16bit or 8bit images?
We spotted some performance issues with very large Tiff files (e.g. 500mb +) in a customer’s account. Upon investigation we discovered that these were high resolution (8000 by 11993px), saved in 16bit colour mode and without LZW compression.
This begs the question - why 16bit, and why no compression.
These are super large and probably unnecessarily so.
These files were saved as 16bit image files which add no perceptible value (for photography where it is unlikely that they will be edited) and immensely increases the file size making most workflows unnecessarily harder/slower.
The main Differences Between 8 Bit and 16 Bit - In a nutshell, it comes down to colour range and detail. 16 bit images offer more colours and deliver smoother tonal transitions. That’s critical if you do a lot of post-processing work because you can adjust your images without losing quality.
However for most photography workflows (e.g. Tourism/Destination marketing) you are capturing a scene and typically would not expect a lot of post processing by users - so 8bit is pretty much what most people use.
Using 8bit images (the default for most cameras and workflows) and adding LZW (lossless) compression (we always recommended for TIFF files), reduces the file sizes from 500+MB, to around 130MB (for the sample images we tested) - still a large file due to resolution but a much more palatable size, for storage, transfer downloading, usage, and cost.
Trade offs
There is always a trade off. In this case think about the likely end use case. Typically you want high quality images saved in size efficient file formats with lossless (e.g. LZW for TIF files) or minimal compression (e.g. High Quality JPG) .
For storing and distributing photographic assets, In most cases an image at A4 size at 300dpi, equivelent to 3508 x 2481 pixels (8.7 megapixels) is good enough and pretty standard across photo libraries and DAMs and Brand Toolits.
Choose 8bit in most cases, but 16bit if you think users will be editing the images, pushing pixels, especially adding gradients etc in post processing.
If you want a much higher resolution for say an A0 poster or a Billboard etc you might want higher resolutions - but if you choose 8bit (the standard) your file will be significantly smaller - typically about half the size. And in the case of these high res use cases the 8bit color image are perfectly acceptable.
Our recommendation for you going forward.
- Generally ask for Photography to be supplied to you are in 8bit format (not 16bit or 32 bit) to reduce filesize.
- Insist all TIFF files are saved with LZW (lossless) compression to reduce filesize.
- Both PNG and TIFF files benefit from lossless compression, meaning that no matter how often you save, open, or resize them, they’ll maintain their quality. (However, TIFF files offer users a choice between lossy and lossless compression, which can help reduce the file size if this is a priority). Generally these days TIF’s are not ideal, PNGs (Lossless) or JPGs at Max quality (still a wee bit lossy) are good enough for 99% of use cases and far more efficient for sharing/working with - especially in digital workflows. So we generally recommend using PNG or JPGs versus TIFs unless there is a specific reason for TIFs.
Happy branding :)
