曾经沧海难为水，除却巫山不是云。If there are no clouds in the mountains, then the mountains are not mountains.If there is no morning fog in the woods, then the woods are not the woods.<ol><li>苍云山/cyan cloud and mountain</li><li>玄远仙境/ Fairy immortal world theme</li></ol>我的Ko-fi网址，里面有一些我做的图片，欢迎来看看。My Ko-fi site, which has some pictures that I made, feel free to come and have a look.<a target="_blank" rel="ugc" href="https://ko-fi.com/asseer">https://ko-fi.com/asseer</a>【Note: 支持性捐助仅适用于经济独立的人群。Supportive donations are only available to those who are financially independent. PS：感谢每一个帮我买可乐的朋友】The same seed comparison plot generated by the XYZ axis script used for the first example plot. It can be clearly seen that under the weight of 0.8, more clouds and more ethereal buildings are generated. The overall composition has undergone major changes. Notice:1. The horizontal X-axis title on the top of the picture replaces the lora call with the same prompt as the title, including the controversial ink painting, and I added a weight of 1.2. A clear difference can also be seen.2. On the vertical Y axis, there are different models, followed by lyriel/the real module of PAseer released by me/RevAnimation/tmndMix/AnythingV5. In the case of different models, significant differences can also be exhibited. And the 2D/2.5D/3D large models all have a tendency to change to the fairy style.Personal opinion:1. Because there is no obvious problem when I tested it myself, I think this module is effective, and the effect is obvious.2. It is also because the tester is only himself, so it is necessary for you to analyze and understand the feedback pictures and the data of the generated pictures, so as to know what difficulties the users have encountered.3. If the model is invalid, or the effect is not obvious, you are welcome to provide your generated data, and I will improve and adjust the generated prompt words that are difficult to generate, and generate posts to show and point out where the problem is. . (Similar to changing homework).4. If it's not obvious, I can't do anything. So, feedback is welcome (assignments are welcome).第一个例图用的XYZ轴脚本生成的相同种子对比图。可以明显看见，在0.8的权重下，生成了更多的云雾，更缥缈的建筑。整体构图都有了较大的变化。注意：1. 在图片上面横向的X轴标题，是将lora调用替换成了和标题一样的提示词，包括了有争议的ink painting，我还加了1.2的权重。也能够看到明显的区别。2. 在纵向的Y轴上，是不同的模型，依次是lyriel/我发布的PAseer的真实模组/RevAnimation/tmndMix/AnythingV5。在不同模型的情况下，也能表现出明显的差异。并且2D/2.5D/3D大模型，都有向仙侠风格改变的趋势。个人见解: 1. 因为自己测试的时候没有出现明显的问题，所以我认为这个模组是有功效的，而且是功效明显的。2. 也是因为测试者只有自己，所以才需要各位反馈图片和生成图片的数据进行分析和理解，才能知道各位使用者遇到了什么困难。3.如果出现模型无效，或者效果不明显，更欢迎各位提供你们的生成数据，并且有时间我会将有困难的生成提示词进行改进和调校，并生成posts进行展示并指出哪里出了问题。（类似于改作业）。4. 如果只说不明显，我是什么也做不了的。所以，欢迎反馈（欢迎交作业）。

Files with SHA256: 01202de3f8f528eeb72ea134eeab7daaeae3dc14954fc8a00314b4de791a9cea