktop with fancy transition effects (ugh!), but such a stupid useless operating system, it isn't actually worth interacting with XD

This could possibly be due to the companies behind Linux (and ultimately FOSS) need cool graphics to sell their products. Or it could be due to Windows "migrants" who want to keep their flashy and inefficient Windows lifestyle in Linux.
 
Because Linux target is Windows and the desktop and not technical competence.

mod3777 I would like to remind everybody that technical discussion of the Linux operating system, even here in the Off Topic board, is off topic still.
 
One day perhaps we will have a beautiful desktop with fancy transition effects (ugh!), but such a stupid useless operating system, it isn't actually worth interacting with XD
.oO(Fisher Price OS)
 
..... I see a worrying trend these days, especially with Linux and more specifically, what Linux users want.....

After my real engineering job went away (the whole company went away) I spent a year selling Apple computers (Apple Solutions Consultant :rolleyes:) I doubled sales at my CompUSA store in one year. For the rare techie customer I could hit a key and switch into a KDE interface and show "unix" but for most customers all I had to do was slowly pull the mouse pointer down to the dock, watch the beautiful icons expand as I hovered over them, click one of the icons and watch it do its cute little jumping up and down while the window opened. Sold!
 
I would say it's OpenBSD.

My first BSD is FreeBSD. After that I played a bit with virtualization but unfortunately my CPU too old and not supports VMX. NetBSD is the only BSD could run on Virtualbox with this trick. So I love NetBSD. Where other OSes refused to work, only NetBSD runs. It reminds me of my loyalty black dogs. But they both gone. I'm also given up on NetBSD. Now I've an upgraded system with full VMX and VT-d capable. I could play with NetBSD further. So bad I realize, the system is blazing fast with 4 cores and 4G ram, even boot up faster FreeBSD with the same configuration, but software is severely lacked. I means binary packages I could install with pkgin. Sometimes pkgsrc.se list the port as available but it corresponding binary package is not found (not yet built? I don't think so. I think it can't be built, so it's unavailable, it's also means the port is indeed broken). You know, packages for firefox, codeblocks, codelite, chromium,... and so many commonly used software are unavailable. Left me with only seamonkey that works and reliable, but the porter forgot to create a desktop entry for it, so I've to start it from the terminal. I tried my best to bring the system to a working state, and I partially succeed with a MATE desktop started via .xinitrc (there's only slim-themes but no slim package, ironically!), seamonkey, geany, libreoffice, ibus. I also not have much success with pkgsrc other than building basic port like bash, joe, nano,... whenever you started to build gui based apps it will be full of problems. Made me wonder, it's the same pkgsrc, how could they deliver binary packages to me, when I can't built the ports myself at all? And I switched to use NetBSD as a server, I not yet want to give up on it. But the lacked of software stopped me again. I want to setup a svn server with NetBSD VM, and I can't found mod_dav_svn anywhere. I changed to make it a NFS server, despite it's only has NFSv3 but if I could find a useage for it it's better than nothing. Then I also realized I've to use brigde interface and to my surprise, how fast it is with NAT how slow and unreliable it is with bridge. So I given up on it completely.

To my sincere, I advised NetBSD to follow the way of SmartOS, become a hypervisor OS and only that. It has many potential: going to have up to date ZFS ported from FreeBSD and the new NVMM. When you can't compete with other on the server side (web servers, mail, file servers,...) and don't have the resource to catch up Linux on the desktop side this is the safest way for you. I heard people said NetBSD also has potential on the embedded side but I don't know anything about embedded so I don't mention it.

Someday I would try NetBSD again and check if I could use it as a hypervisor OS or not.

Back to the topic, I choose OpenBSD because I think it's a better NetBSD than our current NetBSD. DragonFly also has many impressed features but the last time I tried it on Virtualbox (5.6.1?), the HAMMER2 FS just unreliable, after I installed MATE and set everything up in rc.conf I restarted and HAMMER started to fail, if I'm not wrong something about malloc error. But I don't blame the developers, though. They've done incredible job even though they're such a small group.
 
Hello,

In any case you cannot run, install,... FreeBSD, which other option would you have ?
e.g. OpenBSD, NetBSD, ...

which one would be the best alternative ?

thank you!
With best regards
DragonFly actually. More-or-less equal WiFi chipset support (AR93xx/94xx also supported for example), 'em' driver is in it's older stable form, uses modified variant of FreeBSD ports, familiar userland. Just kernel is different, tho that "just" is different a great deal.
At once very familiar ('old pre-fork FreeBSD shows through') and strange OS. Learning curve is less than with other BSD's if you want to start using it right after install. Often, when you can't find DragonFly's examples or docs, FreeBSD's work. Huge downside: no driver support for Nvidia. but Radeon's driver support might even be a bit better. No forums, mailing list is pretty inactive, everything communication-wise seems to flow through IRC. Worth trying out.
 
No forums, mailing list is pretty inactive, everything communication-wise seems to flow through IRC.
This could change in the near future :
The mailing lists are not seeing much if any activity any more. This is
more a generational issue... people kinda prefer web-based forums these
days and younger generations do not use mailing lists at all for group
stuff (not really). Even the devs almost universally use IRC and not
mailing lists for discussions now (its kinda too bad that we don't have a
permanent irc log stored on DFly servers for posterity). So we are looking
into potentially shifting user interaction to a web-based forum
, perhaps
this year, and retiring the mailing lists, leaving just an archive for the
mailing list. Possibly sometime this year, so look for action on that
upcoming.
Source : http://lists.dragonflybsd.org/pipermail/users/2019-July/358226.html
 
Yeah, I have subscription there, I've seen it and have read it. Most follow-up replies were content to proceed without creating web-based forum.
 
Yeah, I have subscription there, I've seen it and have read it. Most follow-up replies were content to proceed without creating web-based forum.
That's very wrong. A user friendly web forums is a big plus for any OS/distribution.
 
I find web based forums hugely easier to use than mailing lists. I would be happy if FreeBSD made more use of them. I'm not a developer, but I would like to be able to browse developer communications more easily. For example my interest in FreeBSD is mainly as a desktop user and I'd like to be able to follow development for the latest processors and graphics. I find it rather difficult to zero in on info about that from the lists.
 
I find web based forums hugely easier to use than mailing lists. I would be happy if FreeBSD made more use of them. I'm not a developer, but I would like to be able to browse developer communications more easily. For example my interest in FreeBSD is mainly as a desktop user and I'd like to be able to follow development for the latest processors and graphics. I find it rather difficult to zero in on info about that from the lists.
No. This time I disagree. I would like a web forums to replace the users related mailing list but not the devs related mailing lists. An automatically mechanic to import messages from the devs related mailing lists into forums threads and posts should be welcome. But I could sure no devs would ever agree with your proposal to replace the devs related mailing lists with web forums. I think devs need their own space. We as users also need our own space. Some devs hanging around users' forums should be great, though ;)

p/s: you're a little off topic here, as we said about DFBSD's mailing lists and the needs of a user friendly web forums for it. FreeBSD already has a great web forums :)
 
Yeah I was going to say that would probably never happen. They started with lists and they're probably here to stay. Just saying it would be better for me, but I'm not the one doing the development. And that's a good point, putting development stuff on a forum would probably cause problems.

One option would be a forum read-only for users where only the developers actually have the ability to post. I did find one site mentioned on the forum that compiles the lists into a forum-like format, but still not as a clean in terms of how a forum looks. Helps though; http://freebsd.1045724.x6.nabble.com/
 
Yeah I was going to say that would probably never happen. They started with lists and they're probably here to stay. Just saying it would be better for me, but I'm not the one doing the development. And that's a good point, putting development stuff on a forum would probably cause problems.

One option would be a forum read-only for users where only the developers actually have the ability to post. I did find one site that compiles the lists into a forum-like format, but still not as a clean in terms of how a forum looks. Helps though; http://freebsd.1045724.x6.nabble.com/
I already proposed a similar idea :) Instead of only the devs allowed to post I think they just done everything on their mailing lists, the conversations then synced to us on web forums.
An automatically mechanic to import messages from the devs related mailing lists into forums threads and posts should be welcome.
 
Back
Top