Jump to content
  • Advertisement
Sign in to follow this  
richgel999

[Release] Advanced DXTn Texture Compression Library Now Available at Google Code

This topic is 2537 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

[font=arial, sans-serif]

[font=arial, sans-serif]I've released a new open-source tool and library that some people here may find useful:[/font][/font]

[font=arial, sans-serif]

[font=arial, sans-serif]Advanced DXTn Texture Compression Library Now Available at Google Code[/font]
http://code.google.com/p/crunch/

[font=arial, sans-serif]crunch/crnlib is an open source (ZLIB license), lossy texture [/font]compression tool and library for developers that distribute and use content in the DXT1/5/N or 3DC/BC5 compressed mipmapped texture formats.

[font=arial, sans-serif]crnlib can compress mipmapped 2D textures and cubemaps to .8-1.25 [/font]bits/texel, and normal maps to 1.75-2 bits/texel. crnlib's quality and performance is competitive to transform based solutions, or other offline/real-time DXTn compressors such as squish or ATI_Compress.

[font=arial, sans-serif]crnlib implements a new form of "clustered" DXTn compression, with a [/font]compressed texture data format that was carefully designed to be very quickly transcodable directly to raw DXTn texture bits with no intermediate recompression step or individual pixel-level operations. The typical single threaded transcode to DXTn rate is equivalent to 100-250 megatexels/sec. Fast random access to individual mipmap levels is supported.[/font]

-Rich

Share this post


Link to post
Share on other sites
Advertisement
Thanks for sharing!
I've only had a quick glance at the project page, but from what I gather, your focus is on achieving great on-disk BPP results (via the intermediate format and/or LZMA-friendly output bytes), rather than focusing on better on-GPU quality when compared to offline compression via squish/nVidia/ATI/etc?

if you're already packing your textures to DXTn you probably don't care or actually need max. quality to begin with...[/quote]...or you're still targetting GPU's with 256MiB of RAM and have no choice but to DXTify everything wink.png

Share this post


Link to post
Share on other sites
Thanks for sharing this, it sounds really cool. Hopefully it can be used as an alternative to the heavyweight transcoding processes currently used for games that use virtual textures.

Share this post


Link to post
Share on other sites

Thanks for sharing!
I've only had a quick glance at the project page, but from what I gather, your focus is on achieving great on-disk BPP results (via the intermediate format and/or LZMA-friendly output bytes), rather than focusing on better on-GPU quality when compared to offline compression via squish/nVidia/ATI/etc?


That's right Hodgman. As far as I know, crunch is the first open-source DXTn compression library that allows you to purposely trade off a small amount of quality (in a hopefully perceptually friendly way) for much smaller disk files, faster downloads, or to free up memory.

It can also do regular (block by block) DXTn compression, like libsquish or ATI_Compress. When used in this way, it usually beats both libraries in a PSNR or RMSE sense, but only by a tiny (visually imperceptible) amount.

-Rich

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

GameDev.net is your game development community. Create an account for your GameDev Portfolio and participate in the largest developer community in the games industry.

Sign me up!