hgm wrote:Good point. Normally root moves that fail low (because there already is a better move in that iteration) would not be printed. How is that different here? As long as there is no move, would you print a PV for every move that fails low? Or just the first? I suppose the whole idea of having an aspiration window is that when the PV move fails low, you will try all other moves first before enlarging the window and redoing the iteration?
I could alter the sorting algorithm such that a new PV could never pass a PV of the same iteration that starts with the same move. Would that solve it?
It seems to me that the most recent PV sent by the engine should always be on top, pushing all others down one row. Easy to read and understand, at least to me....
jm
Nearly all GUI's display the most recent PV at the bottom, which is more logical than putting it on top.
There is 1 GUI doing that and when i'm seeing it, each time i find it confusing.
We tend to read from top to bottom, not from bottom to top, that's why.
I like the most recent on top exactly because that is where I start reading. Usually I am interested in the last PV only.
There also is an objective reason why this works better: When you would append at the bottom, first the bottom would move down while you are filling the display, and only then it would start scrolling (if you are lucky, otherwise it would be out of view). By adding it at the top it is always in the same place.
hgm wrote:I like the most recent on top exactly because that is where I start reading. Usually I am interested in the last PV only.
There also is an objective reason why this works better: When you would append at the bottom, first the bottom would move down while you are filling the display, and only then it would start scrolling (if you are lucky, otherwise it would be out of view). By adding it at the top it is always in the same place.
Just for the record, there are two views to this.
I regularly use "tail -f logfile" to see all the program output that I don't want sent to xboard. That is a traditional scrolling mechanism that has oldest on top, newest on bottom. That is my preference since all programs that run and send output to stdout display their output in that order.
Having a window that behaves differently (the opposite, in fact) is certainly OK. But it would be useful to have a toggle where it can be changed if the user prefers newest at the bottom. Perhaps call what you do "reverse scrolling" or something similar and let me choose to disable it if I want.
Note that the thread is not about whether pv's should be added on the top or on the bottom of the thinking output window, but about the fact that the thinking output window is messed up in case of a fail low at root.
