Loading...
> >

Endlessly ongoing... playing cards/casino chips/coins/banknotes scatter... hiatus time (still)

24

Comments

  • 3dcheapskate3dcheapskate Posts: 2,719
    edited December 2019

    Thanks.The picture I started creating this stuff (cards and casino chips) for is now complete (here's where the picture was originally posted)...

    ...and actually having to use it, I've now got a much clearer idea of what this stuff should be like. A single prop for a bunch of chips (around 50) with some x/y/z scatter morphs and some randomish rotation morphs, but also an individual chip prop - necessary for the two foreground chips. Similar for the deck of cards.

    I've attached the texture images that I used for the chips. I created the basic design in GIMP using solid-colour layers with masks, which makes it easy to change colours. I'll probably do a few different chip designs based on googled photos (as I've said before this set is loosely based on the Casino De Montreal)and include the XCF/PSD.
    (Edit (16.45 UTC Mon 23 Dec): I've added a screenshot tothis post,showing the relevant bits of my GIMP workspace with the masked solid colour layers just in case what I said didn't make sense. I also noticed that oneor two of the chips I've seenphotos of have edge markings that I can't do with the simple top view 'Project from view' UV mappimng I've used for the edges)
    The five-in-one image was for the 50-chips-thrown-in-the-air prop (the prop has 10 chips of each type - I still reckon that for a chucked-in-the-air set of chips the ratio of $1:$5:$25:$100:$500 isn't that important. I made some mistakes in the model, mapping, and image which I need to fix, but didn'tbother me for theparticular render I was doing.
    The single chip image is for the single chip prop. So the texture image can just be swapped for another instead of fiddling with U/V offsets (which would be required if I used the 5-in-1 image)

    My plan for the chips is to use (I think):
    - a simple 24 vertex cylinder based mesh for the 50 chip prop. Good enough for anything that's not really a focal point of the image. Poser smoothing & DS/Poser inbuilt SubD won't work nicely with this.
    - a 32 vertex cylinder with 4 control edges for the single chip prop. Good forforeground, centre of attention. This mesh works okay with Poser smoothing, and with DS/Poser inbuilt SubD, should you want to use those.

     

     

    SixAces for 3b OBJ RED.jpg
    1024 x 1024 - 176K
    $100.png
    1024 x 1024 - 272K
    xcf example.jpg
    1110 x 704 - 149K
    minor problem.jpg
    1398 x 226 - 88K
    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited April 2020

    Decided to forget about the awkward chips with edge markings mentioned in the previous post and stick with my 'project from top view' UV mapping.

    ~ ~ ~

    Anyway, here's another chip texture...

    ...based on somwthing google chucked at me...

    (Those 'real' ones look rather plastic to me. By the way - I know I've got antialiasing problems on my white bits, and that my text isn't bendy the way it should be !)

    ~ ~ ~

    Yes, I am slow. I've been distracted. I'm very easily...what was that ?

    ~ ~ ~ ~

    Edit: Completed the extremely monotonous task of creating a first set of textures for a full set of chips. But not those ones. Different ones.
    As hoped I used masked layers in GIMP for the stuff that only changes colour. The difficult bit was making sure that everything was antialiased correctly - I've got that mostly resolved now, but I'm still getting the odd problem.
    Creating each texture was then just a case of changing the value text, choosing the appropriate colours, and flood-filling the masked layers.

    ...based on another image that google threw at me...

    (Yes, they're the ones with the little mark on the edge that I decided toignore. Do they look rather plastic to you?)

    And here's a little render...

    ...the mesh I used for the render is a 32 vertex cylinder with control edges, with one level of in-Poser SubD applied (I've removed the subDfor the top chip for the screengrab below)

    I've also attached one of the JPG textures...

    The 'project-from-top-view' mapping maps the chip to a 1024pixel diameter circle. The dark grey area at the edge is outside a 1044 (yes, 1024+10+10) diameter circle, and the outer edge of the 1024 pixel diameter texture is extended radially to fill this (just to be sure that the edge/side of the chip picks up the correct colour)

    Comments and suggestions welcome. Even ones that would require me to go back to square one and redo everything !

    Macau50.jpg
    663 x 366 - 39K
    MC Set Textures.jpg
    1122 x 852 - 199K
    googlevomit.jpg
    502 x 421 - 54K
    Crap.jpg
    767 x 767 - 76K
    macau-poker-chips.jpg
    650 x 650 - 142K
    CC V2.0 MC $10000.jpg
    1024 x 1024 - 147K
    crap2.jpg
    681 x 750 - 45K
    Post edited by 3dcheapskate on
  • Faeryl WomynFaeryl Womyn Posts: 3,632

    Looking good, anticipating the completed project.

  • 3dcheapskate3dcheapskate Posts: 2,719
    edited March 2021

    Thanks. smiley

    The main delay (if you discount laziness, apathy, et al) is that I'm trying to set up scatter morphs that allow me to put multiple scatter props (e.g. two packs of cards, plus two sets of 50 casino chips) at the same location, and ensure that none of the scatters intersect. I've had several bad ideas of how best to do this, and several false starts - I've given up a few times. But I keep coming back with new, stupid ideas of how to do this. I think the latest one is probably the best so far, and I'm still plugging away at it.

    It's basically a spreadsheet into which I've entering pseudo-random x/y/z offsets (all integers 1-50 ) for 200 items.I let the spreadsheet calculate the distance between each pair of items (i.e. SQRT((Xa-Xb)^2 + (Ya-Yb)^2 + (Za-Zb)^2), and indicate which are the closest. Then I manually twiddle some of the x/y/z values to try and increase the minimum separattion - I've almost got it up to 5.0...

    I've had to use the '=indirect(address(4,5))' type formulae - that had me banging my head against the air* for a while

     

    P.S. My base models for the decks of cards and bunch of chips have all 52/54 50 items at exactly the same location,so I didn't bother pseudo-randomizing the x offset,just used 1,2,3,4...49,50.
    That may have been a mistake.

    ~ ~ ~

    Edit (after FaerylWomyn's subsequent post):

    First set of  results from my spreadsheet (this is mostly for my own backupin case I lose the values!
    I decided to increase the number of items (casino chips, cards) per prop/set from 50 to 54 - 50 was arbitrary, whereas 54 is the number of items in a standard deck of playing cards (4 suits, 13 cards per suit, plus 2 jokers).
    I also decided to make the X offset values random too, rather than sequential.

    The base mesh for a prop (or 'set') will consist of 54 'items' all located at exactly the same position (centred at the world origin) - so a deck of cards will look like a single card, and a bunch of chips will look like a single chip.
    The 'items' will each fit inside a 0.5 unit radius sphere centred on the origin for any rotation.
    I plan to have a set of 'scatter the 54 items in a pseudo-random manner along the X-axis' type morphs. By pseudo random I mean sort of linear-congruential-generator type random, but done manually. What on earth do I mean by that ? After applying the morph at 100% the items (which were all co-located) will be spaced at exactly 1 unit intervals along the axis, but not in sequential order (look at the tables below - that's what I mean)
    Thus with a single morph applied at 100% none of the 54 items in a single prop will intersect.

    Now...

    I plan to have separate 'Scatter X (A), Scatter Y (A), Scatter Z (A), Scatter X (B) ... Scatter Z (D) morphs for each prop (i.e. prop 1 = set of 54 cards, prop 2 = set of 54 casino hips, prop 3 = set of 54 coins, prop 4 =set of 54 banknotes, etc). There'll also be master parameters 'Scatter (Set A)', 'Scatter (Set B)' etc which simply set the appropriate X/Y/Z scatter morphs to the same value.
    My upgraded spreadsheet plus some manual tinkering has now given me a first set of results that have (assuming everything in the spreadsheet is working correctly, and that I haven't overlooked something important. So a rather big assumption!) a minimum separation of 5 units between any pair out of the 216 items, when four props are loaded and a different 'Scatter(Set #)' morph is applied at 100% to each prop.

    Which should mean that Icould apply 'Scatter (Set A)' at 20% to one prop, 'Scatter (Set B)' at 20% to another, 'Scatter (Set C)' at 20% to another, and 'Scatter (Set D)' at 20% to another, without any pair out of the whole 216 intersecting.

    So that's the theory... indecision

    And here are the values...

    Set A, Item: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
    X Offset: 20 7 40 12 46 16 27 4 6 37 30 19 34 41 5 39 51 15 48 21 11 28 31 54 24 1 45 17 36 42 3 49 43 14 52 9 38 29 22 50 32 8 23 47 25 33 53 13 26 10 2 44 18 35
    Y Offset: 22 34 9 44 33 12 53 48 21 54 5 32 16 28 10 31 38 3 41 51 45 23 47 8 25 40 18 46 4 35 52 49 11 26 30 6 43 24 36 7 19 39 27 13 50 14 37 2 29 15 17 42 20 1
    Z Offset: 34 13 30 3 21 9 41 19 28 6 25 18 40 8 33 17 1 29 53 44 10 50 14 45 26 31 36 4 52 42 15 2 49 24 12 37 54 11 46 20 51 16 35 43 5 22 47 38 23 27 32 7 48 39
                                                                                                                 
    Set B, Item: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
    X Offset: 21 5 29 36 15 28 3 13 43 1 37 20 30 9 25 4 42 14 27 11 38 49 31 6 54 24 47 2 48 19 39 44 10 50 32 53 16 26 40 7 33 46 18 51 12 52 34 22 41 35 17 8 45 23
    Y Offset: 48 9 33 37 41 25 52 18 34 8 45 16 11 50 4 19 17 42 3 24 7 27 12 10 54 21 47 5 40 26 32 49 20 28 39 31 53 2 15 36 1 30 6 44 14 23 38 46 22 51 35 29 13 43
    Z Offset: 34 3 52 45 18 35 47 38 30 54 5 23 11 19 36 15 31 39 8 49 1 53 6 12 32 44 26 22 24 16 46 10 2 40 21 43 17 13 48 37 33 29 51 27 20 42 7 4 28 14 41 50 25 9
                                                                                                                 
    Set C, Item: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
    X Offset: 45 16 39 50 10 17 35 7 23 11 28 15 5 43 26 38 36 8 21 33 6 25 46 53 19 41 14 4 51 27 18 44 34 12 32 42 20 9 54 29 1 24 48 13 47 30 22 2 52 31 3 40 49 37
    Y Offset: 2 27 38 14 34 6 21 1 47 26 43 37 53 33 46 5 25 17 39 42 30 28 8 13 29 23 3 11 15 49 7 52 44 50 20 40 16 4 36 9 22 31 41 48 32 54 10 18 12 45 51 19 35 24
    Z Offset: 23 31 52 8 18 20 4 14 49 30 3 36 24 9 13 44 26 39 12 48 17 7 37 45 6 15 54 25 29 22 2 46 32 51 10 42 38 1 27 19 53 40 33 16 47 28 50 5 34 41 35 21 11 43
                                                                                                                 
    Set D, Item: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
    X Offset: 10 9 42 29 18 11 3 30 25 51 35 40 54 44 16 53 50 20 17 36 34 24 41 8 46 28 6 39 23 19 1 12 52 26 49 45 4 31 15 38 22 48 21 47 27 32 43 13 37 2 7 33 5 14
    Y Offset: 20 18 53 39 1 49 23 29 36 25 42 30 17 2 38 45 4 52 37 15 50 14 33 7 46 5 21 44 3 27 19 34 16 10 8 48 6 35 22 40 11 26 24 54 9 51 31 13 47 32 12 41 28 43
    Z Offset: 43 33 36 17 25 44 6 18 34 4 26 40 8 3 49 15 47 50 1 28 9 19 35 45 48 2 54 37 41 39 24 52 13 23 53 16 30 11 31 10 27 22 5 21 7 42 14 38 20 29 51 12 32 46

     

    So now all I need to do is modify my base meshes, create 12 morph targets, load 'em all up, and try 'em out.
    What do you think the oddsare that it'll work ? ...

     

    *hurts less than a wall or desk, and if you're playing some heavy metal people don't bat an eyelid...

    Note to self since I always forget what the pseudo-random number generation is called - Linear Congruential Generator, Xn+1 = ( (a* Xn) + c) mod m , produces X values in range 0 to (m-1). Trial and error to get parameters that have the desired repeat cycle, i.e. m)

    scatter.jpg
    1086 x 747 - 326K
    Post edited by 3dcheapskate on
  • Faeryl WomynFaeryl Womyn Posts: 3,632
    edited January 2020

    The only method I know of that has been used in the past and having a section for each of the four sides but not close enough to touch the model, and then individual or small groups that can be carefully placed.  Least that is how I have worked with other sets in the past. It was time consuming, especially if there were no small groups and had to make them and save as such, but it worked.

    Now for animation that is a different take all together and partical based physics is the only thing I can think of that might work.

     

    *I have banged my head against walls, keyboards and the like, but always in my head...lol...I have such a wonderful imagination and it hurts less.

    Post edited by Faeryl Womyn on
  • Silver DolphinSilver Dolphin Posts: 1,608

    Looking good. Great work on casino chips.

  • 3dcheapskate3dcheapskate Posts: 2,719
    edited January 2020

    Thanks.

    However, I've just run into a problem that I'd completely overlooked*...

    The UV-mapping was done in Blender from a top view (i.e. looking at the front of the chip) and using 'Project From View (Bounds)'. Now this would obviously make the back face of the chip a mirror image of the front face (i.e. all text back to front), so I simply flipped the UV mapping of the back face to make it the same as the front face.
    No problem there.
    But I overlooked the potential mismatch between the back face and the edge. It didn't happen with my original solid-colour-with-six-white-bits (LtoR symmetrical) when I actually checked the back face, and I obviously didn't check with my red-with-three-yellow-bits (not quite LtoR symmetrical due to a slight manual misalignment) one.

    So what I'd overlooked was that any chip texture must have its outer edge symmetrical about a vertical axis on the texture map. No big problem - simply rotating the outer parts of the Casino Dubbioso texture maps through 30° should resolve it. But I wish I'd spotted that earlier - more unnecessary work ! :FACEPALM:

     

    *a common occurance, which I'm well acquainted with !

     

    ~ ~ ~

     

    Edit: While I'm here - that patterned dark-brown ring on the Dubbioso chips (and the white/grey patterned label plus gold-patterned ring on the previous chip) has a sort of shiny,reflective,hologram-ish type look to it on the photos, and I'm wondering about the best way to reproduce that in DAZ Studio(3Delight/iRay) and Poser (Firefly/Superfly). Any ideas ?

    oh dear !.jpg
    762 x 761 - 75K
    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    Back to the non-intersecting-pseudo-random-scatter stuff - it works !

    surprise
    My base prop mesh is a chip created from a 24 sector cylinder (1 unit diameter, thickness 1/11.47) with triangle fans capping each end (i.e. 50 vertices,72 faces) duplicated 53 times, so 54 chips total per prop.
    All 54 chips are colocated at the origin.
    I created four sets of X, Y, and Z offset morphs using the values for sets A, B, C, and D from the table in the earlier post. I also created a single random rotations morph for test purposes.

    Loading a prop with just one set of X/Y/Z (and the rotation) set to 100% I get what I usually expect from a 'scatter' morph/control - quite widely separated objects.Which is good - I believe this is what's normally wanted from a scatter.

    But I'm not normal- I want a tight scatter with no intersections. So I reduced the X/Y/Z from 100% to 20% (leaving the rotation at 100%, the only value at which it works correctly due to the linear morph doing a rotationthing). It looks okay to me, and theredoesn't appear to be any intersection (difficult to tell from one render, but I was checking while spinning it around, so I'm fairly confident.

     

    Render 6.jpg
    767 x 767 - 9K
    Render 5.jpg
    767 x 767 - 42K
    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited April 2020

    ...and then the acid test. Load three more of the 54-chip props, and set a different X/Y/Z set to 0.2 for each (use same rotation at 100%).
    I also used a different chip texture for each prop.
    So here we have 216 chips in a small space...

    Even more difficult to tell from a single render, but I checked while spinning things aroud and it all looks good to me.

    But do I really want to go through all this morph target creation again for a pack of cards and a wad of bank notes ? (Edit: I think the answer's going to me "No")

    Note: With four props you begin to see the cubic domain in which all the pseudo-randomness is taking place - specifically the last image, which I think was a top view.

    Render 1.jpg
    767 x 767 - 91K
    Render 2.jpg
    767 x 767 - 99K
    Render 3.jpg
    767 x 767 - 93K
    Render 4.jpg
    767 x 767 - 122K
    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    Anyway, I think I'm close* to being ready to release a first batch of casino chips. Cards will have to wait. Banknotes can wait longer.


    Notes to self for things I need to do first (comments welcome):
    - Only one complete set of textures at present (Casino Dubbioso) and I need to rotate the outer part of each texture.
    - Haven't decided on the UV mapping. I now think my original 5 chips on one jpg is a bad idea. I prefer 1 jpg for 1 chip. But for the 54-chip prop do I want all 54 chips the same,or 27 of one/27 another, or 13/14/13/14, or what?
    - The chips are 1 unit diameter in Blender/OBJ, so when Iimport into Poser they're rather large. Do I want to scale them down first ?
    - The X/Y/Z offsets are currently +1 to +54. I think I want thespread tobe around the centre,so I should change them to be -26.5 to +26.5
    - Create a few non-morphing props for piles of chips on a table. Probably use Blender physics if I can remember how.
    - Stacks of chips ? All same value ?
    - Remember to put this all in a 3DCheapskate/CasinoChips/Beta folder, to avoid the problem I've got with all the Poser books.
    - Poser or DAZ Studio first ? I usually do Poser PP2's as they can be imported into DS,but then usually don'tbother with a DUF version.

     

    *relatively** speaking.

    **My closest relative is around 6000 miles away

    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    Google found some more Casino de Montreal chips for me and I've used them as a basis for Casino Douteux...

    These are the only values I've found images of (except for a $2.50 which I'll be doing a version of as Ilike the colours). The colours were mostly picked from images with GIMPs eyedropper thingy. I've attached a screenshot of thumbnails of the reference photos for comparison.

    Question (for Casino-goers): Would the different value chips for a specific casino at a specific point in time all have the same label design (as I've done for the Casino Douteux - except the 50¢ one),or would you see a mix of different ones, like in the reference photos?

     

    I think I also need a bump map since the central label is distinctly depressed (not much, but it's obvious),and the 50 cent chip has six four-leaf clovers embossed around it

    CasinoDouteux.jpg
    871 x 576 - 70K
    Actual CDM Photos.jpg
    831 x 575 - 79K
    Post edited by 3dcheapskate on
  • Faeryl WomynFaeryl Womyn Posts: 3,632

    Looking really great so far.

    I have one question mainly because I have an older computer and don't have nvidia...when loading that many chips for your scene, do you notice any impact on the computer or software, like freezing or slowing down or anything along those lines?

    Oh and I don't worry about anything being Iray as I have both the free and paid versions of the conversions and they work just fine for me.

  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    No,I can't say that I've noticed any impact at all. They're not resource hungry. One of my main goals was to keep this low-poly, so 50 vertices per chip, times 54 chips per prop, times 4 props = 10,800 vertices total for 216 chips. I believe that a single DAZ/Poser figure is about double that (I know there are several threads with the vertex counts of various DAZ figures, like this Genesis Evolution from 1 to 3 one from 2015 which gives Gen(esis) 1 = 19296 vertices, Gen 2 = 21556, Gen 3= 17418 vertices)

    I'm using an 8/9-year old not-quite-bottom-of-the-range Dell Inspiron N5110 laptop - 2.3GHz quad-core with 4GB RAM.
    How does that mean machine compare to yours ? wink

    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    I've just (re)discovered* how to do a rigid body physics simulation in Blender, so I dropped astack of 54 chips to see what would happen. They're all mapped to a single texture - as I said earlier I still haven't decided on how to do the UV mapping.I'm not keen on the five-chips-on-one-image that I originally used, so I'm leaning towards five separate materials,each chip being assigned to one material. That makes it easier for a user to change one or two of the textures, and I can use the same textures as the hi-poly chip (32 side cylinder with control edges, 192 vertices - I think I live in a different dimension from most people)

    Edit: Also probably want to have some stacks of chips

    *I'll add some notes (reminders for myself) tomorrow done.

    Originally discovered how to use Blender rigid body physics here (SM forum)

    1) Blender Tutorial - Quick Rigid Body Fun YouTube videoby BlenderGuru gives me all the basics. The main bits I forgot:
    - Edit mode, select all (A), separate (P) by loose parts to split the 54 chip object into54 separate objects.
    - Use a stretched squashed box for the ground
    - Object mode, select all objects, Ctl-Shft-Alt-C to set all their origins to centres of volume
    - Physics tab, rigid body,set chips active and box passive
    - Tool Shelf > Physics tab > Copy From Active
    - (don't think this was in the video) toincrease simulation length , as well as increasing end frame on timeline don't forget Properties pane > Scene tab > Rigid Body Cache and set end frame there too.
    - Alt-A tocalculate simulation
    - When the chips failto settle down and keep jiggling increase their mass tenfold and red

    2) The screencap video answerby GiantCowFilms to How to bake rigid body physics frame in blender render? gave methe extra bit for exporting the result as an OBJ(Object mode, select all the chips. Tool Shelf, Physics tab, hit Bake To Keyframes, set both start and end frame to the frame number Iwant to export (step=1) and hit OK. Now if you slide the green marker back and forth on the time line the chips don't move. N.B. this obviously destroys the physics simulation.I need to start a new for a new simulation.
    Go to that frame with the green slider and hit Alt-I (in object mode with all chips selected) to delete the keyframe on all the objects.

    Select all the chipsand Join into one object (it took a while before it would let me do this? That was because I got the "active object is not a selected mesh" warning, so make sure all chips are selected and one is the active object)

     

    Render 1.jpg
    767 x 767 - 58K
    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    Just a quick proof of principle for a simple stack of chips - 96 vertices. The thickness of the intermediate chips is wrong (the central texture was done manually, quickly, and sloppily). The top and bottom map as per a single lo-res chip. There's also a simple bump map for the central texture to make it look more like astack of discsrather than a solid cylinder.

    And if I added a morph to raise all the vertices for the top chip, then you could use it as a stack of however many chips you wanted simply by scaling and tiling the central texture vertically to match the morph.

    Yes, that works. smiley
    And I don't really notice the texture repeating four times up the side of the stack,which it does since my scale factor was 0.25.
    And I recall from my books project that in a DS prop I can slave a Surfaces value to a Parameters value, so this can probably be automated, assuming it applies to morph sliders too. In a Poser prop I'd have to resort to Parmatic.

    Of course, for a less tidy stack with chips offset from centre it's probably best/easiest to use separate chip meshes and a constrained Blender physics simulation. But I like this idea.And I should be able to get the vertex count down further to juse just 50 vertices, i.e. a simple 24 vertex cylinder, exactly the same as the lo-res chip.

    CheatStack.jpg
    799 x 563 - 63K
    TallerStack.jpg
    504 x 777 - 70K
    Post edited by 3dcheapskate on
  • Faeryl WomynFaeryl Womyn Posts: 3,632
    edited February 2020

    That's a big stack lol

    Post edited by Faeryl Womyn on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    I think I'd seen some photos of the chip racks trays they use at casinos, so just imagine a few of those stacks lying horizontally.

    I've got a bit bogged down with the stack, mainly due to the bump at the bottom edge of the top chip (and top edge of the bottom chip). Simple to resolve when the morph is set to 100%, and V-scaling for the side texture is set to 1.0, giving 10 (or 11 - still not decided on that) intermediate chips. But at most other morph values (don't forget that the stack height must only change in steps matching the chip thickness, and that the V-scale must be modified in concert with the morph) the problem reappears. A picture wouldprobably explain this better, but I don't want to get bogged down creating a picture to explain how I've got bogged down...

    wink So I've put that to one side to ferment for a while...

    Anyway, I now have my first pre-release of the 54 casino chips flung in the air prop almost* ready. It's currently in Poser prop version only, but it imports (almost**) without a problem into DAZ Studio 4.11. I've also created $2.50, $10, $50 and $500 Casino Douteux textures, the last three using arbitrary (and not particularly good) colour choices on my part. Four different textures are used by the prop, so you can mix and match the chip values you want.

    Anyway, iray and 3delight renders attached - 216 chips as before. Hopefully not long*** now before I upload something...

    * ''almost' is nearer than my previous 'close'

    **except that DAZ Studio doesn't like the ¢ symbol in two of the texture file names,which means that I have tomanually locate the file every timeI load it (Poser Pro 2014 doesn't have any problem with it). Maybe I'll change the 50¢ in the filename to $0.50 or 50c ...

    ***'not long' is relative too

    iray.jpg
    1000 x 1000 - 655K
    3delight.jpg
    1000 x 1000 - 465K
    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited April 2020

    What about the default size of the chip prop ? I created the chips in Blender with a 1 unit diameter, exported OBJs, and imported those into Poser with no scaling - so each chip has an 8' 7.2" diameter.
    Real chips are, according to reivax, 39mm diameter. Not 2621.28mm. So I'd need to scale them down to around 1.5% (giving 39.32mm).

    So I'm wondering what's the best way/place to do that - N.B. not wondering in a technical way, but in a 'what's best for a user' way. A previous freebie, the Goblin Enamel was made to be the correct size when compared to Poser/DAZ figures when it was loaded with all scaling at 100%. But when I used it I found that I always wanted it much larger, a similar size to the Poser/DAZ figures. I'm wondering whether maybe 39cm diameter may be good default size. Scale down by a factor of ten to 10% to get correctly sized chips compared to Poser/DAZ figures, scale up by a factor of 5 to 500% toget chips about the same size as a Poser/DAZ figure.

    On the goblin enamel this scaling also meant that the micropolygon displacement used to create the old/used textures was far, far too small.
    I'm not using displacement for these chips, but I am using bump. And that'll have a similar size/scaling factor which needs to be taken into account. There's also another odd effect I'm getting with the bump - I've only looked at it in Poser so far.
    Of course, it may be better to use normal maps...

    For your delight and delectation, the attached images show a single 54 chip almost-pre-release prop loaded at default settings. But since all 54 chips occupy exactly the same space that confuses the renderer, soI've applied small X/Y/Z morphs  (1% to 5%) to move them slightly apart, but they still intersect (values of at least 20% to100%, depending on whether one, two or three of the translation morphs are applied, are needed to avoid intersection)

    Your* thoughts would be appreciated on this.

    *i.e.anybody reading this

    defaultload.jpg
    1000 x 1000 - 411K
    morphabit.jpg
    1000 x 1000 - 288K
    morphabitmore.jpg
    1000 x 1000 - 247K
    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    Just noticed a minor UV-mapping problem in that previous image. Simple solution I think - just shrink the UV-mapping by a couple of pixels all round so that it doesn't quite reach the edge of the UV map.

    line.jpg
    1026 x 899 - 111K
    Post edited by 3dcheapskate on
  • MarcCCTxMarcCCTx Posts: 924
    edited February 2020

    Try dropping the chips into a box (in Blender) if you want a tidier pile. You might have to remove some edge ones before bringing them back into DS.

    image

    630 books dumped into a box roughly the width of the aisle. The actual shot I used was from a much lower angle so you won't see the books do go into the shelfspace.

    Bookvalanche.jpg
    800 x 450 - 348K
    Post edited by MarcCCTx on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    smileyyes That's part of the plan ! - see this post from just over a week ago

    P.S. Using Blender's rigid body physics definitely has lots of promise for making rows of books with a more realistic look to them (this link to the original post on the Smith Micro forums - https://forum.smithmicro.com/post/64226 - should work until those forums vanish, although it takes a while for the SM forum software to find and display the correct post !), although I haven't got round to releasing any bookrows/stacks created that way - just the normal bookrows/stacks
    Piles of books would be more complicated, because some books should almost certainly fall open...

    P.P.S. I like your picture - I didn't notice the arm for quite a while ! laugh

    Edit: P.P.P.S. Those images in my previous post aren't actually piles of chips. They were just to demonstrate the 54 chips all located at the exact the spot when loaded (first image),moving towards their pseudo-random X/Y/Z positions for the 'chucked-in-the-air' morphs (secondand third images).

    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    Just gave the Blender physics another run...

    Imported the resultant OBJ into Poser Pro 2014 and rendered there in Firefly...

    I think I used the default  'convex hull' on the chips, since I reckoned it should be near enough identical to the actual (convex) mesh. Maybe it isn't, maybe I should try 'mesh' ?Maybeplay with 'collision margin' ?

    physfilter.jpg
    774 x 451 - 38K
    Render 1.jpg
    751 x 557 - 92K
    Render 2.jpg
    751 x 516 - 79K
    Post edited by 3dcheapskate on
  • MilosGulanMilosGulan Posts: 1,959

    That looks very cool :)

  • 3dcheapskate3dcheapskate Posts: 2,719
    edited April 2020

    More fun, this time 3Delight and iRay. The chip stack is working nicely in DAZ Studio/3Delight - you simply set the number of chips using a slider. In DS/iRay the tilingof the intermediate chips doesn't work, but I think that'sbecause I'dneed to do an iRay-specific texture.

    In Poser it's less user-friendly - unless you're  using Parmatic you need to manually set the image tiling as well.

    3dl.jpg
    1000 x 1000 - 671K
    iray.jpg
    1000 x 1000 - 915K
    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    DS Chip-Stack Pre-Release 0.1

    (Feedback and suggestions are very welcome, even negative ones!)

    Pre-Release of the DS version of the chip stack now available for playing with and feedback

    It's attached to this post as a ZIP file - simply unzip to your chosen DS content directory and give it a whirl.

    Notes (may be amended as time goes by):
    - The default materialand materials presets are 3Delight. I haven't looked at iray yetfor this
    - Yes, you could probably do it using a single chip and instancing. But I didn't.
    - the zip has files in /3dcheapskate/TEMPORARY/Casino Chips subfolders under data, presets/materials, props, runtime/textures and scripts. There's also a /3dcheapskate/Generic folder under scripts containing aduplicate of a generic script from my DS books.

    No readme or anything like that, just this picture:

    Things I already know that I need to do with the stack:
    - Poser Parmatic version already exists, but I needto actually test it with parmatic
    - iray mats.
    - shrink UV to 99% so it doesn't quite touch the edges. I might have already done that for this.

    StackPre-ReleaseInfo.jpg
    1920 x 774 - 211K
    zip
    zip
    DS ChipStack Pre-Release 0.1.zip
    388K
    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited February 2020

    Poser 54 Chip-Scatter Adjustable Prop Pre-Release 0.1

    Feedback and suggestions are very welcome, even negative ones!

    Pre-Release of the Poser version of the 54 chip prop now available for playing with and feedback.

    It'sttached to this post as a zip file. Simply unzip to whatever Poser runtime you want.

    Load the prop, which will be found under /Runtime/Libraries/Props/3DCheapskate/TEMPORARY/Casino Chips/ - it will look like a single 8 foot diameter chip, so you may want to scale it down.
    Play with the morphs and see what you think.
    Try loading four props and apply different morphs to each prop
    There are a couple of material presets, which will be found under /Runtime/Libraries/Materials/3DCheapskate/TEMPORARY/Casino Chips/

    DAZ Studio users who understand Poser runtimes can simply load the prop from amapped Poser runtime.
    For some reason the Poser materials (MC6 files) don't seem to work in DS4. I know DS3 couldn't handle MT5s andMC6s, but I thought DS4 could.

    ~ ~ ~ ~

    PP2014 renders/screenshots:
    Four props with the 4 ddifferent mid-air scatters applied

    Four props with the four pile morphs applied N.B. pile A group 2 goes on top of pile A group 1, 3 goes on 2, and 4 goes on 3. 2, 3, and 4 can't be used independently.

     

    I'm already thinking that separate non-morphing props might be better for the piles. With the basics of the Blender rigid-body physics still in my head it shouldbe easy to create a large number of piles of different sizes.

    zip
    zip
    Poser 54 Chip-Scatter Adjustable Prop Pre-Release 0.1.zip
    1M
    midairx4.jpg
    1506 x 842 - 221K
    pilex4.jpg
    1512 x 900 - 206K
    Post edited by 3dcheapskate on
  • 3dcheapskate3dcheapskate Posts: 2,719
    edited April 2020

    Not even tumbleweed...

    ROFL !

    (with the view counters for many threads, including this one, no longer working I have no idea whether anybody's even looking at this thread any more. I wouldn't be surprised if people have lost interest - I know it's nearly 3 months since I started this and there's nothing really to show forit yet.
    But 3 months is nothing in my timescales.
    My projects are like expensive wine - they need to be locked in a dark cellar for a long, long time have to be given adequate time to mature*)

     

    *rather like me really. But seeing that over half a century hasn't done it, I don't think there's much hope of me maturing ! At least I'm not turning to vinegar...

    Post edited by 3dcheapskate on
  • ankromtankromt Posts: 64

    3dcheapskate,

    When I saw these today a couple of ideas popped into my head so I downloaded them. As soon as I my little one gives me a break I will give some feed back(might take a day or two).

    Remember:

    "Never despair, keep pushing on." Sir Thomas Lipton

    Tim

  • Catherine3678abCatherine3678ab Posts: 8,344

    Oh yes we're watching you ;-)

     

  • Catherine3678abCatherine3678ab Posts: 8,344
    edited February 2020

    Okay, so let me get this straight. There's a zip for Daz Studio which only provides a morphing prop, loading to appear as 3. Scattered all over the place are script files and mat presets, etc. The lovely thing with D/S4+ is that we can now put the mat files in with, or in subfolders of, the main figure or prop so everything can be in one place. The mat files do work. The stack morph works.

    Nothing for throwing them all up in the air too???

    oh yes, setting the number of chips just stretches the height morph. Still all one chip.

    Post edited by Catherine3678ab on
Sign In or Register to comment.