[BradT]

Same here. Demo just gives me "Filter Crash", no matter what I feed it.
If I run it a second time without closing it, I get this in the log:
----------
Incorrect number of inputs per feature:
Ensure that each feature has three respective inputs entered: b0, b1, and var.
----------

The inputs are correct.
Restarting the program sends it back to the generic "Filter Crash" state.

[innobright]

Dear Altus denoiser users,
Release 1.4 went out yesterday with the following improvements:
We've improved detail preservation, getting rid of the "oversmoothing" effect many of our customers have complained about. The quality improvement is significant. We're very proud about the quality improvements in this release and you'll be hearing more about it soon.

We've made memory management improvements while handling animations. If you ran out of memory or encountered crashes while filtering animations, we’ve fixed those problems.

Interaction on the command-line is much more user-friendly with improved error reporting. For example, if forgot to specify an AOV or the path to one is wrong, we’ll let you know. We’ll be continuing to improve the UX of our product over the next few releases.

Happy Denoising! Download Altus 1.4 at http://www.innobright.com

Team Innobright

[innobright]

Same here. Demo just gives me "Filter Crash", no matter what I feed it.
If I run it a second time without closing it, I get this in the log:
----------
Incorrect number of inputs per feature:
Ensure that each feature has three respective inputs entered: b0, b1, and var.
----------

The inputs are correct.
Restarting the program sends it back to the generic "Filter Crash" state.

Just noticed this post—if you're still having a problem, do you mind dropping us a log to us at support@innobright.com?

[innobright]

Everyone,

If you were interested in Altus, we just pushed out a beta release of 1.5.

Altus 1.5 has a huge number of changes, so much so that we’re calling 1.5.0 a “beta”. If you’re doing production work we encourage you to stay with 1.4.3. We’ve done a fair amount of internal testing but rather than have you wait for us to finish, we want you to enjoy all the improvements as soon as possible. A later 1.5 point release (e.g. 1.5.x) will be considered “release” and will be available for in August. 1.5.2 is still considered a beta.

Altus finally has both multilayer EXR support as well as stereo (i.e. side-by-side) render support! Using multilayer support is a bit tricky, but with stereo you'll reduce the number of files per frame from as much as 12 or more down to 1.

We now support filtering of any number of AOVs—you are no longer limited to normals, caustics, etc. You are now required to specify both the beauty and position AOVs, otherwise Altus will not run. The command line for using Altus has completely changed to accommodate this, and now makes much more sense and is easier to use. These changes will break existing configurations, scripts, and workflows, so please make these changes before upgrading to the 1.5 release.

On Maxwell user, we now better handle shadow passes.

Our debuted a new GUI back in an Altus 1.4 point release. The new GUI is easier to use and now also supports stereo and multilayer EXRs as well as arbitrary AOVs.

With the release version of 1.5, we’ll be moving CUDA out of beta. As such it is now bundled with our OpenCL binary in the same download on Windows. If you’re using the GUI, it will select the best binary (CUDA or OpenCL) for you.

On Linux, we’ve a universal binary that should run on any CentOS 6.x+/Ubuntu 12.x system.

If you've any feedback, please reply to this thread or e-mail us at support@innobright.com.

Thanks!