Also; What works for one model (setting wise) will probably not work for another. Since we (explicitly) don't send models with the slice data, we don't know what features the model has.
We do look at the data (quite a bit actually) and use that to update / improve our default profiles.
Recommended Posts
ahoeben 2,016
An average does not always make a good value, especially if you realise that there is interplay between settings. And there’s also an interplay between what settings work for what printer types, materials and even specific models.
In short, a LOT of data would need to be combined to train a deep learning network to be useful for predicting profile settings.
Link to post
Share on other sites