Manjaro is no alien operating system to me for I have reviewed it three times in the past. First I reviewed Manjaro ‘Ascella’ XFCE edition back in 2013, then I reviewed Manjaro KDE (4.12.2) edition back in 2014, and then in December 2015 I again reviewed Manjaro KDE edition (Capella) which featured KDE 5.5.1.
I actually tried to review Manjaro Linux a few months ago, but as I pointed out in my Ubuntu 17.04 review, back then the partition table was set up as logical volumes (‘logical volumes’ are mostly in use in server environments due to their ability to let administrators easily expand the storage capacity without having to take the whole system off-line), and unfortunately, the Manjaro installer (‘Calamares’) did not support LVs (it still doesn’t), therefore, I couldn’t review not just Manjaro, but a whole lot of other GNU/Linux distributions as well. So I decided to completely rearrange the whole partition layout and created a traditional partition for testing ‘Linux’ operating systems.
So anyhow, the latest Manjaro release is 17.0.2 and Manjaro developers provides you with three official flavors: XFCE, KDE & GNOME. Manjaro community also releases its own flavors including Cinnamon, Mate, Budgie, Lxqt, LXDE, i3 and Deepin. But I just wanted to put one of the official flavors to the test. While I deeply respect it, I’m not a big XFCE fan. The desktop UI doesn’t look modern enough for my taste. That said, the last time I tested, it was quite impressive and it was mostly suited for lower-end computers. So I had to choose between KDE or the GNOME releases. Well, since I was already using Ubuntu 17.04 which is largely based on GNOME applications, and I’m also hoping to review the recently released Fedora 26, so I just instinctively choose KDE flavor (64-bit edition).
Manjaro KDE 17.0.2 features plasma 5.10.2, Kernel 4.9.34, X-Server 11.0 and wayland 1.13.0 mainly. I have been actually using Manjaro for more than 2 weeks, and so far I’m quite happy with its performance, otherwise I wouldn’t have stick with it in the first place. However, I won’t go into all subtle details of ‘what’s new’ in KDE in this review. This review will actually be more based on the performance of the operating system and its stability.
The reason that I’m going to avoid describing ‘what’s new’ in KDE is largely because I’m not hardcore KDE user, therefore, ‘what’s new’ for me could easily be ‘old news’. Thus, to save the embarrassment, I’ll skip it, but below is an official video from KDE featuring ‘what’s new’ in the KDE 5.10.2 release. It’s about 2 minutes long and should provide you with a brief introduction to this plasma release, so I urge you to view it first:
Before starting the Manjaro 17.0.2 KDE review, below is a brief introduction of the hardware on top of which I tested it:
Intel Core i7-5500U, Hybrid GPU Setup (Intel Broadwell HD Graphics 5500, Nvidia 920M), 4GB RAM DDR3, Hybrid Permanent Storage Setup (Seagate 5400 RPM, 500 GB rotational disk and a Kingston 24 GB SSD), Qualcomm Atheros AR9565 Wireless Adapter, Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller, Realtek ALC3236 Sound Card, LED Display (1366 x 768 resolution, 60 FPS/HZ). It's an Asus laptop (F302LJ-FN024H).
I don’t have performance related data (Boot-up speed, Memory usage after loading the desktop, Power-usage at idle, System responsiveness --more on that later & Shutdown delay) of a recently released KDE distribution. The closest I have is data from my Ubuntu 16.10 flavors comparison review (they were all released in October, 2016). Kubuntu 16.10 is based on KDE plasma 5.7 release, and since its not a too distant cousin of Manjaro’s KDE plasma 5.10.2, I decided to take performance data from it for the comparison. However, since I felt that not including an up-to-date distribution’s performance data is somewhat ‘wrong’, I also decided to use data from the Ubuntu 17.04 (released on April, 2017) for the comparison as well. While it’s true that KDE and GNOME are quite different from each other (philosophically and a practically), but since they’re both desktop environments competing for dominance, then why one should not compare them?
So anyhow, let’s begin the actual Manjaro KDE 17.0.2 review.
‘Calamares’ is a very intuitive installer. It’s easy to use, doesn’t confuse the user with too many complex details, although its stability has been somewhat questionable in my experience. For instance, it has crashed in some of the KDE distributions in the past thus totally terminating the installation, while trying to change the ‘Keyboard’ settings.
However, in recent times I’ve been much luckier and haven’t seen the crash so far. All in all, I’m very satisfied with how it performed in Manjaro 17.0.2 KDE edition.
GRUB Theme & The Boot-Logo…
Manjaro includes a GRUB Theme that matches its Black & Greenish styling as shown below (I took the photo using my cheap android phone, sorry about the quality). There is also a Green Manjaro logo around the right-bottom corner of the screen as well (it’s not present in the screenshot).
And for the boot-logo, well, as you can see, except for the two lines of text that ‘systemd’ outputs (in layman’s terms, ‘systemd’ is the utility that is in charge of the boot-up process) one denoting the version of ‘systemd’ that’s included & the other denoting the status of the file system (whether it was properly mounted/umounted etc), the rest of the screen is pretty much an ugly blank one.
One thing that I’ve always praised when reviewing Manjaro releases (despite what the flavor was) is how good they all look, and the 17.0.2 release is no exception. I don’t know about you, but this one look gorgeous to me! And also, with each new release, KDE’s UI is getting more and more intuitive and simple looking, yet without sacrificing access to useful options & features, unlike GNOME.
So all in all, when coupled with how well Manjaro developers have ‘personified’ the theme & the ever growing intuitive nature of the KDE desktop, for the 2 weeks I’ve been using KDE plasma 5.10.2 (I’m still writing this review on it), I’m starting to develop a sense of love towards KDE. As they say, ‘it grows on you!’ 🙂 .
By default ‘folder view’ is enabled on the desktop, so you can create folders and links right away, without having to change anything as well.
This is certainly not ‘hot news’, but just to point it out (with the hope of criticizing GNOME 😉 ), KDE lets you ‘tweak’ the desktop further more through options such as changing the icon size, adding tool-tips, how they’ll be arranged etc, and yet it gives you access to these features without damaging the simplistic look of the desktop in general.
Fonts too in general look quite sharp and contrast level is also very easy on the eyes. I’m very satisfied with KDE by this time around. Who knows, maybe I’ll completely ditch GNOME or GNOME based distributions in general in the future!
However, one annoying ‘feature’ of KDE that I never liked is that it opens up folders & files with a single mouse click. I find this very problematic while doing things like re-naming files & folders. Sure a user can change this behavior in a matter of seconds, but still, personally I think having ‘double-click to open files & folders’ enabled by default is a much better approach.
I also think that there is still room for ‘cleaning’ the desktop context-menu and the context-menu of the file manager (Dolphin) as well. For instance, I don’t see the point of having ‘Run command’, ‘Add panel’, ‘Open with Dolphin’, ‘Leave’ (isn’t it better to add a shortcut for this on the panel?) on the right-click context menu on the desktop. I think they unnecessarily clutters the menu and make it look bigger which isn’t always better.
I mean sure the marvelously kind KDE (KDE takes a very feminine approach while creating the user-experience) does let you change these in a matter of seconds, and I don’t know about you, but after getting rid of some of these options, have a look at how simple looking the new desktop context-menu is. I could be wrong here because I’m just getting my head around KDE to be honest, but for my needs as everyday desktop user who spends most of his time in front of the web browser, this set-up is the most ideal. Although not exactly identical, the file manager’s context-menu is also affected by the same mind-set as well (eg: what is the point of having ‘compress’ sub-menu when right clicking on an empty space? And while having those ‘root actions’ is quite useful, it can be quite dangerous as well).
What do you hardcore KDE users think?
So anyhow, that’s as far as I’m willing to talk about KDE’s features. As far as the default set of applications included, Firefox 54.0 is the default web browser, VLC 3.0, Cantata 2.0.1 & Kdenlive 17.04.2 are included for video and audio management. Yes VLC and Cantata are capable of playing proprietary multimedia codecs (I don’t have the faintest idea of Kdenlive, other than to say that it’s a video editor) by default. LibreOffice Suite 184.108.40.206 is also included and Octopi 0.8.8.5 is the default software manager (it’s lightweight and very easy to use).
qBittorrent 3.3.13 is the default Bittorrent handler. Krita 3.1.4, Gwenview 17.04.2, Inkscape 0.92 & Skanlite 2.0.1 are included under the ‘Graphics’ section in the main-menu. There are many other applications included that come with KDE application set.
Adobe Flash Playback & Skype…
Google Chrome isn’t included by default and it doesn’t show up in the Octopi software manager either. Luckily however, there is a command-line tool called ‘YAOURT’ and you can use it to install many applications that aren’t shown in Octopi. I used ‘YAOURT’ for installing both Google Chrome (
yaourt -i google-chrome) and Skype (
yaourt -i skypeforlinux). Be aware however that some applications that ‘YAOURT’ installs need to be compiled from scratch, so they’ll take some time to install, depending on the capability of your hardware.
So far Adobe Flash performance has been flawless. I haven’t had a single issue while trying to play videos through it in any of the websites I attempted (not just on YouTube).
Skype installation (I used the latest beta) and the after installation set-up was also flawless. In Ubuntu 17.04 for instance, I had to manually adjust a few settings in order to run Skype properly (for instance, in Ubuntu, Skype’s notification icon didn’t appear in the system-tray area). But since Skype’s UI is written in Qt, I guess that’s why it automatically integrated with the desktop without any issues.
I also made an audio call (for about 15-16 minutes) and everything worked perfectly. I also made a short video call, and in that instance too, from both ends everything worked.
This whole idea of trying to get a sense of how good the operating systems perform is one that I take very seriously. Therefore, in order to make sure the measurements are of high accuracy I always do the following. Other than enabling the auto-user-login & adding the System Monitor application to the task-bar (for measuring the memory usage), I try not to make any other tweaks to the system, although if possible I disable system-update applications from running in the background because when checking for updates etc, they tend to fluctuate the memory usage reading a bit. But on some ‘Linux’ operating systems (Manjaro KDE in general) this is not always possible.
Also before starting to take any measurements, I boot into the newly installed operating system at least 5-6 times for letting things to settle down (let first time ‘wizards’ to be done with their configurations etc etc). And when I’m satisfied that the operating system is now at an ideal state, only then I begin taking the measurements. So let’s dive into the details now shall we.
I use the stop watch app on my Android phone for measuring the boot-up times. I start measuring from the moment I hit Enter at the GRUB screen and stops when the desktop is usable. In other words, I do not necessarily wait till the desktop is fully loaded (in the sense till all the startup applications are loaded). I do this because even if an app or two are still loading, on most occasions, the user can start using the computer.
Anyhow, in my experience, KDE has never been that fast to boot, it surely takes its time. Sure some KDE distributions do it better than other KDE distributions, but in general they aren’t that fast compared to distributions that ship with, say GNOME based desktops in general.
Here too this is the case. As you can see, Manjaro is the slowest to boot where compared to Kubuntu 16.10, it roughly takes 28.7% more time. Compared to Ubuntu 17.04 its even worse where Manjaro takes 81.1% more time, that’s a huge gap! Is this a huge concern? Well, if you don’t normally shutdown your computer whenever you’re going away for a while and only puts it to say into the ‘Sleep mode’, then of course not. However, if you can’t put your computer to sleep, which is unfortunately the case with my laptop under Manjaro (more on that later), then you’re either left with shutting it down or leave it turned ON.
If you however, turn it OFF, then that 50+ seconds of boot-up times is going to be quite annoying, unless of course it’s installed into an SSD (this 52.9 seconds of boot-up delay from a rotational disk).
When measuring the memory usage I make sure the desktop is fully loaded. Even when I know that it’s fully loaded, I still wait for another 15-20 seconds before measuring the memory usage.
As you can see, KDE distributions actually are quite lightweight compared to GNOME these days. Ubuntu 17.04 is the most memory hungry operating system from the three where it consumed 121% more memory compared to Kubuntu 16.10! And 110% more compared to Manjaro KDE 17.0.2!
Between the two KDE operating systems however, there is only a slight difference where compared to Kubuntu 16.10, Manjaro consumed about 5% more memory.
CPU Usage at Idle…
When the operating system is at idle, two processes (‘plasma shell’ & ‘kiwin_x11’) quite often jump in and each consumed 1% of the CPU. But other processes did not interrupt the CPU. All in all, I’m quite satisfied. Why is this important? This is important because if at idle (meaning that user is not interacting with the computer, and has not opened any user applications) the CPU is interrupted unnecessarily by processes (this can happen due to viruses, bugs in programs, poorly optimized applications etc) then it increases the power usage. And also, those wasted CPU time can’t be allocated to other applications that actually trying to do some useful work either.
So basically this observation gives one a reasonably good understanding of how well applications are optimized (in a certain sense). A couple of years ago, ‘Linux’ didn’t do so well in this regard. But these days it simply excels!
Power Usage at Idle…
Before measuring power I set the screen brightness to its maximum (100%) and disable features that could negatively affect it (such as screen dimming, automated screen turning off mechanisms, screensavers, screen lockers etc). As always, Bluetooth is turned OFF but Wi-Fi is turned ON & connected to my wireless router, and an USB mouse is also connected. Then I run ‘powerstat‘ (it’s an excellent application that’s specifically designed to measure the power usage, and its quite accurate).
As you can see, Manjaro 17.0.2 KDE is the clear winner here & both Kubuntu 16.10 and Ubuntu 17.04 scored identical values. Manjaro was roughly 33.8% power efficient compared to the other two, very impressive. Since I have a dual GPU setup (Nvidia & Intel), this could be (just a guess) because Manjaro 17.0.2 KDE includes ‘Bumblebee‘ by default.
‘Bumblebee’ is a utility that has the ability to turn ON or OFF GPUs depending on the workload thus improving performance and saving power as well (it doesn’t work on AMD/ATI GPUs). But Kubuntu nor Ubuntu includes it by default. That said, after installing a power usage optimizer on Ubuntu 17.04 I was able to bring down the power usage to 6 Watts nonetheless.
To put this all into a more simpler context, with the battery fully charged & the screen brightness set to somewhere around 17-18% (I use the laptop only indoors, and the brightness is rarely increased over 40%), mostly using the web browser (including watching videos online and browsing Facebook) & then playing a portion of a movie on VLC, I was able to use it for about 3 hours and 13 minutes until the battery hit 18% charge level.
I have had some issues with my touchpad (Focaltech) under most of the ‘Linux’ operating systems that I tested on this Asus laptop, except under Ubuntu 17.04. The issue is, after putting the laptop to sleep-mode the next time I wake it up, the touchpad stops working. This is a known bug of this Focaltech touchpad and all that it requires is a simple tweak to be enabled while compiling the driver. Since this has been largely reported to Ubuntu, with the latest release (17.04) it has been finally fixed. But sadly, Manjaro 17.0.2 KDE couldn’t make it work after putting the laptop to sleep.
I also couldn’t connect by Samsung Android Phone to KDE through the USB cable. The file manager failed to mount the storage, and I didn’t receive the message on my phone that comes up whenever I plug it in using the USB cable, asking me to allow certain permissions because only after I allow those permissions the file system become readable.
And thus unsurprisingly, KDE file manager displayed an I/O error.
However, KDE includes a GUI called ‘KDE Connect’ (I wrote about this in one of my older Manjaro KDE reviews) which is designed to sync data between the phone and your computer through Wi-Fi. So out of my frustration I fired it up and then following its advice I installed an android app called ‘KDE Connect’ on my phone as well. After that everything worked as expected. Not just sending & receiving files, I could also use the phone’s touchscreen as a mouse to control the KDE desktop which was pretty cool. Truth be told, I couldn’t have cared less even if I couldn’t connect my phone to KDE. But I guess for most people this is an important feature to have, so that’s why I actually decided to bother with ‘KDE Connect’ in the first place.
Other than that, rest of the hardware was correctly recognized by the operating system. It was able to correctly restore the previously set brightness & the status (ON|OFF) of the Bluetooth adapter each time it loaded the desktop as well.
Manjaro also includes an excellent hardware driver installer GUI that basically lets you install both open-source and proprietary drivers for the GPUs & the Network card. I have not installed the proprietary Nvidia GPU driver and have been using the open-source one that Manjaro uses by default. So far everything has worked extremely well.
As mentioned in the beginning, I have been using Manjaro 17.0.2 KDE for more than 2 weeks now and in all this time I have not seen a single application crash! KDE didn’t use to be like that not that long ago in my opinion. But things are looking better and better by each release it seems. Well done!
This is one of the most important tests that I run for getting a sense of how responsive the operating system is, and it’s thoroughly related to the performance of the main storage unit. This is because in many ways, the real strength of an operating system can be measured by observing how it manages its most weakest points.
And when it comes to responsiveness, the main storage device can be considered as the weakest point, because compared to other devices involved, it is one of the slowest components. Therefore, when that device is stressed out heavily, the way in which the operating system responds in order to maintain the responsiveness, is usually an excellent indicator of how well it is optimized to deliver the best possible responsiveness. And the ‘test partition’ into which I install these operating systems is located on a rotational disk. Therefore this is even more important because slower the storage unit gets the more challenging it is going to be for the operating system to be responsive under stress.
So with that in mind what I do is actually very simple. First I try to copy a reasonably large file (1.5 GB -- 2 GB) within two locations in the currently logged in user’s ‘Home’ folder. As soon as it starts to copy then I try to play a multimedia file (usually 720p) through VLC. And then I also try to use the start-menu (if such functionality is available) and search (to put more pressure on the disk) and then try to open couple of programs through it. Sometimes while all this is happening, I try to open a location that includes a reasonably large amount of folders & files using the file manager, but this time I didn’t try that because in the past, except on two or three occasions, KDE had a tendency to be quite sluggish during this test.
The idea is that, if the operating system was able to open most of the programs & played the multimedia file without major interruptions before it finished copying the file, then I consider it to be a responsive OS. I also observe the sensitivity of the cursor (to see whether it loses its sensitivity while using) as it is also an indication of the responsiveness.
So that’s what I did in Manjaro 17.0.2 KDE. What was the outcome then?
(Note: This image is only an illustration)
It was quite good. Manjaro was able to open most of the programs that I tried to open, and VLC played the multimedia file without major issues. However, while I was searching for applications using the start-menu, I noticed that the system just kind of got stuck two or three times (each instance lasted about 1-1.5 seconds). When this happened cursor stopped moving as well. However, even this happened VLC was not that much affected by it. Is that a big issue? Absolutely not.
However, I have observed very similar behavior in some of the other KDE distributions as well. I’m not sure what the real issue here is, though. Maybe whatever is behind the search functionality of the start-menu is set-up in a way so that it gets a high priority (just a wild guess). Since we’re talking about the start-menu let me also say that I’m not that impressed with its searching ability either. I mean (this is just one example) have a look at the below screenshot. Here I’ve searched for ‘libre writer’ and the start-menu displays zero results. If I type just ‘writer’ then it pops up the LibreOffice Writer app.
So anyway, if I had to give the responsiveness a score, I’d give Manjaro 17.0.2 KDE 3.4 stars out of 5 for responsiveness. All in all, I am happy.
As you can see, while Ubuntu 17.04 was the winner. Manjaro 17.0.2 KDE was 118% faster while shutting down compared to Kubuntu 16.10! Manjaro was about 37% slower to shutdown compared to Ubuntu 17.04. But any operating system that shuts itself down with 6 seconds is a fast one, I would say.
My biggest concern over Manjaro 17.0.2 KDE is the 50+ seconds of boot-up times. It’s a concern to me only because I can’t put my laptop computer to sleep due to that Focaltech touchpad issue. So when I take a break (say for about 20-30 minutes), I’m forced to keep the laptop turned ON. If its low on battery, then I’m forced to plug in the charger (yes I’m talking from the two weeks+ experience I’ve had so far with Manjaro). That said, even if you have a troublesome Focaltech touchpad, if you’re going to install it into an SSD then I guess it’s not much of a concern at ll anyway. Other than that, this is a gorgeous looking, responsive, power efficient and one of the stablest KDE distributions I’ve had used so far!
If you’re interested in taking it for a spin, then download it from here. If you want to dive into the community maintained, unofficial flavors, use this link. Oh! and one more thing. If there is no other way around it, you actually can install Manjaro into an LV using a community developed command-line based installer called ‘Manjaro Architect’ (you’ll need a working internet connection). Good luck guys & thank you for reading!