Pulze. Errors all the time. Too sensitive

Hi, I keep getting errors in pulze for ever possible issue, causing disruptions in the production. I’m used to backburner and it renders these scenes fine. However, I was hoping Pulze could be the natural next generation render manager software, as backburner starts to get pretty old and outdated.

There should really be a way to ignore these things, causing them to throw warnings instead of critical errors. When backburner renders them, surely Pulze also should. For an example a corona material somewhere in a file, when using vray, should cause black objects instead of a fatal error. In this specific example with incompatible material, it’s not even being used in the scene, it’s enough that it’s in the material editor.

It gets further complicated when Forensic makes the file to crash when trying to remove these missing plugins. Any advice on going around these errors is appreciated!

Thank you!

1 Like

I agree on the sensitivity of things. Or at least gives us the possibility to have an option to render no matter what.

2 Likes

Agreed. It’s a great productivity tool in principle, but I’m spending way to much time troubleshooting bugs and errors. Starting to run out of patience.

Hey @jon, @gethin.hooper and @DGTLTWNS

Thank you very much for the feedback. Could you guys give us some examples what kind of critical errors you get all the time? Then we can discuss why these things are happening and we can make the required adjustments.

If you would like to share some logs or screenshots feel free to send it to support@pulze.io!
Thanks in advance!

It doesn’t remember the rotation of the HDRI from camera to camera…Please tell me if I’ve missed a button. The only way I can get this to work is if I save out a whole render preset and callback for each camera. Frustrating.

Missing maps and missing plugins, these things are very big issues. I also wonder if any same asset file name also is an issue (this is mostly likely a problem when rendering a master scene that consists of many different xref files, which is almost always). With hundreds of different models, and thousands of maps, there are surely some missing assets or plugins some place, which doesn’t really affect the rendering if skipped. Sini also doesn’t find all these.

The point is that these things gave a mild warning in backburner, but rendered, which is all about time efficiency. We simply don’t have an hour or two to spend in each project to relocate a simple missing map deep in the xref structure, and testing 5-10 submits to the farm just to be sure to find out in the max-log on the main farm host which missing plugins it is this time, because it doesn’t even find all problems at once, just the first one triggered and then critical exit, and voila, no render to post process and deliver to the customer that day.

It’s come to the point that we no longer use pulze in our studio, we use local rendering and/or Vantage instead, which is respectively more time conusing and less reliable, but at least it renders, and hence it makes us able to deliver our projects pretty much on time.

Additionaly, it would be a nice feature having better error reporting in the Pulze GUI, and not having to remote logon the farm and locate max-log and then finding the correct log line every time. Just. Too. Time consuming. I can’t think of any place in the world, despite different living costs and different overhead project budgets, where a company would like to loose money because they have to solve problems that shouldn’t be a problem.

Sorry to be critical in my writing, but you need to fix this quick or else you have lost a customer.

1 Like

Apologies for my ignorance, but it is possible to change the ‘sensitivity’ of errors in Pulze. Can’t you just lower them from critical to warning in the sanity checks? Or are you referring to something else?

The sanity check could be all fine, but pulze gives an error on the job after submitting the scene, for the slightest possible reason. One could think it could have something to do with 3ds max, but these errors didn’t happen before we started using Pulze.

Interesting. I have my fair share of trouble with Pulze RM (though I still do love it) but haven’t had too many error-related issues.

(That being said, since upgrading to the latest version I’m getting a ton of loading errors, Pulze RM won’t even start 25% of the time!!)

Do you mind giving me an example of such errors, just out of curiosity? I ask as I often have missing maps in scenes rendered with RM, but this doesn’t create errors…

I second this, to have an option RENDER NO MATTER WHAT. Sometime as said above we spend more time trouble shooting and optimizing for network rendering that it defeats the purpose. Especially on vray 6 gpu, gets stuck on loading while on local it renders normally on each pc.
I also tried many times with Redshift, its even more sensitive, I dont think they even test redshift, the slightest yellow or non critical error stops the whole job.
We just want something to send and render no matter what, like a super admin mode where it becomes our responsibility to have everything in place but just PLEASE RENDER NO MATTER WHAT. Can we have that ?

Setting the rotation of the HDRI in the scene manager gui now works for me in latest build. Maybe a reinstall did the trick!

Agree on this. A RENDER NO MATTER WHAT kinda setting would be highly appreciated.

Thanks for all the great feedback and suggestions. I won’t go into details why some of the errors or warnings happens currently. We are aware that it could be annoying.

We will definitely include these suggestions in the new version and try to make it as resilient as 3ds Max allows us to be.

Any ETA for this? Right now the render manager is basically unusable for production work because of the way it handles errors. Just an option to let it retry unlimited times after erroring would already be so awesome to have. Such an implementation would be highly appreciated!

Another vote for this from me. There’s pretty much not a single day when we don’t have to manualy allow some nodes to re-join the job (once or twice) until the finally render.

Yeah it’s crazy. I have two scenes that load perfectly in 3ds Max, on two different computers. Everyting is linked locally, all plugins are installed too. They render without problems and without errors from within 3ds Max. But when I want to render them through Render Manager; no way. He just complains that it makes 3ds Max crash and then links me to some forum thread that has nothing to do with it. I was tired of this and formatted my computer so I don’t have any logs. And I’m tired of this BS really, I’m paying for alpa testing basically.

I try to use RM, 6 times in the last year and every-freaking-time there’s a problem or there’s a crash.

I have been waiting on the VPN tutorial so I can render from different locations too, which was promissed a couple of times already. Never happened. Yes I tried getting it work myself.

Unless it’s not getting more robust and some VPN support (like Hamachi, TailScale,…) I have no use for it anymore.

1 Like

Certainly. PULZE thinks we are rendering nannies. You have to employ an employee just to restart errors that don’t cause problems with the images. Please correct this immediately. All Managers have error limit editing before crashing.

1 Like