Just an idea: sometimes when I mistakenly release something what is memory hungry, a watchdog would be useful, and not only for the subdivision. When an older (ie. 1-2 cores) machine is starting to use the swap, sometimes it's nearly impossible even to open a terminal to kill Blender. After x% of memory usage, a watchdog should pause the process (on threading level, or with a trap at the malloc function) and should ask the user what to do, with an option to undo the last edit.
Fazek 2016-04-30 11:53 keltezéssel, Rain Gloom írta: > Yeah, on second thought the warning would get irritating pretty fast. I > vote on Gandalf's too. > > On Sat, Apr 30, 2016 at 9:25 AM, Patrick Boelens <p_boel...@msn.com> wrote: > >> This would be internally inconsistent though, and frankly seems like >> overkill to me. +1 from me on Gandalf's soft/ hard limit suggestion. >> >> Sent from my iPhone >> >> On Apr 30, 2016, at 1:39 AM, Rain Gloom <raingloo...@gmail.com> wrote: >> >>> IMHO a warning would suffice. It shouldn't be hard to compute an estimate >>> of the number of resulting vertices/faces, so the warning could say >>> something like: "This will create <resulting-number-of-vertices> which >>> might slow down your computer." >>> Same thing as asking before overwriting a file or closing without saving. >>> >>> On Thu, Apr 28, 2016 at 6:36 AM, gandalf3 <zzyx...@gmail.com> wrote: >>> >>>> As seen in this SE question >>>> <http://blender.stackexchange.com/q/51525/599>, the current >> hard-maximum >>>> number of subdivision cuts (100) is easily within the realm of what >>>> users would like to use. >>>> >>>> What is the reasoning behind having a hard-maximum at all? I can >>>> understand having a soft-maximum to protect against accidentally setting >>>> the number of cuts too high, but that system is already in place and >>>> works well. >>>> >>>> Would it not be more user-friendly to allow any number to be typed >>>> directly, especially if the user wants a prime number of cuts? >>>> >>>> This maximum apparently didn't exist in 2.74 >>>> _______________________________________________ >>>> Bf-committers mailing list >>>> Bf-committers@blender.org >>>> https://lists.blender.org/mailman/listinfo/bf-committers >>>> >>> _______________________________________________ >>> Bf-committers mailing list >>> Bf-committers@blender.org >>> https://lists.blender.org/mailman/listinfo/bf-committers >>> >> _______________________________________________ >> Bf-committers mailing list >> Bf-committers@blender.org >> https://lists.blender.org/mailman/listinfo/bf-committers >> > _______________________________________________ > Bf-committers mailing list > Bf-committers@blender.org > https://lists.blender.org/mailman/listinfo/bf-committers > > _______________________________________________ Bf-committers mailing list Bf-committers@blender.org https://lists.blender.org/mailman/listinfo/bf-committers