Jean-christophe Boulay
Forum Replies Created
-
Hi Martin,
As you’re finding, real-time meters aren’t ideal when the delivery format having to conform to LKFS levels is not your main deliverable. When it is, you just mix to that level throughout. Most of the tools that will do what you want are far from cheap. A few hardware boxes from Dolby will do this in one go. If that interests you, contact a Dolby rep and they’ll be falling over themselves to sell you one.
Nugen Audio make two software options you might want to look into. LMB (bit.ly/qocx9t) is a stand-alone software solution with batch processing capabilities that will automatically correct the audio mixes you drop into it but costs a bundle, while LM-Correct (bit.ly/w1zlkZ) is an Audiosuite plugin that will process your selection in PT and costs “only” 400$. I’ve tried these out, but only on short-form content. I’m not 100% certain if they include Dialog Intelligence or what the workflow would be for long-form content. Remember that your LKFS measurement should be done on the “anchor element” (dialog) only for anything over 3 minutes long if you’re conforming to ATSC A/85 specs. I recommend getting in touch with Nugen with your situation, their service has been stellar in my experience.
To touch on Peter’s question of LKFS vs LUFS, they’re close but correspond to slightly different norms. LKFS is for ATSC (US & Canada) while LUFS is for EBU (Europe & North Africa). I think there are only slight differences in gating between the two norms.
IHTH,
JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com -
Hi Jeff,
While I don’t want to bite the “are color bars needed” stick, I can say that audio tone requirements will be going away. Or at least the necessity for their presence will.
With the implementation of the new ATSC practices for maintaining audio levels, almost all ingest systems now analyse the audio’s LKFS level and either assign correct dialnorm metadata to it or conform the level to the system’s preset metadata entry. All automatic. You are still expected to deliver material with appropriate dynamic range, as that is not corrected, but there will almost universally be one less human interaction with audio before it airs. One more reason to check those mixes before submitting.
I always try and include a separate color bar file with my first upload to a particular broadcast pipeline and ask that someone checks it to validate the workflow. Whether they do or not is another story. Pointing out the vagueness of the specs to them is always a plus as it’ll help future users and the specs may well have been written by a technically-less-than-literate employee and never been checked by techs.
IHTH,
JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com -
The MBox Mini has CoreAudio drivers on the Mac, so should work with any application that can use those, which is most of them. On the Windows side, you need to have the corresponding version of ProTools installed on the system for the drivers to work well with other applications. It should show up as an ASIO or WDM device. ASIO is better, of course. This should apply to the MBox2 Mini (squarish blue) and the newer MBox Mini (rounded dark grey). I don’t remember there being a first-gen MBox Mini.
Once the drivers work, it’s really up to you what software you use but these boxes are made for PT and should ship with ProTools Express at the minimum or in a bundle with ProTools.
This info is from a contact and I don’t have personal experience of it myself, so your mileage may vary.
IHTH,
JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com -
Hi Alex,
To elaborate on Peter’s answer with FCP open in front of me, in the “Hum Remover” filter you have a few controls. Frequency is not preset and can be scanned from 20Hz to 15kHz. You’ll still want to use either 50 or 60 to remove mains hum, depending on electrical standards where you recorded.
The “Q” setting determines how wide your removal notches will be and can be left to preset to begin with. “Gain” determines by how much you’re reducing the hum so go as low as you can without losing material you actually want to keep. It’s preset to maximum reduction.
Then you have selectable boxes to suppress each harmonic. You’ll probably have to tick the first one or two but beware that higher harmonics will start to cut into your real audio you probably want to keep and come in handy mostly with really nasty overdriven hum.
Once you’ve eliminated the hum, you can try to lower the “Q” setting until it starts rearing its head again, go back up a tad and you’ll have preserved as much of the original audio as possible while killing the hum.IHTH,
JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com -
Hi Pawel,
We do this daily with the Sony XDCam Transfer for FCP plug-in. You can download it from Sony’s Pro site (direct link: https://bit.ly/PekAHy). It’s listed as “PDZK-P1 XDCAM Transfer Software”. Once this is installed, a new option will appear in FCP’s Export menu entry, named “Sony XDCam”, rather appropriately. This will bring up a dialog where you can select the container (MXF in your case) and the settings.
I’ve only ever done this with HD material and you don’t specify anything about format, so your mileage may vary.
IHTH,
JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com -
Hi ajay,
OMF can only transmit mono tracks. Whatever software you start from or go to, what you’re seeing will happen. Logic will pan these mono tracks to center, so each is going to both channels of your stereo output.
The track order should be correct, meaning the first track of a stereo pair will contain the left channel and the second the right channel. You only need to create a stereo track beneath those two mono tracks, select the content of both mono tracks and drag it to the stereo track you created. You’ll have your stereo material back, panned correctly. There is always this extra organization step to do when importing from OMF.
IHTH,
JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com -
Hi Jean-Marc,
Have you tried changing the microphone’s sample rate outside FCP? If not, try closing FCP completely and open the Audio Midi Setup, found in the Applications/Utilities folder. What you see in there will depend on the devices connected, but usually you can set the sample rate for the device through there. That setting may very well overpower the setting in FCP. Then, you may be able to change in FCP, or maybe FCP will be “stuck” in 48kHz.
I’ve never worked with that particular device, but this would be my first guess as I’ve seen many similar situations with other devices. The mic may have come with its own control software also, where the setting could be changed. The manual doesn’t mention this at all, so I doubt it.
There is no way at all your computer configuration is limiting your sample rate. That’s pure bullocks.
IHTH,
JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com -
The actual hardware variations in the current iMac range are not very significant as far as deliverable horsepower is concerned. All the different CPUs are of the same i5 S-badged series which share the same topology, core count, voltage and caches. The only difference is a small ramp in clock speed which has had a smaller effect in the last few CPU generations than it did in the past. The 27-inch models can be upgraded to a hyperthreading-enabled i7 which FCPx should be able to exploit but how much of a difference that will really make, I have no idea.
The graphics power in them is not too dissimilar either. With GPUs ranging from 6750M to 6970M, you get a small ramp up in power along the range but nothing that will really affect an editing system in a meaningful way, especially if your software doesn’t exploit GPU power for floating-point calculations. The difference would be felt mainly by gamers and even then it’s not so big. And Apple isn’t even playing in the current 7000 series at all so you’ll be off peak power anyways.
All iMacs come with the same default memory configuration and all but the cheapest share the same HDD as well, so not much differentiation there either. You’ll probably want to upgrade RAM, but doing that through Apple when purchasing will cost more than getting your memory from another reseller. Installation is a snap. If money’s no big thing, the SSD option is nice but will not affect the editing experience that much. You’ll mostly go “oooh” the first few times you boot up.
Since the internals are not so different, the exterior plays a big part. If you can go for a 27-inch, do. They are beautiful machines. The smaller one will mostly work just as well but that big screen is sweet. If critical audio mixing is important, I recommend an exterior audio interface as the built in one is rather noisy, in my experience.
Personally, without knowing exactly your needs or means, I’d say the cheaper of the 27inch models with some RAM added is a nice sweet spot.
IHTH,
JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com -
Hi Tyson,
While I can’t help specifically with that recorder and software combo, the terminology and workflow can be cleared up a little.
A channel is one mono audio feed. A track is a container for audio, used to keep different sources separate. A track can contain any number of audio channels. For instance, a stereo audio track contains two audio channels, as it is comprised of a left mono feed and a right mono feed. A 5.1 audio track contains six audio channels. On most recorders, you can set up your tracks with differing numbers of channels.
Your problem is that your Scratch software can only read one track. That track can contain many channels. Any subsequent tracks are ignored. If you see two channels, it’s safe to assume his first audio track contained two channels, so it was a stereo track.
One solution to your issue would be to have the recordist assemble his different inputs into one multi-channel track, 5.1 or 7.1 for example. That would just be wrong, though, as that creates a nightmare for the audio guys and editors ingesting these recordings and, let’s be honest, their work has a more direct impact on the final result so should be the priority here.
Many recordists will record their different sources to specific tracks on the recorder but also do a rough mix on set and record that to a stereo track. If your recordist does do that and pipes the rough mix to the first track on the recorder, you should have all the audio that was recorded import into Scratch. You won’t get the separation but I think I understand you don’t really need that. You can output dailies with all the relevant audio and the audio guys get their separate tracks when the whole card is ingested for edit. Also, the editor has a complete audio landscape to work with easily by monitoring only the rough mix but the separate audio tracks follow his edits.
Off the top of my head, I think this would keep the workflow efficient for all departments and get everyone what they need. This may well be what the recordist is already doing, you’ll have to check with him. If I misunderstood the workflow, please elaborate a little.
IHTH,
JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com -
Hi John,
Your first instinct seems to be the best to me as well. If anyone besides an audio pro has a 5.1 audio setup, it’s bound to be attached to a DVD player. Any other medium is a lot less likely and some players may not play nice with stuff like DTS CDs.
Why the DVDs don’t work is what you should concentrate on then, IMO. Right up to getting into DVDSP, your workflow seems spot on. You should have all the files you need in the correct formats. If you were burning from inside DVDSP, you could try using the Build function instead and burn to disk using Toast, for example. I’ve been doing this for all DVDs for years after getting a few weird burns from DVDSP’s internal function. It’ll give you VIDEO_TS and AUDIO_TS folders you can burn using Toast’s “DVD-Video from VIDEO_TS” option under its Video tab, or similar function in whatever other burning software you may be using. The disks should not be burnt as Data DVDs with the VIDEO_TS folder on it; that won’t work.
You can open the VIDEO_TS from the Apple DVD Player app included in Mac OS before burning to check all is well with the build. If the folder tests fine and your disks don’t read, you’ll have troubleshot the problem a bit further as well.
If the problem lies in the DVDSP build, it could be that you have not set First Play on anything, though DVDSP will flag that in bright red while building. Still, I’ve seen it before.
IHTH,
JC Boulay
Technical Director
Audio Z
Montreal, Canada
http://www.audioz.com