Testing Hype With The New M4 Mac Mini

1

Back with the Adventures in Line Drawing, I was having a lot of trouble exporting such large animations with Tumult Hype. I was using a Mac mini 2018. It's not a slouch of a computer, as it was the 6-core version.

Although, the computer is six years old. Exporting video with Final Cut Pro is the main time my computer felt slow. But occasionally, I would create fairly intensive Hype projects. So, does the new M4 Mac mini solve this problem?

Well, I created a video comparing the 2018 Mac mini to the 2024 M4 Mac mini. You can skip to the part about Hype, or you can watch the review of the M4 Mac mini from the beginning.

But basically, I think the new M4 Mac mini is really good. Costco has a significant Black Friday discount on the machine, which is one of the reasons why I upgraded my Mac earlier than expected.

Oh, and the difference with Pixelmator was astonishing! :exploding_head:
No wonder Apple bought it. :smile:

1 Like
2

I saw your video before seeing this post, so I'll just copy my comment here:

I'm surprised to see the Hype export was so much faster! The video export has a static time based delay since there's no way to know when webkit is actually finished rendering a frame, so we just wait a bit. So I'd think that would account for most of the export time and reduce the difference between the two machines. Imagine how fast this could be if we didn't insert a delay!

1 Like
3

If there's a delay, how are the two different results possible? :thinking:

Could it be customized via one of those command line things? :thinking:

If video exporting works better with faster Mac computers, perhaps the delay changes based on the Mac model. It would be tedious to setup a list like that, but that sounds like huge time savings.

4

Got me excited there, sadly UK Costco selling them for the same price as Apple.

5

There's still work it is doing in between, but this is usually minimal, so that's why I'd also :thinking:.

See if you have it modified via:

defaults read com.tumult.Hype4 videoExportFrameDelay

There's no way we would know an appropriate time one way or another, it is more content dependent. I think it would be better to just have the export operate without blocking document edits.

6

The domain/default pair of (/Users/Photics/Library/Containers/com.tumult.Hype4/Data/Library/Preferences/com.tumult.Hype4, videoExportFrameDelay) does not exist

…but if you don't have access to that option, shouldn't the delay be decreased for newer computers?

What would be a value to test?

Because if the value is changed, and there's no difference, then doesn't that suggest something else is going on?

7

There's no great way to test it other than seeing that bug reports of improperly rendered frames cease coming in. It really depends on content and what is going on with the computer at a specific time the frame is rendered. Accuracy is more important than speed when it comes to video export.

I realize this is unsatisfying, but lots of parts of WebKit are unsatisfying :upside_down_face:.