Storyline 2 hangs due to memory issues

Sep 15, 2015


I've been working on a very large course for a couple of months.  I recently upgraded a desktop machine (16Gb memory) and a Surface Pro 3 (8 Gb memory) to Windows 10, and found that Storyline 2 on both machines would work as expected for a few minutes, then start getting slower and lagging behind keyboard entry, and apparently hang. I tried all of the advice on other threads about troubleshooting slow / sluggish performance of Storyline, to no effect. So Storyline looked like it had become useless to me.

On the Surface Pro, I got fed up and left it running after hitting the Save button. It took almost 1 hour for the backup bar at the bottom to appear, then almost another hour for the backup to complete. So Storyline is getting the job done, but so slowly as to be unusable. I turned on the resource monitor under the Task Manager: at rest, with my course open, Storyline was using about 840Mb of committed memory; this blew out to over 3Gb when I hit the Save button.

Issue #1: it looks to me that the approach to backup used by Storyline 2 is grossly inefficient; as though it's building an image in memory then writing it to disk. OK for files that are small in comparison to available memory; terrible for very large files.

In previous versions of Windows, I'd had to fool around with pagefile sizes to resolve performance issues. The pagefile is a place where Windows offloads low volatility data when it starts running out of memory. So I did some Googling on pagefile sizes to get the latest advice, then did some experimenting.

On the Surface Pro, the default setting was to let Windows 10 work out the pagefile size; Windows recommended a size just over 1Gb, which was probably too small since Storyline was chewing up almost 3Gb at peak. So I told Windows to allow a minimum size of 2.5 Gb and a maximum of 5 Gb. Now when I hit the Save button, the backup bar appeared immediately, and the save completed in a few minutes.  And Storyline's memory usage never went over 1.2 Gb. So I did the same on the desktop machine, with the same result. 

Issue #2: Windows 10 is lousy at memory management.

However, if I make some changes to the course before saving, the backup time jumps out and the maximum memory usage jumps up, but nowhere near what they were before. Refer to Issue #1.

So, if you're experiencing Storyline performance issues, maybe check out what's happening with memory using the resource monitor, and if necessary, take control of the pagefile away from Windows and increase its size. Set the minimum to the maximum memory usage you see with the resource monitor, and allow double that for the maximum.

I hope this helps people building larger courses.

3 Replies
Ashley Terwilliger-Pollard

Thanks Geoff for sharing here - and you mentioned reviewing the other troubleshooting elements, so I just wanted to clarify the thoughts or ideas I'd want to ensure you go through. First, you'll want to confirm you're working on local project files as described here, and then look at conducting the repair.  I haven't heard of particular issues with Windows 10 in terms of memory hogging, but I know I've heard a few folks in general not thrilled about Windows 10. Personally I installed it in a VM. 

If you're still running into issues and you need additional help, our Support engineers would be happy to take a look. Can you reach out to them here? 

Geoff Rankins

Thanks Ashley

I've decided to split the large story into 26 smaller stories, and use the "Jump to URL/File" command to connect them to/from a menu/navigation story. I did a dry run with some dummy stories to ensure that the user experience isn't damaged, and sorted out some odd behaviour under Chrome that didn't happen under other browsers.

I've also raised three support tickets - one to do with Storyline's memory management, particularly during Save and Autorecover, and two others arising from issues I experienced when splitting the large story.

This discussion is closed. You can start a new discussion or contact Articulate Support.