Problems with the Canon EOS Movie Plugin-E1
Here’s a down and dirty screen cast showing some problems I’ve been having using the Canon Movie Plugin E1 for Final Cut Pro. I’ve literally recorded this and uploaded it in the last 20 minutes so please forgive the quality, I’m right in the middle of an edit so didn’t have much time.
provider: youtube
url: http://www.youtube.com/v/_wyx5aJ64Sk
src: https://www.youtube-nocookie.com/embed/_wyx5aJ64Sk?feature=oembed&wmode=opaque
src mod: https://www.youtube-nocookie.com/embed/_wyx5aJ64Sk?wmode=opaque
src gen: https://www.youtube-nocookie.com/embed/_wyx5aJ64Sk
Canon EOS Movie Plugin-E1 errors
If you’ve also seen similar issues or don’t suffer from it then please leave a comment, it might help to nail down the cause of the problems.
UPDATE – 26/08/2010
Canon has just announced an update to the EOS E1 plugin which is due for release in September. According to Canon the update adds support for their new 60D body. Of much more interest to me though is the announcement that it will also provide support for multi-core processing. Just in time for my shiny new 12 core!
Lets hope they fix the above bugs too!
Hi.
One solution can be in Logging panel, on the “Name preset”, to select the option “Current name with Counter”, so that if you reimport some clip again the initial name will be the same but with a counter added that will make it different from other versions (for example “MVI_4796 001” or “MVI_4796 0012”, etc.
Thanks al for your help!
M
I still have the same problem. Is canon planing to do something about it? thanks
Like many of you, I think we are all still trying to side-step these bugs. I think I can say some things fairly definitively.
1.) Don’t touch Final Cut during transfer. That causes issues. I was surfing the internet during the operation and out of 50 clips, none were incomplete when I left FCP alone.
2.) If you do, and need to go back and reimport clips, either add i/o points, or click “Clear Logging Autofill Cache” from the preference menu. If you clear the cache, you can re-add the clips and they will be present in their entirety.
I know this is a collection of what a number of you have said, but I think this may narrow the issues down further.
I’ve been using the E1 plugin quite a bit. Similar to a couple of the comments above – I have only had a problem when importing many clips and then doing things in FCP at the same time. In that a case of number of clips did not import but they were clearly marked as such in the E1 plugin. So, I knew I had to go back and import them again.
I have not had any issues with clips doing partial imports.
This is w/1DM4 footage and an ’06 Mac Pro (2x 3 ghz dual core.)
These were the exact problems I had with my latest project using the t2i and the EOS FCP plugin. I thought I had bad takes where the camera failed or it turned off early that I swore I remembered getting. I had to go through clip by clip and match the clip lengths to make sure they transferred properly. A complete headache, but my project would have missed some of the best takes and moments had I not caught the mistake. I was a one man shoot with no slates and doing 6 plus takes per shot, sometimes rolling straight through so it’s easy to just think I screwed up. Getting partial files could have been a disaster. I really hope they fix it, but I may just have to use compressor instead until it becomes a trusted workflow again.
Thanks, Paul. That has been my understanding as well. It always freaks me out when I see an option that I don’t think relates to the operation, but I assume that is there because of the framework for writing a transcode plugin.
I don’t think the issue is related to system load because when this happens to specific clips it happens all the time. For instance if I delete the imported file from my scratch disk and the reference to that file from final cut and totally re-import a fresh copy it will cut it short at the same point, regardless of how much load there is on the system.
I suppose there could be something happening the first time that put incorrect meta data on to the clip.
Darryl,
Could you please explain what the use of the Remove Advanced Pulldown and Duplicate Frames is? Why would you select in when transferring 24p footage from a Canon? Thanks.
Hi Dave.
I may be wrong here as I’m not overly familiar with how pulldown works but I do have a rough understanding.
‘Advanced pulldown and duplicate frames’ relates to footage that has been captured to interlaced tape devices using a specific pattern to try and achieve 24p or other progressive frame rates that differ from the capturing systems native recording method.
I don’t think this actually relates to DSLR footage as it records true progressive frames, ie no pulldown or duplicate frames.
As I say, I may be wrong so please correct me somebody if I’ve misunderstood. I only had to work with tape based interlacing whilst owning my Canon XH-A1.
Hi Paul- I’m not sure how much you read dvinfo so I’m posting this here for you.
I’m glad I’m not the only one. I really think this is a widespread problem and many people are not aware that it is happening because the plugin gives no error indication.
After quite a bit of troubleshooting I think I discovered a workaround:
-In the Log & Transfer window, click the gear icon and select Preferences
-Click the ‘Clear Logging Autofill Cache’ and the ‘OK’
(you should also uncheck Remove Advanced Pulldown and Duplicate Frames if you’re not working with 24p).
Let me know if that works for you as well.
I found I also had that problem, if I did anything else on the computer while it was transferring. I now just leave the computer alone when its transferring, never had the problem again. However I also now transfer the files from the CF card to the hard drive and log and transfer from there, I don’t log & transfer direct from the CF card anymore. I do like the plug-in because it automatically can add a descriptive name to the original file name, which is very useful. I always also now do a quick double check that the durations match-up. I agree Canon should fix this bug.
Hello Paul. Hopping over to your blog from DVinfo.net
Very useful post you’ve made here. Thanks for sharing this.
I find it interesting that the plugin doesn’t seem to break if you set in/out. I hope that is a good work around. Because I mostly do doc work, with a lot of impromptu shots, most of my clips are long rolls with a lot of junk, so my workflow is to:
1) set i/o
2) name clip
3) add to queue
4) set another i/o on same clip
5) name clip
6) add to queue
7) rinse, lather, repeat
Until it can be verified that setting i/o works flawlessly for not truncating media, I will have to add the step of manually adding log notes on the clip length and verifying each transferred clip length against that log note. I think… unless you have another suggestion.
I do like the plugin’s intelligent timecode assignment. It does seem to create accurate code for each transferred segment of the original media. And that can be an added help when organizing or binning the transferred clip with 2nd system sound files. Pluraleyes does a nice job, but I find I’m always adding stuff in and syncing manually. For that reason I’ve been avoiding MPEG Streamclip.
Thanks for taking the time.
I have noticed in the past too that on a couple of occasions when a clip (for whatever reason) failed to import, and I added it to the queue again later on, the imported clip got a different number from the source clip (indeed the first available free number). Very weird, but it seems I’m not the only one. However, I haven’t noticed the pulgin not bringing in the whole clip – but this has got me worried. Not being able to trust such a crucial part of the workflow is killing. Thanks for pointing this out! I will either check all he imported clips manually from now on (I do like the plug-in), or go back tot using Streamclip. I just hope Canon will come with an update which will solve this issue!
Thanks Paul, incredibly useful info, I’ve just finished a big job where we were gathering B roll with a 5DMk2, luckily I opted to use Streamclip, as the B roll was not being logged had i used that plug in I may well have lost some stuff!!