• Content count

  • Joined

  • Last visited

  • Days Won


Everything posted by Wingnut

  1. Wingnut

    Rotation physicsImpostor physicsBody bug

    *nod* Your choice on drinks, but... no gold malts or diamond soup, eh? Just send bill to BJS Forum Comptroller. I'm wondering if the new build was done or if the PG "latest" was updated. I still see the issue. (after I cleared history and did unconditional reloads). ---------------------- @RaananW/others... do you have any information about @adam's debug error? I show... this.BJSOIMO is undefined
  2. @Arte Stick-in-the-mud work-a-holic, you. See, we even got Babs onto the smoking subject. heh. My guess is no. Most of them are heroine users, though. hah. (just kidding, of course). But, perhaps... code-starers... always need SOME kind of vice. Doritoes got a reputation for being the coder's "junk" for a while, in US history. Ranch flavored. Nerd fuel. Oh yeah, I guess we should get back to snaps.
  3. Oh man, that's a beauty! Nice! (hi-5) I have about 10 pipes... mostly Grabow apple-ish bowls (Dukes) with straight stems... filter pipes. A couple Savellini's? too... originally for balsa filters... for which I short-cut standard filters to fit. Exclusively MacBarens tobacco (Denmark)... aromatic blends (girly flavors)... no Perique or Latakia crap... but I'll occasionally run some "leftist" stuff... mostly MacB "Black Ambrosia" cavendish-like stuff, or some Navy Flake. Use to LOVE MacB "Highland Mixture" but it was a limited run and it went away. 1/10 the price of cigs, 3 times the delicious flavor, but a bit more maintenance and lighter usage. Love it! So cool! No wonder we became good acquaintances so easily. Babs is probably saying "Oh no, my two collaborators are on a man-crush tangent."
  4. With pipe? Now what kind of pipe would that be? Surprised me, because i am both a tobacco pipe smoker, and... umm... other things too, sometimes. But, I've been a tobacco pipe smoker for... ever. You, Arte? Sorry for off-topic. :)
  5. *nod* Sometimes, the primary objective... is to drive it all with the keyboard and not need the mouse. This allows things like voxel constructions... to be built very fast. All your offsets are probably... 25%/50%/75%... and you can almost "type" a construction just like you would a sentence. Not sure if that is sought... but if it IS, that opens-up many hot keys and easy-buttons. Even things like... keep hitting control - [ & ] ...until the snap is positioned as wanted. (Keypresses cycle-thru all the possible snap-to scenarios until user finds one that user likes). Any thoughts about activating entire keyboard?
  6. Hi guys! Selected mesh needs an elevator system, both upward and downward, eh? U/D keys. The amount of Y-axis ground/selectedMesh change... could be based upon the height of the nearest non-dragged mesh (the snap-to destination mesh) (actually, its extendsTo or similar). I've only had a half-cup of coffee so far, so I could be blurry, still. More blurry than normal.
  7. It seems to work for me... at least somewhat. Firefox ESR. Earlier, I thought it needed alt-dragging, but I think I was wrong about that. Standard-click any mesh to turn it yellow. It must be yellow before you click again and start dragging. When the dragged item is in a position where it can be dropped, a green proxy box will activate at the snap-to position. Then release the dragging, and the selected box will move to green proxy box position... and stay yellow/selected (green proxy box will disappear). Also, I think it can only be snapped-to a position where it is not overlapping another box. Any errors at the console, guys?
  8. Wingnut

    Bezier Surface I had some momentary thoughts about warped fun-house mirrors, distorting reflections in funny ways. But, that BABYLON.Plane in line 195... refuses to bend. Nice feature/meshmaker, JohnK! Thank you!!!
  9. Wingnut

    Leaderboard sample

    Hi guys. I've been playing... Pretty complicated and rigid... using a row-stackPanel swap-in/out system... to simulate scrolling. MouseWheel scroller is being grumpy yet, but slider works ok. Dialog height and row rectangle height... super sensitive... not easily scalable (currently frozen to 6 items in-view). *shrug* It's something to mess-with. Other/better coders might advance it. I just finished the sort-by-score section. The database (array-of-objects) is built in lines 158-169, with the sorter happening next. I'm not sure if I will continue to advance this or not. The lawn needs mowing, and I need lunch. Are those good excuses to give my brain a break? A nod/wink/hi5 to @The Leftover. Although the sort code is borrowed from web-folk, I thought about some things he's been teaching me... when I entered line 174. He's been teaching me about calling a function with a function as the arg/param. It still feels retarded, but, I'm workin' it into my brain... slowly. A dynamic parameter. weird. Update: PG now at 46. I had the dialog set for proper height... when I had my dev tools open. When I closed my dev tools pane, dialog white space went bad. Better now. I THINK I used to know how to set a GUI control so that it didn't "squish" when I opened my dev tools panel . But, I forgot how to do that. Note: How many of you had a flashback to "Leaderboard Golf" when you read this thread title? :) Twas a popular series, long ago.
  10. *nod* There's a grid-drag-snap. (yawn) Also... umm... *shrug* Party on!
  11. heh. Yeah, I actually INVENTED the stuck-together-when-collide-too-fast feature. I've had some lame ideas... about how to fix this. Mainly, reverse gear. We could record ALL the "previousPositions" that happen during the drag. Then, upon collision, repeatedly check intersectsMesh()... and keep reversing the drag... until intersectsMesh goes false. Pain in the butt, but might work. The real fix... is at core level... preventing the overlap/stuck condition from happening. Might be a good team project, if anyone can manage to get a team together. Should we go core-diving and see what we can screw-up? As you know by reading forum archives, this has been a long-time issue. As you might ALSO know, the BJS built-in ellipsoid collision system... was not designed/intended for mesh-to-mesh collision... only for camera-to-mesh. *shrug* We won't let that keep us from hacking/experimenting, though, right? :)
  12. Wingnut

    Reshape SpotLight with a texture

    Would you feel like editing your first post and adding 'projector' tag, and maybe some other tags (theater, movie, gel, etc)? I can do it, too, if you like/approve. We want to be able to find this... with forum searches. I bookmarked it locally... so hard that i broke my mouse button.
  13. Wingnut

    Reshape SpotLight with a texture

    THAT ... is VERY cool! haha. Truly tasty mad-scientist stuff. Time to open a VR movie theater, because kjue has built the projectors.
  14. Wingnut

    Reshape SpotLight with a texture

    Hey, thanks, Kjue. I was wondering about that. Good info... about the shadowMap renderLists not being layerMask-sensitive.
  15. Wingnut

    Rotation physicsImpostor physicsBody bug

    Thanks Raanan! From the PR... // set the quaternion, ignoring the previously defined (euler) rotation impostor.physicsBody.resetQuaternion(globalQuaternion); // update with delta 0, so the body will receive the new rotation impostor.physicsBody.updatePosition(0); That's cool! Last line... sync the shape with the impostor... with a 0 timeStep (do it now). Perhaps the Oimo author SHOULD have made two update functions, one for position and one for orientation [rotation]. Instead, the Oimo updatePosition() func syncs/updates both position AND orientation. Sweet fix and good learning opportunity for all of us. Well done! Raanan gets free drinks for the next 24 hours!
  16. Wingnut

    Reshape SpotLight with a texture

    Hi guys. Wingnut goofin' around. This requires shadowGenerator. I can't make transplane2.visibility = 0, though. Not so far. Essentially, I'm shining light thru an inverse cookie-cutter. Kludge. This is probably not helpful, but I thought I should show it. Note: This PG was contributed-to... by many people. Line 44 - mat2trans.anisotropicFilteringLevel = 0; not understood by me. Future users may wish to investigate its use/need.
  17. Cool wall-generator code, Thamali! You are driving BJS/JS quite well... considering you are "new". Thx for providing that useful playground demo. @dbawel... squares/rectangles... I could MAYBE help-with. That "indented corner" probably makes this challenge too difficult for my simpleton brain. But, I will keep thinking and looking thru my list of playgrounds. I have a strange personal issue (what's new?). It seems my brain can't code anything... until them soccer kids are removed from that cave. hmm. Sorry to hear about your internet issues, DB. First time I have ever heard about heat causing internet problems. Sucks, but interesting. I bet the ISP(s) is/are getting tons of complaints.
  18. Yep, well said. I hope I never implied that you were a bad person or that you had done evil. I think you are a good person. I only had/have issue with the "sound" and "feel" of that first reply. It feels "hope killing". You know... rock in trail, idea dead, no hope for work-arounds, end of discussion, cope and deal. You know about hippies like me. We hug everything, even the technically deluded/confused.
  19. Hi gang! Newbie Wingnut here again, hope everyone is well. Here begins a massive "chronicle" of "Stupid Stuff That Wingnut Talks About". It will get big and stale, but what the heck. I goof around in 3dMax a bit, and I've seen their massive 'library' of procedural primitives/shapes, and one would think that BJS is destined to have every one of those, plus a few hundred more. Thoughts, anyone? Thanks!
  20. *nod* You are right. Thx. Other emotions could be affecting me. The nation I am part-of... is not being very nice to others, right now. Not sure. I don't bitch about things I perceive as unkind... very often. But the first reply (to a new user) is rather important... to me, to the new user, to the image of the forum. I cannot assume or expect that the first reply is similarly important for other forum helpers. I can only hope.
  21. Sorry Arte, and thx for being a supporter of my concern, recently. I took a stance on something I believe-in, and it blew up in my face. A fruitless effort on my part. It makes me sad. Sometimes I don't understand some things. I hope you're doin' ok. :)
  22. Wingnut

    Rotation physicsImpostor physicsBody bug

    Hi. Does anyone have any (change-of) status reports, on this? Get the BJSOIMO issue first? Work-in-progress? (thx) Update: Not sure if helpful, but I assembled a playground with ALL THREE applicable physics classes... into the playground code, gathered from the latest BJS distribution. All three are active in this PG. PhysicsEngine, physicsImpostor, and OimoJSPlugin classes... all on-board, ready for experimenting. All are VERY fresh and default. Some reports are going to console, reporting the problem. An unwanted difference in signs... for some of the values in the last two reported lines. The problem: After physicsImpostor is added to mesh (lines 1201+ area)... ...physics_mesh.rotationQuaternion (a BJS Quat class) does NOT match physics_mesh.physicsImpostor.physicsBody.orientation (an OimoJS Quat class, likely) It should, I suspect. The act-of adding the physics impostor.... rotated the mesh. It should not do that. Not only boxImpostor. Same problem if/when sphereImpostor is chosen. Wingnut-tested. I think issue might be in lines 126-130 area, or nearby. Line 126 is not MY comment. Another coder, likely rahhhh-nin Lines 187 and 189 are points of interest, too. All assistance welcome, all ideas welcome, all play welcome.
  23. Hi thamali, welcome to the forum. First, I'm going to assume that "footprint" is almost the same as "building foundation". Footprint is a good term, but might be confusing to some readers. Umm... you need to "sum" the bounding-areas of all the walls. I think there is a function nearby... which does that. I'm searching. Help needed from others, please. (I will keep looking, though.) I think, if you meshMerge all the wall meshes... then you can do... var myTapeMeasure = mergedMesh.getBoundingInfo().boundingBox; Then examine the properties on your tape measure object, and you will see some useful values. Stay tuned... better answers are nearby. Does anyone have/know-of a function that ... sums the bounding-volumes of an array of positioned mesh? Thx! Post url/info here, please.
  24. Interesting idea! The only problem... a plane-button won't flow-in with standard GUI controls... things like containers and stackpanels. This plane-button would NOT be a gui control at all, right? It might need to be parented to the camera, so it can stay in a persistent location on the screen. Certainly possible, though. Put an actionManager on the plane, and boom, it's a button. Full power from BJS StandardMaterial... including setting diffuseCOLOR and emissiveColor WHILE a diffuseTexture (with transparent background?) is also in-place. AND, abilities to overlay other colored, partially-transparent planes, as Arte mentioned, but they will need to be set isPickable = false... to keep from blocking clicks targeted to the actionManager on the underlying plane-button). Sort of a "texture color-tinting system", eh? *nod* I think I'm on your frequency, @Arte. Interesting.
  25. HI DR! Yeah, the universal cam, and the arcRotate cam... utilize their targets differently, don't they? A possible good first step... might be... when you switch to arcRotate cam, un-parent all cams (if any are parented), and then parent the universal cam to the arcRotate cam. And when you switch to universal cam, un-parent all cams, and then set arcRotateCam.parent = universalCam. You may need to do arcCam.position = universalCam.position (or opposite) just AFTER un-parenting all cams, and before doing a "fresh" parenting action. Not sure. This MIGHT keep the cam positions the same, no matter which is being used. BUT, this will not set the target in the same direction... for both cams. You are wishing for a "seamless" (un-noticeable) change... when switching-to EITHER type of cam, yes? Not an easy challenge, but interesting. I have never tried this, and I will give this some thought. Setting the arcRotate target DISTANCE-FROM-CAM (the camera's radius)... after switching to arcRotateCam... might be a difficult thing. There IS a possibility... to modify a universal camera to make it ACT like an arcRotateCam. That way, you never really switch cams. Instead, you turn ON/OFF the "universalCamActsLikeAnArcRotateCam" -mode. This idea might not be easy to do. Perhaps other forum people will have ideas and tests, too. Stay tuned.