Jump to content
Sign in to follow this  
nube

Edge of Tomorrow Discussion & Poll - Closed

Edge of Tomorrow  

59 members have voted

  1. 1. Execution?

  2. 2. Recommendation?



Recommended Posts

I got to this one too late to have my vote counted, but I gave it 5 stars anyway.  I played it at "-6" and enjoyed the soundtrack very much.  The dialog noise was the only apparent flaw.  The sound design and bass were superb.  I disagree with those who thought that there was insufficient "quantity" here.  I noticed bass everywhere I expected to, and none of it sounded superfluous.  That's to say, all the bass was convincingly part of an overall sound effect.  It had lots of detail, dynamics, and texture.  Most effects seemed to be more wide-band and impulsive, which may explain why many were not so impressed.  The bigger hits seemed very well extended.  I recall hearing/feeling multiple impressively strong impulses.  I also agree that the upper bass content contributes a lot in this one!

 

I also think this had better bass than Oblivion with the increased extension very apparent.  On the other hand, I liked the Oblivion movie more.  I also liked it more than CATWS, and it got my vote for top LF film of 2014.

Share this post


Link to post
Share on other sites

This soundtrack was a delight, especially the bass was full-range, dynamic and with lots of punch and impact.

 

And it doesn't all sound the same - there are subtle detail in the very lowest frequencies, but also huge dynamic hits that feels more like a brief, sharp gust of wind, and then there is the great smack in the chest from the upper bass.

 

Surround use was also very good, with some very good scenes with overhead sounds.

 

While loudness war and compression and clipping gets a lot of attention these days, this movie clearly shows they can still make great soundtracks.

  • Like 1

Share this post


Link to post
Share on other sites

I am just posting some spec labs to calibrate my system correctly and to not muck up Adam's sub system thread. Here is the opening scene flat at reference t my LP.

 

capt1505140358_zpsfeonprsy.jpg

Share this post


Link to post
Share on other sites

Hey James,

 

Nice work on the SpecLab caps. :)

 

I've always maintained that the color scale offset is not relevant because regardless of that setting the colors are relative. If you really want to "calibrate" your caps, you need to adjust the input and/or the offset until they match Nube's/Max's peak hold graph.

 

Since in this case the 10 Hz peak is unrivaled by anything else in the movie, the peak hold graph gives you the target setting.

 

In the cap you posted above, your 10 Hz peak is at +5dBFS and Nube's is at -7dBFS:

 

b2ac67f9da3aea1fd7ccd06096bc87cb.png

 

Again, the colors will remain relative, but I've always preferred to offset higher so that more of the color scale is utilized. It shows better resolution of intensity as you'll see the loss in intensity resolution if you change your offset value to match Nube's graph.

Share this post


Link to post
Share on other sites

I have a rising response below 20hz due to my room and is about 5 dB high at 6-11hz which does not help for a flat representation but feels great!  I noticed Nube's peak hold but thought since the scene hit about 120 dB with two channels it was over reference and why I did what I did.  Can a signal of -7 hit 120 dB from the LFE plus center?  I just did some other spec labs from popular scenes. I will post them in Adam's thread if he does not mind. 

 

You are right, these are addictive! 

Share this post


Link to post
Share on other sites

7.1 channels of re-directed bass sum to around 126 or 127dB.  So, yes, a -7dBFS signal can result in around 120dB of output required, but it's a worst case scenario.  This is something that maxmercy and others have been saying for some time, and why he's creating his worst case scenario test disc.

Share this post


Link to post
Share on other sites

I am using 5.1 so how should I calibrate it?  Should I let the peak hit -7? 

 

 

Yes. If you want the caps calibrated, you need to comply with the actual encoded levels. Just go to Options in the tool bar and click on spectrograph 1 or 2, I forget which. There will be an Offset feature with an offset number in the box. Change that to a lower number until the peak is -7dBFS on your bar graph.

 

I eliminate the bar graph altogether because that's covered in the peak hold graphs Nube posts. This lets me get more of the spectrograph on the screen.

 

Since you're mic'ing the scene, it will reflect frequency response non linearity at your seat and hot/cold/flat calibration. If you want to get the cap to match Nube's peak hold for EOT, instead of messing with the offset, just adjust your mic input down until the 10 Hz peak hits -7dBFS.

Share this post


Link to post
Share on other sites

But the purple is so cool looking! ;) I have my mic input already at 35 so I will have to adjust the off set which is tab 2.  I know that scene contains LFE plus center channel so all channels are not active. It makes sense though since you can just scroll through Nube's thread about longest and intense scenes and they are not as hot as mine. When I say flat I mean the crossover is flat with the speakers and not running the LFE hot.  My response rises down low so my graphs will always show a hotter under 20hz. 

Share this post


Link to post
Share on other sites

But the purple is so cool looking! ;) I have my mic input already at 35 so I will have to adjust the off set which is tab 2.  I know that scene contains LFE plus center channel so all channels are not active. It makes sense though since you can just scroll through Nube's thread about longest and intense scenes and they are not as hot as mine. When I say flat I mean the crossover is flat with the speakers and not running the LFE hot.  My response rises down low so my graphs will always show a hotter under 20hz. 

 

 

I'm with ya all the way! Using a mic at the seats is what matters. The digital feed graph is the reference and the mic'd version tells you how accurately your hearing it. No flowery ethereal wording that no one understands anyway... just seeing what you're hearing. How you gonna beat that? You ain't! :lol:

Share this post


Link to post
Share on other sites

So I have heard there are more of those opening tones at the end of the credits.... I haven't checked myself yet, but will definitely do so next time I am firing up the HT... can anyone else confirm?

Share this post


Link to post
Share on other sites

slightly random question but I happened to compare a raw digital capture of the EoT scene (mixed down to mono using jriver) vs what my sub actually plays (again using jriver but mixing through my bass mgmt and room correction filters). I happened to notice the "square wave" aspect of the waveform was no longer present.

 

This compares the 10Hz tone 

 

post-1440-0-14106800-1439158843_thumb.jpg

 

I then ran the raw (mixed to mono) digital mix through speclab and zoomed out

 

post-1440-0-79880500-1439158849_thumb.jpg

 

I think it is the C channel content that lends the combined waveform the look of a square wave. 

 

It's probably a minor point as the bulk of the content is in the LFE track but I guess it indicates that just mixing to mono in jriver may not be a completely accurate view of what your sub sees (when looking at the waveform that is). 

Share this post


Link to post
Share on other sites

Mixing to mono does not account for the lowpasses present in bass mgmt, which will make the square waves more saw-toothed.  Both C and LFE have near-square wave content in that scene.

 

 

JSS

Share this post


Link to post
Share on other sites

Mixing to mono does not account for the lowpasses present in bass mgmt, which will make the square waves more saw-toothed.  Both C and LFE have near-square wave content in that scene.

 

 

JSS

This depends on the nature of those filters.  I mention this because IIRC, 3ll3d00d uses linear phase filters for playback that might actually preserve the square wave shape.  On most systems, the lack of phase linearity of the crossovers will cause this deformation.  It remains an open question whether this deformation is actually audible though, assuming there's enough headroom to pass what comes out of these filters.

 

Edit: 3ll3d00d: do you mind clarifying for me which waveform is which?  And also, can you explain what you mean by "what my sub actually plays"?  Are you measuring the sub out or the response in-room?

 

The gist here is that there are many possible results depending on what kind of filters are being used.  And furthermore, even if the crossovers have linear-phase, the room correction filters may not be.  In fact, it would not be ideal if they were linear-phase because the room behavior that they are designed to correct is not linear-phase either.  It is much more common but not universally true that sub bass response in-room will be minimum phase and not linear-phase; hence, the correction filters are likely to be minimum phase, which will distort the square waves accordingly.  If the filters are "perfect", then there's a chance the result in-room will actually be flat and minimum phase, at least if you are listening at precisely the same position that the (single) measurement was taken.  Note that "perfect" is in quotes, because often a "perfect" for a single measurement position will yield quite poor results with small changes in location, so the "perfect" approach is not always best.

Edited by SME

Share this post


Link to post
Share on other sites

This depends on the nature of those filters.  I mention this because IIRC, 3ll3d00d uses linear phase filters for playback that might actually preserve the square wave shape.  On most systems, the lack of phase linearity of the crossovers will cause this deformation.  It remains an open question whether this deformation is actually audible though, assuming there's enough headroom to pass what comes out of these filters.

 

the final filter contains a linear phase crossover, some correction that is definitely minimum phase and some excess phase correction.

 

Edit: 3ll3d00d: do you mind clarifying for me which waveform is which?  And also, can you explain what you mean by "what my sub actually plays"?  Are you measuring the sub out or the response in-room?

the bottom one is "summed to mono with relevant gain adjustments but no filtering" 

the top one is the output from my correction filter

 

by "what my sub actually plays", I mean it is the signal that is sent via the SW out. jriver lets you apply it's dsp engine as part of its "convert format" feature, in this case I just configured it as per my usual listening setup and then opened the resulting wav in audacity. This means it went through the following steps

 

- mix down to 5.1 by jriver

- pre bass management gain adjustment 

- convolution filters applied (which apply the correction and sums the low passes)

- post bass management gain adjustment 

Share this post


Link to post
Share on other sites

the final filter contains a linear phase crossover, some correction that is definitely minimum phase and some excess phase correction.

 

Thanks!  What I said should make perfect sense then.

Share this post


Link to post
Share on other sites

slightly random question but I happened to compare a raw digital capture of the EoT scene (mixed down to mono using jriver) vs what my sub actually plays (again using jriver but mixing through my bass mgmt and room correction filters). I happened to notice the "square wave" aspect of the waveform was no longer present.

 

I posted something a while back similar to this here: http://data-bass.ipbhost.com/index.php?/topic/12-the-low-frequency-content-thread-films-games-music-etc/?p=6599

 

Your "sawtooth" style wave output does look different than what I normally see.  It's peak is at the end of the wave as opposed to the beginning like it what I show in my post.  Maybe that is due to your phase linear filters?  If you are taking that from an analog sub out jack it is to be expected that there is some distortion in the square wave because square waves seem to be hard for most analog audio equipment to reproduce accurately.  There will be a longer rise time than ideal and some ringing.   

 

You can always go to REW and experiment using the generator and try to get the square waves to be square on your sub out with your phase filters. 

Share this post


Link to post
Share on other sites

I suspect the deformation in my waveform is due to the shape of the correction filter more than anything else. I have a big cut at 40Hz to deal with my 1st axial length mode as I don't have space for subs front and back. I also use an NT (Neville Thiele) crossover which has a v steep slope. FWIW here a pic showing a vanilla NT XO (120Hz) overlaid with my correction filter

 

post-1440-0-86305900-1439246771_thumb.png

 

The big cut around 40Hz damps the harmonics of the main tones in that scene. The top capture is what is sent to my sub, the bottom is the result of just running it through a low pass.

 

post-1440-0-90485300-1439247812_thumb.jpg

 

I was also curious to compare the waveforms using different filters so the following graphs show the LFE channel with a 120Hz linear phase NT crossover (top row), my correction filter (middle row) and a 120Hz minimum phase LR4 (bottom row)

 

30Hz

post-1440-0-94729600-1439246862_thumb.png

 

25Hz

post-1440-0-24195800-1439246910_thumb.png

 

20Hz

post-1440-0-05318100-1439246919_thumb.png

 

15Hz

post-1440-0-66430400-1439246927_thumb.png

 

10Hz

post-1440-0-73454600-1439246936_thumb.png

  • Like 1

Share this post


Link to post
Share on other sites

I suspect the deformation in my waveform is due to the shape of the correction filter more than anything else. 

 

That would definitely explain it.

 

JSS

Share this post


Link to post
Share on other sites

11061182_859820164087381_696920796498051

 

Opening of film has deep pulse tones that peak on centre channel and LFE.1.

 

Spectrum lab, monitored on THX 3417 crossover/monitor.

 

11222243_859818657420865_624225581392204

 

Centre channel 

 

11844938_859818680754196_852992948658703

 

^ LFE.1

 

11823074_859825420753522_320629991543110

 

^ chapter 3 the side surrounds carry some lows, right surround.

 

11885646_859826990753365_898974989023478

 

^ One of the back surrounds rear right chapter 3, during the drop scene.

Share this post


Link to post
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.

Loading...
Sign in to follow this  

×
×
  • Create New...