Prim cuts not staying stable

I have been trying to get this addressed for the past month or so. I have left a few bug reports on items when they have problems, but can normally do something else for awhile and come back some days later and find it cleared up... only to have it return days or weeks later.

My problem has been on Prim Cuts. Simple, basic, whole number prim cuts. This morning I have been trying to cut a Prism scaled to 0.01m^3 to 0.22 and 0.78... and it keeps changing to 0.219 and 0.779 or jumping back and forth between 0.78 and 0.779. Noticed this with a Box, also. The cut can be done on a sphere or one of the advanced shapes, but will not hold on a Prim type change.

So I sent in 5 wonderful bugs to cover this. As this was even happening to things I took out of my inventory! Say what you will, but if things in my inventory are not stable, then I will not take them out for use. Luckily everything I wear beyond my clothes is self-made, so I can easily dispose of problem pieces. But if this starts happening to pieces that other people make.... *sigh*

So *please* let Prim cuts, advanced cuts, dimpling and other such things stay stable? An upgrade in SL version should really, really not give the impression that basic, user made things, are not stable. Even if this is just an interface glitch, I need to know it and how it behaves so I can properly make things.

Days when I have energy, attention and wakefulness are rare to me these days... and then to not be able to do something enjoyable with them... *sigh*

I expect that this is just the revenge of the Prims for what I do to the poor things...

Di - I don't have a solution - but maybe I can offer some peace of mind.

First, the bad news: Position and rotation can be slightly unstable. Take three prims and link them together. Then unlink them. Then relink them with a different prim as the root. Keep on doing this over and over and you should see the prims drifting slightly from their original position and rotation.

The good news is that twist, hollow, cut, advanced cut etc etc etc are perfectly stable as far as the prim is concerned. There is an issue that numbers like 0.78 get represented as though they were 0.779 - but once the prim goes to 0.779 it stays there.

The flashing to-and-fro that you see is indeed an interface glitch - and while it's frustrating - be assured that your prim is perfectly safe.

I have one cylinder prim (a window) that is cut to .8, and it keeps jumping to .799. Not a big difference numerically, but it makes a dark line in the glass as it's supposed to be seamless. Strangely, the other prim on the other side doesn't do this.

Bug reported, and I hope everyone else that has this happen does the same.

That is actually outright untrue in my experience.

Since 1.8 i've found that shapes don't like to cut to ANY round numbers other than 0.0,.25,.5,.75 and 1.0 (sometimes not even those). This seems to pop up mostly on cubes, but it can strike at any time on any shape. the misalignment is VERY real. it's not just the interface telling me the wrong information. It wouldn't be as big of a deal but lets say i need to match the inverse of a cut and it's set to .649 . to match that i would have to do .351 . last i checked this is impossible, especially since i'll probably wind up with an error and get .349 or .359 . This does seem to have it's off and on times and sometimes it seems fixable with the right amount of kurmudgeoning the numbers in. But other times i've input the values at least 20 times with the same offsets occuring again and again. grr argh aaargh feh!

Ben - I do thank you... I have seen prim drift, also, and it is very frustratring, especially at small scale. And as for my prim being safe... well I hope it is... but from what I have seen (and it may just be an interface problem) 0.779 and 0.219 to not *look* like 0.78 and 0.22... but then I *am* fussy! There is a difference between nearly straight and straight for some work, and nearly just doesn't look right as the real thing. Probably more mental than anything.... but *still*...

Lora - Thank you! This little bug comes and goes but today it really just got to me... it is the little things that always do so for some reason...

Kenzington - I originally saw this with the cylinder and sphere... then it migrated to the torus and ring (but not tube)... now to the box and prism... and each time I just look at the numbers and see them decide to change on their lonesome... bad prim!

I thank you all!

yeah - the problem is that what the software considers a round number is not what we do. As you've noticed, 1/4, 1/2 and 3/4 work (I've never seen them not). So do 1/8, 5/16, 17/32 etc etc - these are all "round numbers" to the computer.
Yes it is real - what I'm saying, though, is that they're stable. The misalignment takes place immediately - and then stays that way, unlike position/rotation drift.
The workaround here is too change the first cut to something like 0.65625 - which is a "round number" (the editor will let you type in this number - just don't worry that it displays it incorrectly afterwards - the prim knows that it is 21/32). Then when you reach the other cut, you'll find that 0.34375 matches correctly.
hehe - I understand, and feel your pain - but I'm afraid this is one of those "divide 10 by 3, then multiply by 3 to get 9.9999" things that can be solved - but usually at more of a cost than it's worth.