Robbo9999, what software do you use for EQ calcs? Does it require a lot of math background or other expertise?I've not mathematically calculated your changes based on the the Oratory pdf:
But, you're right, the bass is great without any EQ as you can see from the RAW (without EQ) measurement at the above link. To be honest, looking at that pdf, I think you can just leave the bass at stock, so you'd set both Low Shelf Filters to 0dB (ie no effect), and you'd just keep the peak filter at 100Hz (Filter #1) which is just acting to smooth out the transition from upper bass through mid/low bass. So if I were you, based on the measurement, to avoid clipping & whilst achieving the good stock low bass extension then just deactivate the two Low Shelf Filters and keep Filter #1. Nice!
EDIT: that gives you 2db less energy at 50Hz and below - literally just following the stock bass of the headphone rather than EQ'd version. (if that does indeed keep you from clipping your little dongle, it is less energy than the full Oratory EQ).
EDIT #2: otherwise, if that doesn't work & you're still clipping then try the following, do as I said previousy in this message but just add an additional 105Hz Low Shelf Q0.71, and try -1dB on that, and keep decreasing that until you don't detect clipping. I guess it must be very underpowered to clip, but yeah, if you're detecting it for sure then this would help when added onto my changes above......this would provide the smoothest bass transitions through the freuqeuncy range (smoothest curve).
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Drop Dan Clark Audio Aeon Closed X
- Thread starter TheHighContemplator
- Start date
That's a good point, and you really should focus on that if you continue to have bad clipping whilst not having adequate bass, but the changes I was showing you was to preserve a linear flat bass extension to 20Hz, but just at a lower level. If you want to experiment with higher bass levels whilst cutting out the lowest subbass, then you can follow my previous advice but also add a High Pass Filter to cut out the subbass, so that might enable you to get the best of both worlds.
Now, the High Pass Filter you want to use, well start with a 12Hz Q0.75 High Pass Filter which will not effect frequencies above 20Hz, but will remove below 20Hz (this might be too low(probably), but try it), following are pics illustrating the effect of this filter (note that this screenshot was done ages ago before I really knew about EQ properly, but it illustrates the point):
before High Pass Filter:
after High Pass Filter:
That's probably a bit mild for your usage scenario, so I'd go with at least a 17Hz High Pass Q0.75 combined with the recommendations I gave you in the previous post, especially if you don't want to do that negative dB 105Hz Low Shelf I mentioned. So you could keep the stock bass of the headphone and roll it off with this High Pass Filter. You could even experiment with doing what I said in my previous post but instead doing a +1dB Low Shelf at 105Hz and upping the High Pass to say 20Hz High Pass Q0.75, maybe even 25Hz (or higher up to 30Hz max I'd say during which point you're losing the normal wonderful bass capabilities/benefits of EQ'ing headpone bass). The thing to remember though is the Q0.75 is important so that it doesn't change the shape of the frequency response above it - if you put in a higher Q and make the cut-off sharper then it will also boost the frequencies above it "proportionally", so to maintain the overall shape of the bass and provide a slow roll off then you'd use a Q0.75 High Pass.
If you want to experiment with EQ filters, then REW (https://www.roomeqwizard.com/ ) is an absolutely awesome & free program that is ridiculously powerful & flexible if you've spent a lot of time with it!
Last edited:
@JanesJr1 , regarding Jimbob's 2kHz peak theory, sure the preamp is setup to negate that peak, but if you're hearing clipping it's unlikely to be coming from there. Firstly, you'd have set your preamp so that you wouldn't get digital clipping and secondly the music spectrum of most music does not have the greatest peaks at 2kHz, so the most energy suck from your tracks is instead likely to come from the bass area where the dBFS level is more close to full scale (0 dBFS), so there's probably even an argument to be had that you don't even need a negative preamp if you're boosting 2kHz by 5dB, but you'd probably want an auto digital clipping meter engaged in your software to cover you for that eventuality (but also you don't need to remove the negative preamp because you've not complained of low overall volume levels). Additionally, you said you noticed that you had to decrease bass levels to avoid hearing clipping - so that doesn't marry together with the 2kHz peak theory.
Best situation is for you to implement an automatic software digital clipping meter - PEACE extension in EqualiserAPO has one or practically all playback software has one, that way you know you're not getting digital clipping (optional if instead fully covering with negative preamp), and then experiment with the bass levels that we've been talking about. The 2kHz thing doesn't marry up with your observations & the fact that most music has peaks in bass & not at 2kHz......but experiment & see where you're at.
EDIT: also sanity check yourself, are you really hearing clipping. Maybe it's just loose bass or just more bass than you like. I don't know.....just sanity check that you're really hearing what you think you are! Sometimes the bass in some tracks might be recorded with some distortion already present as part of the "artistic sound"....but I'm less certain on this, albeit distortion is often incorporated into some tracks in some areas on some recordings, intentionally or otherwise.
Best situation is for you to implement an automatic software digital clipping meter - PEACE extension in EqualiserAPO has one or practically all playback software has one, that way you know you're not getting digital clipping (optional if instead fully covering with negative preamp), and then experiment with the bass levels that we've been talking about. The 2kHz thing doesn't marry up with your observations & the fact that most music has peaks in bass & not at 2kHz......but experiment & see where you're at.
EDIT: also sanity check yourself, are you really hearing clipping. Maybe it's just loose bass or just more bass than you like. I don't know.....just sanity check that you're really hearing what you think you are! Sometimes the bass in some tracks might be recorded with some distortion already present as part of the "artistic sound"....but I'm less certain on this, albeit distortion is often incorporated into some tracks in some areas on some recordings, intentionally or otherwise.
Last edited:
- Joined
- Oct 25, 2019
- Messages
- 11,098
- Likes
- 14,755
I assumed he meant clipping due to his dongle running out of juice. Not digital clipping due to illegal samples.
It will manifest more on low bass tracks as that would be requiring the highest current draw. You overcome hard clipping by not working the device as hard. I suspect he wants higher volume than the dongle powering the DCA can manage. Hence suggesting losing some of the preamp. But to do that he would have to lose some of the 2k peak. Not everything revolves around eq. Sometimes it's basic electronics
@JanesJr1 a simple test. Disable all eq filters but leave the -5db preamp. Then turn the volume up as loud as you can handle on some of those bassy tracks (or full volume ideally) and see how that sounds. Briefly. I suspect you will hear lots of distortion and possibly drops in volume. If you dont then I am probably wrong and you're not getting clipping on the amp end of things. Unlikely its the drivers clipping but you never know. But you shouldnt be getting digital clipping with that EQ- like Robbo says test it with software that has a clipping indicator (but dont use a clipping limiter , that defeats the point of the test)- if the red light flashes, its digital clipping.
EDIT- just got my Aeon RT out and my Meizu dongle - with a -5dB preamp (no filters) I can max the dongle- if I add a few dB bass shelf I can get it to clip- but that clipping manifests as cuts in the sound not distortion of any part of the spectrum. The s9 (think thats what you are using I believe has more power so clipping might manifest differently on that.
Last edited:
The more I learn about EQ, the more I find out that I don't know about it. (Does that mean I'm coming out of Dunning-Kruger land?)
I will try the PEACE extension, though I kind of like the plain-vanilla EAPO.
Do I understand your comments correctly that the pre-amp's required setting in EAPO isn't adequately keyed to the underlying power requirement as it varies by frequency? I.e. take it seriously at 100 Hz, but less seriously at 2000 Hz, esp. if I have a clipping limiter?
I think it is clipping, because I went back to some very familiar tracks with significant bass. At the worst with volume turned up, when a deep bass line appears, or a deep bass percussion transient hits, most of the sub-bass simply disappears, and the overall SPL drops for all frequencies.
With some EQ adjustment, I will still hit passages where there is some degree of "pumping" of SPL with dropout of the lowest frequencies. Most acoustic, classical, and even rock sounds ok but may edge into hollowness in the lowest frequencies. (I'm laid up at home for a bit, but when I get back to my desk and listen on my desktop amp, I'm expecting I'll hear the clean lows and see how much I have accommodated to the bass attenuation with the dongle.) But big organs, metal, and EDM and other electronica do bring it out.
On the one hand, to solve this problem, I probably need more power, maybe an E1DA dongle or use a different mobile headphone. But honestly, I'm learning something from trying to attenuate the problem with EQ.
Again, thanks for the EQ-101 and specific problem-solving; we all owe ya.
This is helpful I will try it later on today.
At the same time, I don't clearly understand digital clipping vs. not-enough-juice clipping. I hate to keep tapping into you guys to learn the basics. I'm going to look around for reference sources to bone up on these things. But if you already have a good resource in mind, let me know.
I'll get back to you on your acid-test. Thanks!
- Joined
- Oct 25, 2019
- Messages
- 11,098
- Likes
- 14,755
That's exactly what I'm talking about and it's the dongle trying to output more than it can. Whether that is technically the dongle amp clipping I know not.
Yes, generally if your EQ has peaks at 2000Hz it's less likely to digitally clip than if your EQ had the same sized peak in the bass if you set a negative preamp that doesn't fully cover your maximum EQ boost of your Total EQ Curve (highest point on the Total EQ Curve) - because most music is closer to full scale (0 dBFS) in the bass sections of the track. You don't always have to use the full negative preamp anyway, sometimes you can get away with say covering 75% of the boost and then have an automatic clipping meter operational alongside that to avoid & catch any digital clipping on the tracks that aren't fully covered by the "insufficient" negative preamp.
Well it seems that you really have identified the problem, and that it's occuring in the bass. So just play with the different bass EQ strategies myself or others have mentioned. I like the idea of using a High Pass filter to enable you to still get good bass without clipping your dongle, but you'll have to experiment to see if it solves the problem & where best to set that High Pass on the frequency range (try anywhere from 12-30Hz, Q0.75).....it'll be a trade off, you just have to experiment to get the best overall bass performance within the power budget that your dongle can provide.
@JanesJr1 , actually you could try the Linkwitz Riley High Pass Filter that is in PEACE:
It's sharper and still doesn't change the shape of the frequency response above it much. That would be better for you, the Q0.75 High Pass I suggested is too shallow to have a significant effect for you I think.
(EDIT: these are old EQ's I'm picturing here, I don't use these anymore, nor do I use High Pass Filters anymore. Instead I like to use wide peak filters in the bass to shape and then small gain Low Shelf's during listening tests at 105Hz just to fine tune overall bass amount.)
It's sharper and still doesn't change the shape of the frequency response above it much. That would be better for you, the Q0.75 High Pass I suggested is too shallow to have a significant effect for you I think.
(EDIT: these are old EQ's I'm picturing here, I don't use these anymore, nor do I use High Pass Filters anymore. Instead I like to use wide peak filters in the bass to shape and then small gain Low Shelf's during listening tests at 105Hz just to fine tune overall bass amount.)
Last edited:
Wow, was the high-pass filter effective! Much better than my roll-off. I mean a transformation. I know I had partly accommodated to the bass attenuation of the dongle, but I wasn't prepared for the return of not-just-bass, but deep-bass, with a Q.7 high-pass filter at 20 Hz. No SPL pumping or bass dropout, even on bass-heavy rock, metal. Maybe just a shade of bass hollowness on some tough electronica, but it was so subtle I'm not even sure it was there .. true fidelity is a hard call on some of that stuff. Theoretically, a bass-head might miss 1-20 Hz, but that band is the far-frontier of audio generally, and I don't miss it at all with the positive impact of the high-pass on deep bass generally and the portable convenience of the dongle with a low-impedence, current-eating headphone.
This is definitely a tool for Hidizs S9Pro dongle owners trying out the Closed X.
I also am trying out 2-3dB cut of the 2150 Hz peak, partly because it did help headroom some before, and because I want to hear how it affects playback in that band; there are some selections where I think things are a tad hot there. I'll A/B it on/off to see the effect both on bass/headroom and mid/treble tonality.
Thank you, JimBob and Robbo! You helped me solve a problem and got me way up the EQ learning curve. I'll keep it going, and try out PEACE.
P.S. several days later. I now believe that the Hidizs dongle in balanced mode provides enough power for most situations with the Closed X headphones. As detailed later in this thread, some of my listening sessions had limited network bandwidth, which caused attenuated bass and other sound quality problems that resembled inadequate current for low impedence 'phones.
Last edited:
That's good then! So you used High Pass Filter 20Hz Q0.7, that has the following effect when viewed as that filters effect on a flat frequency response:
So it's not really reducing/damaging bass above 40Hz, and between 20-30Hz it's cutting out a reasonable amount, and from 10-20Hz it's cutting out a boat load. Some music has content below 20Hz as some subwoofers aim to hit lower than that sometimes which is content you certainly feel more than hear when listening to speakers & subwoofers. I think a fair amount of music has significant 20-30Hz content, music like Massive Attack which is one of my favourite bands has a good chunk in that area for instance, I don't know if they have much sub 20Hz though as I don't have spectrum analysers to look that low with enough granularity. Well I'm glad that worked for you......are you using the Oratory EQ with this High Pass Filter slapped on or have you tweaked the bass in other ways?
Funny, Massive Attack is one of my electronica reference groups. And it's interesting you say there's a lot in the lowest bass bands with their music, because I was kind of amazed at how much a difference the high-pass filter made. I didn't think there was that much energy down there in most music
Other than the high pass, I left bass with no EQ (not at 100 or 105), and just relied on the native bass response of the 'phones. I might play around more with this; but for now, I'm just happy to have bass back in a big way. I'll try the steep high pass in PEACE when I get a chance to install and learn the s/w.
Last edited:
According to this power calculator:
Headphones Power Calculator
Calculate the power needed to drive your headphone based on sensitivity (db/mw or db/V), impedance (ohms) and loudness (dbSPL)
www.headphonesty.com
104dB is only 20 milliwatts and 40 millamps
109dB is 64 milliwatts and 72 millamps
114dB is 200 milliwatts and 125 millamps
And your Topping A50s was reviewed here of course:
Topping A50s Headphone Amplifier Review
This is a review and detailed measurements of the Topping A50s. It was kindly sent to me by the company for testing. The A50s is already available in limited quantities and costs US $199.99. Despite being small the A50 is rather heavy and sturdy which bodes well for staying on the desk...
www.audiosciencereview.com
Seems like there is plenty of power / current being offered by the Topping, so I don't think that's a thing to describe your observations.
Sorry, I transcribed off the "power" line, rather than the "current" line by mistake. I deleted the misinfo from the thread."
According to this power calculator:
94dB is only requiring 2 milliwatts and 12.6 millampsHeadphones Power Calculator
Calculate the power needed to drive your headphone based on sensitivity (db/mw or db/V), impedance (ohms) and loudness (dbSPL)
104dB is only 20 milliwatts and 40 millamps
109dB is 64 milliwatts and 72 millamps
114dB is 200 milliwatts and 125 millamps
View attachment 193849
And your Topping A50s was reviewed here of course:
Topping A50s Headphone Amplifier Review
This is a review and detailed measurements of the Topping A50s. It was kindly sent to me by the company for testing. The A50s is already available in limited quantities and costs US $199.99. Despite being small the A50 is rather heavy and sturdy which bodes well for staying on the desk...
Seems like there is plenty of power / current being offered by the Topping, so I don't think that's a thing to describe your observations.
However, I am still left with the front-line fact that my under-powered dongle now has more "thwack" (albeit with a high pass filter at 20 Hz) than my desktop amp. The slam-shy desktop sounded just like the starting-to-clip dongle with the same music. I guess I'll have to play with that when I'm back at my desktop and can explore it.
It's tough, it might be your mind playing tricks, you'd have to try to A/B them at around the same volume level.
Yes, I'll verify it when I'm back at the desktop.
Meanwhile, is there some aspect of PC USB power handling that could affect current available to the headphones?
I've an apparent anomaly, and it must have something to do with powering a dongle from a USB port. When I applied the successful high pass filter to attenuate the bass clipping on my work PC (Thinkpad T490) it worked beautifully. When I listen to it now, it still sounds like a successful EQ bandaid for the dongle's current limitations.
However, when I apply exactly the same EQ on my Surface Pro 7, there is still significant bass clipping. This is not a volume-leveling thing; it is true even if I depress the volume on the clipping PC, and increase the volume on the other PC. The pattern is the same on each PC whether I connect USB-C to USB-C, or connect to USB-A with an adapter.
The EQ's were created from the same EAPO text file, and I have compared them more than once to make sure they're exactly the same. The sound device setup in Windows on each machine is identical. Both PC's played at 24/96.
I often switch back and forth between these PC's and it helps explain why at various points in this process of tweaking EQ's, I sometimes thought I had found significant improvement, only to find the next morning the return of clipping. Now, I'm A/B switching the headphone dongle back and forth between PC's, playing the same music, with the volume lower on the clipping PC: deep, sustained organ notes wobble and pump on that PC; while they're loud, clear and smooth as glass on the other.
Is there some USB power limitation that could be at work in the SP7? The SP7's USB ports otherwise seem to work ok. It's a variable I hadn't anticipated. It is not a subtle effect; it is quite striking, but doesn't seem right. What am I missing?
If you're for sure certain in your observations there, then for some reason it seems your Surface Pro is not able to output enough power to your dongle. Different types of USB port have different specs in terms of how much "power" (I say this in quotes because I don't know the unit of variance - is it milliamps?) they can offer, and sometimes there are anomalies like some ports are charging ports that can offer more power than standard.......hmm, I don't know how much variance there can be from manufacturer to manufacturer but there are for sure some minimum standards associated with various types of USB ports. Maybe that's the deal, you're experiencing. I can't help you much with that, Google is your friend as well as knowing specifically the specs for the ports on your specific devices.
david.aj.silva
Member
I've been playing with my ACX eq a bit for about a month, and settled on a 6 band solution based on @oratory1990 FR graph and @Robbo99999 adjustments that sounds quite good to my old ears.
Using RME ADI-2 FS
1- low shelf G 5 F 150 Q 0.7
2- peq G 3 F 1k Q 3
3- peq G 3 F 2k Q 3
4- peq G 3 F 7k Q 3
Bass adjustments
G 0>+5 F 100 Q 0.8
Treble adjustments
G 0>-4 F 6.5k Q 0.7
A few recordings benefit from the B & T adjustments, the rest only the 4 fixed bands.
Thanks to contributors to this thread!
Test music:
Return to Forever: Romantic Warrior, Light as a Feather
Al Dimeola: Opus, Elysium, Casino
Edgar Winter: Jasmine Nightdreams
The Band: Rock of Ages
Jean Luc Ponty: Aurora
Mahavishnu Orchestra: Visions of the Emerald Beyond
Using RME ADI-2 FS
1- low shelf G 5 F 150 Q 0.7
2- peq G 3 F 1k Q 3
3- peq G 3 F 2k Q 3
4- peq G 3 F 7k Q 3
Bass adjustments
G 0>+5 F 100 Q 0.8
Treble adjustments
G 0>-4 F 6.5k Q 0.7
A few recordings benefit from the B & T adjustments, the rest only the 4 fixed bands.
Thanks to contributors to this thread!
Test music:
Return to Forever: Romantic Warrior, Light as a Feather
Al Dimeola: Opus, Elysium, Casino
Edgar Winter: Jasmine Nightdreams
The Band: Rock of Ages
Jean Luc Ponty: Aurora
Mahavishnu Orchestra: Visions of the Emerald Beyond
Last edited:
Did you ever graph up the Total EQ Curve that results from your EQ vs that of my EQ and Oratory's EQ - it might be interesting to see where the differences are? It's ok if you haven't, not many people would have done so, just that's the best way to understand the differences between the various EQ's. But either way it's good that you've got to something that works well on yours.
david.aj.silva
Member
I didn't draw them out, but I did compare the major dips and peaks in the curve and how the corrections from you and Oratory fit. My final dcision was to address only the biggest dips in the midrange / lower treble, and instead of decreasing the midbass bump to incorporate it into the low shelf correction.
In trying to correct the midbass bump and then apply the low shelf correction, I noticed on the well produced RTF album, Stanley Clark's bass would walk up and down and the volume of the notes would swell down low and thin out up top. Also the bass would often take center stage, this does not happen live with that group. Unless improvising out front all instruments are evenly balanced volume wise. No slamming bass leading the way. Anyway, that is what I aimed for, even-ess of bass. With the FR graph as the starting point.
Similar threads
