I seem to have excruciatingly long render times for relatively short lecture videos...

  • 1
  • Question
  • Updated 1 month ago
14 minute average video length. Content is typically PowerPoint screencast with vidcam lecture. Am currently using 2015 MBP 13", Mojave, 3.1 GHz i7, 16GM RAM, Camtasia 2018.

What's the sweet spot for significantly decreasing render times? 

Upgrade to faster CPU MBP (e.g., 8-core i9)? Upgrade to Camtasia 2020? Looking for advice before I pull the trigger on anything expensive.
Photo of Tim Wat

Tim Wat

  • 4 Posts
  • 0 Reply Likes

Posted 1 month ago

  • 1
Photo of JMichaelTX

JMichaelTX

  • 111 Posts
  • 38 Reply Likes
You said "I seem to have excruciatingly long render times".  That is a relative term.  Please give us actual times (minutes).

It will be interesting to see TechSmith's response.  SW upgrades rarely produce major improvements in performance, but it does happen.

IME, RAM can be one of the most cost-effective improvements you can make.  Unfortunately, Apple has made it hard to upgrade RAM in MBP.  If you do a lot of video processing, I would recommend getting an iMac-27, in which it is very easy (and cheap) to upgrade RAM.  Buy the min RAM from Apple, then buy additional RAM from companies with OWC.  Much cheaper.

I'm sure if you do a google on HW for video processing you will find many reviews to help you make an informed choice.
Photo of Tim Wat

Tim Wat

  • 4 Posts
  • 0 Reply Likes
You're absolutely correct - one man's "excruciating"...

A 15:27 minute video took 7:36 to render.
A 16:03 minute video took 7:55 to render.
Then a 19:12 minute video took 10:05 to render.

I am disappointingly aware the latest MBP iterations are very difficult to upgrade (soldered-in-place RAM and SSDs), and am probably committed to a laptop due to my overall work environment and workflow. Was thinking about a refurb MBP or similar, as a brand new shiny one is $$$$.

Was hoping there was some sort of best practices recommendation re: hyper threaded cores vs. maxed-out RAM (and at what point is more RAM diminishing returns - is it 16GB or 32GB?).
Photo of Paul Middlin

Paul Middlin, Employee

  • 1306 Posts
  • 334 Reply Likes
It's super system dependent, but generally speaking:
  • If it's a screencast, faster CPU could make the biggest difference
  • An old GPU (graphics card) could slow you down too
  • RAM probably doesn't matter as much. 
I do recommend trying out a project on Camtasia 2020; make a copy of it (because it will upgrade the project to 2020 format when you open it) and try a rendering speed test of 2020 vs 2018 on your particular hardware. There have been a lot of changes to the rendering technology in Camtasia 2020.
Photo of JMichaelTX

JMichaelTX

  • 111 Posts
  • 38 Reply Likes
RAM probably doesn't matter as much.
That does surprise me, but it does seem to be the consensus of the tech reviewers.
However, RAM does play a role in general video editing.  Here's one review:

How Much RAM Do You Need For Video Editing 

When I bought my iMac-27, I did considerable research.  I would encourage everyone considering the purchase of a new Mac for video editing to do the same.

I ended up with this config, with only 8GB of RAM coming from Apple:
2019 iMac-27 512GB SSD, 3.6Ghz i9 CPU, 40GB RAM
Photo of Tim Wat

Tim Wat

  • 4 Posts
  • 0 Reply Likes
Thank you, Paul. This is very helpful. I'll see how easily I can upgrade to 2020 through my company's IT department.

Thanks for your hardware comments as well; this helps suggest a particular line of thinking for me re: upgrading.

Tim
Photo of Paul Middlin

Paul Middlin, Employee

  • 1306 Posts
  • 334 Reply Likes
WIth the free trial, it's at least worth an experiment!
Photo of Tim Wat

Tim Wat

  • 4 Posts
  • 0 Reply Likes
Well, 2020 isn't thrilling me:

Short video:
2018 render: 2:58
2020 render: 3:51

Long video:
2018 render: 27:24
2020 render: 35.13

For whatever reasons, 2020 is slower to render the exact same files than 2018. 
Photo of Paul Middlin

Paul Middlin, Employee

  • 1306 Posts
  • 334 Reply Likes
yeah, this is why I encouraged an experiment; rendering performance sometimes feels like more an art than science in the sense that your combination of hardware may perform better with one rendering method than another.

For an extreme example, on very old machines we would fall back to a completely software-based rendering scheme (doesn't use the graphics card at all). Normally, that's horrible, but if your graphics card was terrible enough, and depending on content, software rendering could actually be faster.

For others, 2020 could be faster. Nevertheless, we are continuing to research faster performance for future versions.
Photo of paulwilliamengle

paulwilliamengle

  • 421 Posts
  • 155 Reply Likes
I think this is one place where TechSmith could really do a better job on putting together an official resource to help its customers understand how the hardware we purchase interacts with Camtasia. 

Adobe's Premiere Pro and Blackmagic's Davinci Resolve are two separate pieces of software, and my understanding is correct, the former relies more heavily on the CPU during editing and the latter more heavily on the GPU. Assuming for the sake of argument that that's 100% correct, a user could then, theoretically, make informed decisions about prioritizing one piece of hardware over the other (i.e., getting an excellent GPU and average CPU for Davinci or vice versa for Premiere.)  

It might be helpful for a primer for a customer, a la "this is a CPU, this is what a CPU does...and here's how the CPU, GPU and RAM interact with Camtasia. When selecting a computer, keep these following best practices in mind." 

I realize that it's not possible to test every single machine for Camtasia, but I see a lot of posts about beefcake machines getting lackluster returns.