khentamentiu
Member
- Joined
- Oct 3, 2021
- Messages
- 16
- Likes
- 1
seriously try it and discuss
-
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!
[ANANDA] how far off is this EQ
- Thread starter khentamentiu
- Start date
OP
khentamentiu
Member
- Joined
- Oct 3, 2021
- Messages
- 16
- Likes
- 1
- Thread Starter
- #3
"He intentionally leaves resonances untouched, so if you want a less plasticky timbre, you'll need to find the ~5 kHz dip and ~8 kHz peak on your specific pair with a tone generator and use high-Q peak filters to flatten them out. Then again, you may as well get a different headphone if you dislike HifiMan timbre."
- Joined
- Oct 25, 2019
- Messages
- 11,140
- Likes
- 14,815
Which is all well and good, but you have used shelf filters at 5 and 8k, not peak.
Use the graph function on Peace and see what effect enabling and disabling that 5k filter has ;-)
OP
khentamentiu
Member
- Joined
- Oct 3, 2021
- Messages
- 16
- Likes
- 1
- Thread Starter
- #5
seems I can't get the graph working right,
but I changed them to peak and wow, that's great, [I misinterpreted when the advise was to "flatten them out"]
I do feel the harmon base plus these two adjustments might lose a little detail tho -- overall what a great sound
but I changed them to peak and wow, that's great, [I misinterpreted when the advise was to "flatten them out"]
I do feel the harmon base plus these two adjustments might lose a little detail tho -- overall what a great sound
- Joined
- Mar 1, 2020
- Messages
- 82
- Likes
- 327
Well you did put in 7 dB instead of 5.5 dB for the 105 Hz low-shelf filter, so currently you're boosting the bass more than the Harman Target.
OP
khentamentiu
Member
- Joined
- Oct 3, 2021
- Messages
- 16
- Likes
- 1
- Thread Starter
- #8
wow so my EQ might have been onto something huh? glad you found it enjoyable; I felt I had to bring that little tweak to the forefront!
Message sent to Amir. Thanks for the info.
Jose Hidalgo
Addicted to Fun and Learning
Your EQ ?...
It's normal, since you didn't use a tone generator like requested.
The whole purpose of the tone generator is to find the very specific frequencies, unique to your headphone, where the resonances are. Those frequencies aren't exactly 5000 and 8000 Hz, they depend on each headphone.
I guess you just applied peak filters at exactly 5 and 8 KHz, with some random Q factor, hoping for them to magically work. It doesn't work like that. Otherwise Oratory would have added those two filters by default to his EQ preset which you used.
- Joined
- Oct 25, 2019
- Messages
- 11,140
- Likes
- 14,815
Your EQ ?...
It's normal, since you didn't use a tone generator like requested.
The whole purpose of the tone generator is to find the very specific frequencies, unique to your headphone, where the resonances are. Those frequencies aren't exactly 5000 and 8000 Hz, they depend on each headphone.
I guess you just applied peak filters at exactly 5 and 8 KHz, with some random Q factor, hoping for them to magically work. It doesn't work like that. Otherwise Oratory would have added those two filters by default to his EQ preset which you used.
Nope - initially they used SHELF filters at 5 and 8k, since amended to peak I believe, but yes, randomly
Jose Hidalgo
Addicted to Fun and Learning
Yes, I had read the posts : first shelf, then peak after you corrected him.
I think the OP is simply lacking some tech knowledge. But we've all been there, right?
- Joined
- Mar 1, 2020
- Messages
- 82
- Likes
- 327
That's what your EQ does to the frequency response of the Ananda (assuming peak filters for the 5k and 8k filters)
- Joined
- Oct 25, 2019
- Messages
- 11,140
- Likes
- 14,815
Indeed. I'm not sure what the OP was asking for though if it was a tweaked version of an existing Oratory EQ. I suppose validation from another user that the tweaks worked?Yes, I had read the posts : first shelf, then peak after you corrected him.
I think the OP is simply lacking some tech knowledge. But we've all been there, right?
OP
khentamentiu
Member
- Joined
- Oct 3, 2021
- Messages
- 16
- Likes
- 1
- Thread Starter
- #15
Well, why is there so much weird stuff going on with it? Cuz it sounds a lot better than it looks, lol
What further adjustments could be made?
What further adjustments could be made?
- Joined
- Mar 1, 2020
- Messages
- 82
- Likes
- 327
The most important adjustment to make whenever you're using EQ is to set the amount of bass that you prefer.
You can do this by adjusting the gain of the 105 Hz shelf filter.
Start with a baseline of 5.5 dB (that's what most people prefer) and then adjust up or down, depending on what you prefer.
about 2 in 3 people prefer somewhere from 3-5 dB.
About 1 in 5 people prefer somewhere from 0-3 dB.
And about 1 in 6 people prefer over 5 dB, as high as 15 dB.
Whatever you prefer is the correct number (for you).
After that, the second most important adjustment is the region from 10-20 kHz, which depends highly on individual ear anatomy.
That's what the 10 kHz shelving filter is for, adjust its gain to your preference (this will affect "airiness" or "detail" of the sound). Basically anything goes here, I've seen people prefer values anywhere from -20 to +20 dB.
OP
khentamentiu
Member
- Joined
- Oct 3, 2021
- Messages
- 16
- Likes
- 1
- Thread Starter
- #17
Thank you so much for this advice!
In reference to the strange dips and peaks? I think this is what was meant by trying to flatten out the 5k and 8k? That can’t be true either as look at the stock response. Those are built into the headphone’s tuning. Why?
Also did you have a chance to test the q slopes vs the peak filters at those frequencies?
In reference to the strange dips and peaks? I think this is what was meant by trying to flatten out the 5k and 8k? That can’t be true either as look at the stock response. Those are built into the headphone’s tuning. Why?
Also did you have a chance to test the q slopes vs the peak filters at those frequencies?
Last edited:
Fwiw, I agree with all of Oratory's excellent advice above. And would not try to mess around with the peaks and valleys in the higher frequencies, as you've been doing, khentamentiu. Because there is really not much to be gained from doing something like that, in this case. The Harman curve (which is represented by the light gray-green curve on the graph above, and also the ones below) is just a very general approximation curve, that is not totally accurate. So there are some places where your headphone's response should deviate from that target.
Some peaks and valleys in the treble are normal with Oratory's measurements, for example. And there should be notches in his data at around 9-10 kHz, and around 15 kHz in the treble. And probably also at around 20-21 kHz (though it's a bit harder to see that area, since the graphs end at 20 kHz.) Those notches are not caused by the headphones. They are the normal resonant characteristics of Oratory's measurement rig and artificial ears. So don't worry so much about those.
You should actually be more concerned if there is NOT a notch at around 9-10 kHz in Oratory's measurement for your headphone. Because that would probably indicate that there is a resonance, and some sibilance within the headphones themselves.
The important thing is really more the general overall levels of the peaks in the higher frequencies. And you do not necessarily want them to conform exactly to the shape of the Harman curve in that area. It is ok if a few peaks rise a bit above the Harman target above 10 kHz, for example. Because Harman's target response curve is a bit too depressed and rolled off for a neutral response in spots in the higher frequencies above 10k.
Here's a graph showing how some other neutral headphones look compared to the Harman curve, to give you a little better idea of how you might want to tweak the response of your Ananda to look, if you want it to have a neutral response. I've grayed out a few areas on the curves that you should ignore, where the responses of these headphones drop down too low to be neutral.
You will notice that there are fairly pronounced notches at around 9-10 kHz, and at 15 kHz on these headphones as well (just like on your Ananda). And there are peaks on these headphones, which also rise above the Harman curve in the treble above 10 kHz. All of that is completely normal on Oratory's graphs.
You will also notice that these headphones droop down a bit more than the Harman curve at around 1.5 and 2 kHz in the upper mids, which is also not a bad thing. Because alot of loudspeakers used for mastering and producing music content will also have a bit of a dip in their in-room response in that range. Because that is where the midrange drivers and tweeters on the speakers will tend to cross over.
If you want to boost the overall levels of your headphone's response a little bit higher in the bass and treble than what is shown above to make them a bit more pleasant for listening at lower volumes, that's probably ok. But I would start out with something a little closer to the responses of some of the headphones above. And then try to make your adjustments from there.
Some peaks and valleys in the treble are normal with Oratory's measurements, for example. And there should be notches in his data at around 9-10 kHz, and around 15 kHz in the treble. And probably also at around 20-21 kHz (though it's a bit harder to see that area, since the graphs end at 20 kHz.) Those notches are not caused by the headphones. They are the normal resonant characteristics of Oratory's measurement rig and artificial ears. So don't worry so much about those.
You should actually be more concerned if there is NOT a notch at around 9-10 kHz in Oratory's measurement for your headphone. Because that would probably indicate that there is a resonance, and some sibilance within the headphones themselves.
The important thing is really more the general overall levels of the peaks in the higher frequencies. And you do not necessarily want them to conform exactly to the shape of the Harman curve in that area. It is ok if a few peaks rise a bit above the Harman target above 10 kHz, for example. Because Harman's target response curve is a bit too depressed and rolled off for a neutral response in spots in the higher frequencies above 10k.
Here's a graph showing how some other neutral headphones look compared to the Harman curve, to give you a little better idea of how you might want to tweak the response of your Ananda to look, if you want it to have a neutral response. I've grayed out a few areas on the curves that you should ignore, where the responses of these headphones drop down too low to be neutral.
You will notice that there are fairly pronounced notches at around 9-10 kHz, and at 15 kHz on these headphones as well (just like on your Ananda). And there are peaks on these headphones, which also rise above the Harman curve in the treble above 10 kHz. All of that is completely normal on Oratory's graphs.
You will also notice that these headphones droop down a bit more than the Harman curve at around 1.5 and 2 kHz in the upper mids, which is also not a bad thing. Because alot of loudspeakers used for mastering and producing music content will also have a bit of a dip in their in-room response in that range. Because that is where the midrange drivers and tweeters on the speakers will tend to cross over.
If you want to boost the overall levels of your headphone's response a little bit higher in the bass and treble than what is shown above to make them a bit more pleasant for listening at lower volumes, that's probably ok. But I would start out with something a little closer to the responses of some of the headphones above. And then try to make your adjustments from there.
Last edited:
The same four neutral headphones above, with Oratory's raw uncorrected plot of the Ananda (without any EQ-ing applied) also included in black. So you can maybe see a little better where some adjustements could be needed in your headphone's raw response relative to some of the other headphones.
You should also use your own ears/hearing when making these adjustements btw. Because HiFiMan will make some tweaks and adjustments to the pads and some other features on their headphones from to time to time. Which could potentially change their response and tonal balance a bit from what might be depicted above on Oratory's graphs.
Amir has also posted some graphs of the Ananda btw. And his plot makes the headphone look more elevated in the bass than Ora's. Even though they are using similar measurement systems. So I'm not sure why there'd be such a discrepancy there. Perhaps Oratory could possibly shed some light on that. Or maybe HiFiMan made some tweaks not too long ago to improve the low frequency bass response in this headphone?
The response in the rest of the frequency range, in the mids and treble, look pretty similar though.
The above plot came from Amir's Ananda review here btw...
www.audiosciencereview.com
The response in the rest of the frequency range, in the mids and treble, look pretty similar though.
The above plot came from Amir's Ananda review here btw...
Hifiman Ananda Review (headphone)
This is a review and detailed measurements of the Hifiman Ananda open-back, planar magnetic headphone. It was kindly sent to me by a member and costs US $700 on Amazon including Prime shipping (retail $999). This is one massive headphone! As you see, it dwarfs the panther next to it. For...
www.audiosciencereview.com
Last edited:
Similar threads
