Michael Gissing
Forum Replies Created
-
Apart from variations at the regional broadcaster level, there are two international broadcast standards, SMPTE & EBU. Ask your delivery broadcasters. They should all have published specs, many of them online.
SMPTE (NTSC countries)- tone of 1 Khz @ -20dbfs (which means decibels full scale on a digital meter). Peak levels should not exceed -10dbfs although the odd momentary spike to -8dbfs may be acceptable.
EBU (PAL countries)- tone of 1 Khz @ -18dbfs. Peak levels should not exceed -9dbfs. The BBC (bless them) however only accept peaks to -10dbfs.
To achieve this, sound post facilities use high quality peak limiters & compression to contain levels and achieve a balance within those parameters. The “correct” balance of dialog to music & FX cannot be measured or quantified in db. It is a mater of taste & experience and varies wildly according to program style, quality of dialog recording etc. As for Shane’s experience of checking the mix on small TV speakers, that frankly is a waste of time. There are so many variables with cheap small speakers that all you can do is get an idea of how it will sound on those speakers.
A much better idea is to use high quality studio monitor speakers which you will find at any decent sound post facility. Listen to the mix with the VOLUME TURNED DOWN. That is the best test of dialog to music & FX balance as it triggers the Fletcher Munsen characteristics of frequency perception to volume. If the dialog is clear at low volume it will work on most speakers.
Broadcast audio mixing in FCP or STP is not advised. They are useful tools but are usually let down by the fact that they are used in edit suites, not sound suites so both the operator and the monitoring are less than ideal.
Their other issue is that you can’t easily create stem mixes undipped of your music, dialog & FX elements which are standard deliverables. Sound is the area that is poorly understood by many picture oriented people. I am unusual in that I am primarily a sound post person that grades & finishes picture in FCP. Know your limitations and pass the job on to a pro for broadcast sound. I certainly do the same for graphics which is my limitation.
-
If you can find a sound post person, they may be able to help. There are a few devices that help diminish background sound.
You need good speakers, good EQ and a good operator. CEDAR DNS1000 is a dialog denoiser which can be quite effective. Waves make a denoising bundle. The STP tools are mentioned by some here but I have no need to use them as I do all my sound post on other systems.
I’ll bet that as well as poor location choice, your friend also cut corners by not having a pro sound recordist. If guess he doesn’t value the most important element in a film.
-
There are heaps of threads about recapturing from low res offline using Media Manager.
What you need to do is use Media Manager to convert the low res sequence to DV NTSC. If your camera originals are anamorphic 16:9 then that needs to be the selected sequence setting.
With your new Media Managed sequence, recapture your footage. The will be no need to rescale, nothing will be pixelated (apart from normal DV) and no resizing of width either.
Read the manual and do a thorough search of Media Manager threads in this forum to unearth all the information you need. With the price of hard drives these days there is never any reason not to capture DV native. It would be better if your partner remains MIA so this basic mistake is not repeated.
-
I have a Decklink HD PRO card and the downconverts from 1080 50i to SD are not great. Recently I compared the card, a nested re-render in an SD uncompressed timeline and the output of an HDW 1800 SD downconvert. The image quality of the HDW 1800 deck downconvert was a clear winner, with the software re render coming in second and the Decklink third.
All were checked on a variety of monitors from a 1080 Sharp Aquos LCD (45″) to a 14″ Sony CRT Broadcast monitor. Clearly on all monitors, the HDW1800 was the best. As I generally work with HDV or HDCam docos, the simple thing is to output to HDCam and crash record the digi beta at the same time.
That said, I am also specing a new system and the Kona 3 is the preferred interface at this stage.
-
720 x 576 is correct. That’s PAL SD anamorphic and it is meant to look squished.
However, if you want it to look unsquished 16:9 on a computer monitor then the magic numbers are 1024 x 576.
-
[Dale Larsen] “don’t see how to stop that in media manager.”
Did you select “Delete Unused Media”? It sounds scary but it doesn’t actually delete anything from your disk, just consolidates the information in the new sequence.
If you do that and set a handle size, it will just recapture the shots in the final edit with handles, not whole reels.
EDLs will not carry forward a whole lot of info so read the manual, experiment with a small sequence and you will get the hang of the Media Manager settings.
-
With the price of big sata drives these days, why not get a large system drive (500, 750, 1TB). Plenty of room for all the apps. Clone your old drive to the new one and then install FCS2
-
Use Jerry’s method and then in Photoshop you deinterlace.
-
[Michael Sacci] “Sorry but this is a Michaels only thread.”
Are you taking the Micky?
-
Interlaced will give your more motion information.
However, the real issue would be to stabilise in the camera, not later. If an expensive mount is not affordable, at least try the baby steadycam devices or an image stabilising lens.
Post stabilising is a compromise.