Image-shrinker is a simple, easy to use open source tool for shrinking images. Under the hood it uses pngquant, mozjpg, SVGO, and gifsicle. You can also install these tools individually if you need to compress some images. I often use pngquantafter exporting PNGs for web projects from Figma or similar tools. I literally run it like this:.

– Source: dev.to
/
about 1 year ago

Searching more I found https://pngquant.org/ which I could add to my bulk workflow to make most png’s approach the jpeg size.

Source:
over 1 year ago

But this did prompt me to do some searching, and I see https://pngquant.org/ which seems to achieve jpeg like size reduction while maintaining the file as a png. One difference they note is that this method will typically preserve sharp edges better than jpeg (which is probably a strong plus for my type of use case).

Source:
over 1 year ago

Pngquant is also great for shaving filesizes down, but unlike oxipng, it’s explicitly lossy. It’ll reduce colors and even dither, but it will try to keep an image visually similar.
Https://pngquant.org/.

Source:
over 1 year ago

Oxipng, pngquant and svgcleaner — optimizing images.

Source:
over 1 year ago

-https://pngquant.org/ compress png without losing transparency.

Source:
almost 2 years ago

In most cases we’re trying to use svg. In case of png’s, to compress it we use this tool https://pngquant.org/.

Source:
almost 2 years ago

Gif’s uses a limited palette of 256 colors. How do you make that palette. There are many options just here. A user defined palette, a palette based on the average use of colors (perceptually based on the human eye or just based on raw image data), create it using the pngquant library or maybe some machine learning voodoo? And these are just the one I could think of right now:) Following that, should you be able…

Source:
almost 2 years ago

While you are correct, in real world tools like https://pngquant.org/ blur the line.

Source:
almost 2 years ago

I’d like to give a shoutout to pngquant. https://pngquant.org/.

Source:
about 2 years ago

AVIF is the example. It supports palette-based blocks in addition to classic DCT with sophisticated deringing. There’s also https://pngquant.org and https://github.com/richgel999/rdopng.

– Source: Hacker News
/
about 2 years ago

Any LiveArea assets HAVE to be 8-bit indexed AND have to be the sizes listed. Otherwise VPK installation will fail

I like to use pngquant.

Source:
about 2 years ago

If your dataset allows it, I would look at pngquant for “lossy” compression of PNG images by means of color quantization: https://pngquant.org/ This works absolute wonders if you are talking about images with a limited amount of colors and without many gradients. I use it to compress screenshots of websites, which, if you think about it, are mostly large blobs of uniform colors with hard edges. I also use it every…

– Source: Hacker News
/
over 2 years ago

Except that you don’t know that with PNG either, other than for images you’ve transcoded yourself. Plenty of PNG optimization programs use lossy methods (AFAIK mostly color quantization, see pngquant), and you can’t know if an image you got off the internet has been run through such a processor. And besides, you never know what’s been done to the image prior to it being converted to PNG, making it non-lossless.

Source:
over 2 years ago

Generally speaking, absolutely. In this case, no. These were core contributors to AV1, JPG XL, and so on. For example, one of them was Kornelski, the guy who wrote https://gif.ski/ and https://pngquant.org/. They know more about this shit than you and me, and between also explained to me how bad a choice APNG was.

Source:
over 2 years ago

While full PNG is lossless, you can still do lossy 8-bit PNG with pngquant, using a custom palette – and it is an excellent way to optimize a PNG for size while minimizing quality loss. It is not going to win you any awards for medical photography, but if you are producing graphics for the web, pngquant is awesome. It is kinda-like mozjpeg for PNGs.

Source:
over 2 years ago

It really depends on the type of image. With photos you want JPG. With comics and other “clean” images you often can do better with PNG. And if you add a little https://pngquant.org/ you can get your lossy compression with much better quality, and transparency too.

Source:
over 2 years ago

Reminds me of https://pngquant.org/ Same concept just for making images more efficient for PNG.

– Source: Hacker News
/
over 2 years ago

The way I do this is to use PNGquant to reduce the size of the PNG screenshots that are bigger than Discord’s limit. I use Snipaste as my screenshot tool but whatever you use, you can run PNGquant on the saved file and its size will be greatly reduced without affecting its quality. My fullscreen screenshots are 1440p and they end up being almost half their original size so it works great for when one reaches…

Source:
over 2 years ago

Lot of posts here say that dithering is something we did in “good old days”. Dithering is still very much alive, just not as needed on general web sites. But it’s critical when you need to squeeze animation sprites for games or multimedia projects. These sprites typically need transparency, and precise control over which frame is displayed, so standard video codecs can’t be used. So 8-bit palettized PNG sprite…

– Source: Hacker News
/
almost 3 years ago

Or you can use https://pngquant.org/ for free!

Source:
almost 3 years ago