Jump to content

The Bass EQ for Movies Thread


Kvalsvoll

Recommended Posts

I look forward to trying this out!  I'll be able to program it into my prototype Motu/PC-based DSP.  All I really need to do is move the bass management from my AVR to the Motu/PC and then I can just program in the filters.

 

I'm curious as to how the mid bass bump will affect things for me.  I find myself quite satisfied already with the slam I get from my system with most movies.  In fact, I recall thinking in the movie theater: "this is going to slam nicely when I hear it at home if they don't screw up the mix".  Do you think the BD mix was altered to have less mid bass?  Or do you think the theater you went to used a mid-bass bump for effect?  Or do you think there's some kind of acoustic secret sauce in large rooms that makes mid bass hit harder than it does at home without more SPL?  Or do you have no clue and just like the sound of the BD with more mid bass?

 

Don't mind all my questions.  I'm just curious about your thoughts and experience.  FWIW, I run a full-time house curve of about +0 dB @ 20 Hz to -5 dB @ 300 Hz and then flat out to 12 kHz or so.  It just sounds more natural to me that way than running flat or using a treble roll-off.   I'm sure you've noticed one of the nice thing about having good bass absorption in the room is that the extra bass doesn't really add in the way of "bloat" or heavy sound.  It mainly just makes stuff hit harder.  :)

 

No clue how the home mix was done nor by whom.  The theater I went to was both a LieMAX and a conventional theater, presentations were similar, to me, most cinemas have a little more midbass, that's probably because you are above Schroeder Freq in the midbass as opposed to the midrange in a small room, which can lend far more midbass impact in a cinema.  Bigger room=better slam, in general, for the same SPL.  I'm sure you have experienced a well-done outdoor concert.  Best impact you can get.  No cancellations, reverb and modal ringing to be had to mess up an impulse response.  

 

I like the changes I made, which is why I said not to use an aggressive house curve with the BEQ, as I tried it with a little more aggressive curve and there was significant midbass 'bloat'...my 'flat' curve drops from 20-20kHz smoothly (on a log graph) by 7dB.  My more aggressive curves add more <40Hz and midbass, and are mainly for listening at lower SPL levels.

Link to comment
Share on other sites

DO NOT add a house curve to this unless you know your system can handle it.  Adding a steep house curve will bloat the midbass and the score will sound unnatural.  This BEQ assumes a gentle (if any) house curve, maximum of -6 to -10dB downslope from 20Hz to 20kHz, with no aggressive slope-up in the midbass region.

 

What do you consider an aggressive house curve.  Is something based on the Harman curve aggressive?

 

933859b1_SynthesisTarget.jpeg

 

That's not my exact curve, but it's the general idea.  

Link to comment
Share on other sites

I consider that aggressive.  I consider this gentle:

 

bandk.jpg

 

My curve is a lot like this one but the slope continues up until 10-20Hz.  The TFA soundtrack provides the Harman Curve for you, so if you start with the Harman Curve, you can get some real 25-50Hz boom, maybe even too much.

 

 

JSS

Link to comment
Share on other sites

  • 2 months later...

?:

 

So, I have this "movie" with EAC3 audio.

It is encoded for playback with -7dB dialnorm/offset.

 

Decoding for processing, fixing the 40hz cut-off, making for an improvement that only bass-eq addicts can believe.

 

But.

Levels are off, and there is something strange going on with dynamics.

Inspection show dialogue is 7dB louder on center channel, music/l+r is 3dB louder.

Listening to different scenes suggests the center could be reduced -4db to -7dB.

Dialogue is far too LOUD when played at 0dB.

But then I notice background effects are also louder on the fixed decoded version.

 

Something strange is going on here.

It looks like the decoder did not do what the processor does on playback of the original eac3 track.

 

If the whole thing was just 7dB louder, that is not a problem, just reduce level to -7dB.

Or simply ply the whole thing louder - my original intention.

But that does not work well if dialogue level is too LOUD and some kind of dynamic compression is applied in the decoding process.

 

I used ffmpeg(Audacity) for decoding.

 

Any ideas?

Link to comment
Share on other sites

I'm guessing that the audio was decoded with dynamic range compression (DRC) enabled.  Indeed, some Googling leads to hints that DRC is done by the ffmpeg library by default.  So if Audacity is using ffmpeg to decode the audio track, it would have to instruct ffmpeg to not use DRC.  The command line program has an option, so it is almost certain that the library has an option too.  The question then is whether Audacity supports this option and allows you, the user, to override it.

 

While I am surprised to learn that this is the case, I really shouldn't be.  Why not?  Because I've seen it done in other devices, and because it's probably a reasonable default.  I believe the AC3 specs instruct that (for home use) DRC should be enabled by default, but the user should be given the option to disable it unless the component in question will perform poorly without it.  (For example, TV sets used with the built-in speakers.)  So in a sense, the ffmpeg library programmers adhered to the spec by having DRC be enabled by default.  The trouble is that programs that simply use ffmpeg to load AC3 audio are not as concerned about adhering to the AC3 spec; hence, it may never occur to these programmers that DRC even exists, much less that it's an option that should be configurable.  In the case of Audacity, I think applying DRC during the import by default is an error.  Perhaps they would be willing to accept a code patch to fix it.

 

Let me know if you can find an option in Audacity to control DRC for AC3 imports.  If you can't find one, the best alternative may be to run ffmpeg directly on the command line, supplying '--drc_scale 0' as one of the options.

 

BTW this is big news, if anyone has been using Audacity to analyze AC3 encoded tracks, such as those included on the crappy "rental edition" BDs as of late.  This could definitely skew things a lot!

  • Like 3
Link to comment
Share on other sites

I found it, it is exactly like you describe it, @SME.

 

DRC is enabled by default in ffmpeg, and this causes disaster for sound quality - hard, harsh sound, too loud dialogue, loss of the impact I imagined I would get from the massive basseq.

 

The impact on sound quality is much more than just a little louder background effects and a little less brutal booms and thunders.

And when it is applied, there is no way to repair it.

 

I had made 3 movies, with brutal basseq, it took me a while, and sat down the next day to enjoy a real beating in a battle scene.

And then.. WHAT IS THIS!

I had prepared to maybe adjust the level a little to compensate for the dialnorm now missing, but clearly something more was very wrong.

 

Solution is to use ffmpeg command line to extract the audio.

Audacity can not take arguments for ffmpeg, at least I was not able to find out how.

All ffmpeg decoding of dolby ac3/eac3 defaults to using drc, I assume controlled by metadata, so some files can be good, others not. 

This problem is caused by Dolby defaulting all playback software and devices to the worst possible sound option, so that people can have the worst sound experience out of their already quite bad soundbars and alike.

Link to comment
Share on other sites

It looks like eac3to relies on ffmpeg for AC3 decoding, but the discussion in this post ("evaluation of available decoders") implies that they are very conscientious about dynamic range compression and ensuring it is not enabled for decoding.

Link to comment
Share on other sites

  • 2 weeks later...

God Almighty... an image of perfection ladies and gents. Send me that OM file, I wanna see how close I stack up. You still have my email address :D

 

Perfection?  Can you explain what about it makes it more perfect than another choice of house curve?  I'm curious if you have an opinion on this.  Either way, I think I'd find that curve to be too steep.  Mine currently varies by about 6 dB from 20-8000 Hz vs. 12 dB in this case.  Although I actually roll-off a lot more above 8kHz, so I guess we're both about 15 dB over 20-20000 Hz.  Interesting.

Link to comment
Share on other sites

Huh? It was just a one-octave smoothed graph illustrating a comparison to the "preferred" FR shape Max was commenting about.

 

Of course the full un-smoothed response does not look like that.

 

Oh I know that, but the overall slope of the graph is what I basically consider perfectly balanced, and aim to hit anytime I am messing around. Send me the file, I just want to see exactly how close we actually are! Or don't, I'm not the boss of you. 

  • Like 1
Link to comment
Share on other sites

Haha! Well first of all... I'd love to but I can't. That response does not exist anymore. I've changed it already.

 

Do you want a copy of my current FR? And... I have no idea what "file" to send you. :unsure:  I can...umm...take a pic, like I do.

 

 

My second most recent FR is in the link in my sig. Second post. Already changed that too. :P

Link to comment
Share on other sites

Ha!  I have the same problem with my response changing all the time.  I need to update mine too.  It looks better and sounds better too.

 

Really this is off topic, but I wanted to start a conversation about house curves.  I simply don't believe in calibrating flat after hearing the difference.  I think it needs to be acknowledged that house curves are necessary for the best sound.  What I don't understand is to what extent house curve selection is based on a subjective needs (a preference for more bass for example) and how much it's based on objective considerations (psychoacoustics).

 

As a thought experiment, consider a controlled-directivity 2-way.  In the stereotypical implementation, the CD and woofer are crossed over approximately at the point where the dispersion of the horn has widened to match that of the woofer.  For various reasons, the directivity often begins to narrow again right at the crossover point before widening below it, but we'll ignore that for now.  Let's instead just assume that, starting from the bass and going up in frequency, the directivity increases smoothly and monotonically.  This is kind of the idealized realization of the 2-way controlled-directivity design.  Now, we have Joe and Sally.  Joe is installing his speakers flat behind a screen with no toe-in.  Sally installers her speakers in her living room with toe-in.  In the MLP, Sally measures less output from the horns because she is off-axis of them.  To compensate, she uses signal shaping to boost the high frequencies so that her house curve matches Joe's.  But are their speakers really identically calibrated?  Because Sally has boosted the output of her horns, they are putting more high frequency energy into the room.  This in turn impacts tonal balance, despite the fact that she is SPL matched with Joe.

 

So the moral here is that I actually expect house curves to vary, and I believe there are objective considerations to choosing a house curve in addition to purely subjective ones.  If anyone is interested in continuing this conversation, let me know and I'll create a thread.  I think it would be cool if we could figure something out.

  • Like 1
Link to comment
Share on other sites

1st half of http://www.aes-media.org/sections/pnw/ppt/jj/room_correction.pptis a good summary, comes from http://www.aes-media.org/sections/pnw/pnwrecaps/2008/jj_jan08/

 

Actually there was a post on this subject in the last day or so comparing the various target curves (b&k, ITU-R BS.1116-3, EBU 3276, the harmon/olive paper from 2010 & the toole paper from 2015) and they're all v similar. The specs (EBU/ITU) go into more detail on the room & speaker parameters, I don't recall whether the B&K & the Harmon papers gave specific guidance on that point though.

 

In 1994 (updated in 2015), the ITU produced a "Recommendation ITU-R BS.1116-3

  • Like 1
Link to comment
Share on other sites

  • 3 weeks later...

Hello everybody!

 

I'm new to this forum and would like to thank you all for all your contributions I have learned a lot reading the different threads.

I have come across a piece of equipment that some of you might want to look into. I didn't know where to post this but as it deals with correcting the signal I figured it might not be misplaced here. Please correct me if I'm wrong.

 

So the equipment in question is called Ultrabass Pro EX 1200 from Behringer. I got it second hand on the french CraigsList. What it does basically is that it takes the lowest frequency of the signal at any given moment and digitally brings it an octave lower or 2, depending on the settings.

 

I should specify that I apply the correction of the signal only to my Crowson actuators. I live in a small appartment and wouldn't benefit much from the device given the lack of space for low acoustic bass to develop.

 

Nevertheless, I have had satisfying results. I have not run it with many movies yet but, for example, Empire Strikes Back and Return of the Jedi benefit considerably from it. The Asteroid Field in Empire Strikes Back has a whole new dimension to it. Same for the whole third act in Return of the Jedi. The rumble at the beginning of Star Wars ep III is somewhat cleaner, but it comes at the expense of some other previously noticeable small effects. Again, I can only talk of what is felt with the actuators.

 

As for some other movies who lack bass, (The Hobbit : An Unexpected Journey might come to mind...), well...the stone giants scene is a little better and feels a tiny bit more immersive. But still very VERY far from its potential. Avengers : Age of Ultron I have found to be more than improved. Kung Fu Panda got a little bulkier as well.

 

That is about the extent of my tests so far but I was enthusiasted enough to think it might interest some of you. I got it for 75 E plus 25 for XLR to RCA cables so I figured it wasn't that much of an expense, even if only to conduct some experiments. This ended up yielding very interesting results so I thought I'd share it on this website.

 

I hope that was worth a read.

 

Thanks again to everybody.

 

Cheers!

Link to comment
Share on other sites

The Ultrabass processor creates new audio data by synthesizing content an octave or two below what is on a track.  It is the only way to make some tracks  ANY better (like The Hobbit, where the data was never there).  It can smear transients, but if only used in Crowsons, that is a perfect application.

 

A smaller room is usually better for lower bass extension from a given subwoofer system.  But usually not neighbor-friendly.

 

JSS 

Link to comment
Share on other sites

  • 2 weeks later...

I actually generate the BEQ curves with S, and calculate Q to post here.  I can post both on future BEQ.  I sometimes use steeper S than 1, though.  JRiver should allow values greater than 1, as the filters are not 'unstable' as they claim.

 

JSS

the next build of MC22 allows S<=5 (which equates to Q~=1.73), there might be some instability in the output if the corner frequency is set to 5Hz but it seems ok above this (not sure if measurement issue on my part)

  • Like 1
Link to comment
Share on other sites

the next build of MC22 allows S<=5 (which equates to Q~=1.73), there might be some instability in the output if the corner frequency is set to 5Hz but it seems ok above this (not sure if measurement issue on my part)

 

Awesome.  Now I have a reason to upgrade the MC22.

Link to comment
Share on other sites

for completeness, a high Q ULF LS has some ripple in jriver. It equates to a minimum corner frequency of ~8Hz for completely predictable behaviour. I'm not sure if this is to be expected or not, probably irrelevant anyway but I thought I'd mention it (can't attach a pic due to space issues, not sure if http://yabb.jriver.com/interact/index.php?action=dlattach;topic=106708.0;attach=22139;imageis visible either)

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...