Update d-adaptation/notes.md

d-adaptation/notes.md

@@ -21,6 +21,7 @@ Lower dims show good performance. Need much larger test to check for accuracy be
 
 ## Resolution
 512 resolution is good so far
+640 resolution is good. did 1.5x the number of steps as compared to 512 (640/512)^2 is about 1.5. Yet to conclude if this is a good rule to follow. UMP has 640 resolution sample.
 
 ## Steps
 600 to 800 steps have been the primary focus of experimentation. 
@@ -29,4 +30,7 @@ Over 1000 has not shown much improvement.
 
 ## 2.X models
 To be tested. 
-Candidate base models: wd1.5, replicant, subtly
+Candidate base models: wd1.5, replicant, subtly
+
+## Noise offset.
+toyxyz has noted that using high noise offset (higher than 0.2 it seems) with d-adaptation creates unusable results. It starts looking better than lower learning rates but even unet 0.5, text 0.25 with noise offset of 0.75 does not look usable.