My review of TrueOS


Somebody recently started a thread on the FreeBSD forums asking for peoples opinions on TrueOS:

So I thought I’d write a short review / summary of my experiences over the last few months of having used it as my main OS which I’ll paste here too, to save you the extra click :wink:

I’ve been using TrueOS as my main OS for the last few months. I like it and I want it to succeed but I cannot recommend it over FreeBSD for servers or Linux/Windows as a desktop OS. Not yet anyway.


TrueOS is an easier way to get started with FreeBSD than vanilla FreeBSD. GhostBSD is still based upon FreeBSD 10.x and doesn’t support ZFS. GhostBSD 11 has some experimental ZFS support but its devs don’t recommend it be used. Hence GhostBSD is ancient and is of no interest to me whatsoever because ZFS is pretty much the only reason I use FreeBSD and TrueOS.

Better (Intel) GPU support out-of-the-box - My laptop has an Intel Haswell GPU. Whilst thats pretty old now, its not properly supported in FreeBSD 11 nor 12 - I cannot achieve 4K output under either. TrueOS has drm-next integrated into its kernel and so this allows me to get the full 4K res out of my display. This is the main reason I’m running it over vanilla FreeBSD.

Boot env install. I think this is the best thing to happen to OS installation during my lifetime. I hope FreeBSD, OpenIndiana and all other ZFS-friendly OSs adopt TrueOS ability to install into a boot env.


I’m very concerned about TrueOS support or lack of it. I have opened 7 github tickets for TrueOS specific issues over the last two months. Only one of them got a reply - that PCDM autologin was broke. They said they fixed and closed the ticket but myself and other users have found that it is still broken and we’ve had no reply since. One of the other unreplied issues involves the TrueOS installer trashing my FreeBSD 11 install and making it unbootable - its been over a month since I reported that and I’ve not had a single response from the devs yet. I realise these are complex probs and may take time to fix and whilst It would be unreasonable to expect an instant fix the TrueOS team should make every effort to at least reply to all valid bug reports within a few days of people opening them or else people are likely to give up on the project and they won’t have any testers or users.

Lumina. I understand why they want to create a (Free)BSD centric desktop and I would like to see it succeed but its just not ready yet. The Lumina version shipped in current TrueOS releases is buggy (and ugly) as all hell and I simply find it unusable. I can’t even play a video fullscreen properly. Once you swap it out for a more mature desktop such as MATE, which is easily done thankfully, you get a usable desktop OS but first impressions count and Lumina is not putting their best foot forward. If the TrueOS/Lumina team are reading this - PLEASE don’t waste your time on creating any Lumina apps! There is too much work needed to be done on the actual Lumina desktop to be wasting time creating new, Lumina specific text editors, media players etc.

Poor default apps. TrueOS now ships with Qupzilla as the default browser, to save space on the install iso. Bad idea! I used it for 10 minutes until I encountered serious probs and was forced to install Firefox. I would honestly rather see TrueOS ship with no browser than an unusable one. In general I find most of the apps shipped with TrueOS not to be fit for purpose and I have to install more suitable alternatives.

On top of this, as someone has already mentioned, many users (including myself) have been unable to install any TrueOS releases since March/April. Again, this is something I have created a TrueOS github ticket for but it has been ignored so far.

TrueOS holds much promise but they need to up their support game and start listening to their users more.


Not to diminish your experience, but perhaps some of the “bad” is related to expectations; as in “exactly what are you expecting from an install of TrueOS”? The equivalent of an Ubuntu Linux install? The equivalent of a Windows install?
My reason for asking is that my experience is different because my expectations are likely different.
Lumina: apps designed for it make for a better “desktop environment” experience. Gnome, Mate, KDE, Unity all have their own versions of apps such as text editors, media players, etc. Again, this is where my expectations differ because I don’t boot into graphical mode, I boot to command line and then manually startx with the window manager of my choice (WindowMaker) and I use the apps I want to use. The current version of Lumina that ships is still using FluxBox as the WM (I believe) so some problems could be related to that.

Default apps: This is difficult because everyone has their own idea as to the “best” ones to use and exactly what apps should be included in a distribution? Have you ever installed OpenBSD or DragonFlyBSD? If not, “embrace the CLI and the pkg command” to get the desktop environment of your choice. I like that paradigm because there is no bloat by default (unlike Ubuntu, Fedora, etc).

Support: not a trivial thing. A lot is based on the size of the team, is this a full or part time gig, what else is going on (the past few months have been loaded with conferences and gatherings of BSD and other folks). If one has the ability to submit patches or workarounds, that eases the load on the main developers. If not, then one has to be willing to try different things as asked either by the devs or if you’ve posted topics here by folks with ideas. An example is your trashed FreeBSD 11 install: did you start a topic here (I’m asking because I don’t recall seeing one and can’t find one with a quick search)? Were you trying to dual boot FreeBSD11 and TrueOS? Those details make a lot of difference and unless someone has a similar setup (down to at least the same machine and bios settings) to try and reproduce, they can only give ideas (look a the threads relating to ASUS machines).

Again, I’m not trying to trivialize or “hand wave” your experience away, just trying to understand.


Hi mer

What was I expecting from TrueOS? I want TrueOS to be at least as good as but ideally better than the popular, more user friendly Linux distros such as Ubuntu, Mint and opensuse etc. I realise its not entirely fair to expect feature parity as Canonical and RedHat have many more people working on their distros than iX on TrueOS so in that respect, TrueOS is doing very well, all things considered.

I hope I didn’t come across as being anti-Lumina because I’m not. There should be at least one X/Wayland desktop that targets the BSDs as the primary platform. I hear things have improved here a bit in the GNOME camp at least but I still prefer MATE. In my opinion, Lumina has too many shortcoming currently and it shouldn’t be the default yet.

MATE would make the best default IMO as it offers a nice balance between features and being lightweight whilst also being well tested and mature. The fact that Lumina currently uses Fluxbox is actually a good thing IMO. I know Fluxbox is rock-solid so I can be sure that non of my current Lumina issues are the fault of Fluxbox. I do not anticipate its replacement will improve the situation, at least not for some time after it arrives.

I have installed NetBSD and DragonFlyBSD in the past yes, as well as many other OSs such as Haiku, macOS, OpenIndiana etc. I think OpenBSD and NetBSD are too outdated. NetBSD is only just getting USB3 FFS! I think DragonFly is interesting, esp. HAMMER2 but really I’m only seriously interested in using FreeBSD and TrueOS thanks to their proper upstream ZFS support and better HW support than Illumos.

As for the default apps, I suppose I could choose the TrueOS server install mode and then install my desktop and desired apps myself. I could do that yes, but my parents or anyone new to UNIX wouldn’t and so they will base their opinion of TRueOS on the defaults. I would rather the install iso be 500MB bigger and include Firefox by default rater than save a few hundred MB and ship qupzilla which barely works.

You ask about my trashed FreeBSD install - that is covered in this github ticket:

I was told that the github issues page is the official place to post bugs, and that I have done in much detail. I shouldn’t have to be chasing the devs up on gitter or on here (discourse) just to get acknowledgement.

I realise TrueOS is only a small team but they should do people like myself who write detailed bug reports and feature requests the favour of at least acknowledging all submissions within a few days of them being posted, even if its just to say. “Thanks! I’ll try and look at this soon”


Good Points:

  • Way easier to use for a newcomer that can later, slowly learn more, and more advance things.

  • Always uses the most updated software

  • Direct access to the developers, who actually make a real effort to help us users. Try getting Linus Torvalds to answer your forum questions.

  • Bugs get fixed very quickly. I posted an answer on the PC-BSD forum suggesting that the user use the full name rather than the username and see if they could log in. The OP opened a bug report and it was fixed the next day.

  • It comes with a desktop specifically designed for *BSD not a Linux DE that needs extensive hacks to get working on *BSD’s

  • TrueOS helps FreeBSD tremendously by using Current which means their Release version of FreeBSD will be far better because of the blood TrueOS spilled on Current. And since TrueOS is based on FreeBSD it will also be better. It’s a vicious (cross pollination) cycle

  • It’s not Windows

  • It’s not Linux

Bad Points:

  • Tries to compete with Linux to the point where people believe this is or at least should work like their favorite Linux “Distro.” when in reality is a very (very) customized version of FreeBSD

  • Tries very hard to help people boot multiple operating systems. You don’t see Windows or Linux putting so much effort into getting their users to a different operating system.

  • Does not have a good Bhyve manager (None of the BSD’s do though) to keep people from needing to dual boot and being able to just run the 1 or 2 windows or Linux apps Bhyve on TrueOS rather than VirtualBox.

The last one is more of a suggestion to the iX big wigs holding the purse. If you guys are reading this, I think that if you can put some money into that last idea. TrueOS would have a Killer app and people would have no other choice but to use TrueOS because it will offer something no one else does. I know you’ve been burned by trying to bring Linux apps like docker to iX but Bhyve is not a Linux app.


I am very aware that TrueOS is based upon FreeBSD and I know how FreeBSD and Linux differ, plus their respective pros and cons. Seeing as TrueOS is running exactly the same open source apps as most Linux distros, it is not unfair or unexpected to compare TrueOS to Ubuntu, Fedora or any other distro. Such comparisons are both logical and inevitable and it is for each individual user to decide which OS works best for their purposes.

We should all try to avoid OS flame wars. As much as I’d love to go all FreeBSD / TrueOS I can’t and that’s mainly because Bhyve lags greatly behind KVM and Xen. VGA passthrough doesn’t work at all yet under Bhyve, there is no USB pasthrough yet, PCI passthrough only works on a few systems and so on. There are several good reasons there is no Bhyve-based alternative to Proxmox yet.

Saying “It’s not Windows” and “Its not Linux” as plus points sounds childish - especially considering how much code the Linux and BSD apps share. I’m talking about the apps people use, not kernel/base utils. Linux isn’t going away any time soon and that is GOOD! Competition and alternatives are good.

Another reason I’m hesitant to recommend TrueOS/FreeBSD, apart from its inferior GPU and virt support vs Linux, is that systemd-based Linux boots over 10x faster on my machine than openrc based TrueOS. That is a big deal for most laptop users.


I’m not completely against Linux, because I also use an Ubuntu Phone, and an Ubuntu Tablet. And the only reason I use them is because it’s not, Android and it’s not IOS. Such a shame that they gave up on their mobile efforts, it was really the only thing I liked Linux on phones and tablets.



There’s not many ubuntu phone or tablet users out there!

Have you ever tried sailfish?



I think have to separate the topics by the content as it was on the PCBSD community (Desktop questions/Serves question) because a lot of topics are same and the new members (and some old too :slight_smile: ) lazy look all over.

But have to make some new topics with advices or with links (for example: ZFS tuning how to, Fester boot up how to, sysctl setup how to, How to download safely the installer files and write and check these, Dual booting, Setup in virtual box, System update, Drver intall, Electronic books from BSD, Useful BSD sites etc ) for the better support.


Congratulations for the last of the Good Points! :smile:

Maybe: :smile:


I tried to install sailfish on a Google Nexus 5, but I failed at it. I ended up installing Ubuntu on it. Before that I had the BQ aquaris E5 HD that came with Ubuntu installed out of the box I forgot it at McDonald’s when I came back to see if it was there it was gone. They don’t make those anymore :cry: After that I bought the BQ Aquaris M10 tablet that also came with Ubuntu installed out of the box. I still have that one. Now Ubuntu has removed any trace of mobile from their website and said they will not work on Mobile anymore. I think they want to focus on robotics now, self driving cars and the like. Oh well…


But you know about UBports right?


Yes, but mostly because their was a link to them on the Ubuntu website (it has been removed). That’s where the link would take you when you wanted to download the images to flash your phone. I know the guys their want to run their own thing, and they can because it’s an open source project, but without canonical resources it will be very hard.


I agree on the feedback of “Thanks, I’ll take a look”. As just a user, not a dev or anyone official, I can’t help.

So the FreeBSD 11 thing was over in git; that gets the devs attention, but not the community at large. I know it sounds a cop out, but a new topic here linking to the github issue may get more exposure. Yes, I know “chicken, egg”.

Now we get back to “expectations”. Most of the popular, more friendly Linux distros have had a lot more “lifetime” than TrueOS. Kind of like expecting Windows 3.3 to compare with Windows 7. :slight_smile:

Feedback is always good, just make sure it’s not “My 1976 Chevy Chevette can’t keep up with a 2010 Chevy Corvettt LC7”


My Counter Review of TrueOS

I’ve been using PC-BSD and TrueOS as my daily driver for a while now. I’ve used FreeBSD for various things and have a couple of arm boards I’m testing 12-Current on.

The truth is software is HARD. What the guys at ix are trying to do is unprecedented as far as I know. They are trying to produce a rolling release FreeBSD variant from the HEAD revision at the same time as produce their own UI. I don’t know if anyone noticed, but the has been a couple of companies that have lost their shirts attempting something like this.

FreeBSD is broken, get used to it.
FreeBSD RELEASE revisions are hard enough to work with sometimes. External applications are always in a fluctuating state of broken in FreeBSD. It’s not hard to find a port thats out of date. That’s what I love about FreeBSD. Personally, I dug in and got MonoDevelop to work on 10.3. The port has since been fixed and updated, but that’s because there was some of us that showed interest and kept working at it.

From what I can tell of the complaints from newcomers, it’s that they can’t get it to install on newer computers. I feel that pain. I sympathise. However I stuck with TrueOS and FreeBSD on and off over the years and it’s paid off (Look at me ma! I’m playing with kernel code!).

What am I saying: FreeBSD is a great platform to drill in and learn about your favourite applications. If you want to learn about open source Unix (oops, can I say that?) Give TrueOS a try. The TrueOS team have created a reasonably stable platform that works on most systems. They provide a complete computer operating system in two variants: Desktop and Server (no desktop) edition. TrueOS is a rolling release edition, meaning you get all the updates to drivers and utilities long before their back ported to RELEASE editions of FreeBSD. That means in the future they’ll be able to run on current hardware sooner. And as an added bonus, you get to help test and improve your favourite operating system!

Lumina is a fantastic desktop. I fight with it whenever time (read kids) permits. I’ve been using it off and on since beta days. I can’t believe how mature it’s become. The fact that someone would even bother to compare Lumina to Mint is a testament to the hard work of the TrueOS team, regardless of their opinion.

Sometimes though it frustrates me. I have KDE installed and keep threatening my computer that I’ll switch desktops if it doesn’t play nice. That’s the nice thing about TrueOS, I can switch the desktops by simply downloading the package from app cafe. But threatening my computer happens less so now because I’ve been using Lumina regularly. The hot keys are different from what I’m expecting (omg ctrl+arrow!), I find the user experience things less than fluid. But then every now and then I find a little tidbit. “Oh, look, I can tab and type through Insight File manager without having to reach for my mouse”! “Oh,wait ctrl+pgup ctrol+pgdown is tab in ALL my windows. That means I don’t have to contort my left hand to cycle using ctrl+tab.”

And then I remembered something. Crap, I had to go through all of that with KDE. I had to fight through that with Windows 8. I know this story: It’s called learning something new and it sucks.

What am I saying: Lumina is a perfectly usable desktop with a definite work flow if you can find it. I have trouble finding things sometimes (Network manager is in Utility but Firewall is in Control Panel?), and I still hit ctrl+right arrow (argh!). I ask openly if there is documentation around the user experience (i.e. all the hotkeys) as I’ve been to lazy to go find it. However, it doesn’t crash, and does a good job of not being intrusive (I’m looking at you KDE). All the basic functionality of an OS desktop is there. Including an update manager…

Updates and Rolling Release And Conclusion
Did I mention that software is hard? I’ve experienced a couple of difficulties with software updates over the past couple of months. I’ve been able to work through them with the guys on this forum, but I’ve got the bullet wounds that it took to get there. It’s frustrating when things like updates don’t work, especially for people not familiar with FreeBSD and TrueOS. However there’s only ONE way to get the release system to be better: use it. The more people using it the more problems that can be found and fixed.

Likewise, the head revision is an absolute gong show from what I can tell. In arm right now pkg is broken on the IMX6 image and I keep getting a lock order reversal that kills ports! Neither the TrueOS nor the FreeBSD teams have ever had to keep head stable before. In my opinion it’s going to take a while before it gets better. But it gets better faster if we all keep working at it. Without TrueOS, how are we ever going to get there? I’d love a committer opinion on that.

What am I saying: TrueOS is a fantastic new rolling release variant of FreeBSD. If you want to play Netflix and have a nice easy free desktop to use, I say go grab a Mint image. If your interested in working with the most current FreeBSD software,help evolve a brand new UI and want to be involved in making an open source OS better, TrueOS is a great choice.


I’ve been a 95% FreeBSD desktop user for about 15-17 years. I wipe my drives frequently, and re-install the OS. Why? I consider hard drives to be like underwear. Once in a while, it’s a good idea to launder. On FreeBSD, I do it all manually, and it’s a little tedious. Yes - I sometimes keep a ghost image (but that get’s out of date). So, for the past few months, I put the TrueOS disk in the drive, start the install and walk away. Have a coffee, come back and it’s ready to go.

I like Lumina, for its light weight. I don’t use Linux (much) because it’s no longer light weight (kind of the same metaphor as the underwear applies for me). Putting Mate onto a BSD distribution would make me feel like it’s laundry day every day. No thanks. Nothing against Mate - just an opinion that it’s too “not light weight” for me. Everything’s relative.


The reason to ship with Qupzilla has more to do with ISO build times which affect testing. Thunderbird alone takes 2 hours, and nothing else can build while it builds. Its a 1 hour vs 6 hour build time for ISO packages to chase installer regressions between 3 seperate developers with limited time.


Thunderbird takes 2 hours to build!?! I presume you’re not building it on P4’s either! Whoah!

Evolution is my desktop mail client of choice but I don’t want to get into a holy war over that.


Why has nobody replied to any of my github tickets? All of the following have had no replies: (Kris did briefly reply to this just to say “works for me” but he hasn’t answered any of my questions)

I hope you appreciate I’ve put a lot of work into making those as detailed as they are and that I’m upset they haven’t even been (properly) acknowledged yet.


I know your tickets have not been overlooked intentionally. Actually your feedback as well as another review was brought up postivley in a meeting that triggered default application discussions. I also suggested maybe chrome, and evolution as alternatives in addition to a community poll providing a range of choices. There is good reason to keep Qupzilla though. Although I fully agree it is not nearly as functional as others, and I do not use it myself. I suspect someone may address the tickets soon. Sorry about that.


Hi jmaloney

That’s good to hear!

I look forward to taking part in a default app discussion/vote and helping get those tickets resolved.



You do realize the comparison you are making there right?
Gnome 2.0 released in 2002
Gnome 3.0 released in 2011
Mate 1.0 released in 2011 with the discarded Gnome 2 code base.

You’re comparing 15 years of development of countless developers to a completely fresh code base that’s been written in approx 3 years of development by one person (Ken) and a year or so of other contributers (myself and others) Lumina 1.0 was released in 2016.

How is that in anyway a fair comparison. Yes, Lumina is not at the same point as Mate, KDE or Gnome. How could anyone expect that it would be?

To be clear, I’m not speaking ill of Mate! I know Martin Wimpress and I’ve spoken with him countless times about a myriad of topics. What he has done with the Gnome code base is fantastic, and I look forward to its continued development. But Wimpy inherited a fully stable and mature code base… he didn’t have to re-write the whole thing from scratch.

KDE 5.0 was released in 2014 with a huge team of developers, and they had been working on it prior to its release, and most people are just now considering KDE stable for daily use. With an entire team of developers, it’s taken them three years to get to a stable state.
Gnome 3.0 was considered by many in the Linux community unstable until the 3.12 version which was also released in 2014. So again, you’re talking about a team of developers taking three years to stabilize a desktop.

What Ken has accomplished by himself is quite honestly amazing, and he did it while also developing a ton of other things at the same time.

I apologize if my response comes off rather defensive, but the way you worded your post is quite honestly extremely disrespectful. The TrueOS team is an extremely small team compared to other Operating systems, yes we do take a ton of work and effort from FreeBSD, but we are going in a totally different direction from them, so we’re the ones finding the landmines. The reason things certain things have not been accomplished is quite simply a matter of limited resources. There are only a few people who are actively coding for the project. (I’m not talking about drive by contributers, we love everyone who does that) What we do, if you can pardon the cheesy analogy, is a labor of love. We do appreciate feedback and constructive criticism, but hearing that something is ‘unusable’ and that we shouldn’t be working on $THING because someone feels that our code sucks really hurts. We need more bodies, we need more people contributing code, we need more people willing to spend time and effort helping us make things better. I was a Linux dev for quite a few years before migrating to PC-BSD. A year ago I didn’t know Qt… so I started learning it so I could pitch in. If you (or anyone else) feels that a particular piece of software needs work… learn Qt and jump in. We’d love to have you help!

As for default applications, the main problem is GTK. There are some applications where the options for Qt apps are minimal or non-existant. The GTK developers dont care about breaking backwards compatibility,and they are only focused on the Linux world. Once we start pulling in GTK stuff we end up in a domino effect of dependencies that means we have to pull in tons of things… including HAL which has known remote code exploits. HAL isn’t going to get updated because it’s been abandoned on Linux because it’s been replaced by udev, which has been swallowed up by systemd. The only way to get the new non-vulnerable code Linux runs is for us to bring in systemd… which I can tell you is never going to happen.

There are very sound technical and engineering decisions for why we have chosen to focus on Qt applications and are trying to avoid any GTK application if we can.

I really hope you didn’t mean to be hostile in what you said in your previous posts… so I wanted to explain to you where we are coming from and how your comments come across. How would you feel if you had a pet that you loved and cared for and I came up and kicked it and said ‘your dog sucks.’ That’s kinda the punch to the gut that your post gives. Please understand that we have not made the decisions we have made randomly… we came to those decisions after a lot of thought and debate and looking forward at where the software world will be in 5 years – and how we need to act now to be prepared.