>In the patch bellow I don't change the current vthreshold system, I just add a new input (-1) to desable it,
>If you set vthreshold -1 the vthreshold = 90 deg, and you never switch to NS, you stay on LRUD, that all. In fact it is the same to set vthreshold 91 deg (I tested this solution but there were bad border line effect) >It solve my issue, I explained may, 22nd ([Therion] Vertical issue) in the list. http://article.gmane.org/gmane.comp.gis.therion/3143/match=vertical >C.S. Cawa It is good to see this issue being looked at and good to have working solutions produced. (I have not actually tried your patch, sorry). If my assumptions are correct it might not be the best approach for the evolution of Therion, although will certainly resolve the most obvious problem that turns up in my models. My understanding of how vthreshold behaves... As the gradient of a survey shot increases, therion switches from the assumption that passages are measured LRUD to an assumption that they are measured LR-front-back and then at 90 deg to NSEW. At present the user has some control of this, but not complete control. From the therion book (and slightly rephrased by me)... 1. vthreshold <number> <units> threshold for interpreting LRUD readings as left-right-front-back reading perpendicular to the shot. 2. If passages are near horizontal (inclination < vthreshold), LR is perpendicular to the shot and UD is vertical. 3. If passages are near vertical (inclination > vthreshold), UD becomes perpendicular to the shot - otherwise passages would not look very good, their height would be 'squashed'. 4. In the case of vertical shots, UD is interpreted as north-south dimension from the station to allow tube-like modelling of verticals (and LR becomes EW). The feature described in sentences 2 and 3 is appropriate, and gives the Therion user complete control to allow them to match the behaviour of their model to the style of booking used in the cave. It should be retained independent of any changes we make to vertical survey leg behaviour. It is only the feature in sentence 4 that is problematic. What is required is a toggle for this feature that allows independent control of 2 & 3. Eg. Perhaps this is a better long term development direction... vthreshold <number> <units> [<NSEW | LRFB>] where the last item is a toggle, for vertical shots only, between the existing behaviour and where vertical shot dimensions are NSEW and a new behaviour where vertical shot dimensions are left-right-front-back. For this to work the orientation of the dimensions would be determined from a compass bearing entered with the vertical survey leg. Cheers Bruce