I want to use Tritium for transcoding like 85 GiB of HEVC/AAC to AV1/Opus.

Mostly personal 1080p footage from older phones I had. Anything special I should keep in mind? Best practices?

I am currently considering RF 35 with preset 3. The size reduction is like 44% in one tested video. The small quality loss can be seen if I compare frames side by side but is very hard to pickup on a 27 inch monitor. Is this normal?

Installed some of the notable ones on it. All of them have the same problem. Even GCam.

They don't record videos with the Camera2 API. The video always is checkered like old TVs lol.

The original Camera API works but is dog shit. Like the frame rate drops to 17 fps or 25.

Please suggest a few solutions.

I tried and converted an 8 GiB ISO to ZSO format using OPL's own ziso.py script. It shrinked down to 4.5 GiB. And it runs fine on a flash drive. It is Midnight Club 3 remix edition.

I wonder how good or bad is the current ZSO compatibility on OPL. Anyone who uses zso games regularly?

I just tried formatting a usb drive with GPT (instead of ancient MBR) and OPL really played the game on it!!!

The drive has exFAT filesystem. That was supported a long time ago. But never knew GPT partitioning was made available. Has anyone tried it?

Now only if they add modern and powerful UFS filesystem support so I can ditch the old exFAT... that would be impressive.

Digital buyers make fun of the protestors. However they don't realize that:

Your bank balance will age like milk when Sony sells 79.99$ games only on PS Store without any discounts for 2 years after the games' launch date. ☠️💀

Because no competition in the second hand market, right?

And after two years, a discount will come. How much? 20% price reduction 🙂. So you will pay 64$ for an old game that has no maps, artwork, cover art, extras, etc.

And the cherry on top? Sony will find a lousy reason to ban your account. You will lose 3000+ dollars worth of games 🥰🥰.

You simply rent the game until they close the store for that console. You don't OWN it. Blu-Ray Discs on the other hand have a lifetime of minimum 50 years.

As a result, far fewer people will be able to afford games, the community will shrink rapidly, and gaming will become an ultra luxurious hobby. 🙃

I recently discovered a tool called rhash. Not sure how reliable or popular it is though. I am planning to also use rhash to verify the sums in the future whenever I need.

Decided to generate --sha3-512 hash sidecar files for my movie mkv files. I decided to use the --bsd format. It looks like this:

`SHA3-512 (filename.mkv) = 64ffb...`

It looks clean and cool to my eyes though. But is it widely used? Or is it bad practice?

I got a second hand Blue WD HDD. It is CMR. From 2018.

I fully tested it. Including extended 2-hour SMART self test. It had no errors. Sequential read is 94-194 MB/s (depending on if it's reading from the edge vs inner part). To me, it seems to be in good condition. Not sure though.

And a read latency test using HDDScan, shows some numbers:

. <5 ms 1778015

. <10 ms 129541

. <20 ms 69

. <50 ms 102

. <150 ms 4

. <500 ms 8

. >500 ms 1

. Bads 0

Note: each unit above represents 1024 sectors at 512 bytes.

How good or bad is it? Is it similar to a new HDD? I do not have much HDD knowledge. This is my first 3.5" drive.

Following my previous post, I want to ask the experts if the default AV1-HDR fork with VQ is enough for animation content (e.g. Star Wars the Clone Wars) with no additional command line options?

Or should I add a specific option to make the result even better?

I tried many times on Aurora (based on Fedora Kinoite). This new update v2.23.11 does not install at all.

It used to update itself flawlessly and automatically restart the system to apply the new system image. Now after downloading the update, it does not restart automatically.

Anyone facing this issue on an atomic distro?

I got the Blu-Ray remux of the 1st season of the Clone Wars 2008. 22 episodes at 42 GB. I transcoded the first episode from VC1 to AV1 for checking the quality of AV1. I transcoded from AC3 Dolby Digital Audio 5.1 to Opus 5.1 too. I used Handbrake.

SVT-HDR.

RF 25 at preset 3. VQ tune.

1.8 GiB down to 447 MiB. ~75% reduction.

Checked the VMAF scores on nmkoder too. VMAF 98.12, SSIM 0.995 and PSNR 50.07.

How is this? Good? Or bad? I don't notice much of a difference looking at the same frame side by side.

Edit: I experimented a bit more.

• Remux: 1928 MB (100.00%), VMAF 100
• RF20: 599 MB (31.06%), VMAF 98.40
• RF25: 468 MB (24.27%), VMAF 98.12
• RF30: 368 MB. (19.08%), VMAF 97.69

Hope it helps.

I want to use it as RAM disk for temporary file storage. I extract a lot of compressed files and then compress them again. This severely reduces the health of my SSD. I'm doing this on Fedora Kinoite (if it matters).

The compressed files are mostly under 4 GB.

Is it ok to put the compressed files on shm and extract them there and then move them to the destination?

Handbrake super HQ 2160p 4k av1 surround preset has these advance options typed in:

Enable variance boost 1

Enable qm 1

Ac bias 1

Tf strength 1

Qp scale compress strength 1

Sharpness 1

Keyint 10s

&#x200B;

Are these necessary for 1080p phone camera footage that has lots of ISO noise especially in night time videos?

&#x200B;

I have 100 GB of these videos that I may transcode to AV1 SVT.

&#x200B;

I use preset 3 and SVT 10-bit with RF25.

&#x200B;

For example, they make my file size 29 MiB (og hevc file is 34 MiB) whereas clearing them and running without them makes it 20.9 MiB which is good compression but textures too soft in my opinion.

&#x200B;

Do I need denoise and film grain too?

I thought my mid tier laptop CPU could not handle AV1 SVT SW encoding. I tried both preset 0 and 10 which are the extremes of the spectrum. And both work!

0 is basically impractical though. And 10 is FAST. The size difference is around 9%. Is it worth it to waste 9% space for extremely fast encoding? I say yes.

Vmaf is higher with 10 bit. 8 bit is good too. The source I tested with is 8-bit hevc.

I used the Nmkoder app on Windows to compare different presets. Vmaf is above 96 even with RF 30. It is crazy!!! 50% file size reduction with vmaf and psnr and ssim scores being so high.

I want to have my separate standard user for work stuff. Aurora linux has this command ujust that cannot write to the linuxbrew directory because it is not in sudoers. Only the admin account can.

What is the workaround here? Some of these ujust commands have sudo in them. That will prevent me from having a non-sudo work account, I guess?

I am confused a bit. Do I get it from the website? Is there a way to download the DX release? Or do I have to get the base release and after installation convert to DX mode?

I was wondering about the actual number of official full game releases in the north america region. Does anyone know the exact count?

The Wikipedia page shows ~2340 NA titles. I have seen other numbers like 1831 too.

Note: I'm not counting prototypes and demos.

I experimented with 7z quite a bit. I compressed an iso file aggressively using single thread compression and also 8-thread compression.

Of course, the compression time using a single thread took way too long. But I care mostly about decompression characteristics of these two approaches.

Result: ~15 seconds for 8 thread archive, ~44 seconds for 1 thread archive.

Basically, the structure of the archives changes based on the number of compression worker threads used.

7z spawned 12 threads when decompressing the 8-thread archive. However it only spawned 1 thread when decompressing 1-thread archive. I guess that is by design.

Does someone approve this based on their experience of using 7z?

Edit: Based on my tests, the difference is huge. More than 50% time saving when decompressing a 8-thread archive compared to a single thread one on my machine. It is totally worth it.