If I recall, Blaine looked into the Edition XS non min phase claims (I believe SBAF is where some of that originated) and found that it was an issue to do with how the measurement was taken. Don’t quote me on that because it was some time ago, but I have yet to find any non-min phase behavior with those types of headphones. It could be the case, but I have yet to see it.
The older HE-5LE and HE-560 are more pronounced with bad CSD behavior but they are old and you may never have experienced them. Audeze never seemed to have those issues. Or DCA since the E2. The Arya U doesn’t seem to either. Maybe
I’d have to see the data to confirm those. Because they may look like they have CSD issues in the CSD, but because time domain is proportional to FR in min phase devices, you’d see that due to the FR anyway. So I’d want to see the excess GD data on those to confirm, or examples where the CSD wall persists despite fixing the resonance in FR.
Probably true. EAPO supports variable frequency GEQ’s though with many bands. Literally 100’s, if need be, I think.
Since I mostly use the FR of other headphones to design my EQ curves, UsyTrace’s Points Per Octave setting often determines the resolutions of the trace (GEQ) curves.
If you want to capture alot of fine high Q detail in the treble, then you’ll need a higher PPO or resolution setting. But you may not really need it for the whole curve. With smoothed curves, you can use lower PPO settings.
The default PPO of 48 in UsyTrace will create a curve with 480 bands or points, if the curve is the usual 10 octaves (20-20k Hz). On a well smoothed FR curve, a PPO closer to 12 might be sufficient.
Approaches like I described above don’t preclude the use of PEQ though. I think Super* Review’s videos cover how to do this in squiglink. I appreciate some of the different flexibility that GEQ’s allow though in the EQ design stage.
Curves downloaded directly from squiglink also appear to have a resolution of 48 bands/points per octave btw. So their resolution is pretty high, and I’m not sure if that can be adjusted.
I’ve had better results with convolution EQ settings over PEQ settings. That is what I primarily use now. PGGB-RT with convolution settings subjectively works better than any of the PEW tools I’ve tried. Too many times, applying EQ winds up sounding un-natural.
There’s an obvious bias against closed back headphones here. With EQ applied, some of the higher quality closed back headphones sound more enjoyable (to me) than many of the higher ranked open back headphones. Don’t care for the “house sound” of some of the brands listed.
I’m out of my depth on the subject of convolution filters, though I believe EAPO’s Configuration Editor also supports this type of filter, in addition to GEQ and PEQ. Seems like it might be more useful for adding various sound effects. But I haven’t played around with it. And if you’re getting good results, more power to ya!
Fwiw, here is one discussion I found on the subject on Oratory1990’s reddit…
The main reason I prefer to use GEQ though is that I find it less cumbersome than PEQs. That may have somethin to do with the EAPO interface though, which is a bit clunky in some ways. Coming from a graphics background, I like the immediate feedback I get with GEQ. And it’s ability to create more complex freeform shapes.
EAPO allows you to use whatever kinds of filters you want in the same project though. So you could theoretically use all three approaches (Convolution, GEQ, and PEQ) at once to create your EQ, if you want to.
The GEQ’s I use could also be converted over to PEQ filters. (And presumably also Convolution filters.) I have only experimented with this a little in EAPO. And it seems like Squiglink might have some easier, or at least more automated tools for doing the conversions than EAPO. I suppose it depends on how adept you are at setting Q, gain, and frequency values though. Fewer is generally better I think.
I played around with Equalizer APO and ultimately it wasn’t effective for me. Maybe there was a mismatch with the signal path and it’s interaction with Windows on my laptop, but I did not like the sound. It sounded like the soul was sucked out. Another issue I have is the necessity of listening on a laptop. I picked up a small, super inexpensive SLMS 100 CDP I use with my desktop rig.
I came into adulthood in the analog era and have a large CD collection. While, I have most of them ripped into a NAS I use in the house with Sonos, I’m limited to the digital files I have saved on my laptop. I don’t currently have a streamer on my desktop so I can’t access my hundreds (maybe thousands) of CDs from my desktop.
I think it’s generational, really. I’m very much an album listener., not so much a tracks or playlist guy. I also don’t stream a huge amount as I prefer to own rather than rent my music. I just feel if I’ve streamed an album a few times, the artists deserves a few buck from me. I have a much bigger physical media collection than I have digital music on my laptop. But, I still want EQ options.
So…….
I recently bought a RME-ADI 2 FS DAC/AMP. I’m using it for it’s DAC and DSP features bypassing the amp feeding into external headphone amps (Violectric 280, GSX-Mini, Schiit Jot2 and Lyr+)
I’ve saved presets for each of my headphones (20 of them all together) and it works swimmingly. In some cases I have multiple presets for the same headphones for different music genres as I can go from extreme metal to Mozart sonatas to Bob Dylan to Miles Davis. I think RME flexibility is brilliant.
I also have a Bifrost 2/64 that I use with their analog knob twisting Lokius. The two give me opposite ends of EQ customization. The RME presets give me precision dental tools with a more analytical, but leaner sound. The Bifforst is nice in that it’s a meatier, more analog-y sound to me…more robust but less precise.
I’m using XLR switches to mix and match the two DACS with the four amps. It can be maddening at times but the payoff is often extremely rewarding when I hit the aural jackpot. Overall, it’s been quite fun futzing variables (2 DACS, 4 amps 6 headphones) each with their own chactter dialing in my sound preferences for various recordings, genres and EQ presets.
It sounds like you’ve got a great setup, Mark. I also had some reluctance about software-based EQs. And it took some time before I really started to feel more comfortable with the EAPO interface. Alot of folks will also download the Peace GUI for it, because it’s supposed to be easier to use. But to this day, I have still not tried it.
The main thing that won me over was the power, flexibility, surgical precision, and of course also the price on EAPO.
I understand why some folks don’t like it though. For starters, the interface can be quite clunky and difficult to use in some ways. It took quite a bit of time for me just to figure out how to properly scale graphs in the interface, for example, so I could see what I was doing better.
And I think alot of folks don’t understand some of its other ins and outs that well. And how bit depths and sample rates need to be set to get the best results, for example. This is something that also took some time and experimentation for me to figure out.
There are things you can do in EAPO that you cannot in a simple parametric or analog EQ though. So that is one of the reasons I continue to use it.
Fewer is generally better I think.