**You wouldn't use AGC on your recording why would you want to implement it on your mix after it is completed? **

So we agree about not imposing additional compression on an already good mix.

Amen.

"So we agree about not imposing additional compression on an already good mix."

ABSOLUTELY!!!

If you check my threads on this subject here they have ALL advocated setting the DRC to 'None'

Whether that is done in the AC-3 Studio or AC-3 Pro encoder is really up to the experience of the end user. However I also agree with you that using the AC-3 Studio encoder is a simpler, and therefore better, choice for less experienced users. Choosing the preset with AGC Off is the best choice there.

I personally like to have a little more choice and use the AC-3 Pro encoder so I can set some of the other parameters in the bitstream info such as Surround Mode, Downmix etc but that's just me.

New Hope Media

Does the Dolby Pro AC3 codec have a Auto Gain Control clause? Gain is a entirely different thing to dynamic range compression.

At the very least you should have the RF mode set to "music light". This is because in setups where the audio path goes through a RF modulator circuit the audio can over modulate and cause major distortion. This can result in lots of returned/recalled product.

It is worth noting that practically all decent broadcasters specify as a basic technical requirement the restriction of dynamic range to quite conservative levels.

But again, DRC is extremely pertinent to this question. The Dolby Pro codec is one that takes quite some navigating. Often specialists are brought in just to handle the single task of encoding in that codec correctly. It is very easy to ruin the whole mix with just one clause being set in the encoder. I sometimes wonder how many surround mixes that arerecorded in a genuine space are compromised by the settings that hard encode pro-logic compatibility, and hence shift the rears out of phase.

Robert

Up until now I've been nice but if you are going to continually post drivel you need to be told that you are.

Where in heaven is the DVD that is encoded with Vegas likely to be played into a system using RF modulators?

No broadcaster wants to receive a pre-encoded Dolby Digital stream on DVD for broadcast.

Dolby Digital is designed as a encode once and decode once format and the broadcaster needs to encode at the transmission stage when the digital signal, vision and sound, is muxed.

Because broadcasters encode at the transmission stage this means they require either discreet channels unencoded or Dolby E which is the transport medium in most broadcast environments for multichannel material destined for Dolby Digital transmission.

I've worked in broadcast TV as an audio post production specialist for thirty years, written the audio broadcast specifications for a major cable TV carrier in Australia and run courses on audio post production for both commercial and public broadcasters.

***"At the very least you should have the RF mode set to "music light".***

Do you seriously want your mix compressed a DRC system over which you have little or no control?

AGC and DRC aren't totally different in fact they are very similar in the end result. AGC is generally a fast attack slow release compressor/limiter. The DRC in the Dolby AC3 encoder/decoder may be more sophisticated than the AGC available in domestic audio equipment but in the end you have no control over how it is going to compress your mix other than choosing a small range of DRC types Film Light, Music Light etc.

****It is worth noting that practically all decent broadcasters specify as a basic technical requirement the restriction of dynamic range to quite conservative levels.****

Yes but they expect the mix to have been produced to those standards by the operator not by some automated process in the AC3 encoder/decoder.

Broadcasters do have analogue and digital broadcast limiters in place over their transmission chains using devices like Optimod etc that I personally abhor for the results they produce.

In the case of transmitting Dolby Digital the broadcasters themselves will set the AC3 encoder parameters at the point of transmission so the only type of Dolby Digital signal your likely to output with Vegas is going onto a DVD for replay in a DVD player NOT through a broadcast chain.

****I sometimes wonder how many surround mixes that arerecorded in a genuine space are compromised by the settings that hard encode pro-logic compatibility, and hence shift the rears out of phase.****

You obviously don't realise that this parameter is set for systems where the Dolby Digital mix is being down mixed into a system where no Dolby Digital compatible amp is present. If the mix is played into a Dolby Digital compatible amp the down mix doesn't happen and hence the phase shift required by Dolby Surround (Prologic) isn't applied.

Robert please check your facts before posting on this thread again.

Regards

New Hope Media

The mark of a true professional is being able to say unashamed, "I don't know."

Offering uninformed advice based on conjecture and fantasy is reckless and irresponsible, because newcomers are generally unable to thresh the wheat from the chaff.

At lease Newhope and most others can have intelligent disagreements based on fact and experience.

Let me interrupt this debate for a quick sec and ask an opinion or 2....

My HC3 cam shoots stereo (of course) at a data rate of 384k. As I mentioned above I usually take this stereo track and create a DD5.1 track from it. It is recommended that the AC3 data rate is normally 448 or above... Now because my original track was at 348, should it be left at that when creating the final AC3 or should it be upped to 448?

If it were me, I'd render at 448 just because it's the norm for DVD-video and DVD-audio and ATSC. Upsampling from 384 won't hurt your audio, won't help it either.
If you are rendering for BD, I believe it is 640.
;?)

Newhope,

I did not write suggesting that broadcasters would use DVD streams for broadcast. I was pointing out how broadcasters use AC3 themselves in delivery.

In both broadcast and in DVD player environments due consideration has to be given for legacy and unusual components in the interest of broad compatibility. For instance, I have a DVD player with an RF output in my house right now. It is a DVD/VHS combi player. So there is a clear example of where a DVD may be played through RF.

Furthermore, it is important to clarify where we are talking about tags and where we are talking about actual alterations the signal. No matter what you set your Line mode and RF mode DRC tags to it will not alter the actual encoded audio. If you are working to spec then you should have faith that the RF line mode DRC tag will only be applied across the RF output. There is no reason not to apply the RF mode line tag (and probably the RF overmodualtion tag too). It is essential to insure against poor audio across the RF output. This will have no effect on line mode.

Finally, there are two ways to handle Pro-logic compatibility. The default in the Pro AC3 codec presets is to shift the phase on the rears at the encoding stage. This does make an audible difference on playback in ALL systems, even in full dolby digital surround systems. It is not a tag. If you have recorded a real space with a mic for each surround speaker, this will mess up your mix.

The alternative mode sets a tag for pro-logic compatibility. The audio is encoded with the correct phase. The phase is shifted on the rears by the player when it is down mixed to stereo. I would recommend this for most mixes.

If we are talking about getting facts right maybe we should not be referring to things such as AGC and DRC as if they were the same thing. As far as I can see, my facts are pretty solid. I'd thank you to not talk so condescending to me. I do not think you would do that to my face, and you should not do that while cowering behind your computer screen either.

"If you are rendering for BD, I believe it is 640."

Really!?
I didn't know that .... just so happens that I am rendering for Blu Ray.

Thanks musicvid

Wow, I didn't think that it was going to end up being such a debate.

Anyway… I have been trying many things. Burned up ten DVD’s or so and this is where I am. Not 100% happy but...

Using AC3 Studio did not make any difference from Pro, actually it was worse. DRC set to none did not make any difference ether. What I have been reading is that it is a Meta tag for the user to match the program to their environment, i.e. if it is midnight they could put on the DRC so that the explosions doesn't wake everybody up and if they are on an jet liner they could use it so they could hear the crickets chirp during the quite part as well as hear the explosion all at the same volume level. I think the DRC depends on how you have your equipment set up; determines if it is used or not. I just don’t know but it doesn’t seem to make much difference on my project though.

Moving on...

The only thing that has helped is to put the DialNorm to -15db moved the LEF tracks down -6dbs and try to adjust the timeline items if any areas sounds loud after it is burned by moving the peak for those areas away from 0db to more like -3 dbs . Like I said it's not perfect but at least my windows are not blown out from going from a commercial DVD to a burned DVD.

It doesn't seem right to do it that way. Maybe my source audio is over compressed to begin with, or the logic of having everything up to 0 db isn’t the best thing to… particularly if the source material is overly compressed to begin with. I’ll keep trying different things out but I’m running out of steam…

Yes, it is mandatory the DVD players are able to play Dolby Digital 5.1 at 448Kbps. Blu-ray players are mandated to be able to play AC3 5.1 at 640Kbps.

I would like to confirm this limitation based on what I’m experiencing. For a DVD, I can only use AC3 5.1 or AC3 stereo at bit rate of 448 Kbps or 192 Kbps respectively and not anything out side of that like 3.1 with a lower than 448 bit rate without it being re-rendered for DVDA pro 5?

****Maybe my source audio is over compressed to begin with, or the logic of having everything up to 0 db isn’t the best thing to…****

You shouldn't be trying to have your 'average levels up at 0dB just the maximum program peaks. It appears that you have moved the average level of your mix up closer to the maximum permitted level of 0db.

You are, in effect, doing what operators mixing television commercials do. They use heavy compression and limiting to raise the average levels of their mix while still keeping the peaks under the maximum permitted level. This is why TV commericals sound louder than program, or historically have until broadcasters and governments started introducing regulations to reduce this trend.

If you have to change your dialnorm down to -15 then this points directly to the problem. Your average levels are too high.

I'd try mixing your program with less compression and averaging the dialogue (assuming it has dialogue and not just music based) between -24dB for quiter dialogue to maximum peaks of -10db. You can then allow an sounds that are meant to be occasionally extremely loud, explosions etc (depending on your content) to peak up to close to 0dB.

Alternatively if don't want to remix ,and your mix is going to stay as compressed as it appears to be from the information in your last post, try dropping the maximum peaks for the whole mix down to typical broadcast standards for non Dolby Digital program peaks which is -10dB.

I would then set the dialnorm back to around -25 to -27.

My reasoning is that all you are doing by setting the dialnorm to -15 is dropping the overall level of your mix by 12db in the AC-3 encode.
If you are going to do that you may as well do it during the mix prior to AC-3 encode.

The first alernative of remixing and reducing your overall compression will give you a mix with more dynamic range, which is one of the potential benefits of Dolby Digital tracks.

However this total dynamic range is intended primarly for controlled listening environments such as movie theatres. As you have pointed out the DRC settings in the AC3 encoder are meant for the end user in situations where the listening envirnment isn't as quiet as movie theatre. I personally avoid having them set and allow the end user to choose to implement them in their own system if they want.

You need to make the choice based on your potential end audience and design your mix to meet the needs of the program and its potential end use or replay conditions.

So if a movie theatre isn't where you see the program being replayed then tailor the mix and it's dynamics to suit.

New Hope Media

Robert
****I do not think you would do that to my face, and you should not do that while cowering behind your computer screen either. ****

There you go again posting before you know the facts.... I have no problem disagreeing with people to their face. Just don't have the opportunity in your case....but you keep adding DRC to your encodes if you want thats the beauty of having choice.

New Hope Media

Thanks newhope. I have been wondering about that from the start, and I agree it doesn't seem right about what I have to do with the dialnorm.

It does almost seem like watching an hour long commercial. LOL Just kidding it's not that bad, but close.

You put this condition in there that leads me to my next question.

"(assuming it has dialogue and not just music based)”

My project is probably about 90% music, 9% dialogue and 1% FX' like explosions, etc...

If the dialog should be -24db to -10 db what should the music levels range be?

Another question is. If the source is already compressed, can it be decompressed?

Thanks,

**If the dialog should be -24db to -10 db what should the music levels range be?**
There is no "should be," just some general parameters controlled by your hearing and taste in music. Maybe -18dB to -6dB average (RMS) would be a good starting range? There are all kinds of music, including a few that barely deserve the designation.

**Another question is. If the source is already compressed, can it be decompressed?**

Not in Vegas AFAIK. There are some "unlimiter" plugins out there, but they're not terribly effective from what I've read.
Sounds like you're not using your own material here, but something that's already been processed.

Why don't you make a few short (10 sec.) clips of your original source music, dialog, and effects available (WITHOUT any processing, of course), and let some experienced eyes and ears have a look-see. You might be surprised what you can find out. And no, pushing everything up to 0dB average program level is not the way to go. In order to render dynamic range you have to mix dynamic range. I thought we covered that some time ago.

"If the dialog should be -24db to -10 db what should the music levels range be?"

There is no specific level for the relationship between music and dialog. This is a situation where the ears come into play.

And, by the way, there is no way that the average level can be 0db without the program having excruciating amounts of distortion.

***And, by the way, there is no way that the average level can be 0db without the program having excruciating amounts of distortion.***

I don't think anyone meant the average was at 0dB but the average was closer to 0dB than a normal (sic commercial DVD) mix when directly compared.

I totally agree with your answer about using the ears to judge the balance between dialogue, music and effects. It is the only way you can do it.

In the end your audience doesn't use meters to appreciate your mix they are using their ears.

A quick note on the music content though. A lot of prerecorded music, commercial, production etc comes heavily compressed already.

So if you are sourcing music off CDs, and not mixing it yourself from original recordings, you may find that the compression in the music, particularly 'popular' genres is unavoidable.

New Hope Media

Like I've said from the beginning, this appears to be a levels issue rather than an encoder issue. You need to fix your mix. Take -15db out of each of the master bus channels and that should resolve the issue while preserving your actual relative mix.

I'd like to inject a somewhat-related experience.

I recorded an outdoor singing performance in surround sound on a very windy day. The first prototype DVD made was with just the two front channels, and all was well. The final version, however, included the rear two channels also (encoded as DolbyDigital 2/2). These rear channels contained a significant amount of wind noise and to my horror the final DVD audio was a disaster, with the singers nearly muted every time the wind level rose.

It turns out that my rear audio channels on the Vegas timeline showed levels significantly above 0db on those wind peaks. I discovered the Dolby encoder in Vegas would try to salvage the final result by reducing the levels of ALL tracks, including the front ones, during the wind peaks; in effect, DD had a built-in AGC "feature" that was barely documented, in my view.

I went back to the Vegas timeline, put a hard limiter on the rear audio tracks, and then re-encoded. This fixed the problem, but it meant I had to destroy the original batch of finished DVDs and recall those that had already gone out. Two hours of content and only 3 minutes of crummy sound towards the end, but the moral of the story is to not assume anything and carefully proof the master before turning the production crank.