Crumplepop makes plug-ins for Final Cut X, as well as for Final Cut 7 & 6. I picked up a bunch of their plugins for Final Cut X a while back. The effects they apply have a great look, but that’s not really what this review is about. To see what plug-ins they offer and before/after demos, you can have a look at their website. In this review, I want to look at the impact of using one of their plugins, Finisher, has on the FCP edit and render workflow, as this may have a more relevant impact on whether or not you choose to use this plugin.
When I started using Finisher and ToneGrade from CrumplePop, I found that dropping them on my footage in FCP X slowed down the background rendering to the point that I couldn’t work anymore. So, I arranged my workflow such that I only dropped the effects on the footage at the end of the editing process, right before sending the footage to compressor for rendering to whatever format I needed. This wasn’t really Crumplepop’s fault, though; I suspect that it had to do with issues in FCP X itself. I also had the opportunity to evaluate the Sapphire Edge plugins (for a review of Sapphire look here) for FCP X from Genarts back when they were in beta, I and I saw some similar issues with early versions of the plugins where render times were greatly increased – but they put a lot of work into reducing those render times before release. Long story made short, Apple released some updates to FCP X, and the issue has gone away – dropping third-party plugins such as Finisher on the timeline in FCP X no longer have an adverse effect on background rendering. But, there was still an issue – a game breaker that made me stop using Finisher and ToneGrade altogether. Note that this issue was unique to these two plugins – I have used other plugins from CrumplePop without issue.
The issue arises not in FCP X, but rather in Compressor 4. Again, this might not be entirely CrumplePop’s fault, but it does prevent me from using these two plugins for most of my work, which is unfortunate because I really like the effects they provide.
First off, in Compressor, when I drag and drop my settings onto any clip for the first time, Compressor normally hangs for a fraction of a second. I’m not sure why it does this, but it has nothing to do with any plugin – it happens on normal footage, and it only happens when I drop in the first setting. I suspect this might be related to the same update from Apple that fixed the background render time issue – it didn’t used to work like this. All the subsequent settings I drop in, there is no ‘hang-time’. However, when I drop the first setting onto a clip that has Finisher applied, the hang-time increases. For a clip that is over an hour long, this hang-time is really long. I found that for a clip that is 3 minutes, 15 seconds long, the hang time was about 17 seconds.
Secondly, with the Finisher plugin applied, the render time to export footage from Compressor is greatly increased. Now keep in mind, your render times using any plugins are probably going to increase in compressor, but I thought some test metrics might be useful in evaluating whether it’s worth it to use a given plugin, which is just what I did here for the Finisher plugin. I did also complain about the performance issues to Crumplepop, and they provided me with a pre-release version of the Finisher plugin they are working on that is supposed to improve on the performance. Below I show the results of rendering a 3 minute, fifteen second clip in Compressor 4 using the new version of Finisher, the old version of Finisher, and without the Finisher plugin applied.
For all three cases, I used the default Apple settings for (a) HD 1080p for upload to Youtube and (H264) and (b) HD for Apple Devices (10 Mbps). The total footage time was 3:15 (3 minutes 15 seconds), of which the speedtest version of Finsher (the development copy provided by Crumplepop) was applied to 1 minute, 48 seconds of the footage. I then went back into FCPX and deleted just the Finisher from the footage, re-rendered with the same settings as before. One note, the hang-time described above, was greatly reduced without the plugin applied – compressor only pauses for a moment when dropping the initial setting, when there is no finisher applied. I then quit FCPX, uninstalled the new version of finisher, and restored the old finisher version, and ran the render in compressor again after applying the old finisher to the same clips as before.
Here are the render times:
|No Finisher applied||Finisher, SpeedTest version||Finisher, Original version|
|render time hrs:min:seconds|
|.mov file (for YouTube settings, Compressor defaults)||0:11:49||0:43:48||0:40:30|
|.m4v file (apple devices 10mbps||0:5:18||0:33:21||0:39:57|
I don’t really see any improvement using the new version of Finisher; if anything, the performance seems to be slightly worse. So, in summary, applying these plugins comes at a cost, in render time. I’m running all this on a Mac Pro (2 X 2.66 GHz 6-Core Xeon processors, 32 GB RAM). Is it worth the wait, especially if you’re rendering hours of footage? That really depends on you, and your use case. Here are a few screenshots from the test footage above for comparison:
|Raw footage||Color Graded (using FCP X only, no third-party plugins)||Finisher Applied|
On a final note, Crumplepop has stated that they are working on finding a solution for the render time issue.