If you mean shearing, you can: The original mesh is a cubeface wrote:I think the last transformation you can´t do with matrixes...
Blender instances "patch" for the unofficial exporter
SW: Octane 3.05 | Linux Mint 18.1 64bit | Blender 2.78 HW: EVGA GTX 1070 | i5 2500K | 16GB RAM Drivers: 375.26
cgmo.net
cgmo.net
Ah ok, thanks.
face
face
Win10 Pro, Driver 378.78, Softimage 2015SP2 & Octane 3.05 RC1,
64GB Ram, i7-6950X, GTX1080TI 11GB
http://vimeo.com/user2509578
64GB Ram, i7-6950X, GTX1080TI 11GB
http://vimeo.com/user2509578
I am still getting scaling issues.
My duplicated object has a scale of 1x1x1
My particles system has the object size as 1 with zero randomness
When I export the particle system I have to scale the duplicated object at least 20 times in order to get it back to the size it was inside of Blender.
Looking at the matrices, the scaling looks quite small (is it right that the file is not a CSV?)
-0.000622 -0.036512 0.016325 -2.50315 -0.039994 0.000681 -0 -2.97594 -0.000278 -0.016322 -0.036517 -1.12284
0.003593 0.002179 0.039779 0.362398 -0.020737 0.034205 0 -2.10696 -0.034016 -0.020622 0.004202 -3.43214
0.029992 0.025276 -0.007848 3.02667 -0.025777 0.030587 0 -2.61555 0.006001 0.005057 0.039223 0.607507
0.024423 0.026323 -0.017624 2.46328 -0.029323 0.027206 0 -2.97613 0.011987 0.012919 0.035908 1.20822
Script version 1.18 FYI
My duplicated object has a scale of 1x1x1
My particles system has the object size as 1 with zero randomness
When I export the particle system I have to scale the duplicated object at least 20 times in order to get it back to the size it was inside of Blender.
Looking at the matrices, the scaling looks quite small (is it right that the file is not a CSV?)
-0.000622 -0.036512 0.016325 -2.50315 -0.039994 0.000681 -0 -2.97594 -0.000278 -0.016322 -0.036517 -1.12284
0.003593 0.002179 0.039779 0.362398 -0.020737 0.034205 0 -2.10696 -0.034016 -0.020622 0.004202 -3.43214
0.029992 0.025276 -0.007848 3.02667 -0.025777 0.030587 0 -2.61555 0.006001 0.005057 0.039223 0.607507
0.024423 0.026323 -0.017624 2.46328 -0.029323 0.027206 0 -2.97613 0.011987 0.012919 0.035908 1.20822
Script version 1.18 FYI
(HW) Intel i7 2600k, 16GB DDR3, MSI 560GTX ti (2GB) x 3
(SW) Octane (1.50) Blender (2.70) (exporter 2.02)
(OS) Windows 7(64)
(SW) Octane (1.50) Blender (2.70) (exporter 2.02)
(OS) Windows 7(64)
@Steve, have you touched any other PS settings that affect their scale? (there is a whole lot of them)
Without an example .blend that reproduces this, I cant really help.
Without an example .blend that reproduces this, I cant really help.
SW: Octane 3.05 | Linux Mint 18.1 64bit | Blender 2.78 HW: EVGA GTX 1070 | i5 2500K | 16GB RAM Drivers: 375.26
cgmo.net
cgmo.net
Ahh, found it. For some reason you are using the particle normal velocity to scale the particles. So in my example I set the normal to be very small because I wanted the particles to touch the object from which they were being projected. This has the effect of scaling the particles wayyyyyy down. When I set the normal to 1 then they appear the right size. Guess what, set the normal to 2 and the particles become twice the size.
Why should the normal velocity affect the particle size?
I am far from an expert but I think it is down to the scaling of the particles by the hair length. Some sort of check to see whether the PS is hair is required. If it is an emitter then no scaling should be performed.
rot = Quaternion.to_matrix(p.rotation).to_4x4()
scale = Matrix.Scale(p.size, 4) * Matrix.Scale(pset.hair_length, 4)
t = loc * rot * scale
t = emitter.matrix_world * t
Why should the normal velocity affect the particle size?
I am far from an expert but I think it is down to the scaling of the particles by the hair length. Some sort of check to see whether the PS is hair is required. If it is an emitter then no scaling should be performed.
rot = Quaternion.to_matrix(p.rotation).to_4x4()
scale = Matrix.Scale(p.size, 4) * Matrix.Scale(pset.hair_length, 4)
t = loc * rot * scale
t = emitter.matrix_world * t
(HW) Intel i7 2600k, 16GB DDR3, MSI 560GTX ti (2GB) x 3
(SW) Octane (1.50) Blender (2.70) (exporter 2.02)
(OS) Windows 7(64)
(SW) Octane (1.50) Blender (2.70) (exporter 2.02)
(OS) Windows 7(64)
In case of hair PS anything in the velocity tab also affects scale. Normal is the only param. supported currently. The scaling by hair length correction must be there to get the correct object scale for hair length. But in case of emitter PS the params. in the velocity tab affects position not scale.steveps3 wrote: Why should the normal velocity affect the particle size?
Yes. Tough, it's not enough to just remove the scaling by hair length term, position must be also adjusted. If you are not interested in posiiton, for now you can replace that line with:steveps3 wrote:I am far from an expert but I think it is down to the scaling of the particles by the hair length. Some sort of check to see whether the PS is hair is required. If it is an emitter then no scaling should be performed.
Code: Select all
scale = Matrix.Scale(p.size, 4) * Matrix.Scale(pset.hair_length, 4) if pset.type == "HAIR" else Matrix.Scale(p.size, 4)
SW: Octane 3.05 | Linux Mint 18.1 64bit | Blender 2.78 HW: EVGA GTX 1070 | i5 2500K | 16GB RAM Drivers: 375.26
cgmo.net
cgmo.net
Well I am starting to get some results. Blender seems to be really struggling with all the particle systems. I currently have 6 of them with a total of around 55,000 particles.
There are some odd grassless patches that I need to look into but so far so good.
The pink things will be buildings by the way.
There are some odd grassless patches that I need to look into but so far so good.
The pink things will be buildings by the way.
(HW) Intel i7 2600k, 16GB DDR3, MSI 560GTX ti (2GB) x 3
(SW) Octane (1.50) Blender (2.70) (exporter 2.02)
(OS) Windows 7(64)
(SW) Octane (1.50) Blender (2.70) (exporter 2.02)
(OS) Windows 7(64)
I've just spotted a bug in the exporter script. It only exports the percentage of the particles that you have set to be visible. I currently have about 200,000 particles in my model and I've just reduced the visible down to 1% just so that I can see where they are but maintain a responsive system. Now when I export I only get the 1% of the particles exported. This means that I have to up the percentage to 100% just to export the particles and then reduce the figure again.
Any chance of a tweak to the script??
Any chance of a tweak to the script??
(HW) Intel i7 2600k, 16GB DDR3, MSI 560GTX ti (2GB) x 3
(SW) Octane (1.50) Blender (2.70) (exporter 2.02)
(OS) Windows 7(64)
(SW) Octane (1.50) Blender (2.70) (exporter 2.02)
(OS) Windows 7(64)
It is the principle of the script to export only visible things, so I think that it is up to you to set up the number of visible particles at 100%.
French Blender user - CPU : intel Quad QX9650 at 3GHz - 8GB of RAM - Windows 7 Pro 64 bits. Display GPU : GeForce GTX 480 (2 Samsung 2443BW-1920x1600 monitors). External GPUs : two EVGA GTX 580 3GB in a Cubix GPU-Xpander Pro 2. NVidia Driver : 368.22.
Really? What is the point of that. If I make all of the particles systems visible then blender is on the verge of crashing. It sems much more sensible for it to export all of the particles that are alive. They are only invisible in view mode. If I were to use the internal renderer then they would get rendered so why shouldn't they get exported?
(HW) Intel i7 2600k, 16GB DDR3, MSI 560GTX ti (2GB) x 3
(SW) Octane (1.50) Blender (2.70) (exporter 2.02)
(OS) Windows 7(64)
(SW) Octane (1.50) Blender (2.70) (exporter 2.02)
(OS) Windows 7(64)