If your drive is the bottleneck, this will make things worse. If you want to proceed:

You’re already using ffmpeg to get the sequence of frames, correct? You can add the -ss and -t flags to give a start time and a duration. Generate a list of offsets by dividing the length of video by the number of processes you want, and feed them through gnu parallel to your ffmpeg command.

My first thought was similar - there might be some hardware acceleration happening for the jpgs that isn’t for the other formats, resulting in a CPU bottleneck. A modern harddrive over USB3.0 should be capable of hundreds of megabits to several gigabits per second. It seems unlikely that’s your bottleneck (though you can feel free to share stats and correct the assumption if this is incorrect - if your pngs are in the 40 megabyte range, your 3.5 per second would be pretty taxing).

If you are seeing only 1 CPU core at 100%, perhaps you could split the video clip, and process multiple clips in parallel?

Considering that you are not using their software, was the laptop worth the premium you paid for it, vs buying from Clevo directly?

I figured the hardware and software coming from the same vendor would yield the best results, and wanted to support a company that supports right-to-repair, and Linux in general. But ultimately I found Pop!_OS buggy and had performance issues, so I’m not using their OS, and their firmware is causing issues with my SSD, so I’d like to be off of it as well (but was told "there’s no process for reverting to the proprietary firmware“ for the specific model I have). I could have bought a Clevo directly, saving hundreds of dollars, and probably had a better working machine.

Would not recommend System76. I’ve had many issues with my machine (primarily software, related to their buggy custom firmware, and Pop!_OS, until I ditched that for stock Ubuntu). Their support has been terrible - rather similar to OP’s, actually. I’ve had the laptop for about 2.5 years, and I’m checking practically daily for something to replace it.

You might get lucky. Based on https://support.system76.com/articles/system-firmware/, it doesn’t seem to be all models. Note however that the list is out-of-date; my galp5 is not listed, but does not work. Fortunately, I found this out doing a RAM check, and not a firmware upgrade.

My usb-c ports can be a little touchy, too. The SD card slot is also really bad - the card has to be positioned perfectly to slide in, or it jams. I’m also upset that the usb-c port can only be used for charging after a full boot. It cannot be used to perform firmware updates, or even to do a ram test. This means day-to-day, usb-c can be used, but I have to keep track of the barrel charger, just in case. This, of course, was not specified on the product details page (nor, I think, that only one of the two usb-c ports could be used for charging - it’s possible I overlooked that, but still frustrating on an expensive laptop that lists usb-c charging as a feature).

I currently have a System76 laptop, and sincerely regret my purchase. When I purchased it, the Framework was not out yet - I wanted to support a company that supports right-to-repair, and figured since they controlled the hardware, firmware, and software (Pop!_OS), it would be a good, stable experience. It has not been, and support has generally been poor. I know other people have had better experiences than I have, but personally, I won’t be buying from them again.

I haven’t personally used Purism, but former co-workers spoke really poorly of them. They were trying to buy a big batch for work, and said the build quality was awful. Additionally: https://youtu.be/wKegmu0V75s

The battery life and speakers will certainly be model dependent. The quality of the machine I received and the lackluster support, given the price I paid, are what I find most frustrating. The computer would be fine for ~$600, but I paid over $1000. I paid a premium expecting System76 to hold themselves to a high standard, and so far, they’ve let me down in multiple ways.

I do recognize with a different model, the experience could be 180°, but if buying from them is a roll-of-the-dice, for me personally, that’s enough to buy from someone else next time.

I would not recommend them. I bought a Galago Pro in 2020, and it’s been a huge disappointment. Pop!_OS was very buggy, and their support was not helpful. I ultimately installed Ubuntu, and it’s now significantly more stable, but I’m left asking the question “why did I pay a premium for a clevo, when I’m not getting anything out of the custom software or support?”

Even with Ubuntu, it’s not a good laptop. The speakers are worse than my phone, a fully charged battery will die completely in less than a day when the laptop is suspended, it runs unbelievably hot. As a developer who depends on this machine for daily work, it’s been intensely frustrating.

Yeah, gadgetbtidge was my first thought as well. I’ve never used it, but in theory it would allow you to control devices without the proprietary app. See the link below for supported devices:

https://gadgetbridge.org/

If you’re ok with Jitsi, and you already use Brave, note that Jitsi is baked in. See https://brave.com/talk/ and check the “Who provides the Brave Talk service?” question:

The Brave Talk service is provided in partnership with 8x8. And the service is built on the open-source Jitsi platform.

(Note that 8x8 owns Jitsi)

See also the “How are my calls with Brave Talk encrypted?” question:

To start, all video and audio data transferred through Brave Talk is encrypted via transport layer encryption. This is similar to how many websites use HTTPS to ensure your traffic can’t be captured on public networks (e.g. coffee shop WiFi).

The video and audio from your call are transmitted to other participants with the help of a Video Bridge server that’s run by Brave’s partner, 8x8. When you enable Video Bridge Encryption in Security Options, your browser exchanges keys with other call participants, and these keys are used to encrypt the video and audio streams. Only people with keys can see your calls. Assuming honest but curious behavior, neither Brave nor its partner, 8x8, have this key by default.

However, there are some important limits to Video Bridge Encryption. If you want to include a phone participant in your call, have more than 20 participants, or want to include users with incompatible browsers (Safari, most iOS browsers, and browsers based on Chromium version 83 or below), this encryption setting will not work. If you record a call, 8x8’s servers will receive a set of keys to decrypt the video/audio stream in order to process and store that recording. Brave will continue to improve Brave Talk’s encryption properties and work to remove some of these limitations.

Read a more detailed description of Jitsi encryption (the open source basis for Brave Talk).

It’s possible to hit issues, especially if different distros are using different major versions of desktop environments or applications, but in practice, I don’t think it’s something that really needs to be worried about.

If you were to upgrade/fresh install, and copy your home folder over, you’d have the same experience - it’s not much safer than sharing the home partition, except that you’re (hopefully) doing that less. You could still easily go from distro A using version 2 of something, to distro B using version 3, and then decide you don’t like it and try to roll back to distro A. If in the process your config was upgraded in place (as opposed to a new, versioned config being made*), you could have problems rolling back.

With configs, you can usually just delete them (or, less destructively, rename them, in case you decide you want them back), and let the application make a new default one for you. With other files (e.g. databases), you might be in more trouble. But a good application will tell you before doing an upgrade like that, and give you a chance to backup the original before upgrading in place. When asked, it’s probably a good idea to take a backup (and not just for this distro hoping case).

*For any developers reading this, this is the correct way to upgrade a config. Don’t be destructive. Don’t upgrade in place. Make a copy, upgrade the copy, and include a version in the file name. You can always tell the user, so they can remove the file if they want, but let them make the choice. If you can’t (e.g. the database scenario, which could be large), tell the user before doing anything, so they can choose whether or not to backup.

You mean with config files stored in your home directory? Or something else?

When you install, whatever you install, partition your drive so that /home is it’s own partition. Then if/when you reinstall, distrohop, whatever, you don’t have to worry about copying over your data. Just use the same /home partition, and format the others. You can actually use this to try multiple distros at the same time - you can install them in different partitions, but have every install use the same /home partition. This is a nice way to test new distros without blowing away your stable install.

Now, for my distro recommendation - Ubuntu gets a lot of hate, but honestly, after 15+ years of Linux, and having tried Mint, Fedora, Arch, Manjaro, and many others, I always end up back on Ubuntu. It’s easy, it’s stable, and it stays out of my way.

The defaults are good, but you can customize as much as you want, and they offer a minimal install (as of 23.10, it is the default) which comes with very few applications, so you can start clean and choose all the applications you want.

Unless you are excited to tinker, I’d really recommend starting simple. Personality, I just want the OS to facilitate my other activities, and I otherwise want to forget about it. Ubuntu is pretty good for that.

I went through this process for the first time recently. I opened the RFP issue, and got a response from Izzy very quickly. They were very helpful and responsive through the whole process. I was nervous it would be a slow tedious process when I started, but it turned out to be pretty quick and easy, largely thanks to Izzy’s help.

Worth noting, they support reproducible builds, which allows developers to sign with their own key:

https://f-droid.org/docs/Reproducible_Builds/

I would definitely recommend going this route if you’re starting with a new app. Having the binary on GitHub (or wherever you’d otherwise publish) match exactly the binary on F-Droid is really good for assuring people nothing in your repo was tampered with during the build process (i.e. that the binary was built from the public code, and nothing else).

It should not take extra work to do this. The project generated by Android Studio should already be reproducible. As long as you don’t change the build setup and break reproducibility yourself, it’ll “just work.” When you submit to F-Droid, just be sure to let them know you want to go the reproducible route (if you make the PR yourself, it’s a flag in the YAML file).

Hmm, strange. The last comment from Dan (second to last comment on the thread) makes it sound like their is another thread, and other users with the same problem:

Whew, I found the other thread and am happy to see that others are encountering this too, and it’s not some super duper weird thing with just my computer…

I tried searching a bit to see if I could find it quickly, but didn’t turn anything up. Maybe if you comment there, though, they could link you to the other thread, and they might have more info.

That sounds like a threading issue. If the app tries to run a task on its main thread, and that task takes a long time (in particular, longer than expected), it could cause the UI to lock up.

Do mouse interactions still work? Does anything on the UI update at all? If not, I’d bet on a task getting stuck on the main thread.

Note that this doesn’t have to be an intense task - you may not see a CPU/network/disk spike. It could be a deadlock scenario, where multiple threads are waiting for the same resources, and each locks some, but not all of the resources. None can move forward, no work is done, everything just hangs waiting for resources locked by other threads.

I’m on a laptop with hybrid Nvidia/Intel graphics, and Wayland has been working fine for me. I typically run in “on-demand” mode, but I’ve used both strictly Intel and strictly Nvidia modes as well, and it’s been fine.

I think the only real issue I’ve had is that Splitgate refuses to launch in Wayland, so I switch to X if I want to play - general computing works fine, native apps have had no issues, and all the other games I’ve played have launched without issue.

The Nvidia GPU is a 1650 TI, and I’m on the Nvidia 535 driver.

See https://youtu.be/GCVJsz7EODA and https://youtu.be/V82lHNsSPww

There are a few problems, but I believe the biggest issue is that .zip and .mov are valid and common file extensions, and it’s common for people to write something like ‘example dot zip’ or ‘attachment dot mov’ in emails, tweets, etc. Things like email clients have features where they automatically convert text that looks like a web address into clickable links. So now, retroactively, all those emails etc suddenly have a link, where they used to just have text, and the domains that are equivalent to those previously benign file names are being purchased by nefarious actors to exploit people unaware of the issue.