/technology/ - Technology

>>1932
>>1944
I had the opposite problem, I got fatter while working from home. I'd buy McDonald's almost every day

>>1945
what did you order?
you font feel smelly and gross after that? not enough to deter you from repeating? McDonalds makes you smelly. its why fats stnky. gross.

>>1975
Big macs and mc doubles and milkshakes, but I stopped doing that, now I eat normal food

I have 32gb of DDR4 that I got for peanuts, but I really wanted 64gb of DDR5

If you want ram prices to go down you should start abducting the children of SK Hynix, Samsung and Micron executives.

Im starting to get blackpilled on the whole topic. yes there are cracks in the ai bubble or whatever, but they have always been there. yet it keeps getting worse. and despite all the slop, ai also offers incredible things.
too bad i didnt prioritize ram or a good motherboard back in 2021 when i bought my current pc. its perfectly fine for gaming and all, but i doubt i will be able to buy a new one for the same money anytime soon, or maybe ever again.

>>1985
KEK

>>1963
He's biased void. There's plenty of arguments in support of JXL. Maybe more people would come to your site if you allowed JXL as well. It's all they bitch about on /g/ so the demand is there.

File: caniuseit.webp (420 KB, 3244x1444)

>>1949 (OP)
>JXL is now supported on Chrome and Firefox.
>@void do you have future plans to add JXL support to your site?
I was looking into the availability and it's still quite low, I think we might need to let it mature a bit.
Since chrome now fully supports it, I think availability might increase rather "fast"

>>1974
I see.

I told /g/ the news. Thanks for taking the time to respond but we'll be going to the fuze imageboard for JXL. You should make a thread on /g/ when you add JXL doe, it will help the image format grow.

File: pin reply.webp (17 KB, 445x523)

>>1977
Yeah it's all good anon
By the way if you go to the slop version of the thread at https://slop.umigalaxy.com/explore/technology/1949 you can use the Pin Reply option and pin a reply you make with the link to the fuze imageboard.
That way you can redirect other anons interested in JXL there in the meanwhile

are only tight posters allowed here?

>>1861
Depends if you're square.

>>1866
i'm cute
also lite mode is really cool

Why did you update the firmware in the first place? I use an LG C4 and it's not even connected to the internet, literally turns on to the HDMI input like a monitor.

>>1719
It auto updates and I connected it for netflix or something. A couple months ago I disconnected and it's much better now

>>1009 (OP)
i updated my phone to android 15 and everything got worse. i feel like android has been dying a slow death since 10

Man it looks like you've been working hard on this, good job

File: 1773979075189791.webm (728 KB, 640x480)

>>1707
Thanks man, I'm planning on mirroring this to Kissu and Reddit soon now that this is usable. I daydream about a Powershell version of this but I'll worry about that next year lol.
Maybe a pro at bash on Kissu/Reddit can help me auto target a CVVDP score, I've read that you need to understand binary search for that and I'm cooked there.

I also made the decision to nerf AVIF and use speed 4 instead of 0. Partly because I feel bad for JXL getting slaughtered by it but mostly because speed 0 is incredibly slow, very few will ever use it.

>>1708
Cool, you can post the kissu and reddit threads here as well

File: ambush.webp (104 KB, 1080x853)

>>1709
They're live. I didn't mention this place on reddit to avoid spooking them, I'll try that when I go through the 12 images and reach a conclusion on the winners and diminishing returns stuff.

https://www.reddit.com/r/compression/comments/1wmxmwd/v009_image_codec_cvvdp_bash_script/

https://kissu.moe/maho/thread/9196

I think I burned myself out last time by focusing on both the results and the flicker tests at the same time so I've decided to focus on the former first and then the latter.

File: IMG_1.webp (758 KB, 1920x1080)

IMG 1 CVVDP 9.0 and 9.5 results

>>1652
>How it feels to fail a Speech check

File: 20260914_192527~3.webp (3.29 MB, 4032x3024)

guns!

>>1662
Always appreciated

File: Untitled24.webp (111 KB, 1920x1040)

!USE AT YOUR OWN RISK! !USE AT YOUR OWN RISK!

Here's the windows .exe fssimu2, you're going to need all of those .dll inside when placing them in PATH btw, I don't know how to compile them together unfortunately.
All I did was modify build.zig/dependencies from the fssimu2 project files and generate a windows executable through MSYS2 and bundle all the .dll files it needs in a .zip.
However I can't guarantee how reliable/stable this thing will be since it's fssimu2 hasn't even hit version 1.0 yet. I also got a false positive on an A/V scan.

https://files.catbox.moe/iot2sr.zip

!USE AT YOUR OWN RISK! !USE AT YOUR OWN RISK!

File: 4178.webp (44 KB, 640x480)

For those who don't know when AV1 was released it was only beating H.264 at low quality. As the SVT-AV1 project improved it also started beating H.264 at medium quality but until recently there wasn't much benefit going from H.264 to AV1 if your goal was high quality. So that's what the Essential fork is trying to fix. In the future we may even see a fork for transparent quality (ie literally can't tell source and AV1/H264 encode apart).

Measuring what we'd subjectively describe as high video quality has also been a guessing game until recently with the introduction of SSIMULACRA2. Before that we had PSNR/SSIM with very poor subjective opinion correlation, VMAF with okay-ish subjective opinion correlation. SSIMULACRA2 AFAIK currently offers a much stronger subjective opinion correlation though I'm not sure if it's the strongest (see IW-SSIM)

Good job anon, I won't be able to buy any new hardware for years I've been converting my stuff to av1

>>1660
It took a while to become usable though. I still remember back in 2020 people complaining about quality sometimes being worse for the same H.264 bitrate. Just way too much smoothing in general.