Idealize not working on 1035

[Susume]

The idealize tool is apparently not working on 1035. Whether idealizing the whole protein or just part of it, the ideality of the affected segments either improves very little or gets worse - it never goes to zero.

Shared a sample pose with scientists, score 24911, named "Idealize not working" - if you idealize the protein at this point the total ideality score (summed across all segments) gets worse, from -293 to -301.77. You can use this script to print the subscore totals: http://fold.it/portal/recipe/37521

Win 7 64 devprev.

[spmm]

Is it the same effect on low and medium wiggle precision? High is not available on 1035 I recall.

[Susume]

This is about the idealize tool losing ideality points even without wiggling. But I have also tried my usual method of idealizing and wiggling at both auto and medium on this protein and failed to gain anything by it.