or Connect
AVS › AVS Forum › Audio › Receivers, Amps, and Processors › 0dBFS + 10dB Handling in BFD And Pro Amps
New Posts  All Forums:Forum Nav:

0dBFS + 10dB Handling in BFD And Pro Amps

post #1 of 13
Thread Starter 
Can anyone explain how BFD and Pro Amps handle any signal gretaer than 0dBFS (105dB) when LFE is mixed 10dB hotter than normal signal? What confuses me is if mains are doing 105dB at dolby reference when the dynamic peaks kick in, LFE should be at 115dB in the recording mix of movie. So, if BFD and Pro Amps clip any signal that is greater than 105dB then how can subwoofers playback at insane SPL levels i.e. 120-130dB??

I posted it in DIY Speakers & Subs section, but not getting definitive answer. I hope somebody here can clarify the confusion. Thanx
post #2 of 13
LFE is not recorded at 115dB when the movie is mixed. LFE is recorded 10dB low to avoid clipping, with the expectation that it will be boosted 10dB later on by the processor. That happens at the amplification stage.

Here's a link to a full explanation in the first post of this sticky thread in the Audio Theory forum:
http://www.avsforum.com/t/748147/lfe-subwoofers-and-interconnects-explained/0_50
Edited by BIslander - 5/9/13 at 1:36am
post #3 of 13
Quote:
Originally Posted by braveheart123 View Post

Can anyone explain how BFD and Pro Amps handle any signal gretaer than 0dBFS (105dB) when LFE is mixed 10dB hotter than normal signal? What confuses me is if mains are doing 105dB at dolby reference when the dynamic peaks kick in, LFE should be at 115dB in the recording mix of movie. So, if BFD and Pro Amps clip any signal that is greater than 105dB then how can subwoofers playback at insane SPL levels i.e. 120-130dB??

I posted it in DIY Speakers & Subs section, but not getting definitive answer. I hope somebody here can clarify the confusion. Thanx

0dbfs refers to the maximum signal the player should send the AVR. The avr then sends a pre-out voltage down stream based on the 0dbfs signal from the player. The 0dbfs signal does not go to amps and EQs they only see input voltages. This voltage is determined by how you have your gain structure setup.

IOW you should have it setup so that when the player sends a 0dbfs signal to the avr that the ave doesn't send too much preout voltage downstream causing clipping.
post #4 of 13
another way to see it is that 0 dBFS is the maximum digital level you can capture. Above 0 dBFS causes digital clipping, and the max level will still be 0dBFS, just with a messed up frequency content. Because digital has this kind of absolute ceiling, levels are referenced to 0 dBFS.

0dBFS is the max you can encode on CDs, DVDs and BluRays. It also is the max that you can encode on digital recorders or digital mixers. I think it's fair to think of 0dBFS as that singla level where the top of the waveform is all 1s digitally, whether it's 16 ones on a CD or 24 ones on a BD or inside a recording machine.* In other words, it is mathematiclaly impossible to encode the LFE channel so that it reaches a peak of +10 dBFS. Just like it's impossible to put a pound and a half of salt in a one pound box. So because the LFE channel might have to keep up with, at least theoretically, 5 other channels at 0dBFS for brief peaks, the LFE channel is mixed and encoded with the understanding that it will be turned up 10 dB more than the other channels when played back. That gives you an additional 10 dB of headroom in teh LFE channel, but the max encoded to the channel is still 0dBFS. It just gets the expected/required 10 dB boost on playback (and during recording/mixing/mastering, so that levels come out correctly.)

*Binary is weird looking. It is the same thing as saying I have a system that can capture numerical values but it has only three places. 000 is the lowerst it can go. 999 is the higheset it can go. I can't encode 1000 because I don't have four places. Digital systems are like this. Their max is defined by the number of bits . . .
post #5 of 13
Thread Starter 
Quote:
Originally Posted by JHAz View Post

another way to see it is that 0 dBFS is the maximum digital level you can capture. Above 0 dBFS causes digital clipping, and the max level will still be 0dBFS, just with a messed up frequency content. Because digital has this kind of absolute ceiling, levels are referenced to 0 dBFS.

0dBFS is the max you can encode on CDs, DVDs and BluRays. It also is the max that you can encode on digital recorders or digital mixers. I think it's fair to think of 0dBFS as that singla level where the top of the waveform is all 1s digitally, whether it's 16 ones on a CD or 24 ones on a BD or inside a recording machine.* In other words, it is mathematiclaly impossible to encode the LFE channel so that it reaches a peak of +10 dBFS. Just like it's impossible to put a pound and a half of salt in a one pound box. So because the LFE channel might have to keep up with, at least theoretically, 5 other channels at 0dBFS for brief peaks, the LFE channel is mixed and encoded with the understanding that it will be turned up 10 dB more than the other channels when played back. That gives you an additional 10 dB of headroom in teh LFE channel, but the max encoded to the channel is still 0dBFS. It just gets the expected/required 10 dB boost on playback (and during recording/mixing/mastering, so that levels come out correctly.)

*Binary is weird looking. It is the same thing as saying I have a system that can capture numerical values but it has only three places. 000 is the lowerst it can go. 999 is the higheset it can go. I can't encode 1000 because I don't have four places. Digital systems are like this. Their max is defined by the number of bits . . .

That's a revelation for me guys, thanx for clarification. I was under the wrong impression that LFE is mixed 10dB hotter in the movie content. I was so wrong.

So, you guys are saying -10dB LFE offset is already accounted for in movie content during recording to avoid clipping in the signal processing stage in AVR and compensation is not applied up untill the amplification stage. I have iNuke 6000 DSP and um using it to power a ported DIY sub. My AVR is denon 3313. I take the pre-out signal (LFE+Redirected Bass) from avr and feed it to iNuke. How does iNuke know that the input signal is LFE+Redirected Bass and it has to apply 10dB boost at the amplification stage??
I hope um making sense or am I completely crosseyed??

I am asking this coz I manage to clip iNuke at -13dB (Reference Scale) on the avr master volume when dynamic peaks are playing. I know the gain structure and how it works and iNuke is properly gain matched. Sub trim level on avr is at -7.5. The problem is if I reduce it further in avr, I lose all the thump. And if I continue to keep the same sub trim level in avr, it gets clipped. What wrong am I doing??
Edited by braveheart123 - 5/9/13 at 2:06pm
post #6 of 13
BH,

I know exactly what your up against with your sub trim. When i had mine set that low the was very on/off and not very powerful with little resolution.

I assume you have the gain knobs wide open on the inuke, try raising the trim on the subs on the avr to close to 0 and attenuate the signal with the gain knobs on the inuke to keep the clip lights out. This has worked better for me.
post #7 of 13
Quote:
Originally Posted by braveheart123 View Post

My AVR is denon 3313. I take the pre-out signal (LFE+Redirected Bass) from avr and feed it to iNuke. How does iNuke know that the input signal is LFE+Redirected Bass and it has to apply 10dB boost at the amplification stage??
It can't know. So, you have to tell it. The bass managed output is likely down 15dB, btw - the original 10dB for LFE, plus another 5dB reduction to accomodate redirected bass from the full range channels.

Assuming that I understand the approach you are taking, you may have to adjust the sub level based on the content you are playing. The sub will need a 10-15dB boost when the source is 5.1, which has a genuine LFE channel. But, it should not be boosted when the source is stereo and the sub output is the result of bass management performed by your receiver.
post #8 of 13
My understanding is the sub boost takes place at the processor's pre-amp stage, so the pre-out is already at the correct level. That's why AVRs generally let you boost the sub +10 or +15 manually for the analogue MCH inputs.

In any case Audyssey takes care of it during calibration even if you use separates. There is no need to do anything manually to boost it for digital inputs as long as the final results are acceptable. A sub trim of -7.5 indicates the sub volume is too high to start with. Reduce the volume so the trim is nearer 0. If the sub sounds too weak that's probably because you are used to hearing it too hot before.
post #9 of 13
Quote:
Originally Posted by Kilian.ca View Post

My understanding is the sub boost takes place at the processor's pre-amp stage, so the pre-out is already at the correct level. That's why AVRs generally let you boost the sub +10 or +15 manually for the analogue MCH inputs.
Yes, if the AVR has already done the boost so that the pre-out is at the correct level, then it should all work easily.
post #10 of 13
Thread Starter 
Quote:
Originally Posted by BIslander View Post

Yes, if the AVR has already done the boost so that the pre-out is at the correct level, then it should all work easily.

That's strange. You guys mentioned -10dB offset for LFE is compensated in the amplification stage and not at the pre amp.
Quote:
It can't know. So, you have to tell it. The bass managed output is likely down 15dB, btw - the original 10dB for LFE, plus another 5dB reduction to accomodate redirected bass from the full range channels.

How do I tell iNuke to add 10-15dB boost to the input signal...... Or do I have to raise the sub trim level by 10-15dB in the Speakers Level in the AVR and then back off the level attenuators on iNuke to where it is a tad below clipping?
post #11 of 13
The 'amplification stage' does not literally mean just the power amp. Volume change is done in the pre-amp. If you change the trim in the iNuke it's behaving like a pre-amp for volume control in that sense.

Read the above linked thread on LFE again, esp. under the sections "In the Home" and "LFE digital output from players". Note where it says 'automatic'.

The important thing to note is: if you feed the AVR with a digital MCH signal the pre-amp processor will by correct design automatically do the sub boost. If it doesn't then you do it manually. The Denon will do it and outputs the correct level at the pre-out. I dunno why BIslander said the sub pre-out is low.

If you feed it with analogue MCH the processor will not and you have to do it manually. If you don't do it at the AVR you can also do it further down the chain at the sub.

Where the boost is done doesn't matter too much as long as the final sub output is the correct level, which Audyssey will measure during calibration and attempt to correct it as needed.
post #12 of 13
Quote:
Originally Posted by Kilian.ca View Post

I dunno why BIslander said the sub pre-out is low.
It appears I was mistaken about that.
post #13 of 13
Thread Starter 
Thanks all you guys for your help and prompt responses. My amp is now pumping out ferocious amount of power without clipping up untill -5dB on avr master volume (Reference Scale).
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Receivers, Amps, and Processors
AVS › AVS Forum › Audio › Receivers, Amps, and Processors › 0dBFS + 10dB Handling in BFD And Pro Amps