Capture One fail

A RAW processor is fairly sophisticated and complex software. The people who are using such software range from complete novices to experts. There are dozens, sometimes hundreds of parameters to adjust.

To make it all accessible, programmers spend a great deal of time to come up with defaults. The purpose of defaults is mainly two-fold. One, they are part of the user interface and meant to guide a user through the options by providing reasonable starting points. Two, they are designed to adequately cover the majority of cases. This latter aspect is very important. If software isn't designed such that the majority of the cases can't be handled with a minimum of fuss, then it is doomed. People don't have the time nor the patience to figure out how to adjust the parameters if they have to do it for most of their images. Particularly not when there is competing software that actually can handle the majority of cases.

Defaults have become so important that, in good software, one strays from the defaults only when absolutely necessary, i.e., under exceptional circumstances. Thus, often enough, there is more effort spent on defining proper defaults than on coding the underlying algorithms.

Now, defaults are very useful for judging software, but they don't really say much about its true capabilities. IMO, the only way to adequately judge RAW processors is to have experts process a bunch of reference images with the goal of optimizing certain objective, pre-determined aspects that are of importance (color accuracy, sharpness, noise, etc., the usual stuff). One needs to have a 'shootout' with true experts, perhaps even the software designers themselves.

And that's what the software manufacturers do. Their testers do nothing but process images, then take the knowledge gained about how to use the software, and parametrize that knowledge. And that parametrized knowledge ends up in form of defaults.