Sorry Arch (EndeavourOS), it's not working out any more and hello Fedora

With all due respect, Jaromir, maybe you think you're doing some good with your BSD crusade here, but you're really not achieving anything of value. You could try to show a user of some non-free operating system that it would be for their own good to try a free one instead, but no, you're here on a site called Gaming on Linux, trying to convince people that they'd be better off without Linux.

Not that you'd be able to convince a man dying of thirst to take a sip of water with your needlessly confrontational style of "evangelism". It should be obvious by now that you're pushing people away instead of drawing them in.

The article on cstan.io was written on December 1, 2021. Are you going to claim that these kinds of problems are all magically solved in a few months? According to his benchmark, the snap performed 6,175 times slower in MotionMark.

As an example application, Firefox 94.x was started in an Ubuntu 21.10 VM with 2 CPUs and 2 GB RAM and tested via BrowserBench.org. VirtualBox 6.1 and an Intel i7-8850H CPU were used on the host side.

Translation by DeepL, emphasis by me: Doesn't VirtualBox require a specific driver to allow some kind of hardware acceleration?

I played Dota 2 for many years on both windows and Linux, I finally had more than 2000 hours of play after 5 years. When I ran Dota 2 on FreeBSD, I found that on average I had better latency than on the other two platforms. I also noticed that my FPS were +- 10% higher, and I suddenly had no more hiccups when browsing through the menus. I think there may be someone who reminds the Linux community of these observations.

But most important of all, Dota 2 was finally stable. With Linux and windows I often had the problem that the connection to the server was lost, and that I then had to continue playing in a disadvantaged position. In FreeBSD, this has not happened once.

I think it's anything but meaningless info for gaming.

So the thing is we are already facing some compatibility issues by using Linux, and of course striving to improve the situation. Switching to FreeBSD (I only presume) means a whole lot more incompatibility, for not much gain.

Maybe when Linux has 50% market share, FreeBSD could become the contending niche for a change.

That said, it's definitely important to point out where Linux is not as performant as possible. However, comments in GoL are not the right place for them. If you do care, maybe prepare a decent report with logs, put up a blog post or document, then point the right people (like the Linux kernel mailing list) to that.

Otherwise, "BSD worked better for me" is not a convincing sentence by itself.

And yes, I do see the irony.

Pretty much the exact same situation, isn't it?

What? No, it's obviously not the same situation at all. But I do admire the sheer size of your ego, thinking you've earned the right to compare yourself to Socrates for simply choosing to use different software than the rest of us.

Also, you ignored my point. You're doing more harm than good to your cause here, and making it even harder to take BSD (and its advocates) seriously.

Oh and if you're just trolling, have fun I guess. We all have our kinks and I'm not one to judge.

Since all the tests have been done in this VM, it seems strange to me that this would be the explanation. Flatpak didn't have the same impact so it was probably just Snap being slow.

I leave it up to you to philosophize about what could go wrong if the driver is missing & why the driver could be missing in one case.

To be clear: I don't give a rats ass about snap and I believe there are valid points (closed source etc.) against it (but I'm just not using Ubuntu or Ubuntu based distributions). But the benchmark looks simply wrong. Even if there is something which degrades the graphics performance in the snap (a bug, a wrong configuration), any unbiased person would question these results or at least mention it.

With that being said: I'm out, have fun with your operating system but leave out the FUD from your arguments.

Think, for example, of Bill Gates, or the high number of Microsoft executives who were involved in sex traffic.

Wow. That escalated nicely. And all because of the imperfections of snaps... Let's see when Godwin's law can be applied.

The comparison was never about myself and Socrates. But it was about the similarity between you and Socrates' murderers (or simply the average person).

The significant difference between me--let's say as the average person (of whom I might not be very representative)--and a murderer is the bit where, at least according to accounts of arguable accuracy, they actually killed the guy because they didn't understand his wisdom and/or disagreed with his politics.

Get over your ego please and realise that the personage of Socrates and the actions of his murderers bear no similarity to the situation here. I have no wish to murder you or anyone else, nor do I object to the way you behave because I lack the intelligence to appreciate the brilliance of your message. You're not getting the response you want because you're acting like a nuisance, not because you're too smart for the rest of us.

Maybe pause for a minute and reflect on your contribution to this thread. Unless you lack all self-awareness, you should be able to see that what you're doing is reminiscent of the stereotypical Windows fanboy jumping into a conversation about a Linux port on a Steam discussion board. Please, either find it in yourself to be constructive or find a better venue for your decidedly unproductive antics.

Read that last sentence, again, and extrapolate.

Sure, I'll start extrapolating as soon as you stop throwing up bad analogies and completely missing the point.

This feels futile but I'll try one last TL;DR: You will not achieve anything useful by repeatedly hitting us over the head with your preferred non-Linux OS on a site called Gaming on Linux.

It is telling that Ubuntu users do so few benchmarks, that they do not yet have the slightest knowledge whether it is 15%-10%-5% slower or faster than deb packages in the current implementation.

I did a benchmark. But you claimed it to "be fake", because it didn't meet your... well, expectations. But indeed I do few benchmarks, I prefer to work (or game) on my machine. And if "absolute browser speed" would be my thing, I'd probably use Chromium. Even the snap version is faster by quite a margin in these benchmarks than any version of Firefox.


Last edited by Tuxee on 10 April 2022 at 11:21 pm UTC

So, I've said this before - it infuriates me that such a tiny niche community as Linux will find ways to shit all over other portions of the same niche community.

As a general reminder, distro-wars are not tolerated on GOL. Read the rules, which are linked just slightly above every comment box you type into.

Please stop with the thinly veiled contempt for other people's choices and experiences - define your distro by the things you love about it, and sing about those things to the world. Stop, stop, stop tearing down others.

As someone very new to all things Linux, I ended up landing on the recently released Nobara Project by Glorious Eggroll. Sounds like it simplifies some of the on-boarding versus traditional Fedora spins. https://nobaraproject.org/

Small edit: That being said, it's not my daily driver. I've got a drive reserved for it to run games that have better performance on Linux, particularly but not limited to Elden Ring currently.

Well, when you are new to Linux I would definitely stick to one of the mainstream distros like Fedora, Mint, Manjaro or Ubuntu. You will inevitably run into "problems" (probably not problems per se, but things that are just... different) and having forums and documentation for your distribution (and not just a "quite similar one") helps tremendously.

jm2c

I actually initially started with Manjaro KDE, and while I liked the environment well enough, the performance wasn't what I had been hoping for, and worse, presumably due to some mistake in how I had done things in setting the installer up to split my drive to dual boot Windows and Linux, I ended up having to wipe the entire drive. A learning experience, at least, and after correcting a mistake with how I initially installed Nobara, I've had no issues with it that I didn't have with Manjaro, all of which I was able to work through with the assistance of a more experienced friend, and I find the performance much better, with and without launch arguments for individual games. I do appreciate the input though!

I really like Debian, OpenSuse TW, and recently, Manjaro.

All three really showcase some of the best things I like about Linux and have a few cool things to add as well.

Not that there aren't others I've tried, or other worthy distros to use, those just happen to be the ones I'm most fond of.

I'm using Debian for my home server, same install for almost ten years ( I'm not really sure how to tell since everything in the server has been replaced a few times). I love it because it's stupid simple.

I recently switched to Manjaro for my PC and laptop after several years of OpenSuse. No major problems with OpenSuse, but I was surprised at how fast and snappy Manjaro ran on my old laptop. XFCE is my preferred DE, so I like that it's the default on Manjaro.

I did a benchmark. But you claimed it to "be fake", because it didn't meet your... well, expectations. But indeed I do few benchmarks, I prefer to work (or game) on my machine. And if "absolute browser speed" would be my thing, I'd probably use Chromium. Even the snap version is faster by quite a margin in these benchmarks than any version of Firefox.

I just have a hard time believing your result, because it doesn't match the motionmark result of the two articles, where motionmark is getting significantly slower on the Snap packages.

Sigh. Whatever...

I have already given three links where people talk about incredibly bad results of the Snap packages. Here are two other
If I have time, I'm going to do some extensive benchmarks myself and compare these Snap apps on Ubuntu with the performance of the exact same apps in Debian and Arch Linux. I'm probably going to make some unprecedented discoveries.

Don Quixote rides again...
Just a heads up: Comparing the same browser on different distributions will be completely pointless.

https://www.phoronix.com/scan.php?page=article&item=linux-distros-50cpu&num=2

Regarding Firefox, this is my result in Speedometer 2.0 with a very old i3-3240 and without a dedicated GPU on FreeBSD: https://i.ibb.co/tJTnbZq/Screenshot-2022-03-25-14-57-10.png
And I get this result in Chromium on the same system: https://i.ibb.co/9vkMVg6/Screenshot-2022-03-25-10-38-48.png

So the difference is 4.94% in FreeBSD. Speedometer is seen by both the Firefox team and the Chromium development team as the benchmark that has the most relevance in reality.

I can probably make Firefox go as fast as Chromium on FreeBSD: https://firefox-source-docs.mozilla.org/build/buildsystem/pgo.html

You can ask yourself, why is Firefox on my Linux system so much slower than Chromium, when Firefox is almost as fast as Chromium for people who use FreeBSD, and why is Chromium also faster on FreeBSD than on my Linux system?

No. I won't. I am definitely not inclined to choose FreeBSD as my OS of choice because some random dude on a mission suggested that Firefox might then run just as fast as Chromium. It has been said before: This site is about gaming on Linux.
Why not head over to Phoronix, where benchmarking is king and FreeBSD gets its fair share of attention. However, the results there are still bleak.

stuff

If you're capable of an objective look at *BSD and comparing to some aspect of Linux, perhaps write an article and submit to GOL instead of ranting in the comments here. Clearly you are capable of writing well formed sentences and can provide sources. Hopefully you'll take this opportunity to enlighten our community.

The dilemma flatpak or snaps is illusional and misleading. Do you prefer playing games installed in your PC or streaming them from servers like Google Stadia? This is the question you have to ask yourself before it is too late.

Things are very simple. If you use flatpaks and snaps you are killing the Desktop in the long run, which means you are killing the Personal Computer as we knew it for 40 years, and the dominant corporate Linux companies (Red Hat, Ubuntu) who only care for IoT, cloud computing, their $$$, have brainwashed you perfectly.

A perfect example for the latter is reading OP going to a distro that updates itself biannualy! in order to... use flatpaks! there (snaps the other corporate equivalent)

Package managers following UNIX/GNU/Linux tradition were written with and respected the KISS principle, containers are adding unnecessary complexity with tons of issues and all this is being done by corporate companies, who are using fanbois of trademarks as a battering ram.

If you keep being lazy using the Windows/Apple paradigms, if your argument is "that's how Google with Android does it", if you keep playing their game with containers which were initially intended for servers (& immutable devices) but they found ways into "convincing" you -by youtube influencers and general ignorance, especially using Linux newcomers who are presented with Snap versions of Firefox and Chromium by default-, soon your PC will be just another console, I can see already those youtube influencers and probably OP in a year talking about how great Fedora Silverblue and the equivalents are, Shells (your Personal Computer in the Cloud) is already here.


Last edited by sudoer on 11 April 2022 at 12:53 pm UTC

The dilemma flatpak or snaps is illusional and misleading. Do you prefer playing games at your PC or streaming games from servers like Google Stadia? This is the question you have to ask yourself before it is too late.

Things are very simple. If you use flatpaks and snaps you are killing the Desktop in the long run, which means you are killing the Personal Computer as we knew it for 40 years, and the dominant corporate Linux companies (Red Hat, Ubuntu) who only care for IoT, cloud computing, their $$$, have brainwashed you perfectly.

A perfect example for the latter is reading OP going to a distro that updates itself biannualy! in order to... use flatpaks! there (snaps the other corporate equivalent)

Package managers following UNIX/GNU/Linux tradition were written with and respected the KISS principle, containers are adding unnecessary complexity with tons of issues and all this is being done by corporate companies, who are using fanbois of trademarks as a battering ram.

If you keep being lazy using the Windows/Apple paradigms, if your argument is "that's how Google with Android does it", if you keep playing their game with containers which were initially intended for servers (& immutable devices) but they found ways into "convincing" you -by youtube influencers and general ignorance, especially using Linux newcomers who are presented with Snap versions of Firefox and Chromium by default-, soon your PC will be just another console, I can see already those youtube influencers and probably OP in a year talking about how great Fedora Silverblue and the equivalents are, Shells (your Personal Computer in the Cloud) is already here.

You've got a pretty nice slippery slope there. The only thing your assertions are missing is causal linkage from one to the next, but I'm sure that it's not necessary when you can substitute it with hyperbole. :P

Things are very simple. If you use flatpaks and snaps you are killing the Desktop in the long run, which means you are killing the Personal Computer as we knew it for 40 years, and the dominant corporate Linux companies (Red Hat, Ubuntu) who only care for IoT, cloud computing, their $$$, have brainwashed you perfectly.

This is probably one of the most ridiculous things I've read. Stop it.

A perfect example for the latter is reading OP going to a distro that updates itself biannualy! in order to... use flatpaks! there (snaps the other corporate equivalent)

Uh, what? You're now absolutely just making things up. I didn't move to use Flatpaks, I noted them as a nice to have. If you keep doing these kinds of posts, I will put you into the moderation queue. Stop it.

Things are very simple. If you use flatpaks and snaps you are killing the Desktop in the long run, which means you are killing the Personal Computer as we knew it for 40 years, and the dominant corporate Linux companies (Red Hat, Ubuntu) who only care for IoT, cloud computing, their $$$, have brainwashed you perfectly.

This is probably one of the most ridiculous things I've read. Stop it.

Nothing ridiculous there and the rest of the post that explains it epigrammatically. Have you seen Canonical promoting the desktop (= the PC) like they did in 2004? Have you read an Ubuntu blog since 2017, when Shuttleworth announced Growing Ubuntu for cloud and IoT, rather than phone and convergence? Everything is about IoT and Edge Computing and how Snaps are perfect for both.

2017:

The choice, ultimately, is to invest in the areas which are contributing to the growth of the company. Those are Ubuntu itself, for desktops, servers and VMs, our cloud infrastructure products (OpenStack and Kubernetes) our cloud operations capabilities (MAAS, LXD, Juju, BootStack), and our IoT story in snaps and Ubuntu Core.

2019:

18th April, 2019: Canonical today announced the release of Ubuntu 19.04, focused on open infrastructure deployments, the developer desktop, IoT, and cloud to edge software distribution.

Control decisions move to the edge with smart appliances based on Ubuntu, enabling edge-centric business models. Amazon published Greengrass for IoT on Ubuntu, as well as launching the AWS DeepRacer developer-centric model for autonomous ground vehicle community development, also running Ubuntu. The Edge X stack and a wide range of industrial control capabilities are now available for integration on Ubuntu based devices, with long term security updates. Multiple smart display solutions are also available as off-the-shelf components in the snap store.

Where does all the money flow from? Not from the desktop (PC), the desktop (PC) is getting irrelevant day by day, to be more precise, it's getting transformed by corporate Linux companies from the powerhouse running "the real thing" locally, and of which you had full control with GNU/Linux into a more and more closed, dumb terminal, which will communicate with the servers and the mainframe via a browser -it's happening already with Software as a Service and Platform as a Service- and if you endorse snap (and competitor's version flatpak) "that was originally released for cloud applications[2] but was later ported to work for Internet of Things devices[3][4] and desktop[5][6] applications"https://en.wikipedia.org/wiki/Snap_(software) you are just accelerating the transition without knowing it, that's the whole point. And what is so bad with the transition? That soon enough, in 10-20 years when your Internet would be down, you would have absolutely no power whatsoever with your black box connected to the powerful mainframe/server.

And what else is bad? From the IoT link:

There are a number of concerns about the risks in the growth of IoT technologies and products, especially in the areas of privacy and security, and consequently, industry and governmental moves to address these concerns have begun, including the development of international and local standards, guidelines, and regulatory frameworks.

We all now how good regulations vs. Big Companies work right? Sorry, too late for tears after that.

You've got a pretty nice slippery slope there. The only thing your assertions are missing is causal linkage from one to the next, but I'm sure that it's not necessary when you can substitute it with hyperbole. :P

You are already on the bandwagon as I see by your own reasons, it's OK.


Last edited by sudoer on 11 April 2022 at 5:31 pm UTC

stuff

I'm not sure what the hell you're saying here about Flatpaks and Snaps being bad, but I get your point about cloud gaming, I suppose.

But from the enterprise perspective (I've been in tech for nearly 30 years and a geek for coming up five decades) that stuff is so cyclic. Back in the 70's and 80's everything was a terminal for big server clusters. Then we all went big desktops in the 90's. Then in the early 2000's, it was all Citrix and VMware Horizon on big clusters again. For the past ten to fifteen years, the focus has been PC's again, and now recently all this cloud gaming stuff has come along. Back and forth. Over and over.

I suppose it's a bit worse generally now in terms of "owning" stuff - music is on Spotify, films are on Netflix, and now we have games on Stadia or GFN. There's subscriptions for EA, or Xbox. Hell, you can even get a Kindle sub for "free" books these days.

But none of that stops you owning things if you want to. I have friends who still buy CDs to burn to FLAC and buy Blurays and DVDs to watch. People still buy books - actual books.

It's a stretch (and super insulting) to be calling people brainwashed or lazy just because, unlike you it seems, they actually enjoy using their PCs however they want.

the desktop (PC) is getting irrelevant day by day, to be more precise, it's getting transformed from the powerhouse running "the real thing" locally, and of which you had full control with GNU/Linux into a more and more closed, dumb terminal, which will communicate with the servers and the mainframe via a browser

This was happening before Snaps and their usage or lack thereof does not affect cloudification one bit. Most of mainstream cloud activity is happening around OCI and Kubernetes, someone installing a Flatpak on their system has absolutely zero bearing on that.

it's happening already with Software as a Service and Platform as a Service

It's good to see you found Wikipedia. SaaS and PaaS are indeed a thing. We've also got IaaS and FaaS, just so that we can fill out the vocabulary.

if you endorse snap (and competitor's version flatpak) that was originally released for cloud applications[2] but was later ported to work for Internet of Things devices and desktop applications you are just accelerating the transition without knowing it, that's the whole point.

Not even a hypothetical causal link is established. Usage of Snaps is fairly marginal in cloud and usage of Flatpaks in there is essentially non-existent. Not to mention using desktop applications from Snap or Flatpak does basically nothing to contribute to these techs being used on the cloud, because you don't run desktop applications on the cloud. You run services with web frontends. And you can do that probably easier with OCI images and Kubernetes deployments than by installing a bunch of Snaps on some server boxes.

If you are worried about the cloudification of software, you are actively doing a disservice to your cause by advocating for actions that will have zero effect on cloudification. If you want to stop cloudification, you need to either make software resistant to cloudification (heavy use of AGPL or non-OSI stuff like SSPL) or you need to provide an incentive/disincentive structure that makes creation of desktop software more appealing than cloud-oriented software. On the disincentive side you can create legal constructs like stronger data protection legislation to attack the profit margins of cloud companies. Or on the incentive side you can make creation and distribution of desktop software so easy, simple and profitable that it becomes the de facto mean to ship stuff. That makes things like Flatpak necessary, by the way, because making your software available across the various distributions is a mess otherwise. And if you want to popularize the third party desktop application as the standard, then having those sandboxing features also becomes quite important, especially when it comes to proprietary software. Unless you want to make it so that all software will be FOSS, in which case I applaud your cause and wish you well on your legislative efforts.

You are already on the bandwagon as I see by your own reasons, it's OK.

Fun memes. Except I am probably the most outspoken member of the GOL editorial team against cloud gaming and for the creation of a thriving, native Linux software ecosystem. But going by your raving and ranting, you probably consider anyone who doesn't agree with you "brainwashed", so really I arguing my points for the audience, rather than you.


Last edited by Samsai on 11 April 2022 at 7:29 pm UTC

I made a similar transition from Gentoo; a hard drive died and I decided rather than the laborious install I would install Ubuntu and call it a day.

Curiously, I started using Kubuntu for a similar reason. I was actually in the middle of a college project when a Gentoo update decided to render my computer unbootable (yes, I know, I shouldn't have updated a mission critical computer that was working just fine), and since I didn't have a week to spend recompiling the whole thing from the ground up (this was about 15 years ago), I installed Kubuntu and was back in business in about 20 minutes.

I was very happy with Kubuntu for a very long time but recently switched to Manjora for reasons I don't entirely remember. I think I just wanted to get away from the lengthy upgrade process every six months, and I liked the idea of always having the most up to software available. Manjora has been rock solid, too (not that Kubuntu was ever unstable).


Last edited by Mountain Man on 12 April 2022 at 2:14 am UTC