How to render a Blender scene on AWS Deadline Cloud

Amazon Web Services (AWS) Deadline Cloud is a fully managed service that streamlines render management for teams creating computer-generated 2D/3D graphics and visual effects for films, TV shows, games, industrial design, and architecture. Deadline Cloud comes with a fully supported integration for Blender. It is the quickest way to render Blender scenes on AWS while taking advantage of AWS scale and security. With the Blender integration, users can submit their first renders on Deadline Cloud in just minutes and directly from inside Blender.

This blog shows how to set up AWS Deadline Cloud and render a scene from Blender. We’ll cover:

• Setting up a Deadline Cloud farm
• Creating a first user account
• Installing the Deadline Cloud monitor desktop application
• Installing the Blender integration with Deadline Cloud
• Submitting a scene from Blender to Deadline Cloud
• Downloading the rendered frames

Set up a Deadline Cloud Farm

1. Create an AWS account if you do not already have one. Once you’re signed in, go to the AWS Deadline Cloud console page.

Figure 1: AWS Deadline Cloud Console: Viewing the landing page.

2. Click Set up Deadline Cloud to start the quick start process. On the Welcome to AWS Deadline Cloud setup page, click Next.
3. On the Set up monitor page under Monitor name, choose a name for your monitor. The monitor is the application you’ll use to check job statuses, progress, and logs. The monitor’s name sets the URL you’ll use to access the monitor. Click Next.

Figure 2: AWS Deadline Cloud Console: Setting up the monitor.

4. On the Define farm details page:

• You can change the default farm name under Name if you like. The farm contains all Deadline Cloud resources you’ll use including your fleet and queue. A farm is commonly shared across a company.
• Under Groups and users, configure your initial group and user. Under Farm access level, select the default option for Owner (recommended) so the initial user has full access to the farm.

Figure 3: AWS Deadline Cloud Console: Creating the initial user and group.

• Click Skip to Review and Create.

5. On the final review page, confirm your selections by clicking Create farm.

A Deadline Cloud onboarding pop-up window will appear showing the progress of your resources as they are created. The process may take up to a minute. Keep the pop-up window open until the process finishes.

Figure 4: AWS Deadline Cloud Console: Waiting for resources to be created.

When the resources have finished creating, a message will show within the pop-up window, “Deadline Cloud onboarding complete”. Leave the pop-up window open for the next steps.

Figure 5: AWS Deadline Cloud Console: Onboarding is complete at the end of the quick start flow.

Set up the first user account

The Deadline Cloud onboarding pop-up window contains a blue information area stating: Reset password for each user so that they can receive an email invitation to join your monitor.

1. Click the Reset password for users button.
2. Select the user we created (Figure 3) from the list and click the Reset Password button.

Figure 6: AWS Deadline Cloud Console: Resetting the password for the initial user.

3. The browser will open a new tab with the console for AWS Identity and Access Management (IAM) Identity Center to view the user you selected. In the IAM Identity Center console, click Reset password.

Figure 7: AWS IAM Identity Center Console: Resetting the password for the initial user.

4. A window for Reset password for user “user name” will pop up with two resetting options. Select the Send an email to the user with instructions for resetting the password option and click Reset password. You can now close the IAM Identity Center console tab.

Figure 8: AWS IAM Identity Center Console: Requesting an email to reset the user’s password.

5. The email address you set for your user should receive an email with the subject Reset your password for AWS IAM Identity Center. In the email, click the Reset password link.
6. The link will open your browser to the corresponding page to set a new password. Set and confirm your password and click Set new password. The page will say Password reset success. You can now close this page.

Set up Deadline Cloud Monitor

1. Back in the Deadline Cloud console, in the left side menu, click on Downloads.
2. Under Deadline Cloud monitor, select the appropriate download for your OS then click Download. When the installer finishes downloading, open it.

Figure 9: AWS Deadline Cloud Console: Downloading the Deadline Cloud monitor installer.

3. Complete the installation process.
4. Open Deadline Cloud Monitor.

Figure 10: Deadline Cloud monitor installation is complete.

5. After a cutscene plays, the Find your monitor page will open and ask for Your customer Deadline Cloud monitor URL. Type in the monitor URL that you chose during setup. To find the URL again, go to the Deadline Cloud console, click Dashboard on the left menu, click Monitor details, and then copy the URL. Click Next.

Figure 11: AWS Deadline Cloud Console: Finding the monitor URL.

Figure 12: Deadline Cloud monitor: Setting the monitor URL.

6. On the Create your profile page, you may edit the name of your workstation profile. This profile represents your workstation AWS credentials and is different from your user profile. Click Next.
7. On the Review and launch page, click Create and launch.

Figure 13: Deadline Cloud monitor: Confirming profile settings before launching the monitor.

8. Deadline Cloud Monitor will open a sign in page. Enter the username and password you chose earlier and click Sign in.

Figure 14: Signing in as the initial user in the browser.

9. Your browser may ask for permission to open Deadline Cloud monitor. Click Open. You are now signed in and your Deadline Cloud farm is fully configured and ready to run your workloads.

Figure 15: Deadline Cloud monitor: Viewing the farm.

Install the Blender integration

Next, we’ll install the Blender integration for Deadline Cloud so we can submit render jobs to Deadline Cloud directly from Blender. If you don’t already have a supported version of Blender installed, download and install it. You can see which versions of Blender are currently supported in the submitters documentation. Note: Mac OS does not have a submitter installer. If you’re comfortable with a more technical install process, you can find manual install instructions for the Blender integration on this GitHub page.

1. In Deadline Cloud Monitor, there will be a blue message at the top of the screen for Missing Deadline CLI. Click its Go to download page button.
2. Click the Download button, select your operating system, then click the link next to Submitter to download the submitter installer. A message will pop up as the installer downloads. When the download finishes, click the message’s Open button.

Figure 16: Deadline Cloud monitor: Downloading the submitter installer.

3. Follow the installer instructions clicking Next to continue. On the Select Components page, check the box for Deadline Cloud for Blender and then click Next. When the install process finishes, click Finish.

Figure 17: Submitter installer: Checking the box to install the Blender integration.

4. Next, we’ll configure Blender to find the new add on. Open Blender, click the Edit menu, then select Preferences. In the Blender Preferences window, select the File Paths tab on the left side menu. Under Script Directories click Add and then choose the directory: `%USERPROFILE%\DeadlineCloudSubmitter\Submitters\Blender\python\

Figure 18: Blender: In the Blender Preferences window, adding the submitter path to the list of script directories.

5. Close Blender, then reopen it. Again, click on the Edit menu and select Preferences. Click on the Add-ons tab in the left side menu. Check the box next to Deadline Cloud for Blender. Close the Blender Preferences window.

Figure 19: Blender: In the Blender Preferences window, enabling the Deadline Cloud for Blender add-on.

Now, click on the Render menu. There should be a new option for Submit to AWS Deadline Cloud within the drop-down list. The plugin is now installed and you’re ready to render.

Figure 20: Blender: Viewing the new menu option for Submit to AWS Deadline Cloud under the Render menu.

Submit a job

Open a Blender scene to render. Blender offers sample scenes that are useful for testing. We’ll use the Blender splash artwork “Blender 4.0 – Gaku” in this example.

1. In Blender with your scene open, click Render and select from the drop-down list Submit to AWS Deadline Cloud.

2. When a window prompts you to install GUI dependencies, click OK. The install will take about 30 seconds and is only necessary the first time you run the submitter.

Figure 21: Blender: Agreeing to install the submitter’s GUI dependencies.

3. Now you’ll see the main submitter window with various options. On this first use, we need to set some initial settings.

Figure 22: Blender: Viewing the submitter options. The farm and queue are not yet set.

1. Click the Settings button located on the bottom of main Submitter window. Then next to Default farm and Default queue, select the farm and queue you created during setup.

Figure 23: Blender: In the submitter settings, configuring the default farm and queue.

2. Click OK to close the Settings window. You do not need to edit these settings on you next submissions.

Figure 24: Blender: In the submitter, viewing the job settings which are now set correctly.

3. Back in the Submit to AWS Deadline Cloud window, you can configure your job before you submit it. Under Job specific settings, you can also see the Output Path where the final frames will be created. When you’re happy with the settings, click the Submit button.
4. A window will pop up informing you that assets related to the job will be uploaded. Click OK.

Figure 25: Blender: The submitter says that it will upload files as part of the job submission.

5. You’ll see a window with progress bars as the assets are uploaded.

Figure 26: Blender: The submitter is hashing and uploading files.

6. When the submission finishes, click OK to close the window.

Figure 27: Blender: The job is submitted successfully.

We’ll now switch to Deadline Cloud Monitor to see the job we submitted and watch its progress. Open Deadline Cloud Monitor, click the farm you created, and then select your queue. You should see the job you submitted in the queue. Using the console’s quick start configuration, your fleet will automatically grow when there are jobs pending in the queue and shrink when the work is complete—you’ll only pay for the capacity you use.

Figure 28: Deadline Cloud Monitor: The job that was just submitted is visible in the queue.

In our example you can see this job has one step and three tasks. Two of the tasks have started running. Your job will start in the Ready status while waiting for a worker, then will transition to Starting and Running as it’s picked up. Lastly, it will move to Succeeded when the render is complete.

Download the result

When the job finishes, we’re ready to download the rendered frames.

1. In Deadline Cloud Monitor, in the queue’s jobs list, right click the job you submitted and click Download output.

Figure 29: Deadline Cloud Monitor: Right clicking on the finished job to download the job’s output.

2. A banner will pop up saying Starting download, then a window stating Confirm download location will ask you where to download the files. By default, the output location will be in the output directory specified in your Blender scene file. Edit the location if you need to, then click Ok.

Figure 30: Deadline Cloud Monitor: Confirming the download location.

3. A progress bar will track the download’s progress.
4. When the download completes, go to its location and find the rendered frames. We can open one and see that the job rendered successfully on AWS Deadline Cloud.

Figure 31: Viewing the downloaded files for the rendered frames.

Figure 32: Viewing the rendered image from one of the frames.

Conclusion

In this blog, we set up a farm in AWS Deadline Cloud, installed the required desktop software, submitted a job from Blender, and downloaded the rendered frames back to our workstation. We also saw how our fleet grew and shrank in response to the job we submitted, so we only paid for the capacity we were using.

Contact an AWS Representative to know how we can help accelerate your business.

Further reading

Now that you have submitted your first sample job, continue learning about AWS Deadline Cloud features in order to integrate them into your workflows:

• Deadline Cloud has integrations with other software, such as Autodesk Arnold for Maya, Autodesk Maya, SideFX Houdini, and Maxon Cinema 4D. See the Deadline Cloud documentation on submitters for more information.
• You can add additional users to your farm so they can also submit and view jobs in the farm you set up. See the Deadline Cloud documentation on managing users for instructions.
• In addition to the pre-built integrations, you can create your own submitters, script your own jobs and use your own software. Explore AWS Deadline Cloud open-source samples on GitHub to learn to build and customize integrations.