LZ4 VS basis_universal

LZ4 is a bit more complicated, but seems faster: https://github.com/lz4/lz4/blob/dev/doc/lz4_Block_format.md

According to these https://github.com/lz4/lz4 values you need around ten (10) quite modern cores in parallel to accomplish around 8GB/s.

The popular NoSQL database Cassandra utilizes a compression algorithm called LZ4 to reduce the footprint of data at rest. LZ4 is characterized by very fast compression speed at the cost of a higher compression ratio. This is a design choice that allows Cassandra to maintain high write throughput while also benefiting from compression in some capacity.

Yeah, sure, when you have monster core counts. on regular systems, not so much, here's from their own github page. it achieves, eh, 5GB/s on memory to memory transfers, i.e. best case scenario. so, uh, no? i'm not even sure it's any better than the CPU decompressor one Nvidia used.

> The downside of lz4 is that it can’t be configured to run at higher & slower compression ratios.

lz4 has some level of configurability? https://github.com/lz4/lz4/blob/v1.9.4/lib/lz4frame.h#L194

There's also LZ4_HC.

Since nobody mentioned it, I'll add lz4 (https://github.com/lz4/lz4).

Get your bootloader unlocked via Download mode and then obtain your stock firmware, preferably for your current region https://samfw.com (Download mode: CARRIER_CODE). Get the boot image from AP with 7zip, unpack from LZ4 with https://github.com/lz4/lz4/releases (drag and drop), patch with Magisk https://github.com/topjohnwu/magisk/releases/latest, grab the new image, name it "boot.img" and pack it into a .tar with 7zip and flash to AP with odin https://odindownload.com

After some investigations and experiments, I found out that it was the PNG compression (well, decompression I should say) that took a while. So I've made some experiments using the LZ4 compression library, which is focused on decompression speed, and it turned out to be an excellent solution!

Root with magisk: whether you use OneUI ≤3 or 4, patch the specific image needed for it (pre 4: boot, after 4: recovery) and flash it to the device. Boot it and enjoy root. https://github.com/lz4/lz4/releases can help extracting it from the AP tarball.

The author of the OP also develops Basis, which is designed to take advantage of GPU texture compression formats. GPU formats have a fixed bitrate which makes them efficient to sample but not the most efficient to transmit, so Basis layers on a second level of compression that unpacks into a GPU texture.

https://github.com/BinomialLLC/basis_universal

You might find https://github.com/BinomialLLC/basis_universal interesting :)

Erm, aren't both WebP and PNG rather useless for games? How do you convert those formats into one of the hardware-compressed texture formats consumed by the GPU (like BCx, ETC or ASTC)? Without those you're wasting a ton of GPU memory bandwidth when sampling textures.

(there are some alternatives, like https://github.com/BinomialLLC/basis_universal, or http://www.radgametools.com/oodletexture.htm)

In addition to this, texture compression is optimized for random access and has to use a fixed amount of bytes per each block, but there is a possibility to further compress this during the delivery. This is called the supercompression and Basis Universal is a good example of this. JPEG XL is also based on tiles for parallel decoding, but those tiles are too large (128x128 minimum, compare with 12x12 maximum of ASTC) so it is not yet suitable for GPU as it stands.

Im interested to see what image format specific compressors go gpu. JpegXL, AVIF, WebP... who wants to show up & throw down? Or even just fastpng?

Meanwhile we dont really hear or regard many of the gpu-oriented compression techs. TIL Basis/KTX2 is itself zstd compressed (formerly LZ apparently?). https://github.com/BinomialLLC/basis_universal

You could check out Binomial's basis_universal. After Google bought them, they open sourced it and allowed use for free.

Another alternative you should consider is https://github.com/BinomialLLC/basis_universal

Simple, open-source, standardized and supports every GPU-oriented format around. And, built-in library support for https://github.com/BinomialLLC/basis_universal

Yep. That lib is great. And, similarly https://github.com/nfrechette/acl and https://github.com/BinomialLLC/basis_universal