But you're talking Music Videos for All God's Chillen'
Not quite. This particular church is Messianic English Protestant and the word Chillen' is definitely racist and implies a connotation not acceptable within their doctrine.
Not quite. This particular church is Messianic English Protestant and the word Chillen' is definitely racist and implies a connotation not acceptable within their doctrine.
Sorry... I am at a loss -- I just don't know how to respond to that comment.
Audio sync is only a fallback for missing time code sync & lock. Audio sync is rarely as frame accurate as a TC lock from a generator or a master camera.
There are two reasons for this: sound travels slower than light, and every camera has a slightly different frame and audio clock.
Cameras at different distances from a stage or speaker will each pick up sound at a slightly different time. It's not a problem if ALL cameras get the same master audio feed OR if you can use the sound from each individual camera whenever you switch shots.
But what often happens is you have one camera that gets clean sound from the board or source and the rest get whatever from on camera mics. So when you line up by sound all the cameras are several frames different and the words or music don't quite sync up.
Additionally, cameras, especially pro-sumer or low end pro often have slightly different clock timings, particularly with respect to audio samples. This can be seen in long takes (30+mins) where the audio and image appear to drift in and out of sync. This plays havoc with audio based line ups and they never stay lined up.
When you sync lock with a TC generator, every camera is on exactly the same clock and every frame and audio sample happens at the exact same instant. There is no question of lining up frames.
Having shot both ways, the a TC generator is always preferable. And yes, I have Plural Eyes and clapboards.
Yeah, I can see where a synchronized TC is superior,
However, for a lot less money and fuss, a sync sound can get you pretty close -- then if you can easily nudge the angles into sync a (a subframe at a time) you should be good to go.
I tried jam syncing 4 cameras with FCPX (Pany AVCHD, Sony HandyCam Mini Cassette, iPhone 4 and iPad 2.
It took about 5 minutes to get acceptable (to me) manual sync.
How accurate must the sync be (for 99% of multicam shoots), as you are cutting between cameras, possibly with a transition effect -- how close is close enough?
How accurate must the sync be (for 99% of multicam shoots), as you are cutting between cameras, possibly with a transition effect -- how close is close enough?
I think his point is that it needs to be dead on since you are most likely getting your audio from the booth not from the cameras. If the lip sync is off by a half frame it will be noticeable.
I believe it is time for Apple to get close enough!
Yeah, they should be able to make a decent go at it. Their file management doesn't help given that it creates shortcuts to referenced files, but still doable - they'd just need to create all those references when importing an XML file.
Even if they just get the in and out points of all the edits and put placeholders where it can't replicate what FCP did, it would help.
I think a lot of decisions Apple made in FCPX were quite sensible and the ones that aren't will naturally be some good marketing for their competition.
It's annoying for example that they default the import command to import from a camera but they have full keyboard customisation so you can switch it around so that command-i imports files.
Having limited FPS and resolution options will cause some people problems but they cover the most common resolutions and frame sizes. Nobody is shooting at 1024 x 768 so if you want to make footage for an iPad, work in 720p with safe zone overlays and then crop it on output. If you want 12fps video for a low bandwidth screen capture video, work in 25/30fps and export to what is needed.
What can be very annoying is the constant rendering. In FCP, you could work in DVC Pro HD or a host of different codecs natively. FCPX wants to render everything into ProRes - even ProRes files that don't match the project frame size. So say you have ProRes files that were cropped above a 720p frame size from 1080p, to allow for stabilisation, it tries to render the whole thing to ProRes again at exactly your frame size.
Not only this, it re-renders every change. If you take a small clip into FCPX in a 720p project, scale it up, let it render, scale it down, let it render and so on, it keeps render files for every change so you can end up with GBs of render files just piling up on your drive compared to tiny source clips. This is obviously to allow you to undo changes but still, if your files are in ProRes, they shouldn't need to be rendered every time for every change.
Imagine if you have 90 minutes of source footage that has a tiny black bar on top and you just want to scale it up and move it up to crop off the bar, it will put the entire timeline into orange and try to render the entire 90 minutes to disk and if the event goes on your internal, so do the render files.
Some people like native editing for everything, which some of the competition has and its ok but you can edit ProRes on any machine if you have fast enough drives and this isn't the case with codecs like AVCHD:
The fact that Apple developed ProRes shows they understand how best to deal with video files in a way that works for everyone. I personally would love if every file in the world was ProRes for capture and editing and H.264 for delivery and I get the impression Apple does too but it's never going to happen as they don't offer ProRes encoders for anyone but us special Mac users and that's where it won't get widely adopted.
Yeah, they should be able to make a decent go at it. Their file management doesn't help given that it creates shortcuts to referenced files, but still doable - they'd just need to create all those references when importing an XML file.
Even if they just get the in and out points of all the edits and put placeholders where it can't replicate what FCP did, it would help.
Absolutely! Then, they could show off the FCPX "To Do" markers to annotate what the import couldn't handle -- and the user could skip directly among clips needing attention.
Another thing the importer could do is pull in the render file corresponding to each clip that it couldn't handle, set corresponding in and out points -- then put it in the storyline above the clip, in a disabled state.
That way the user could enable it, see how it looked in FCP7, disable it, then try to accomplish an equivalent result with FCPX. In the project I manually imported, there were some Title Generators that aren't available in FCPX yet. I just got the render file, trimmed it to match the missing effect clip and was good to go.
Quote:
I think a lot of decisions Apple made in FCPX were quite sensible and the ones that aren't will naturally be some good marketing for their competition.
It's annoying for example that they default the import command to import from a camera but they have full keyboard customisation so you can switch it around so that command-i imports files.
Having limited FPS and resolution options will cause some people problems but they cover the most common resolutions and frame sizes. Nobody is shooting at 1024 x 768 so if you want to make footage for an iPad, work in 720p with safe zone overlays and then crop it on output. If you want 12fps video for a low bandwidth screen capture video, work in 25/30fps and export to what is needed.
What can be very annoying is the constant rendering. In FCP, you could work in DVC Pro HD or a host of different codecs natively. FCPX wants to render everything into ProRes - even ProRes files that don't match the project frame size. So say you have ProRes files that were cropped above a 720p frame size from 1080p, to allow for stabilisation, it tries to render the whole thing to ProRes again at exactly your frame size.
Not only this, it re-renders every change. If you take a small clip into FCPX in a 720p project, scale it up, let it render, scale it down, let it render and so on, it keeps render files for every change so you can end up with GBs of render files just piling up on your drive compared to tiny source clips. This is obviously to allow you to undo changes but still, if your files are in ProRes, they shouldn't need to be rendered every time for every change.
Whoa! How did you discover that? That sure can eat up a lot of disk space in a short amount of time -- especially critical on a portable.
I fooled around with this a bit and:
-- it doesn't render while wou are making the changes -- only after you stop for 2 seconds
-- it doesn't appear to re-render if you go to a setting where it has already rendered
But, I suspect that Apple will want to give the knowledgeable user some control over this: temporarily turn rendering off, set a longer interval, etc.
Quote:
Imagine if you have 90 minutes of source footage that has a tiny black bar on top and you just want to scale it up and move it up to crop off the bar, it will put the entire timeline into orange and try to render the entire 90 minutes to disk and if the event goes on your internal, so do the render files.
Some people like native editing for everything, which some of the competition has and its ok but you can edit ProRes on any machine if you have fast enough drives and this isn't the case with codecs like AVCHD:
The fact that Apple developed ProRes shows they understand how best to deal with video files in a way that works for everyone. I personally would love if every file in the world was ProRes for capture and editing and H.264 for delivery and I get the impression Apple does too but it's never going to happen as they don't offer ProRes encoders for anyone but us special Mac users and that's where it won't get widely adopted.
Yeah! There are other things to be aware of. I transcoded clips from my Panny AVCHD on import. A 3-minute clip was imported as is at 306.5 MB AVC1 (I think that AVC1 is a new codec designation for FCPX -- maybe AVFoundation).
The transcoded clip came in at 2.85 GB ProRes 422.
So, your event library has the originals and any transcodes.
I go above and beyond to find things to complain about. I actually just can't stand storage space being wasted so I make sure to know what programs are doing when they use their automated processing. You can watch the folders being created in your project render folder when a render starts.
Quote:
Originally Posted by Dick Applebaum
That sure can eat up a lot of disk space in a short amount of time -- especially critical on a portable.
Indeed it can and it also slows things down unnecessarily because writing to drive storage is most likely to slow down your system. It only does it when you aren't actively doing things in FCPX but it also does it when FCPX is in the background. This isn't good when it comes to Flash storage either like on a MBA.
Quote:
Originally Posted by Dick Applebaum
Yeah! There are other things to be aware of. I transcoded clips from my Panny AVCHD on import. A 3-minute clip was imported as is at 306.5 MB AVC1 (I think that AVC1 is a new codec designation for FCPX -- maybe AVFoundation).
The transcoded clip came in at 2.85 GB ProRes 422.
So, your event library has the originals and any transcodes.
It also keeps the proxy files there. Proxy files are quite good as they don't use as much space. If you use ProRes LT and work in that, it should save a decent amount of storage. It all comes down to bitrate. Consumer AVCHD cameras will give you maybe 9Mbps video so if you transcode to ProRes, which is 50-100Mbps, you are going to increase the storage use dramatically.
An interesting thing about AVCHD is that they are really just MPEG-4 clips in a different container (.mts), although there will likely be variants. You can actually rewrap them into a .MP4 container and they will play in Quicktime natively. There's an app here called rewrap2m4v:
You were wrong about this as you are now by repeating it. It's clear from Apples response and intent before releasing the product that the focus is as firmly on professionals as on anyone else interested in editing.
The point is that pro editing doesn't have to mean that it's awkward and difficult to do without thousands of hours learning it. It can be pro and intuitive and elegant at the same time. This makes it accessible for most people using iMovie and pro editors at the same time and unifies the editing software into one codebase.
J.
Pro is about how you cut, and who you cut for. iMovie was easy too, but I don't remember anyone talking about being paid to cut with it. You might as well open a photo retouching business using iPhoto. Or do desktop publishing in TextEdit. Simple.
As I understand if you are running Leopard or Snow Leopard Server you can use VMWare Fusion or equivalent to run Mac OS X Server as guest OSes. Correct me if I'm wrong...
"This License allows you to install and use one copy of the Mac OS X Server software (the "Mac OS X Server Software") on a single Apple-labeled computer. You may also install and use other copies of Mac OS X Server Software on the same Apple-labeled computer, provided that you acquire an individual and valid license from Apple for each of these other copies of Mac OS X Server Software."
I think the updated agreements say you can't run Snow Leopard Client as guests on a Snow Leopard Client Mac. Just Servers on Server. AFAIK. Not sure about running Snow Leopard Server as guests on a Snow Leopard Client Mac or vice versa. Any specific info on this, anyone?
Maybe this will all change in Lion. Any info on that?
Comments
But you're talking Music Videos for All God's Chillen'
Not quite. This particular church is Messianic English Protestant and the word Chillen' is definitely racist and implies a connotation not acceptable within their doctrine.
Not quite. This particular church is Messianic English Protestant and the word Chillen' is definitely racist and implies a connotation not acceptable within their doctrine.
Sorry... I am at a loss -- I just don't know how to respond to that comment.
Sorry... I am at a loss -- I just don't know how to respond to that comment.
No worries, just an FYI
Audio sync is only a fallback for missing time code sync & lock. Audio sync is rarely as frame accurate as a TC lock from a generator or a master camera.
There are two reasons for this: sound travels slower than light, and every camera has a slightly different frame and audio clock.
Cameras at different distances from a stage or speaker will each pick up sound at a slightly different time. It's not a problem if ALL cameras get the same master audio feed OR if you can use the sound from each individual camera whenever you switch shots.
But what often happens is you have one camera that gets clean sound from the board or source and the rest get whatever from on camera mics. So when you line up by sound all the cameras are several frames different and the words or music don't quite sync up.
Additionally, cameras, especially pro-sumer or low end pro often have slightly different clock timings, particularly with respect to audio samples. This can be seen in long takes (30+mins) where the audio and image appear to drift in and out of sync. This plays havoc with audio based line ups and they never stay lined up.
When you sync lock with a TC generator, every camera is on exactly the same clock and every frame and audio sample happens at the exact same instant. There is no question of lining up frames.
Having shot both ways, the a TC generator is always preferable. And yes, I have Plural Eyes and clapboards.
Yeah, I can see where a synchronized TC is superior,
However, for a lot less money and fuss, a sync sound can get you pretty close -- then if you can easily nudge the angles into sync a (a subframe at a time) you should be good to go.
I tried jam syncing 4 cameras with FCPX (Pany AVCHD, Sony HandyCam Mini Cassette, iPhone 4 and iPad 2.
It took about 5 minutes to get acceptable (to me) manual sync.
How accurate must the sync be (for 99% of multicam shoots), as you are cutting between cameras, possibly with a transition effect -- how close is close enough?
How accurate must the sync be (for 99% of multicam shoots), as you are cutting between cameras, possibly with a transition effect -- how close is close enough?
I think his point is that it needs to be dead on since you are most likely getting your audio from the booth not from the cameras. If the lip sync is off by a half frame it will be noticeable.
The engineer says: "I can get close enough".
I believe it is time for Apple to get close enough!
Yeah, they should be able to make a decent go at it. Their file management doesn't help given that it creates shortcuts to referenced files, but still doable - they'd just need to create all those references when importing an XML file.
Even if they just get the in and out points of all the edits and put placeholders where it can't replicate what FCP did, it would help.
I think a lot of decisions Apple made in FCPX were quite sensible and the ones that aren't will naturally be some good marketing for their competition.
It's annoying for example that they default the import command to import from a camera but they have full keyboard customisation so you can switch it around so that command-i imports files.
Having limited FPS and resolution options will cause some people problems but they cover the most common resolutions and frame sizes. Nobody is shooting at 1024 x 768 so if you want to make footage for an iPad, work in 720p with safe zone overlays and then crop it on output. If you want 12fps video for a low bandwidth screen capture video, work in 25/30fps and export to what is needed.
What can be very annoying is the constant rendering. In FCP, you could work in DVC Pro HD or a host of different codecs natively. FCPX wants to render everything into ProRes - even ProRes files that don't match the project frame size. So say you have ProRes files that were cropped above a 720p frame size from 1080p, to allow for stabilisation, it tries to render the whole thing to ProRes again at exactly your frame size.
Not only this, it re-renders every change. If you take a small clip into FCPX in a 720p project, scale it up, let it render, scale it down, let it render and so on, it keeps render files for every change so you can end up with GBs of render files just piling up on your drive compared to tiny source clips. This is obviously to allow you to undo changes but still, if your files are in ProRes, they shouldn't need to be rendered every time for every change.
Imagine if you have 90 minutes of source footage that has a tiny black bar on top and you just want to scale it up and move it up to crop off the bar, it will put the entire timeline into orange and try to render the entire 90 minutes to disk and if the event goes on your internal, so do the render files.
Some people like native editing for everything, which some of the competition has and its ok but you can edit ProRes on any machine if you have fast enough drives and this isn't the case with codecs like AVCHD:
http://forums.adobe.com/message/3631146?tstart=0
The fact that Apple developed ProRes shows they understand how best to deal with video files in a way that works for everyone. I personally would love if every file in the world was ProRes for capture and editing and H.264 for delivery and I get the impression Apple does too but it's never going to happen as they don't offer ProRes encoders for anyone but us special Mac users and that's where it won't get widely adopted.
Yeah, they should be able to make a decent go at it. Their file management doesn't help given that it creates shortcuts to referenced files, but still doable - they'd just need to create all those references when importing an XML file.
Even if they just get the in and out points of all the edits and put placeholders where it can't replicate what FCP did, it would help.
Absolutely! Then, they could show off the FCPX "To Do" markers to annotate what the import couldn't handle -- and the user could skip directly among clips needing attention.
Another thing the importer could do is pull in the render file corresponding to each clip that it couldn't handle, set corresponding in and out points -- then put it in the storyline above the clip, in a disabled state.
That way the user could enable it, see how it looked in FCP7, disable it, then try to accomplish an equivalent result with FCPX. In the project I manually imported, there were some Title Generators that aren't available in FCPX yet. I just got the render file, trimmed it to match the missing effect clip and was good to go.
I think a lot of decisions Apple made in FCPX were quite sensible and the ones that aren't will naturally be some good marketing for their competition.
It's annoying for example that they default the import command to import from a camera but they have full keyboard customisation so you can switch it around so that command-i imports files.
Having limited FPS and resolution options will cause some people problems but they cover the most common resolutions and frame sizes. Nobody is shooting at 1024 x 768 so if you want to make footage for an iPad, work in 720p with safe zone overlays and then crop it on output. If you want 12fps video for a low bandwidth screen capture video, work in 25/30fps and export to what is needed.
What can be very annoying is the constant rendering. In FCP, you could work in DVC Pro HD or a host of different codecs natively. FCPX wants to render everything into ProRes - even ProRes files that don't match the project frame size. So say you have ProRes files that were cropped above a 720p frame size from 1080p, to allow for stabilisation, it tries to render the whole thing to ProRes again at exactly your frame size.
Not only this, it re-renders every change. If you take a small clip into FCPX in a 720p project, scale it up, let it render, scale it down, let it render and so on, it keeps render files for every change so you can end up with GBs of render files just piling up on your drive compared to tiny source clips. This is obviously to allow you to undo changes but still, if your files are in ProRes, they shouldn't need to be rendered every time for every change.
Whoa! How did you discover that? That sure can eat up a lot of disk space in a short amount of time -- especially critical on a portable.
I fooled around with this a bit and:
-- it doesn't render while wou are making the changes -- only after you stop for 2 seconds
-- it doesn't appear to re-render if you go to a setting where it has already rendered
But, I suspect that Apple will want to give the knowledgeable user some control over this: temporarily turn rendering off, set a longer interval, etc.
Imagine if you have 90 minutes of source footage that has a tiny black bar on top and you just want to scale it up and move it up to crop off the bar, it will put the entire timeline into orange and try to render the entire 90 minutes to disk and if the event goes on your internal, so do the render files.
Some people like native editing for everything, which some of the competition has and its ok but you can edit ProRes on any machine if you have fast enough drives and this isn't the case with codecs like AVCHD:
http://forums.adobe.com/message/3631146?tstart=0
The fact that Apple developed ProRes shows they understand how best to deal with video files in a way that works for everyone. I personally would love if every file in the world was ProRes for capture and editing and H.264 for delivery and I get the impression Apple does too but it's never going to happen as they don't offer ProRes encoders for anyone but us special Mac users and that's where it won't get widely adopted.
Yeah! There are other things to be aware of. I transcoded clips from my Panny AVCHD on import. A 3-minute clip was imported as is at 306.5 MB AVC1 (I think that AVC1 is a new codec designation for FCPX -- maybe AVFoundation).
The transcoded clip came in at 2.85 GB ProRes 422.
So, your event library has the originals and any transcodes.
Whoa! How did you discover that?
I go above and beyond to find things to complain about.
That sure can eat up a lot of disk space in a short amount of time -- especially critical on a portable.
Indeed it can and it also slows things down unnecessarily because writing to drive storage is most likely to slow down your system. It only does it when you aren't actively doing things in FCPX but it also does it when FCPX is in the background. This isn't good when it comes to Flash storage either like on a MBA.
Yeah! There are other things to be aware of. I transcoded clips from my Panny AVCHD on import. A 3-minute clip was imported as is at 306.5 MB AVC1 (I think that AVC1 is a new codec designation for FCPX -- maybe AVFoundation).
The transcoded clip came in at 2.85 GB ProRes 422.
So, your event library has the originals and any transcodes.
It also keeps the proxy files there. Proxy files are quite good as they don't use as much space. If you use ProRes LT and work in that, it should save a decent amount of storage. It all comes down to bitrate. Consumer AVCHD cameras will give you maybe 9Mbps video so if you transcode to ProRes, which is 50-100Mbps, you are going to increase the storage use dramatically.
An interesting thing about AVCHD is that they are really just MPEG-4 clips in a different container (.mts), although there will likely be variants. You can actually rewrap them into a .MP4 container and they will play in Quicktime natively. There's an app here called rewrap2m4v:
https://public.me.com/catservant
You have to install ClipGrab from here:
http://clipgrab.de/en
Then you just drop a .mts onto the rewrap script and it creates a .m4v that you can open in Quicktime. It's handy for previews.
You were wrong about this as you are now by repeating it. It's clear from Apples response and intent before releasing the product that the focus is as firmly on professionals as on anyone else interested in editing.
The point is that pro editing doesn't have to mean that it's awkward and difficult to do without thousands of hours learning it. It can be pro and intuitive and elegant at the same time. This makes it accessible for most people using iMovie and pro editors at the same time and unifies the editing software into one codebase.
J.
Pro is about how you cut, and who you cut for. iMovie was easy too, but I don't remember anyone talking about being paid to cut with it. You might as well open a photo retouching business using iPhoto. Or do desktop publishing in TextEdit. Simple.
With Lion, you will be able to run multiple instances of OS X on a single computer.
But you can do that now...
Legally?
Leopard Server had it way back in 2008:
http://blogs.vmware.com/vmtn/2008/01...l-leopa-1.html
As I understand if you are running Leopard or Snow Leopard Server you can use VMWare Fusion or equivalent to run Mac OS X Server as guest OSes. Correct me if I'm wrong...
Back in 2007:
http://db.tidbits.com/article/9277
"This License allows you to install and use one copy of the Mac OS X Server software (the "Mac OS X Server Software") on a single Apple-labeled computer. You may also install and use other copies of Mac OS X Server Software on the same Apple-labeled computer, provided that you acquire an individual and valid license from Apple for each of these other copies of Mac OS X Server Software."
I think the updated agreements say you can't run Snow Leopard Client as guests on a Snow Leopard Client Mac. Just Servers on Server. AFAIK. Not sure about running Snow Leopard Server as guests on a Snow Leopard Client Mac or vice versa. Any specific info on this, anyone?
Maybe this will all change in Lion. Any info on that?
Actually why was this topic in this thread? LOL