Unity Community


Page 18 of 19 FirstFirst ... 816171819 LastLast
Results 341 to 360 of 363

  1. Posts
    370
    Hi,
    This is awesome tool for making fluid types effects. I want to buy but i have some doubt so pleases mind telling me can possible with this tool .I want to make like this effects in 3d.

    https://www.youtube.com/watch?v=C5tOEBmBAHg

    Best wishes,
    play_edu


  2. Posts
    848
    Quote Originally Posted by MPanknin View Post
    Hi and thanks for getting back to me.

    I'll definitely check out the free version but I'd like to ask one more thing in advance. Do Fluvio particles interact with lighting?

    http://assets.digitaltutors.com/dtv_...ender_1040.jpg

    We want to achieve something similar to the above but need the greenish gas/smoke to be fully dynamic. Is this something, Fluvio can be used for?
    Fluvio Standard and Pro include a bunch of shaders, many of which do react to lighting and cast shadows. All editions of Fluvio also allow you to make your own custom shaders.

    Quote Originally Posted by Play_Edu View Post
    Hi,
    This is awesome tool for making fluid types effects. I want to buy but i have some doubt so pleases mind telling me can possible with this tool .I want to make like this effects in 3d.

    https://www.youtube.com/watch?v=C5tOEBmBAHg

    Best wishes,
    play_edu
    Sure! This could be done with a quick color change of a fluid's particle system (using Unity's API for particle systems).


  3. Posts
    232
    Has something changed in Unity since 4.1.5f ?

    I upgraded to 4.3.4 and the collision does not happen when my collider has "trigger" enabled, did that change in Unity particle system in general ?

    I try to make Fluvio collide with a trigger collider, which was happening in Unity 4.1.5f and now does not seem to


  4. Posts
    848
    Quote Originally Posted by nasos_333 View Post
    Has something changed in Unity since 4.1.5f ?

    I upgraded to 4.3.4 and the collision does not happen when my collider has "trigger" enabled, did that change in Unity particle system in general ?

    I try to make Fluvio collide with a trigger collider, which was happening in Unity 4.1.5f and now does not seem to
    That must have changed with particle systems in general - we've never directly supported fluids colliding with triggers, but Shuriken collisions may have worked with them before.


  5. Posts
    232
    Quote Originally Posted by Thinksquirrel View Post
    That must have changed with particle systems in general - we've never directly supported fluids colliding with triggers, but Shuriken collisions may have worked with them before.
    Ok found it after searching all changelog

    After Unity 4.2 they removed collision with trigger

    "Shuriken: Fix collision module such that trigger objects do not cause particle collisions."


  6. Posts
    1
    Got this nasty error when exporting to Windows Phone 8, this happens with every project, which has Fluvio exported, including the example project. I am using the latest Unity version and just downloaded the free Fluvio Asset.
    Could you please look at this problem and perhaps find a solution.

    Error building Player: Exception: Failed to run reference rewriter with command "--target="Temp/StagingArea\Fluvio.Runtime.dll" --framework="C:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\WindowsPhone\v8.0,C :\Program Files (x86)\Microsoft SDKs\Windows Phone\v8.0\Libraries" --platform="C:\Program Files (x86)\Windows Phone Kits\8.0\Windows MetaData\Windows.winmd" --support="Temp/StagingArea/WinRTLegacy.dll" --winmdrefs="" --system=System --dbg=pdb --alt=System.Net;System.Net.Sockets,System.Net;Syste m.Xml.Serialization;System.ComponentModel,System.W indows;System.Threading,mscorlib". [Temp/StagingArea\Fluvio.Runtime.dll]
    Error: method `!!1[] System.Array::ConvertAll<ThinksquirrelSoftware.Flu vio.Plugins.FluidPlugin,T>(!!0[],System.Converter`2<!!0,!!1>)` doesn't exist in target framework. It is referenced from Fluvio.Runtime.dll at System.Collections.Generic.IEnumerable`1<T> ThinksquirrelSoftware.Fluvio.Fluid::GetPlugins().
    Error: type `System.Runtime.Serialization.SerializationInfo` doesn't exist in target framework. It is referenced from Fluvio.Runtime.dll at System.Void ThinksquirrelSoftware.FluvioInternal.AssertExcepti on::.ctor(System.Runtime.Serialization.Serializati onInfo,System.Runtime.Serialization.StreamingConte xt).
    Error: type `System.Runtime.Serialization.SerializationInfo` doesn't exist in target framework. It is referenced from Fluvio.Runtime.dll at System.Void ThinksquirrelSoftware.FluvioInternal.AssertExcepti on::.ctor(System.Runtime.Serialization.Serializati onInfo,System.Runtime.Serialization.StreamingConte xt).
    Error: method `System.Void System.Exception::.ctor(System.Runtime.Serializati on.SerializationInfo,System.Runtime.Serialization. StreamingContext)` doesn't exist in target framework. It is referenced from Fluvio.Runtime.dll at System.Void ThinksquirrelSoftware.FluvioInternal.AssertExcepti on::.ctor(System.Runtime.Serialization.Serializati onInfo,System.Runtime.Serialization.StreamingConte xt).
    Catastrophic failure while running rrw: System.NullReferenceException: Object reference not set to an instance of an object.
    at Unity.ReferenceRewriter.RewriteMethodSpecMemberRef s.Visit(MethodDefinition method)
    at Unity.ReferenceRewriter.MethodDefinitionDispatcher .Dispatch(TypeDefinition type)
    Last edited by estpaul45; 02-21-2014 at 03:04 AM.


  7. Posts
    848
    Hey estpaul45,

    The current version of Fluvio doesn't support Windows 8 platforms. We're working on it - expect the next update to fully support the platform.

    Thanks for the heads up though, we didn't catch this one yet. I've logged it as a TODO.


  8. Posts
    83
    Hi Thinksquirrel,

    I'm currently trialling Fluvio free edition before deciding if it is suitable to purchase a higher version. I am having issues which I'm hoping you could help answer. I'm attempting to have a 3d body of liquid in a container, which could then be manipulated.

    I have a container, which contains the liquid in place as expected and collision is working as intended (if it's not obvious this is a hollowed out ring):
    Name:  Screen Shot 2014-03-03 at 22.04.39.png
Views: 141
Size:  109.9 KB

    However, when I rotate the container back and forth, even just slightly then the particles very quickly go though the object. Are there some values I could play with to improve this?

    Also I'd like to ask, when settled it does look like the fluid particles remain separate, do you have a system for particles joining together to form a bigger body( of water for example)?

    Thanks!
    Last edited by HazeTI; 03-03-2014 at 02:23 PM.


  9. Posts
    848
    Quote Originally Posted by HazeTI View Post
    Hi Thinksquirrel,

    I'm currently trialling Fluvio free edition before deciding if it is suitable to purchase a higher version. I am having issues which I'm hoping you could help answer. I'm attempting to have a 3d body of liquid in a container, which could then be manipulated.

    I have a container, which contains the liquid in place as expected and collision is working as intended (if it's not obvious this is a hollowed out ring):
    Name:  Screen Shot 2014-03-03 at 22.04.39.png
Views: 141
Size:  109.9 KB

    However, when I rotate the container back and forth, even just slightly then the particles very quickly go though the object. Are there some values I could play with to improve this?

    Also I'd like to ask, when settled it does look like the fluid particles remain separate, do you have a system for particles joining together to form a bigger body( of water for example)?

    Thanks!
    Hey Haze,

    I recommend reading this thread on our forums:
    https://thinksquirrel.com/forum/topi...nting-emitters

    In short - we're working on it. There is a workaround in that thread that you could try, though.

    As for particles joining together, Pro and Standard have a screen-space metaball effect (for Unity Pro only) that can do this.
    Last edited by Thinksquirrel; 03-03-2014 at 03:07 PM. Reason: Fixed link


  10. Posts
    83
    Thanks for the reply!
    Just a FYI update:
    I noticed on that thread you linked to that I should really be using shuriken physics. So I switched to that, unfortunately the particles leaked through the object even when it wasn't moving. I did notice that when I switched from a mesh collider to a box collider the leaking stopped for a non moving object but still had the leaks when the object rotate, even slowly. I guess the shuriken physics just isn't good enough for mesh colliders? Does that sound right?

    I've not tried the parenting or using the Get/Set you mentioned in the thread, I'll try that next with some must simpler collision and let you know how I get on.
    Cheers


  11. Posts
    848
    Quote Originally Posted by HazeTI View Post
    Thanks for the reply!
    Just a FYI update:
    I noticed on that thread you linked to that I should really be using shuriken physics. So I switched to that, unfortunately the particles leaked through the object even when it wasn't moving. I did notice that when I switched from a mesh collider to a box collider the leaking stopped for a non moving object but still had the leaks when the object rotate, even slowly. I guess the shuriken physics just isn't good enough for mesh colliders? Does that sound right?

    I've not tried the parenting or using the Get/Set you mentioned in the thread, I'll try that next with some must simpler collision and let you know how I get on.
    Cheers
    Yea, the issue with Shuriken collisions seems to be something I don't have much control over. It seems like particle collisions in Unity are optimized (and leaky) even on "High" settings. Concave shapes and overlapping colliders end up being the trouble spots. There are ways to work around it - you can destroy particles that escape the object's area and create new ones - but that makes it difficult to keep a fluid stable and at rest.

    I'm still working on potential fixes, but they're in the concept stage right now.


  12. Posts
    1,550
    I'll check soon but, have you got the marching cubes mesher and gpu acceleration implemented yet? I'm doing something for a friend and you already have an entire system at work, but those features are somewhat make or break things for what we are trying to do. If not already in, do you have a timescale?


  13. Posts
    848
    Quote Originally Posted by lazygunn View Post
    I'll check soon but, have you got the marching cubes mesher and gpu acceleration implemented yet? I'm doing something for a friend and you already have an entire system at work, but those features are somewhat make or break things for what we are trying to do. If not already in, do you have a timescale?
    Not in yet. Not comfortable giving a timescale just yet, but the work we have here is more out of the R&D phase. The latest internal build just restructured the solver to make it a lot easier to swap out, in fact.

    What sort of platform would you be looking at, as far as GPU acceleration goes? We've looked at the pros and cons of a few different options now, including both DX11 and OpenCL.


  14. Posts
    1,550
    The platforms question is a good one, the concept itself is simple as this stage, you have a cube with realistic water sloshing about (and sph was the only solution i found relevant to the job, at least until positional based fluids and set adopted at typically glacial rates) and, its a quick reaction test to avoid water being leaked. This simple idea is without any shadow of a doubt going to be expanded, graphics wise but also gameplay-wise as the numbers of discrete volumes and size of these may increase. I'm an artist and owned Fluvio for a hell of a long time (relatively) but could never find a satisfying use for it as the represented fluid fell well short of what I felt I required

    I offered to soup up my friends game, and when people say 'sure!' that means I tend to go all out, already pegged a directcompute based model even without having the material to actually construct that, simply because I know Unity gets along with it and I couldn't imagine a process fast enough to mesh these volumes based on the cpu, and further, calculate the solution at satisfying speeds to underline in impact intended. So yes the initial target would be DirectX11, even if just to get a specifically striking looking game aesthetic produced. I wasn't even aware OpenCL was usable with Unity3D - if you know i'm wrong I would, on a tangent to this post, very much appreciate material related to implementing OpenCL code

    There would be a lot of GPU things going on, it would initially be an impressive showcase of a few gpu technologies I own, all DirectCompute, so thats the environment I would have most expected a GPU solution to the process for.

    It will be great fun adding some impressive but well judged effects to his game for the most capable of GPUs and when it comes to meshing, there's probably going to have to be a line at which you have to change how the fluids are rendered dramatically, and my intention was to have the product more or less scale automatically, and as my friend did show keen-ness on an android build, it would definitely require a different technique and scaling down the simulation

    After reading what seemed like endless papers it seemed like a very good simulation and mesh could be produced using GPU techniques, dropping the GPU aceleration ran into rather ugly rough surfaces and I didn't go anywhere near mobiles. I saw a technique somewhere called sphsplaatting, which sounded lovely but I then couldn't find any of it to read anywhere, i've come to the conclusion that some plaforms maybe be able to run aspects of whats intended to be a very highly polished and honed GPU-centric version, again using Dx11, but cut down where it's necessary, like still have a visually aceptable smooth water surface and invevitably a less precise simulation

    Mobile phones might imply an entirely new approach to presentation, from the GPU based pbr modelled exercise (I need this selfishly as i'm getting very rusty with my modellingand its a good opportunity to catch up), to a much more abstract view possibly visualising particles as view aligned elipses shaded by screen depth, local height, local density and so on, whatever suits i suppose! - inteestingly some of the pbr based models and materials could be retained thanks to the efforts of skyshop and c

    That was a lot of talk about things maybe missing the point somewhat, but I was hoping for a fairly quick turnarund of features announced last year iirc and it's becoming a pity because I do want to use this for something and implementing my own sph system, or convincing my pal to write one seems like reinventing the wheel really! - And we did really want to get cracking on this

    I'm assuming without a pro license I wont have access to the relevant data for implementing my own (or friend's own) surfacing solution, and absolutely up shit creek without a paddle when it comes to moving processing to the GPU (my pal writes voip stuff they put in satellites, he's a bit unclear on directcompute, and i am flotsam on the sea as soon as I try to comprehend all that stuff outside the bits that make sense to me as an artist

    Sorry for rant but would definitely like to hear back if you have any suggestions or answers or just 'you said something silly' responses, it's I suppose just getting a touch frustrating having to stall a bunch of things i'm working towards because promised features arent materialising as soon as one would hope

    That said, and it may be a fruitless request but a very thankfully heeded, if you have any sort of beta program or package that would allow me to test what you have, should it be in a state you found acceptable, i'd happily volunteer some of my time and can then iteratively update the game as improvements are made - this would tie me into fluvio, which is something I don't mind as i've been wanting a use for of it for ages, and maybe feedback might prove useful

    Cheers for reading anyways


  15. Posts
    848
    Thanks for the insight! I agree that the speed of development hasn't been fast enough - internally we're working on scaling up the company (I'm the only developer right now so workload has been spread out a bit thin between all of our products) in order to devote more development resources to Fluvio.

    While the GPU stuff isn't ready just yet, the next update will include a fully multithreaded solver, fluid plugins, and API (the new solver is about a 10x performance boost from what we have currently).

    Also, I'm likely going to open up the solver pipeline a bit more, to allow replacement of the solver used for both Standard and Pro, without needing to touch the source.

    One of the biggest blockers with DX11 are fluid plugins. They would likely need to be rewritten as compute shaders along with the solver and renderer, or they will become the bottleneck (data would need to be transported back from the GPU every frame).


  16. Posts
    1,550
    Without too much focus on the improvements to the solver, although that does sound great and I have to congratulate such performance improvements, particularly as we'd be expecting multithreaded multicore machines to be running what is to be intially something very flashy at first anyways, will you have an answer to the surfacing of the fluid in some form of a smoothed mesh or does it remain an open issue?

    Given the fact this is a product I already own it makes no sense to me to make something to do the same thing but with nowhere near the competence of fluvio, so soon i'll be looking at it's renderer (very soon) and looking for options. While i'm not too keen on how ive seen it rendered so far, and if surfacing comes short of what i'd hope for currently, perhaps I can think up something novel that might look quite nice in it's abstraction (I'll assume the current API gives me access to particle positions and such if needed)

    If you have any personal suggestions or insight into possibilities i'd love to hear them but perhaps i'll get shader forge open and if possible use that to play around with the shading of the particles until something i really like turns up. I'll have a go at any suggestions too, and as it's been so long since ive used it anways, maybe there's something already there that will do just fine

    Thanks for the response anyways and i look forwards to the updates!


  17. Posts
    38
    I have seen some examples where Fluvio 2 is used to create particle effects that works like steams of water and then that water is contained in some jar made of meshes and colliders. But what I need is a tank that fills and drains very quickly, and the pipe from where the water goes in and out of the tank is always below the surface. Is this possible with Fluvio 2?

    Also, are there any more examples available than the 2 that come with the plugin? In any case, very nice work!


  18. Posts
    1

    Fluvio 2 Paint Preview

    Hello Thinksquirrel,

    I had bought your Fluvio 2 professional edition and I would like to test and run your demo Fluvio 2 Paint Preview for learning purposes but your example project doesn't include it with such splash effects.

    Since I am fairly new to Unity, Is it possible for you to upload the example project so that I can play around with it?

    Thanks


  19. Posts
    848
    Quote Originally Posted by lazygunn View Post
    Without too much focus on the improvements to the solver, although that does sound great and I have to congratulate such performance improvements, particularly as we'd be expecting multithreaded multicore machines to be running what is to be intially something very flashy at first anyways, will you have an answer to the surfacing of the fluid in some form of a smoothed mesh or does it remain an open issue?

    Given the fact this is a product I already own it makes no sense to me to make something to do the same thing but with nowhere near the competence of fluvio, so soon i'll be looking at it's renderer (very soon) and looking for options. While i'm not too keen on how ive seen it rendered so far, and if surfacing comes short of what i'd hope for currently, perhaps I can think up something novel that might look quite nice in it's abstraction (I'll assume the current API gives me access to particle positions and such if needed)

    If you have any personal suggestions or insight into possibilities i'd love to hear them but perhaps i'll get shader forge open and if possible use that to play around with the shading of the particles until something i really like turns up. I'll have a go at any suggestions too, and as it's been so long since ive used it anways, maybe there's something already there that will do just fine

    Thanks for the response anyways and i look forwards to the updates!
    Rendering a fluid mesh will come with GPU simulation (which I've decided will be DX11-based for now - hopefully Unity will expand their Compute Shader API to include OpenCL in the future). I have a working example with this already on the CPU, but generating a sufficient resolution fluid mesh has proven to be too slow for my liking and for any real-world use (even with multithreading) - so I haven't released it for now.

    You are correct about the API - you can get positions and velocities easily with it. The next update (with the multithreaded solver) will also include multithreaded plugins and a parallel API so you can do things like this (C#):

    Code:  
    1. fluid.ForEachParticle((ref FluidParticle p) =>
    2. {
    3.     p.position = DoStuffWithPosition(ref p);
    4.     p.velocity = someVelocityValue;
    5.     // etc.
    6. });

    This method will automatically parallelize your code. I'm trying out the prospect of being able to do something like this with Compute Shaders to arbitrarily execute your own parallel code on the GPU when using a GPU solver (no idea if this is possible even, so no guarantees).

    As for Shader Forge, I do recommend trying it out and seeing what effects you can get. The current fluid shader is just a standard shader for the particle system and any particle shader made in Shader Forge will work - you may be able to get a more volumetric effect by tweaking values somewhat or procedurally calculating some normals.

    Quote Originally Posted by Assault View Post
    I have seen some examples where Fluvio 2 is used to create particle effects that works like steams of water and then that water is contained in some jar made of meshes and colliders. But what I need is a tank that fills and drains very quickly, and the pipe from where the water goes in and out of the tank is always below the surface. Is this possible with Fluvio 2?

    Also, are there any more examples available than the 2 that come with the plugin? In any case, very nice work!
    This would be possible, the area where the pipe is could just have an emitter + attractor, with each toggled off and on.

    As for examples, we currently have the 2D and 3D example only. We plan to have some more (likely in a separate package, to keep file sizes down).

    Quote Originally Posted by hkiehs View Post
    Hello Thinksquirrel,

    I had bought your Fluvio 2 professional edition and I would like to test and run your demo Fluvio 2 Paint Preview for learning purposes but your example project doesn't include it with such splash effects.

    Since I am fairly new to Unity, Is it possible for you to upload the example project so that I can play around with it?

    Thanks
    The paint example is one of our most requested ones - it wasn't converted to later 2.0 version. I can confirm that a version of it will be coming back soon.
    Last edited by Thinksquirrel; 03-15-2014 at 10:15 AM.


  20. Posts
    1,550
    I dont mind dx11 at all, yes OpenCL would be a boon but since i decided im not going to support relics and dx12 is coming with dx11 so slowly adopted which just makes think this conservative legacy support for older graphics apis just extenuates the continuing conservative support (Im referring to dx9 specifically i suppose). I somewhat hope i'll never have to look what i consider legacy solutions in the eye ever again and however that affects adoption of whatever i create I do not give a fig - im sick of conservative moneyhoarders hamstringing a very powerful media

    Well I cant bug you for dates but i hope your gpu work comes along quickly and satisfyingly as i'm very keen on the surfacing, but while that comes I can at least get some hopefully very pretty results with some shader fiddling and the api example was perfect, should get me where i want very quickly

    As an aside I cant really imagine a cpu surfacer being nearly fast enough with it having so much already on it including the simulation itself, i'll just look forwards to the GPU flavour for the win/mac/linux targets and in the project i'm using this i think I can have quite a lot of fun finding a way to represent the volume in a very pretty, abstract way for mobiles

    If you're at all interested i'll post screenies to show how i get along

Page 18 of 19 FirstFirst ... 816171819 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •