Wikipedia editor
| This is a Wikipedia user page.
This is not an encyclopedia article or the talk page for an encyclopedia article. If you find this page on any site other than Wikipedia, you are viewing a mirror site. Be aware that the page may be outdated and that the user whom this page is about may have no personal affiliation with any site other than Wikipedia. The original page is located at https://en.wikipedia.org/wiki/User:Noir~enwiki. |
Hello. Bonjour. Ciao. こんにちは。
I suppose I'm what you could call an "all-around" contributor. I'm here to help with anything I can lay my hands to, and always enjoy learning new, better ways to go about things. Comments, advice, help, etc. are of course more than welcome on my talk page.
One of my major projects right now is to try to losslessly compress the highest-visibility PNG files I can find. This helps articles load faster and cuts server costs.
After some experimentation, I've found that the best chain of programs (linked in a small bash script) to achieve maximum compression is:
- PNGOUT (very nice zlib implementation and custom DEFLATE algorithm; best compression ratio at the cost of CPU)
- advpng -z4 (maximum AdvanceCOMP restructuring)
- advdef -z4 (maximum AdvanceCOMP compression, using 7zip)
- OptiPNG (sometimes catches a few extra bytes after the above)
- DeflOpt (fast, and almost always catches a handful of bytes on small images, sometimes kilobytes-worth on larger ones)
I haven't come up with a notation for listing my compression process like Tene has; usually I just manually list what I've done with a file. Where appropriate, sometimes I'll index an image's color and truncate its palette to save even more space, but generally I stick to pure lossless methods.
As of 1/2009, the Wikipedia logo itself was actually last compressed on my computer, using this method.
By the way, I'm still waiting to hear back from Ben Jos Walbeehm regarding his GCC-compiled binary of DeflOpt. If you happen to have access to the Linux executable, please leave me a message. Wine is great, but it's definitely slower than native code.
Update: An anonymous contributor was kind enough to provide me with the link! The elusive Linux version can be found here.
