/vt/ - Virtual Youtubers

>>52694324 (Me)
>>52683686
>Makes sense, but some times it can feel like spending a ton time mixing something that ultimately if not put side by side you can't really tell it's a new model.
Following up on this as something came to mind after my original post but sometimes that is the point for some model merges. If you are trying to merge a concept from one model to another, then the perfect merge would be one where only that concept comes over while the overall artstyle remains the same. Which goes back to my original point in that not every merge is meant to visibly look different. Of course your standard artstyle merges are doing that, but some merges just look to drag a concept from a model that doesn't produce very good images (due to it being overtrained on a single concept) to one that produces fantastic images, but is missing the desired concept.
Following up on this, you can actually measure the two different approaches with two models I made, the two most popular BBW models are BBW GAME, and BBW Aurora Borealis. The former was one that I spent a lot of time merging models into in order to try and fix a concept model into being a usable model. The Latter was a merge that I did in order to bring the BEEEGG concept from a bbw model to the Aurora Model, which produces incredible output images by itself. Aurora Borealis has a watered down version of the Beeg concept, it also requires you to be very specific of what fats you want in order for them to appear. Whereas with GAME, you just put in body tags, and then you go beeg. At the end of the day, I still consider AB to be the better model, even though it can't do the concept as well as GAME, but both are very good models and have their strengths and weaknesses.

TLDR: At the end of the day, not every merge is going to be noticeably different as that may not be what they are going for.