Jump to content
Kvalsvoll

The Bass EQ for Movies Thread

Recommended Posts

Hi Mike.  I use JRiver to implement BEQ filters, and it works really well.  Some movies, like Pacific Rim and The Avengers, are a completely different experience with BEQ (assuming you have a sub that can dig deeper than 30 Hz).

 

I create a zone for each film I want to BEQ, and then use the zone switch feature in JRiver to automatically switch zones when I start the move.  I'm not near my HTPC right now, and I've had quite a bit to drink (celebrating the Cubs' victory in the NLDS against the Cardinals), but if you're having a hard time figuring out how to implement it, I might be able to put together a short HowTo in the next couple of days.

That would be cool.  If it's not too complex, or cumbersome to implement, I would do that for a few films.  I will look for your "how to" in the next few days.

Share this post


Link to post
Share on other sites

A low shelf with this Q is not supported directly by jriver. It's low shelf is described in terms of Q but, as I understand it, it really mean S (shelf factor) and it limits this to a max of 1. 

 

I'm just getting back to this thread (sorry for the delay, MikeDude!), and just noticed this exchange.  Can you explain?  Does this mean that all the BEQ that I've done for JRiver using the low shelf Q values in this thread are invalid?

Share this post


Link to post
Share on other sites

I think so, but I am not familiar enough with JRiver to know for sure.  I post Q values as most DSPs use Q values.  S is used when calculating biquad coefficients for a Low Shelf, and some software (like SoX), use the S (Slope) factor for shelf filters instead of Q.

 

Your 'invalid' corrections would simply not be as steep, as a Slope of 1 = Q of 0.707

 

 

JSS

Share this post


Link to post
Share on other sites

I'm just getting back to this thread (sorry for the delay, MikeDude!), and just noticed this exchange.  Can you explain?  Does this mean that all the BEQ that I've done for JRiver using the low shelf Q values in this thread are invalid?

the jriver devs haven't commented one way or the other but my testing says they say Q on the dialog but it is implemented as S. I wouldn't say this invalidates it but it does mean the shape of the filter is not going to be exactly as intended. If you have a way to calculate and sum filters offline then it is easy enough to create a different set of filters that jriver does support and will fit the provided target more closely. I could do it though it might take me a while to get round to it.

Share this post


Link to post
Share on other sites

the jriver devs haven't commented one way or the other but my testing says they say Q on the dialog but it is implemented as S. 

 

How would one go about testing this to know for sure?

 

I went back and listened to a bit of Pacific Rim with and then without the modifications recommended by maxmercy.  Even if the slope I'm using is different than intended, the result of a flawed BEQ is significantly more pleasing and engaging than the bass response on disc.

 

To make sure I understand correctly, assuming you're correct about Q in the JRiver interface really being S, the correct way to calculate S would be the following formula:

 

S=1/(((1/Q)^2)-2)/(A+(1/A))+1) - I didn't bother to factor this out myself, so I'm just going to assume you did it correctly  :) 

where 

A = 10^(dBgain/40)

 

I guess I'll have to use Excel to simplify this, unless you know of any great online calculators that simplify this process.

Share this post


Link to post
Share on other sites

How would one go about testing this to know for sure?

 

run a sweep through a jriver zone configured appropriately and loop the output back into REW (or similar) then compare against a theoretical shelf filter. This gives you

 

post-1440-0-24946900-1446904689_thumb.jpg

 

these are all 50Hz +5dB low shelves for which S=1 is equivalent to Q = 1/sqrt(2) ~= 0.707 

 

green: jriver Q=0.5

red (hidden behind blue): jriver Q=1

blue: acourate Q=0.707

brown: acourate Q=1

 

blue is almost exactly the same as red, the minor variation is because I used 0.707 rather than 1/sqrt(2) :)

 

This proves that jriver means S when it says Q when it comes to a shelf filter. I'm not sure of a tool that lets you sum filters on paper other than acourate, I guess it would be possible to write one. 

 

I don't know of a calculator to convert between Q and S either but easy enough to plug into excel or similar.

  • Like 1

Share this post


Link to post
Share on other sites

S=1/(((1/Q)^2)-2)/(A+(1/A))+1) - I didn't bother to factor this out myself, so I'm just going to assume you did it correctly  :) 

where 

A = 10^(dBgain/40)

 

I guess I'll have to use Excel to simplify this, unless you know of any great online calculators that simplify this process.

 

Excel barked at that formula for S.  There was an extra parenthesis, and for the life of me, I couldn't get it to result in ~1.7.  Here's what I came up with:

 

S = (A^2*Q^2+Q^2)/(A^2*Q^2-2*A*Q^2+A+Q^2)

 

It comes back with a value of ~1.7, so it matches what you found.

Share this post


Link to post
Share on other sites

I find excel, i.e. me writing equations in excel, terrible at that sort of equation unless you break it down into many cells. Funnily enough though I used what you posted to calculate that post earlier and it worked OK :)

Share this post


Link to post
Share on other sites

:(

 

It looks like JRiver's 1.0 limitation for Q (or really S, but Q in the UI) means many of these BassEQ settings can't be properly implemented.  I had previously created all of mine in JRiver 19, before they implemented notification that values over 1.0 are ignored.  I've recently upgraded to JRiver 20, and while it imported my old BEQ filters, including ones with a Q>1, if I try to edit it, I get an error message notifying me of the problem.  

 

One of the best features in JRiver is how easily it allowed me to implement BEQ.  This is a blow.

Share this post


Link to post
Share on other sites

:(

 

It looks like JRiver's 1.0 limitation for Q (or really S, but Q in the UI) means many of these BassEQ settings can't be properly implemented.  I had previously created all of mine in JRiver 19, before they implemented notification that values over 1.0 are ignored.  I've recently upgraded to JRiver 20, and while it imported my old BEQ filters, including ones with a Q>1, if I try to edit it, I get an error message notifying me of the problem.  

 

One of the best features in JRiver is how easily it allowed me to implement BEQ.  This is a blow.

 

the problem is just that you can't copy the params straight into jriver and go for it, you'd have to spend time porting the published values to something that jriver can handle. Here's an example

 

post-1440-0-55632700-1446983755_thumb.png

 

the red line (in acourate) is the BEQ filter for the Pacific Rim LFE channel, you can see it's basically a fairly steep shelf filter that adds 28dB by about 6Hz and starts ramping up at ~40Hz

the green line is a LS with S=1 (Q=0.707) at 15Hz with +14dB

brown is the result of stacking 2 greens (to give 28dB total boost)

 

the REW trace is showing what happens when you run 1 and then 2 of those LS filters through jriver which shows that it can implement that filter

 

IMV that's a pretty close approximation to the intended curve and a much simpler filter to enter. If you want to fine tune it to hit the original BEQ filter more closely then you would just stack up some small notch filters along the curve to push it up/down as appropriate. It's not that hard to do if you can loopback jriver into REW as you can just iterate over it until you get a result that matches the target. Obviously it's a lot easier if you have a feel for how different filter shapes will sum up but that's just practice.

 

The only stumbling block is working out what the actual intended filter shape was. You could do this using the minidsp spreadsheet by plugging in the values and then copying out the values for dB by freq into another worksheet, summing those values up to get the final filter shape and then working out a jriver compatible filter. In fact you could do this all on paper without the loopback at all by plugging such values into a spreadsheet. I imagine this would be quite easy to do actually (albeit probably rather tedious work).

Share this post


Link to post
Share on other sites

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

  • Like 1

Share this post


Link to post
Share on other sites

Bølgen (The Wave) - Norwegian disaster movie about a giant wave causing massive destruction.

 

Opportunity to make some REAL bass here, I would say..

 

post-181-0-52172200-1448762860_thumb.jpg

 

Comes with 2 sound tracks, DTS-HD which is kind on the ears but also with no bass, and TrueHD Atmos which has bass and clipping.

I have not watched it yet, only a few scenes to do the bass-eq.

 

Original lfe TrueHD/Atmos:

 

post-181-0-28902200-1448763153_thumb.jpg

 

(See the main ULF content thread for waveform pictures.)

 

 

A relatively gentle lift below 30Hz to raise the 15-20Hz a bit, and recover what was lost in the 15Hz filtering:

post-181-0-93332700-1448817709_thumb.jpg

 

Now, this is more like it.

When the wave comes, it should not be a subtle shake in the floor and some nice midbass punch, this event is supposed to shake the whole house, violently. 

 

I may have overdone a little, but I think it would be worse the other way - originally the wave feels like splashing in a bathtub, and that is clearly much worse than having a slightly too destructive and violent disaster wave sound experience that resembles what you see on the screen.

 

 

Bass-EQ spectrum plots

 

post-181-0-21592600-1448820572_thumb.jpgpost-181-0-23469000-1448820602_thumb.jpgpost-181-0-96396600-1448820628_thumb.jpg

 

 

Two soundtracks and the general status of the movie content providers

 

As mentioned, the film comes with two soundtracks.

There is no information on the cover or anywhere else that I could find about those soundtracks, and it defaults to the DTS with no bass.

While watching the movie you can of course switch between the tracks to determine which one you prefer, but when you sit down to watch the movie, you don't want to experiment with soundtracks and setting.

The information for those soundtracks must be presented beforehand, so that the audience can decide BEFORE starting the movie which soundtrack is most suitable.

 

This is fault #1 - lack of information for soundtracks.

 

Then there is the general status for technical knowledge on sound reproduction among content providers.

 

Here, the DTS track can be suspected to be meant for commercial cinemas - it is not too loud to be played at 0dB, and all bass below 30-40hz is simply gone. 

 

The Atmos track is very different, it is LOUDer, there is significant clipping in all LCR channels, and the bass is now quite good and reaches quite deep, though I suspect the level is reduced a bit below 30hz - not by much, only 1-2dB, and it is filtered below 15hz ("No one can hear that low, anyway.."). 

This is something that they would likely create by starting out with a reduced master volume level ("They don't listen that loud.."), and having a monitoring system more similar to a high-quality home set-up with some modest subwoofers. 

The need to push levels too far and cause severe distortion due to clipping comes from the lowered master volume setting - because it now sounds a litte too quiet in volume, they compensate by pushing levels upwards, and they probably can't hear the clipping distortion due to lack of resolution on the monitoring system.

The filtering and reduced level in the low bass is due to limitation in the subwoofer system in use - they will experience more impact and better sound when removing content below the systems capacity both in level and low frequency extension.

Once you have reached the limits of capacity, it will not be more impactful or powerful when turning it up any further.

On a proper system this will of course not be the case, as there will be much more headroom available.

 

Whether the Atmos is made like this for the home market, or there is actually also a theatrical similar to this one - louder, and with more proper bass - is something only the producers know.

 

Fault #2 is not using decent sound reproduction equipment in the studios.

 

If they had the overall sound quality, resolution, dynamics, wide frequency range, then the soundtrack would be much better and no bass-eq would be necessary.

It would not be clipped with neutered low bass, and it is very likely that especially the low frequency sound effects would have been better, because they would be able to actually hear what they are working on.

 

 

BassEQ

 

BassEQ parameters for Shelf-filter:

 

lfe:

low-shelf 16hz 1.6 +6dB
low-shelf 16hz 1.6 +6dB
(x2.)

high-shelf 36hz 1.4 +0.5dB
 

Note:

If you prefer a slightly smaller wave, just leave out one of the 16hz gains. 

Also, this filter is quite modest and you will be able to do it in any parametric DSP or JRiver.

The clue is to apply a filter that rises frequencies below and around 20Hz, and the 6 or 12dB max boost will then occur at a much lower frequency, around 10Hz.

The high-shelf is merely cosmetic to maintain mib-bass level when the corrected lfe is summed in bass management.

Here the filter is applied to lfe, it will work just fine to apply the filter on the subwoofer output on a bass-managed system.

Share this post


Link to post
Share on other sites

Nice....what does the DTS track look like? PvA wise?  I'll take a higher rolloff without clipping than a lower rolloff with any day.  

 

JSS

Share this post


Link to post
Share on other sites

The DTS is rolled off closer to 40hz on all channels, to retrieve anything useful from this is very difficult, and the TrueHD/Atmos sounds reasonable out-of-the-box.

There are other differences as well - missing 2 rear surround channels, overall surround is perhaps missing some of the atmos effects (The 7.1 has overhead sounds), dynamic range may be different.

 

I suspect the clipping will not be very noticeable, the waveforms are no worse than M&C, just to put it in the right context.

And we all liked M&C, didn't we?

 

I have to screen the whole movie to really know for sure how it sounds, looking forward to that.

Share this post


Link to post
Share on other sites

@maxmercy, I just checked the DTS vs 7.1 tracks, you definitely do not want the DTS.

 

The DTS is dynamically compressed and all bass is gone, what is left of the upper bass is reduced in level.

To me it seems like this DTS track is intended for mobile phone speakers, you know - "..because that's what they have at home."

Quiet dialogue scenes are louder, while the wave and action scenes are very reduced in level.

This does not help much for dialogue clarity in action scenes, though, because they applied the compressor on the final bus, reducing the dialogue as well as the blasts and sound effects.

 

But I really don't care about that, what is excellent is that they provide a "sound-bar" track - the DTS, and a real sound track - the 7.1, with great sound and excellent dynamics and bass.

Only thing missing now is INFORMATION ON THE DISC so that people can manually select the proper 7.1 sound track.

 

I think it is the wave that causes the clipping.

I guess the wave was too big even for 24 bit audio, after all, it is only 144dB dynamic range..

When you look at the size of that wave, it makes sense, sort of.

And it lasts for a long time.

With bass-eq the whole house shakes, like it should.

 

So, look forward to getting this when they eventually release it in other parts of the world, the story itself is different form the standard Hollywood disasters, there are wonderful scenery from the norwegian coastal mountain landscape, and - with bass-eq - house-shaking bass.

Share this post


Link to post
Share on other sites

I always tend to prefer the theatrical mix when I have a choice.  More dynamic range in general, more standardized setup, and not mixed for a TV/soundbar. or HTIB satellite system w/ less than competent LCRS.  Mixing for home is such a compromise due to the variety of playback systems.  I wish the production houses would just put both tracks on every BD.

 

JSS

  • Like 1

Share this post


Link to post
Share on other sites

Watched Bølgen today, sounds quite good, never too loud and nice surround presentation.

 

Ended up using the +6dB moderate bass-eq track.

I noticed clipping in one scene only, when the wave comes rolling.

But I can not stop wondering whether the ulf effects on the wave would be much better if they could actually hear what they were doing in the studio.

Share this post


Link to post
Share on other sites

The Force Awakens BEQ:

post-20-0-34394700-1460076461.png

 

BEQ Correction (implement prior to AVR, with HTPC software or nanoAVR):

 

LCRS:

 

1. Low shelf - 15Hz, Slope 1 (Q of 0.707), Gain +6dB

2. Low Shelf - 15Hz, Slope 1 (Q of 0.707), Gain +3dB

3. Parametric EQ - 85Hz, Q 1.12 (Bandwidth 1.25 Octaves), Gain +4dB

4. Overall Gain -7dB

 

LFE:

 

1. Low Shelf - 15Hz, Slope 1, (Q of 0.707), Gain +6dB

2. Low Shelf - 15Hz, Slope 1, (Q of 0.707), Gain +6dB

3. Low Shelf - 15Hz, Slope 1, (Q of 0.707), Gain +3dB

4. Parametric EQ - 70Hz, Q 1.12 (Bandwidth 1.25 Octaves), Gain +4dB

5. Overall Gain -7dB.

 

What I did was give the track more room-shuddering effects as well as more midbass slam while minimizing the boost to the 30-40Hz bandwidth which was over-utillized, IMO.  I like the result, but beauty is in the eye of the beholder.

 

These changes will mean that Reference Level playback will occur at +7dBMV.  The track does not reach over 126.5dB peak level at this level of playback, and max RMS level is 119dB (over 1/8th sec), so if your system can handle Reference Level with very demanding material, it can handle this track at +7dBMV, provided your AVR does not clip the signal.

 

The track gains over a dB in Dynamics, Extension and Level become 5-Star Level if played back at +7dB from your typical listening level.  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.

 

JSS

  • Like 2

Share this post


Link to post
Share on other sites

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.  :)

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×