<h2>ComfyUI: <a target="_blank" rel="ugc" href="https://github.com/comfyanonymous/ComfyUI">LINK</a></h2>Github repo + nodes description: <a target="_blank" rel="ugc" href="https://github.com/Derfuu/Derfuu_ComfyUI_ModdedNodes#nodes-descriptions">LINK</a>Leave suggestions and errors if you meet themWhat's new in 0.5.0:<ul><li>CombiningArea scaler</li><li>More user-friendly ui names</li><li>ALL nodes description moved to <a target="_blank" rel="ugc" href="https://github.com/Derfuu/Derfuu_ComfyUI_ModdedNodes#nodes-descriptions">GitHUB</a></li><li>Tuples and so on moved to their own directory in UI</li></ul><h3>Simple* introduction:</h3><ul><li>Automate calculation depending on image sizes or something you want</li><li>easier(or not) editing multiple values of various nodes</li><li>Math</li><li>Modded scalers</li></ul>Installing: unzip files in ComfyUI/custom_nodes folderShould look like this:<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/2c994aaf-e598-4ce9-69af-c5ee2067f000/width=525/2c994aaf-e598-4ce9-69af-c5ee2067f000" />For example (v0.5.0) there is an example how scaled ConditioningArea can improve image after scaled latent combining:Only LatentCombine:<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0c1fbcb4-0f2b-4342-19b1-5affeed15f00/width=525/0c1fbcb4-0f2b-4342-19b1-5affeed15f00" />Combining preview:<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0ad127f1-03fc-49fc-1e2c-a28aff8f0500/width=525/0ad127f1-03fc-49fc-1e2c-a28aff8f0500" />LatentCombine with scaled ConditioningArea (640*360 to 1360*768):<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/62c6c3e9-62d1-4629-39f0-e32b53974500/width=525/62c6c3e9-62d1-4629-39f0-e32b53974500" />Example of workflow i made for this located in: /Derfuu_ComfyUI_ModdedNodes/workflow_examples/<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/642d4f97-aa94-435d-0b21-ae2394bcd300/width=525/642d4f97-aa94-435d-0b21-ae2394bcd300" />model: hPANTYHOSENEKO (sorry, couldn't find link)negative promp: embedding:<a target="_blank" rel="ugc" href="https://civarchive.com/models/11772/verybadimagenegative">verybadimagenegative6400</a><h3>TROUBLESHOOTING:</h3>If there are troubles with different sizes, aside from *64, this may solve problem: found on <a target="_blank" rel="ugc" href="https://github.com/CompVis/stable-diffusion/issues/60#issuecomment-1240294667">GitHUB</a>This code is at the end of this file: /ComfyUI/comfy/ldm/modules/diffusionmodules/openaimodules.py<h3></h3>NOTES#2:<ul><li>Debug nodes counts as OUTPUT nodes and can be used withowt image preview or save nodes to get results</li></ul>P.S.:<ul><li>All fixes wou can find or post on github, i look there too</li><li>If you catch error like: Calculated padded input size per channel: (2 x 82). Kernel size: (3 x 3). Kernel size can't be greater than actual input size. This MAY be because of too high or low offset you give to node</li></ul>

ComfyUI Derfuu Math and Modded Nodes

comfyuiDerfuuMathAnd_derfuuNodes01.zip

Derfuu_Nodes_01

Files with SHA256: 5bf12f3c6ff68da6c54c6c231c2c20eb8085fd9987304df1da0dafc6f0d0de84