Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by MishkaKatyusha

  1. its really quite simple. its no the uploader's problem of crudeness its the antiquated physics coding in secondlife any physics engine worth more than two cents should be able to draw a collision model of exacting detail from the surface of an object exactly.in a decent physics engine there is no such thing as a secondary model for physics shape because what you see and what you bump into are the same thing(of course this isnt talking about texturing)
  2. well first things first,im a little new to house building,but relatively experienced in graphical object making. the one thing i cant understand is what is so morosely wrong with second life's collision analyzer. most of the things ive worked with in the past took there collision surface straight from the surface of the object itself. far as i can tell,it seems one of two things on here: 1.you can upload your own physics shape,which actually seems abit crude unless your desperate to save space,which im not 2.second life's default collision surface seems to be permanently stuck in "one dimension mode" ,which is a term i sue to describe antiquated physics engines that cant calculate beyond an object that is in reality one dimensional,like say a cube or a sphere with a shape but only one continuous surface but nevertheless.i assume given the rich variety of houses on here a way to trick or soundly defeat the antiquated collision calculations in second life has been found. since im not familiar with these tricks,nor looking to save space in the prim count or upload cost. i could use a straightfoward,step-by,-step explanation of how to make it so i can actually draw the collision shape from the model itself,which is basically a roughed up tube (im open to using a simple tube model as a physics shape if the aforementioned problem can be solved).for reference here is a picture (i almost forgot to add,im well aware of the different between convex hull and prim under physics shape in the features tab of object edit mode.but if i use the object model itself as the physics shape,the same one as initially chosen in the upload tab,i either get an impenetrable object or a phantom object not set to phantom when i set it to prim,and ive already tryed making a simple shape or using a completely different model as the physics shape but the same problem still presents itself,i get an impenetrable object when set to prim.and i need to walk on the inside surface) 
  3. thank you thank you ^-^ what you did was exactly what i wanted,algebraic coordinates for rotation axis turned out 90,90,10 and a little adjustment to the height worked flawlessly,ty again ^-^
  4. ahem,this is very nice.as far as i can deduce your attempting to tell me how to convert those quaternion mathmatics into algebraic coordinates. but im afraid i didnt entirely understand what you said.as i explained inthe first post,im not terribly good with the scripting (or was it the second post?) which part of my sleep script do i put this in >.> and how to do so without getting a syntax error,that requires abit more experience in lsl than a first timer(ish) can manage)
  5. well,logicaly it should be possible to do a variety of things with such a system. lasers,missile barrages.a massive playing field using the same mechanics of the game in question,with the same object (size increase of course),where the different peices are star fighters. there are alot of things you can do with a little creativity and some time id definitly love to take a try at it
  6. i almost forgot,the larger of the two hemispheares is the root prim of the two object linkset
  7. welp,lemme show you my problem following is the code,with the problem highlighted in bold,and a screenshot showing what that does.this bed is two linked hemispheares  // Dhyana Writer 2009integer broadcast = 20;// Left Side Sleeping Position//vector target = <-1.06585, 0.71774, 0.18293>;//rotation targetRot = <0.50028, -0.49972, -0.50028, 0.49972>;// A More relaxed sleeping positionvector target = <-0.900, 0.36328, -0.23000>;rotation targetRot = <0.00001, -0.53456, -0.06875, 0.51038>;integer debugRotation = FALSE;key sitAgent = NULL_KEY;integer gotPermission = FALSE;integer time = 0;default{ state_entry() { llSetSitText( "Sleep" ); llSitTarget( target, targetRot ); if ( debugRotation ) { llListen( 1977, "Rotation Broadcaster", NULL_KEY, "" ); llListen( 1978, "Rotation Broadcaster", NULL_KEY, "" ); } } listen(integer channel, string name, key id, string message ) { if ( channel == 1977 ) targetRot = (rotation) message; else target = (vector) message; llSitTarget( target, targetRot ); if ( time == 0 ) llSay(0, (string) targetRot + ", " + (string)target ); time = (time +1) % 50; } changed(integer change) { if (change & CHANGED_LINK) { key agent = llAvatarOnSitTarget(); if ( sitAgent == NULL_KEY && agent != NULL_KEY ) { // Someone new sitting down sitAgent = agent; llRequestPermissions(sitAgent,PERMISSION_TRIGGER_ANIMATION); } else if ( sitAgent != NULL_KEY && agent == NULL_KEY) { // sitting down person got up - wake up if ( gotPermission ) llStopAnimation("sleep"); // Reset the script because we don't have a way of releasing permissions llResetScript(); } } } run_time_permissions(integer parm) { if(parm == PERMISSION_TRIGGER_ANIMATION) { gotPermission = TRUE; llStartAnimation("sleep pose 6"); } } }
  8. welp,first an introduction of sorts. ive been going from one place to the other on the web for the past 15 years.i love computer stuff.I do 3d graphics,story writing,animations,video animations,sound editing,music,and the list goes on. and i also love code,im not terribly good at programming mind you,but i know how to alter things when needed,and have a thorough grounding in logic,also algebra. bit of a problem though.now im on here.ive created a bed which is essentially two linked hemispheares (one for the bed part,the other acting as a sort of pillow).ive bought a script and some animations to go with it.being very mindful to only pick the resellable ones (since that is my intention) but the one thing which has befuddled me so far is "targetrot".ive already worked out quite easily how vector target works,that was easy. but targetrot does not seem to follow at all the logical algebraic representation of axis of movement.the best i can tell each set of numbers in targetrot is some kind of nonsensical looping,spiraling path. i know that it has something to do with the object itself. whats bad is though,the object movement and rotation is just fine,represented with algebraic axis standard. but since obviously targetrot isnt,this leads to a massive mismatch where it isnt at all clear what those numbers will do when you alter them. so my question boils down to: how to use targetrot to make sure my sleeping on side script and the animation has my character sleeping on her side on my bed. (im also wondering,is it possibly a bad animation which is messing up things?)
  • Create New...