HD600: My custom bass extension curves.
Jun 21, 2016 at 7:32 PM Post #91 of 123
Originally Posted by _gl /img/forum/go_quote.gif
 
Try putting the currve after the StereoConv, technically that's more correct as the EQ correction should be last.
 
Also try Sox linear phase = 45-46% Bob, I find it's slightly less 'hard' without smearing detail (it starts smearing to my ears below that on the Easy UES, though on resolving cans like the HD600 it's less of an issue).


Will do!  Thanks!
 
 
As an aside I've been a crossfeed user since 1993 with Tyll Hersten's Supreme headphone amp's circuit.  Things have come quite a way since then. 
 
Best of the bunch (for FB2K) in order of my preference currently are:
 
foo_dsp_stereoconv v 1.0.1 (w/Joe Bloggs HRFT 44.1 wav files v0.1; using SoX v 0.8.3  & _gl #4 HD600 curve @ 44.1 in convolver v 0.3)
 
foo_dsp_xfeed  v 0.1.3 (set with my head diameter, etc.); using SoX v 0.8.3 & _gl HD600 #4 curve @ 88.2 in convolver v 0.3
 
foo_dsp_stereoconv v 1.0.1 (w/Joe Bloggs' 30 degree 44.1 wav files v 1.1); using SoX v 0.8.3 & _gl HD600 curve #4 @ 44.1 in convolver v 0.3
 
foo_dsp_meiercf v 1.1 (with settings anywhere between 15 and 25); using SoX v 0.8.3 & _gl HD600 #4 curve @ 88.2 in convolver v 0.3
 
Aug 3, 2016 at 1:38 PM Post #92 of 123
Great thread, glad it was suggested to me by Bern. So I am using Curve 4 in JRiver. Just reread where the TS confirms that he does drop the gain by -4db to allow headroom. Makes sense. I'm just comparing this technique versus using  low-shelf filter to boost centered on 80Hz and boosting around 3db.
 
Aug 3, 2016 at 3:44 PM Post #93 of 123
Thanks Sonic.
 
The reason for the 4db drop is that the curves boost the lower bass a lot, so when you get something loud and bassy in a track, it can overload.  The volume drop buys you some headroom for that.
 
You might find that it's not enough (depends on the track).  Foobar seems to have a limiter built in that stops the harsh digital distortion if it still goes over, not sure what jRiver does.
 
Re. comparing, you can use the "-4db" curve to compare.  That's just the original sound with the same -4dB drop, so it's matched with the curves.
 
Aug 3, 2016 at 3:49 PM Post #94 of 123
Re. comparing, you can use the "-4db" curve to compare.  That's just the original sound with the same -4dB drop, so it's matched with the curves.

 
Argh, I was thinking of my other corrections (for the Easy UES), I haven't included the 'Dry -4db.wav' impulse in the HD600 ones yet.  Here it is:
 
 
 
Unlike the curves, it works at all sample rates.
 
Aug 3, 2016 at 4:06 PM Post #95 of 123
  Thanks Sonic.
 
The reason for the 4db drop is that the curves boost the lower bass a lot, so when you get something loud and bassy in a track, it can overload.  The volume drop buys you some headroom for that.
 
You might find that it's not enough (depends on the track).  Foobar seems to have a limiter built in that stops the harsh digital distortion if it still goes over, not sure what jRiver does.
 
Re. comparing, you can use the "-4db" curve to compare.  That's just the original sound with the same -4dB drop, so it's matched with the curves.

I like what the curves do, I have tried 3 and 4. The treble seems to drop back and things in general seem to get a little more recessed. Is this due to some actual treble cuts in the curves, or is it simply due to the increased bass? Anyway, it is obvious you did some really solid work as the results sound very clean and quite natural. I'm going to keep using the curves for a time, then switch over to the low-shelf approach and see what feels like the best fit. I wish my OTL was here as that is the amp the HD 600 will be paired with almost exclusively.
 
Aug 4, 2016 at 7:16 AM Post #96 of 123
  I like what the curves do, I have tried 3 and 4. The treble seems to drop back and things in general seem to get a little more recessed. Is this due to some actual treble cuts in the curves, or is it simply due to the increased bass?

 
It doesn't sound that way to me.  It might be that you have to get used to the extra bass (technically it's not really extra, rather restored to what it should be), when something else gets more prominent, the rest will initially feel recessed.  But it may also be the volume drop, that confuses us massively.  For example, even a tiny volume increase sounds 'better' to us in A/B, even though nothing has changed except volume.  So comparing with the volume matched Dry impulse I posted should help there.
 
Let me know how you get on.
 
Aug 4, 2016 at 7:27 AM Post #97 of 123
   
It doesn't sound that way to me.  It might be that you have to get used to the extra bass (technically it's not really extra, rather restored to what it should be), when something else gets more prominent, the rest will initially feel recessed.  But it may also be the volume drop, that confuses us massively.  For example, even a tiny volume increase sounds 'better' to us in A/B, even though nothing has changed except volume.  So comparing with the volume matched Dry impulse I posted should help there.
 
Let me know how you get on.

Yes, I suspect it is just the psychoacoustics as you are suggesting. I will be using the HD 600 when my La Figaro 339 arrives today or tomorrow. I do think that I like the HD 600 better with your curves, either 3 or 4, both seemed nice and rather similar. Thanks for the great work. I wish this could be done with my ONKYO A800 which I think has crazy potential.
 
Aug 4, 2016 at 7:41 AM Post #98 of 123
  Yes, I suspect it is just the psychoacoustics as you are suggesting. I will be using the HD 600 when my La Figaro 339 arrives today or tomorrow. I do think that I like the HD 600 better with your curves, either 3 or 4, both seemed nice and rather similar. Thanks for the great work. I wish this could be done with my ONKYO A800 which I think has crazy potential.

 
It's not really hard to do as long as you can find a flexible EQ where you can really shape the curve.  More generic shelving EQ can sound fine, but is not accurate enough to really 'correct' headphones.  I found the smallest curve adjustments made an impact, and it's really in the fine tuning and your ears.
 
Ideally you'd measure your cans and create the correction from that.  But it turns out that doesn't necessarily translate perfectly to another person's copy, there seems to be a bit of variation between copies.  Otherwise it's just down to your ear.  Foobar for example has a VST addon that allows you to use VST plugins, there's quite a few free good VST EQs out there.  I bet JRiver has VST support somehow.
 
Aug 4, 2016 at 12:38 PM Post #99 of 123
   
It's not really hard to do as long as you can find a flexible EQ where you can really shape the curve.  More generic shelving EQ can sound fine, but is not accurate enough to really 'correct' headphones.  I found the smallest curve adjustments made an impact, and it's really in the fine tuning and your ears.
 
Ideally you'd measure your cans and create the correction from that.  But it turns out that doesn't necessarily translate perfectly to another person's copy, there seems to be a bit of variation between copies.  Otherwise it's just down to your ear.  Foobar for example has a VST addon that allows you to use VST plugins, there's quite a few free good VST EQs out there.  I bet JRiver has VST support somehow.

So what would you suggest by way of decent free EQs? Ideally it would incorporate a feature that allows you to see how the adjustments effect frequency response. I don't even mind paying about $30, but what I'm finding is that the paid programs out there are of course for mixing and not for my purpose. I think it is important to see the curves and even better as I said to see how the frequency response is effected. Any thoughts are welcome (from anybody of course). DSP isn't my thing, but I think I need to start learning about this very important subject.
 
Aug 4, 2016 at 12:59 PM Post #100 of 123
  So what would you suggest by way of decent free EQs? Ideally it would incorporate a feature that allows you to see how the adjustments effect frequency response. I don't even mind paying about $30, but what I'm finding is that the paid programs out there are of course for mixing and not for my purpose. I think it is important to see the curves and even better as I said to see how the frequency response is effected. Any thoughts are welcome (from anybody of course). DSP isn't my thing, but I think I need to start learning about this very important subject.

 
I don't use free ones, but one option is the free plugins form the Reaper DAW, they should work in any VST host and they seem to have one or two curve-editing EQs in there (and lots of other goodies).  Their GUIs aren't as pretty as paid stuff, but their DSP is usually really good: http://www.reaper.fm/reaplugs/
 
Aug 4, 2016 at 4:52 PM Post #101 of 123
   
I don't use free ones, but one option is the free plugins form the Reaper DAW, they should work in any VST host and they seem to have one or two curve-editing EQs in there (and lots of other goodies).  Their GUIs aren't as pretty as paid stuff, but their DSP is usually really good: http://www.reaper.fm/reaplugs/

Yes I downloaded that but it didn't seem to install, but I think I know what I was doing wrong so I'll try again. So, you take a song or a few and develop an EQ correction that you feel improves the headphone. Then you save that curve which can be used by the Convolver plugin in JRiver. Is that essentially a simplified version of what I will be doing? Thanks mate.
 
Aug 4, 2016 at 5:44 PM Post #102 of 123
  Yes I downloaded that but it didn't seem to install, but I think I know what I was doing wrong so I'll try again. So, you take a song or a few and develop an EQ correction that you feel improves the headphone. Then you save that curve which can be used by the Convolver plugin in JRiver. Is that essentially a simplified version of what I will be doing? Thanks mate.

 
Unfortunately making impulses is more involved, you basically have to render a special single-sample file through the EQ, and there's quite a bit of editing involved afterwards (even worse if you want to support lots of sample rates).  So that needs proper audio editors & time : ).
 
Instead you can save your EQ's settings as a preset file (all VSTs support that) and share that with others.  if the EQ is free they can use it with their own copy.
 

Users who are viewing this thread

Back
Top