You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have a workflow that worked fine using a single send/receive pair (ID0), but when I added a second pair (ID1) then the behaviour seems to be broken and the incorrect images are sent. It's hard to explain, but ID1 seems to use the image created from the previous run of the workflow.
It seems to go wrong at this point in the image below, where ID0 is showing the correct image, but it's passing the image from the last run into the node. You can see the output doesn't match the input faces.
The one on the left here is the newly created image, and the one on the right was from the previous run.
The text was updated successfully, but these errors were encountered:
This is definitely a problem between the input and output of the receive node. You can see that the correct image has been passed to the receiver, but the output (input b, on the left in the comparer) is still showing the previous image.
I think I found out what is happening. Instead of creating the initial image first, the image receiver is loading the image it already has, from the last wf run, into the facedetailer before the first ksampler has done anything. Then the ksampler creates the initial image, passes that to the receiver, but the facedetailer has already run, so ignores the new image.
Basically, the load image behaviour of the receiver is being acted on before ksampler.
I have a workflow that worked fine using a single send/receive pair (ID0), but when I added a second pair (ID1) then the behaviour seems to be broken and the incorrect images are sent. It's hard to explain, but ID1 seems to use the image created from the previous run of the workflow.
It seems to go wrong at this point in the image below, where ID0 is showing the correct image, but it's passing the image from the last run into the node. You can see the output doesn't match the input faces.
The one on the left here is the newly created image, and the one on the right was from the previous run.
The text was updated successfully, but these errors were encountered: