Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - TomG

Pages: 1 ... 265 266 [267] 268 269 ... 365
3991
[C4D] General Discussion / Re: Purchase interior scenes
« on: 2018-08-03, 16:34:57 »
You can check the sites listed on our resources page, who provide models and may also provide complete scenes:
https://corona-renderer.com/resources/models


3992
[Max] General Discussion / Re: Velocity Element Glitches
« on: 2018-08-03, 15:30:12 »
A quick look in the CIE doesn't show any discontinuities in the data there, it all seems to change smoothly as expected. May want to check RSMB's help pages too, and engage their support or community in taking a look at the EXR. Also, check out their help pages at https://revisionfx.com/faq/motion_vector/ (perhaps there's something up with alpha for instance, as RSMB uses that to decide whether there is "no object" there to blur; or the setting of the Max Displacement value in RSMB). As far as I can tell, everything from Corona is as expected... but whether it's "as RSMB needs" is a different question :)


3993
[Max] General Discussion / Re: Velocity Element Glitches
« on: 2018-08-03, 14:48:02 »
Cheers! Will take a look when I can, hopefully not too long.

3994
[Max] General Discussion / Re: Velocity Element Glitches
« on: 2018-08-03, 14:36:40 »
TY for the extra info! Everything still sounds ok there, at least as far as I could tell. Since it's the same frame each time, can you send over the 32 bit with the velocity pass in it? Along with the result from RSMB once blur is applied. I'd be interested to see what the 32 bit values are in the EXR saved from Corona, at the point where the RSMB-calculated blur "goes wrong". You can use the private uploader if you can't share here (https://corona-renderer.com/upload)

3995
[Max] General Discussion / Re: Velocity Element Glitches
« on: 2018-08-03, 13:53:22 »
Settings look correct, in as much as I know (as RSMB is a paid plugin, and we don't have a copy to test with directly). Have you also checked in with RSMB support?

Some thoughts - what kind of animation is it, just regular camera and object transforms (no geometry deformation / soft body / other more unusual stuff)? Is Camera and Object motion blur enabled for the scene? The files were saved out as 32 bit? In the problem files, what does the data look like in the velocity pass (a right click in the VFB will let you check the pixel data, so you can see if it is indeed suddenly changing to something unexpected at those points where the resulting blur is different - if it is, then could be something to do with how Corona is writing the data; if it isn't, could be to do with how RSMB is processing it). Can you share an EXR of a problem frame? Does the problem happen on the same frames each time you process it with RSMB, or on different frames (again, could be a clue on whether it's something in the data, or something with how it's being handled)? What kind of camera is it being rendered from? What version of Max, and of Corona?

Thanks!

3996
Anyone from the Corona team going to Half Rez this year?

Nothing on the plan, as yet at any rate.

3997

Side note, will there be any Corona fans going to Siggraph this year in Vancouver? I'll be there any I'd love to meet anyone who is a Corona for C4D user.

- Shawn

Well the Corona Core team will be there, in the form of Ondra and Jaroslav - be sure to say hi (even though they aren't from the C4D team, no playing favourites!)

3998
Unfortunately, we'd need more details before being able to look into a fix. For example, is this all scenes, or just one specific scene? Can you upload the scene (there's the private uploader if it can't be shared publicly - https://corona-renderer.com/upload). What are the specs of the different machines? For rendering, do you mean rendering locally on each machine (running C4D and loading the scene and rendering), or was this sent across some sort of network rendering solution (e.g TeamRender)? We'd need to know which passes you were using too, but uploading the scene would give all that sort of information.

3999
Ah, as an image - I was looking for it as the actual log doc :)

4000
Meantime, the DR logs would be important of course, to see what DR is doing at the time. You can use the private uploader for those if they can't be shared here (https://corona-renderer.com/upload)

Thanks!

4001
I still suspect this was due to an update in Windows 10, that is preventing the Win 10 machines communicating with the other machines (which would mean it isn't a Corona bug). For sure on not installing every daily build, but this particular one has the particular solution that you are looking for (and not very much else so it is pretty safe). Also, that's always where any fixes come first, in a daily build, as it has to be tested first to make sure it fixes things and doesn't cause any other problems.

It would be interesting to know if it resolves it - whether a case arises for their being a hotfix to Corona 2 or not would be dependent on whether this fix actually remedies that issue (mind you, if it does fix it, working with that daily build would mean not waiting for any other potential next steps).

4002
I found that at some point, auto detecting the nodes stopped working, due I believe to some change to Windows in some update to it (I suspect related to the remove of the Workgroup or some such thing). The solution for me was not to use the detect nodes, but to enter their names or ip addresses manually.

You don't mention whether you were use the search LAN to find the nodes, or if you'd added them manually. One thing to try is the latest daily build, there were changes there ("DR is now able to autodiscover slaves connected through all network interfaces." from the changelog at https://forum.corona-renderer.com/index.php?topic=20852) that seem to have undone whatever it is that Windows did, and I can go back to just using the Search LAN and letting Corona add the nodes by itself.

4003
[C4D] Resolved Bugs / Re: Team Render 300% Slower?
« on: 2018-07-30, 20:22:47 »
Teamrender with latest daily build seems back to slower result than local (single) render. Known issue?

Still not able to replicate this here, in one test, it was 3m 11s for local render, 1m 52s for team render using 3 nodes, to same number of passes (10, in this case). With the network overhead in such a small number of passes, and fact that the two nodes are slower than the main machine, this is about the expected improvement (rather than 3 times faster)

Some questions:

- Is this TR to Picture Viewer, or TR through the web manager? My test was to picture viewer.
- Is this PC, or Mac?
- Is this every scene, or just particular scenes?
- Is this for rendering to passes, or noise limit?

Thanks!

4004
[Max] Bug Reporting / Re: strange IES spotlight
« on: 2018-07-30, 20:07:54 »
It's not happening for me with the IES profiles provided with Corona (and using a CoronaLight, haven't tested other light types) - I am getting the expected results.

On the second question, simply drop a CoronaBitmap into the Texmap slot in the Color options for the CoronaLight (and check the button next to Texmap). Depending on how sharp you want the projection to be, you may need to increase Directionality (but be aware that higher Directionality means more noise, so longer render times) - in the example below, so that the pattern was visible I raised Directionality to 0.9, which is very high :)


4005
[Max] Corona Goodies - User Contributions / Re: Rock VOL.1
« on: 2018-07-30, 18:25:29 »
And just for a moment I thought it might be guitars, Marshall stacks, and 80s hair styles! ;)

Pages: 1 ... 265 266 [267] 268 269 ... 365