Straight from the Gh1 to your door

trouble importing clips into FCP 6.0.5

trouble importing clips into FCP 6.0.5

I get "Unsupported media or invalid structure" warning message. Any suggestions?

THX.
 
FCP's kind of a lost cause unless you have all the directories that go around the .MTS clips. Then you still end up transcoding to ProRes or AIC or whatever. Cineform Neo Scene is a pretty nice workaround for now, if I end up getting a GH1 I'll probably buy it. Either that or see what Apple does in the next FCP release
 
Hang tight. Was in town tonight shooting under the general ambiance of Sakae (Nagoya, Japan)... on the LCD on review most shots seemed pretty nice. Will upload on Wednesday, if Daniel's server space is still available. For all you FCP users that need the 'Private Folder', I will upload as such. In most cases I was at 1600 ISO, wide, AWB (just playing for now), FHD and SH (1080 and 720) shot in Manual mode...sometimes AFS, AFC and M but will give you all the appropriate info when I upload. Please note that these will be straight from the cam without any editing or transcoding (despite the fact that I may change the file names for recognition)
From what I can tell (and this has been reiterated a few times) it seems that this little cam is quite exceptional for 123,000 yen...the form factor truly makes me smile and that's just the start.
 
I have a theory. I think that the camera is dropping frames and messing up the 3:2 cadence. The interlacing is all over the place and definitely part of the reason for some of the macroblocking. Hmm, I hope this gets fixed, but I bet it wont. Also, this may be caused by using SDHC media that isn't fast enough to write the data. Keep in mind, even though SDHC class 6 is fast, when it is completely full, it has to rewrite blocks and this will be slower than when the card is completely empty. At least this is something to try before dismissing this as a cmaera fault.
 
Oedipax said:
FCP's kind of a lost cause unless you have all the directories that go around the .MTS clips. Then you still end up transcoding to ProRes or AIC or whatever. Cineform Neo Scene is a pretty nice workaround for now, if I end up getting a GH1 I'll probably buy it. Either that or see what Apple does in the next FCP release
Click to expand...

Interesting. So you use NeoScene to create the ProRes file and then bring that into FCP?

I have Premiere CS4 installed on my Mac, but I've never used it. Maybe I'll have to now? (Apple needs to stop pouring all their manpower into the iPhone and instead give some lovin' to Final Cut Studio!!!).
 
plasmasmp said:
I have a theory. I think that the camera is dropping frames and messing up the 3:2 cadence. The interlacing is all over the place and definitely part of the reason for some of the macroblocking. Hmm, I hope this gets fixed, but I bet it wont. Also, this may be caused by using SDHC media that isn't fast enough to write the data. Keep in mind, even though SDHC class 6 is fast, when it is completely full, it has to rewrite blocks and this will be slower than when the card is completely empty. At least this is something to try before dismissing this as a cmaera fault.
Click to expand...

You might have a good point on the SDHC card speed, as this doesn't seem to be a consistent problem, maybe some further tests on faster/higher spec cards could throw some light on this.
 
plasmasmp said:
I have a theory. I think that the camera is dropping frames and messing up the 3:2 cadence. The interlacing is all over the place and definitely part of the reason for some of the macroblocking. Hmm, I hope this gets fixed, but I bet it wont.
Click to expand...

I just need to point this out as being THEORY AND NOT FACT. Sorry, but statements like this have a tendency of being quoted enough times and then being taken as fact. No frames are being dropped. I have found that to properly reduce the pulldown I've only been successful from importing the original "Private" file into final cut with prores, and then removing pulldown with either Compressor or JES Deinterlacer. I've actually found that JES looks a little better than compressor.

Removing pulldown from just the MTS clip with no metadata from the original folder has been damn near impossible, but with the original folders intact it works as it should. You wouldn't import P2 by stripping the folders apart, and you need to take the same approach to AVC-HD. You remove that original file from the folder structure and you lose all the metadata for it.
 
SLoNiCK said:
I've noticed a lot of interlace artifacts even after pulldown in "countryside...1080". Does it mean that coder is unaware of interlaced structure and simply compress video on frame-by-frame, not field-by-field basis?
Click to expand...
It is not a surprise at all. Interlaced encoding shouldn't be used for a progressive output -- that's for sure!

Take a glance on the Adam Wilt's findings posted here:

http://provideocoalition.com/index.php/awilt/story/review_canon_vixia_hf11_avchd_camcorder/P1/



For this reason, I would avoid any camera using this method as same as new Canon VIXIA lineup, except HV40. Even based on the old MPEG-2 HDV.
 
Check out Adam Wilt's comments (scrolling down the page) on such similar advanced pulldown removal methods. It simply doesn't work it out as we wish. Interlaced and progressive are distinct worlds, as simple as that.
 
So I watched the 1080/720 Countryside and Train Jello Tests on my 102" screen.

First of all I want to make a quick comment- it's asinine to jump to any conclusions or make any definitive assessment of camera performance based on the 'countryside' clips! No overall assessment can be made on the GH1 from these clips (Sadly, it seems that someone did just that on a different forum I read today).

Ok, back to my opinion- which as it turns out is essentially the same as Ian-T said earlier; playing at normal speed this artifact is NO worse than another long GOP codec, found in some cameras commonly available around $5K USD. Which certainly looks better than the MPEG4 codec faults in my HD DirecTV DVR!

Watching all this stuff at 60i/60p for the first time, the artifacts didn't make me fall out of my chair and clamor that either the 1080 or 720 was unusable. The GH1 has so many pluses in the image quality, ergenomic and (especially) price categories, over other frequently mentioned cameras, I think this will be a non-issue for the vast majority of GH1 owners, especially those that know how to shoot.

Yes, the same artifacts are present in the Countryside 720 just as in the 1080 version. In case anyone has any doubt, pause when bright/contrasty foliage passes close to the cabin window, and step through frame by frame- same codec faults.

Watching a portion of each "Countryside" clip (720/1080) frame by frame during a few seconds with a lot of motion- that is- handheld, seeking focus, exposure compensating, extremely fast moving (close to camera) foliage, was the most egregious. When all that high-frequency detail zooms past the camera, there's really not much that can be done. However, the good news is that is that it's EXTREMELY unlikely someone would choose to shoot/playback this way for creative purposes- that is stepping through material shot like this- frame by frame.

I'm slightly more concerned about the codec faults on the wider shots with less up-close foliage. However, based on what I'm seeing during these moments, I think that all the rapid scenic change combined with continuous "auto" mode adjustments during the shot significantly contribute to the overtaxing of the codec. It'll be interesting to do a small test to compare codec tollerence in manual vs. auto. I've also noticed "auto" mode contributing to codec problems in HDV.

I watched the "Lemon Soju" clip 0006 on the big screen, while not particularly detailed- you get some idea of what's happening with the running commentary track. As further eveidence of my theory, each time the camera operator suddenly changes shutter speed/exposure during the 'out the window' shot, there are momentary codec faults in some areas of the image- subtler than the train obviously; but still present. These same codec faults are not immediately apparent when the camera is just experiencing handheld motion and concurrent motion from the subjectmoving across the frame. Testing could reveal if "auto" mode is a contributing factor of codec fault, as I said- I suspect it is.

A couple questions maybe SonicStates can answer which could help a little to understand what's happening. On the 1080 Countryside- Shutter speed looks to be around 250? Was this camera in Auto, AP or SP? OIS turned on or off? Any idea about the the detail/sharpness/ISO settings?

Anyone have any other examples like this? I'm interested to see them.

I meeting up with someone who has a GH1 and plan on doing a few quick shots this weekend. Probably no codec "testing" going on, just beauty shots and that's fine by me.

I've seen some original .mts files that haven't (actually can't) been posted on-line and I'm feeling confident that when the GH1 is shot poorly, it looks like other cameras/formats shot poorly- crap. However, when it's shot well, the camera yields result similar to MUCH, MUCH more expensive HD cameras, so the GH1 is probably the greatest bang for buck camera on the planet right now, and takes fantastic stills to boot. When GH1 material is transferred to 35mm, I think the feel will be quite similar to a S16 blow-up, although this remains to be seen. I'm working with two professional facilities in Hollywood right now to facilitate a quick D.I./35mm film-out test. I'll keep this forum posted as results roll in.


I.

Illya Friedman
Hot Rod Cameras
 
Illya Friedman said:
A couple questions maybe SonicStates can answer which could help a little to understand what's happening. On the 1080 Countryside- Shutter speed looks to be around 250? Was this camera in Auto, AP or SP? OIS turned on or off? Any idea about the the detail/sharpness/ISO settings?

Illya Friedman
Hot Rod Cameras
Click to expand...

I must confess that as this was an outing with my wife I kind of pointed out the window for the few seconds that you see and pressed the button. It was in manual mode, the OIS was off (I'm relatively certain), the white balance was set to auto, F stop and ISO I can't remember but the Intelligent contrast was off.
I have uploaded some more 'raw' footage for people to have a look at and this time I can provide you with the details...just waiting for Daniel L, who has provided the server space and bandwidth so generously, to give me the green light and then it's all yours. Shouldn't be too long....night time footage in the PRIVATE folder as requested.
 
Fantastic report Illya please keep us up to date on what you find. This is encouraging news. The film transfer will be such a benefit that i hope you do this soon.

Sonicstates thanks for the all the effort in the uploads.. Please make sure Illya knows when you've uploaded the new files.
 
Illya Friedman said:
So I watched the 1080/720 Countryside and Train Jello Tests on my 102" screen.

First of all I want to make a quick comment- it's asinine to jump to any conclusions or make any definitive assessment of camera performance based on the 'countryside' clips! No overall assessment can be made on the GH1 from these clips (Sadly, it seems that someone did just that on a different forum I read today).

Ok, back to my opinion- which as it turns out is essentially the same as Ian-T said earlier; playing at normal speed this artifact is NO worse than another long GOP codec, found in some cameras commonly available around $5K USD. Which certainly looks better than the MPEG4 codec faults in my HD DirecTV DVR!

Watching all this stuff at 60i/60p for the first time, the artifacts didn't make me fall out of my chair and clamor that either the 1080 or 720 was unusable. The GH1 has so many pluses in the image quality, ergenomic and (especially) price categories, over other frequently mentioned cameras, I think this will be a non-issue for the vast majority of GH1 owners, especially those that know how to shoot.

Yes, the same artifacts are present in the Countryside 720 just as in the 1080 version. In case anyone has any doubt, pause when bright/contrasty foliage passes close to the cabin window, and step through frame by frame- same codec faults.

Watching a portion of each "Countryside" clip (720/1080) frame by frame during a few seconds with a lot of motion- that is- handheld, seeking focus, exposure compensating, extremely fast moving (close to camera) foliage, was the most egregious. When all that high-frequency detail zooms past the camera, there's really not much that can be done. However, the good news is that is that it's EXTREMELY unlikely someone would choose to shoot/playback this way for creative purposes- that is stepping through material shot like this- frame by frame.

I'm slightly more concerned about the codec faults on the wider shots with less up-close foliage. However, based on what I'm seeing during these moments, I think that all the rapid scenic change combined with continuous "auto" mode adjustments during the shot significantly contribute to the overtaxing of the codec. It'll be interesting to do a small test to compare codec tollerence in manual vs. auto. I've also noticed "auto" mode contributing to codec problems in HDV.

I watched the "Lemon Soju" clip 0006 on the big screen, while not particularly detailed- you get some idea of what's happening with the running commentary track. As further eveidence of my theory, each time the camera operator suddenly changes shutter speed/exposure during the 'out the window' shot, there are momentary codec faults in some areas of the image- subtler than the train obviously; but still present. These same codec faults are not immediately apparent when the camera is just experiencing handheld motion and concurrent motion from the subjectmoving across the frame. Testing could reveal if "auto" mode is a contributing factor of codec fault, as I said- I suspect it is.

A couple questions maybe SonicStates can answer which could help a little to understand what's happening. On the 1080 Countryside- Shutter speed looks to be around 250? Was this camera in Auto, AP or SP? OIS turned on or off? Any idea about the the detail/sharpness/ISO settings?

Anyone have any other examples like this? I'm interested to see them.

I meeting up with someone who has a GH1 and plan on doing a few quick shots this weekend. Probably no codec "testing" going on, just beauty shots and that's fine by me.

I've seen some original .mts files that haven't (actually can't) been posted on-line and I'm feeling confident that when the GH1 is shot poorly, it looks like other cameras/formats shot poorly- crap. However, when it's shot well, the camera yields result similar to MUCH, MUCH more expensive HD cameras, so the GH1 is probably the greatest bang for buck camera on the planet right now, and takes fantastic stills to boot. When GH1 material is transferred to 35mm, I think the feel will be quite similar to a S16 blow-up, although this remains to be seen. I'm working with two professional facilities in Hollywood right now to facilitate a quick D.I./35mm film-out test. I'll keep this forum posted as results roll in.


I.

Illya Friedman
Hot Rod Cameras
Click to expand...
Your optimism is encouraging. I've seen some really interesting stuff coming from Philip Bloom, for example.

But going with any film-out test, do NOT forget to test the codec in every AVERAGE situation the codec faces out there. Not only those kind where it is possible to state this camera is able for the job.

If you won't be minimally scientific with your experiments, what for they serve after all?

I'm referring independent tests for sure, not good marketing on brand advertisement for camera sales.
 
Emanuel said:
Your optimism is encouraging. I've seen some really interesting stuff coming from Philip Bloom, for example.

But going with any film-out test, do NOT forget to test the codec in every AVERAGE situation the codec faces out there. Not only those kind where it is possible to state this camera is able for the job.
Click to expand...

Emanuel,

I am a bit of a veteran of film-out tests. When I ran the Camera Rental department at Moviola in 2002, I did one of the earliest comprehensive prosumer/professional digital camera to 35mm transfer test @ CFI. 9x cameras across several formats including SD/HD and 16mm D.I. to 35mm test. Since that time I've been involved in one way or another with about a dozen film out tests- in all modern formats, including uncompressed 4K.

I.

Illya Friedman
Hot Rod Cameras
 
I should say to clarify is was 16mm two ways actually- and neither the way that people think of today as D.I. 16mm was actually telecine to HDCam and back to 35mm, and photochemical/optical.

I.
 
Emanuel said:
Your optimism is encouraging. I've seen some really interesting stuff coming from Philip Bloom, for example.

But going with any film-out test, do NOT forget to test the codec in every AVERAGE situation the codec faces out there. Not only those kind where it is possible to state this camera is able for the job.

If you won't be minimally scientific with your experiments, what for they serve after all?

I'm referring independent tests for sure, not good marketing on brand advertisement for camera sales.
Click to expand...

I agree wholeheartedly with you Emanuel. Do everything, not just static camera shots.

Illiya I also completely respect your opinion and I'm very happy that someone has an opposite to mines, it helps people to know that one person can't be "right", only opinion so 1080/24 will really suffice for some.

I'm absolutely, positively okay with that and happy about it. It'll be really great to see filmout results, not only, but as well HF100, A1 and all the comparisons that are set to come next month when it hits the states
 
IIlya some of us know you are more than capable of doing the right tests without having to be told how. And again I look forward to seeing what you come up with great anticipation. Please don't be distracted by every 2 center.
 
You must log in or register to reply here.
Back
Top