Complete crash.

[Ivaylo Katev]

well, about the particles on the grid walls, i found what is it, it's the air friction, the drag outside the grid is the correct one, the drag inside is wrong (too low). just set splash and mist drag zero. value 1 means 7 meters per sec, that is too low for 100m container. sending the build after few minutes

[Ivaylo Katev]

the build is sent, waiting for the result

[oophus]

the build is sent, waiting for the result

With the sent plugin, i get an phoenix fd error when opening the test scene.

"Get gpu_viewport[0]:NOT FOUND"

EDIT: When pressing OK I get into the scene, but after selecting the phoenix fd grid, the same error jumps up again, and it goes in a loop whenever I press OK it turns up again. So I had to forceclose max.

[Ivaylo Katev]

that is strange, but should not lead to crash, it informs us for mistyped parameter names (the strange part is that is not mistyped)
what happens after pressing ok?

[oophus]

that is strange, but should not lead to crash, it informs us for mistyped parameter names (the strange part is that is not mistyped)
what happens after pressing ok?

After pressing OK, the scene opens. But when I select the phoenix fd grid, the error comes back up, and "nothing" happens when I press OK. It disappears for a millisecond and shows up again, and it goes like that infinitally untill I have to forceclose max.

[Ivaylo Katev]

ok, svetlin told me what is the reason, i will send you a new build

[oophus]

ok, svetlin told me what is the reason, i will send you a new build

If it helps. I've got an nVidia GTX 980 card with driver version 10.18.13.5330

[oophus]

Ok, the new build is ok with the error, and I'm doing a low-res sim now, seeing as I've ran into the crash most often at a low res sim.

Still going strong, and the particles do seam to work now with a lower air-resistance.

Thumbs up so far! (knock on wood ^^)

EDIT: The mesher is still a bit weird.

On these pictures I've tried to take the "surflevel" up and down to fix that weird angeled"wall" it seams like, but to illustrate I've put it down to 0.1 here.


The grid is small at 2m, but the same weird "wall" was seen in the previous sim as well at 12m grid size.

EDIT2: I managed to get to to crash.
I put the substeps down from 4 to 1, and it crashed at first frame.

Log file:
PhoenixFD.zip

[Ivaylo Katev]

i really doubt to be hardware related, the crash is in a part that has nothing common with the video card. btw the second build is sent

[Ivaylo Katev]

well, the air resistance may change the particle behavior and prevent the crash, can you restore it as before, just for the test?

[oophus]

well, the air resistance may change the particle behavior and prevent the crash, can you restore it as before, just for the test?

See the previous post in EDIT2. I got it to crash, and I've attached the log file.

Also, I dont remember what the original splash and mist drag properties are. Should I just try and simulate with a high number?

The fact that air resistence is different inside and outside of the grid is a bug, or something to do with the missing UI? Seams weird that they should not follow eachother.

[Ivaylo Katev]

i was already afraid that the issue is compiler dependent, we are getting closer and closer, wait for the next build in few minutes

[oophus]

i was already afraid that the issue is compiler dependent, we are getting closer and closer, wait for the next build in few minutes

I upped the air resistence to 10 on both mist and splash, and it simmed fine, except for the particle behaviour outside the container.

I stopped the sim, took down the "steps per frame" to 2, and it crashed again, so I guess that is to blame then. Will try to set it up to 10 and see a bit, and then back to 1.

EDIT: It crashed with SPF of 10.

EDIT: Took the air resistence back to 0, and left the SPF at 4. Crashed again.

[Ivaylo Katev]

apparently the air resistance is not important, the crash is in the foam, btw.
the most suspicious part is the foam patterns, i really have many sims with patterns, but in this scene should have something specific, perhaps in the geometry. whatever, the built is sent, expecting the result

[oophus]

apparently the air resistance is not important, the crash is in the foam, btw.
the most suspicious part is the foam patterns, i really have many sims with patterns, but in this scene should have something specific, perhaps in the geometry. whatever, the built is sent, expecting the result

I opened the scene with build 3, and just took down the grid size from 12m to 2m and it crashed right away after I pressed simmulate.

Here is the log file:
PhoenixFD.zip

EDIT:

I am now trying to disable the foam, and only have splash/mist on, and it calculates fine. At frame 20 atm.

EDIT2: Enabling foam again, and disabling the foam patters. It calculates fine. At frame 20 again.

EDIT3: Noticed the foam size is 10 meters. ^^ So I tried to enable the strength pattern again to 1, and the size to 10m. Now it calculates fine.. I dont understand this..

EDIt3: I stopped and put the foam patterns to 1 meter, and it crashed.

EDIT4: You mention the geometry. There is a strange result in the geometry as shown earlier. It looks like the mesh hits an slightly angled wall, and gets cut off.

EDIT5: Put the patterns down to 0. It calculates fine again.

Edit6: Put the pattern to 2, and it crashes.

EDIT7: Put the pattern to 0, and it crashes here also.. So I don't think its the patterns?

Latest log file:
PhoenixFD.zip