3 ms·
I'm surprised this doesn't include Google Guetzli (https://github.com/google/guetzli https://github.com/google/guetzli) in the comparison. In my experience tryi
by karamanolev 6y ago
I'm surprised this doesn't include Google Guetzli (https://github.com/google/guetzli https://github.com/google/guetzli) in the comparison. In my experience trying to optimize product images for an eCommerce site, this provided the best compression. Yes, it's ridiculously slow, but for encode-once-transmit-often scenarios, it's perfectly usable.
- acqq 6y agoDid anybody really find any reason to use that? https://www.pixelz.com/blog/guetzli-mozjpeg-comparison/ https://www.pixelz.com/blog/guetzli-mozjpeg-comparison/ "MozJPEG files had fewer bytes 6 out of 8 times MozJPEG and Guetzli were visually indistinguishable MozJPEG encoded literally hundreds to a thousand times faster than Guetzli MozJPEG supports progressive loading"
- karamanolev 6y agoI am compressing 200x200px and 800x800px images. For those sizes, while the encoding speed is slow for real-time use cases, for product images it's completely irrelevant - anything under 1 minute/image gets a pass and lower, generally, isn't better (or worse). I'm looking at absolute image quality/byte and it's even more niche than that - it's bytes to pass a threshold of acceptable quality (85 on Guetzli IIRC). I don't care about very low or very high qualities. In terms of absolute bytes to pass this threshold, Guetzli felt the best when I was doing the research (~ mid-2017). I don't have any hard data to back this though - I did the experiment with 5-6 different product images, drew the conclusion and started using Guetzli.