In the interest of integrity, if there has been one thing I have learnt, everything is subject to change. And opinions, including mine, also change. Ears get tired after long days of listening and in the bright light of another day after a good nights rest, we hear things differently. Same - even after a few months, we hear certain things better.
It would be nice to live in a world, where we can change our minds, and say so, and that revision in our thinking is respected, and our honesty/integrity, is appreciated, but that is not always the case, sadly.
It was great to have come across AutoEQ, but I've changed my mind about it. Too many opportunities for inaccuracies. No certainty of how close your own copy of the headphone/IEM matches the specific copy you choose in AutoEQ's database, not sure what the algorithm is doing. Unfortunately with audio, any change is likely to be welcome to the ear, cos it's different and sounds refreshingly different. Question is - is it correct. With AutoEQ, which is a interesting idea, it fails in the fundamentals, similar to how any solution based on a database of measurements, will be flawed.
So this is not a criticism of AutoEQ itself, but that whole approach, and same applies to the Oratory based corrections. We also have the challenge of vetting the algorithms, of each of these methods. How correct are they? Pretty hard to tell. We are in uncertain territory.
@GaryH - sent you a far more detailed PM, explaining the assertions I have made here. And why although it also has similar issues, cos it is also based on a database of measurements, the immediacy of using a plugin like Morphit by Toneboosters, will deliver certain advantages, over AutoEQ or Oratory based corrections.
Thinking through the possible inaccuracies, I had to stop using AutoEQ, a few weeks ago. In another few years - hopefully measurement rigs of sufficient accuracy will become affordable, and we can do what we already do with speakers in a room, measure and correct the exact device we have in front of us.
If one thinks about this critically, no one in the speaker world would correct their speaker based on a pre-done measurement of another speaker. So why should we do this with headphones/IEMs? The approach of using a pre-measured result, is therefore, inherently flawed. Any use of such approaches, MUST have this caveat in mind.
I have had an interesting result with an alternative solution which you will find at PEQdB.com. Won't bother to discuss details here, cos its all easy to figure out on their web site. It also has some kind of database, at least of user preferences, and its algorithm is also not known to me. Nevetheless when trying this out I got the following interesting result
I pulled these into REW so I could see them side by side.
The green curve (see image below) is one of the measurements of my IEM - the ARTTI T10, from the squig.link database, measured by ToneDeafMonk, and not a measurement of my own copy of this IEM. The red curve is the correction generated by PEQdB.com
While the correction, generated by PEQdB.com seems to have also introduced my own personal listening preferences, into the correction, its somewhat uncanny to note, how there is a correlation between the correction, and an example of the measurement of another copy of my IEM.
At this time, I can say, I like the end result of the correction, better than the uncorrected result. Certain sibilances and harshness in the ARTTI T10, appear to have been reduced, and the end result is smoother. The bass has obviously been attenuated, I could hear this from listening, and it clearly shows in the correction graph.
I was already quite satisfied with the Non EQ'd result of the T10 with the eartips which I normally use on them, after tip rolling to discover what I thought was the most balanced. So the result of using PEQdB.com, was what I think at this time, is an improvement on an already excellent listening device, to make it sound that bit better.
Caveat. I did not have a similar positive experience attempting to correct another IEM, the CCA Polaris. But I also have not enjoyed listening to the CCA Polaris, without EQ, and any attempts by me to improve its sound by me creating EQ filters on my own from listening, had poor quality results, and I do not use this IEM for this reason, just not one I enjoy listening to. With or without EQ.
I think with PEQdB.com, the listening device itself, and how well it resolves audio, will have an impact on how one responds to the listening tests in PEQdB.com, and the effectiveness of the results. The ARRT T10 is a more resolving listening device than the CCA Polaris, to my ears, and most likely that is why I am having a more favourable result, with that IEM and the correction effort by PEQdB.com.
Last edited:
