I was going to poke fun, but let my computer do the walking - with a hat tip to Schiit and @generic
He once claimed, “No EQ—divine tune!”
Scoffed filters from dusk until high noon.
Now Lokius near,
He tweaks without fear,
And grins at his past like a buffoon.
Sorry, not sorry. And clearly no bad intent (Aqualung). But I do like Lokius.
Yes. It’s easier to rip the file and upscale via PGGB-RT. Keeps the file size reasonable, and provides some level of improvement. The amount of improvement does vary, but at least there is improvement.
It’s worth stating that tools like PGGB/HQPlayer/Mscaler are not the same as or a substitute for EQ.
These tools give you better nyquist reconstruction, meaning you’re playing back a more accurate representation of the information contained within the digital sampled data, but they do not change the FR/tuning of the content itself other than at the highest frequencies.
I use pggb/hqplayer myself, and do find them beneficial, but if for instance you have a headphone with too much 5khz these tools aren’t going to fix that.
Nyquist for 24/48 is 24 kHz, which is well above my hearing range at age 63. That’s why I have some doubts about the benefits of such tools for my own use. Not sayin it won’t do somethin for others though.
Maybe some of my bones can still “hear” that high. 
I agree with this. There is more to doing EQ well than meets the eye, I’m afraid.
I know you understand this. But EQ doesn’t really do anything to the headphone’s response. It changes the timbre/tonality of the content playing through the headphones. Hopefully in a way that compensates for the timbral/tonal/euphonic imbalances in the headphone’s response.
And it can also fix some of the tonal imbalances in a recording.
It is not a perfect system though, because not all headphones (or speakers) are minimum phase at all frequencies.
This is true. However, I think there are some listeners who have unique EQ profiles for everything they listen to.
I prefer to use the same EQ profile for everything on a pair of headphones.
“Tone controls” can’t hurt though. And can also be added in a program like EAPO, using separate filters or GEQs. I will sometimes use a simple 2-band GEQ for this purpose, with controls at 20 Hz and 20 kHz, for example. Or sometimes a 7-band GEQ, with controls at 20, 63, 200, 630, 2k, 6.3k and 20k Hz, for some basic tonal/timbral tweaking.
This is probably true. And it is unfortunate what the loudness wars have done (and continue to do) to a good proportion of the music we listen to.
I don’t agree with this. There is a lot of full range audio content that isn’t music related. And you will need a system with good dynamic range to appreciate this type of content. THX certification requires 85 dB plus another 20 dB for headroom in the midrange (and a bit more for LF). So a total of 105-110 dB for transients. That’s probably a bit louder than an average listener would use in a home though.
Hearing loss can occur at much lower levels than 120 dB, with sustained listening.
When we talk about dynamic range though, we are not talking about the average or sustained listening levels, which are more likely to be around 60-80 dB, or possibly lower if you prefer to listen more quietly. We are talking about the full dynamic range of the system, including headroom for transients (transitory peaks in the music or content). Which can be 10 to 20 dB higher than the average sustained listening levels.
it is also a mistake to assume that you need good dynamic range only for the brief and louder transients. Audiophiles will often target dynamic ranges that are close to the threshold of pain of 120 dB because having such a wide range can potentially make some quieter music recordings easier for them to hear and enjoy.
Hearing loss can be another reason for wanting a higher dynamic range in your system (though I don’t recommend using it for this purpose).
It’s unclear where you’re going with this or why you continue to knit pick, but when you apply parametric EQ, the changes it makes to the input signal changes the frequency response. This can be confirmed by taking an FR measurement of the transducer after EQ is applied. The changes to the signal/FR are what change the perceived timbre and tonality of the transducer.
I think the Sennheiser HD 650 sounds excellent with no EQ on the majority of recordings. Probably one of the most neutral headphones at any price.
Peace. Out.
Errrrrrrr… 
Nope, too much mid bass. Way too much mid bass and not enough oomph in the vocal range.
Heh, good point. Sage advise as always.
When you alter the timbre or tonality of a recording going through your speakers or headphones with an EQ, you might perceive it as… A) the tonal response of the recording has changed, or B) the tonal response of the speaker or headphone has changed. Neither of these are really correct though.
The first assumption is probably somewhat closer to the truth though, because the audio signal that represents the recording is being changed or altered by the signal processing in the EQ. Which might make it seem as though the headphones or speakers have a different sound… This is just an illusion though. An illusion that many audiophiles enjoy taking advantage of to try to make their audio systems sound a little better.
Not according to the Soundworks measurements. In fact, just the opposite of the statement.
Then Soundworks is full of cr*p. Your ears matter. Measurements are secondary. Always. This is perception science, not subjectivism.
From a science perspective, the Sonarworks article is a lot more accurate and backed with evidence than one’s opinion. Check out the measurements. The HD 650 measures quite a bit better , or closer to neutral, than many headphones available.
Listener’s preferences are all over the map, and often don’t line up with measurements regarding neutrality. Horses for courses.
From a science perspective observation comes first.
Sorry this annoys me, if the majority of people who describe the sound of a headphone agree what sounds neutral (not that I think they do), that’s what “neutral” means, that’s how language works, and if your doing “science” you can try and quantify that with measurements.
You can arbitrarily define some state, lets say flat free field response in a speaker, as neutral, but that’s just a convention, and using the same word for it is confusing and prevents useful discussion, call it something else like “flat response” or “harmon response”, or “objectively neutral”.
This is one of the fundamental problems with discussing how things sound, people assume “neutral” is some sort of a reference, but people do not define neutral the same way.
In amps for example flat frequency responses can be perceived as warm or cold or mid forwards….
What sonarworks is trying to do is normalize frequency response to some “norm”, so an engineer can listen through a number of different devices and hopefully be comparing apples to apples, it’s not about EQ’ing to something that is “correct” in some way, they’ve picked an arbitrary target based on nearfield monitors in recording studios, and they are just trying to make everything else consistent with it.
if you happen to like that response, great, but it wasn’t what they were trying to do.
FWIW I went through the it should sound like what it does in the studio thing decades ago, and from personal experience most studios do not sound particularly good. Engineers aren’t targeting what they are listening on they are using it as a tool to understand how it will sound in a variety of playback environments.
I particularly dislike the use of the term “neutral” when picking somewhat arbitrary targets, because as a word it implies a “correctness” that isn’t really what it is.
Sonarworks is measuring headphones to a standard the touts a minimal amount of frequency deviation from 20 to 20 KHz. . Their definition of neutral frequency response is as good as any other reference. The headphones I’ve owned or auditioned that are supported by Sonarworks measurements have demonstrated (to me) that their EQ works well.
I’ll stand by the assertion that the HD 650 is one of the more neutral headphones as measured on the market. That assertion has no connection to what a given listener prefers. Many well regarded headphones, such as the Abyss 1266, has some significant frequency deviations.
We’ve had this discussion several times over the last number of years. The root assumptions start with epistemology (@Resolve) and the philosophy of science. Humans tend to act from direct or naive realism, whereby they consider “stuff outside” to be absolute and factual. Yet, no human has a God’s Eye perspective on the universe or anything – including mathematics and physics – it’s all invented, refined, and learned models of the outside world. We’ve created exceptionally great physics models that resulted in moon rockets, satellites, and the Oceangate Titan 
. Still, there’s an inherent uncertainty because every device we ever make and interpret is rooted in human observations (@Polygonhell). We all must learn how our sensory systems work and then learn to interpret models, be they math, electronic, chemical, or biological (human).
I expressly rejected subjectivism or “one’s opinion,” and I referenced perception science. This follows 100+ years of university labs and gobs of audio research. (Link to a free 463 page Sensation and Perception PDF textbook). From page 22/463 at the link, see the figure showing how aspects of human perception deviate from linear functions. Copied below:
Human biology and perception is arbitrary – but seems to be associated with survival/evolutionary needs. Human hearing potential follows from the need to perceive human voices, and consider the value of hearing screams for help and warnings of threats. Evolutionary analyses can be speculative, but there’s no better way to account for our arbitrary deviations. As such, many modeling standards ARE NOT “as good as any other reference.” The best and most relevant standards are those that are calibrated to human norms for perception. (Yes, preferences indeed may have only a modest correlation with nominal neutrality and people label things “neutral” when the enjoy “warm” or “bright” products.)
Human perception routinely results in imprecise bell curves that form a “norm” or the standard. Here’s an example of the height distributions for adult males and females, but a LOT of human body factors follows this pattern. While the most frequent female height is about 65”, there’s lots of variation.
When shopping for my first serious headphones I demoed the HD 650 but bought the HD 600. This followed from the overly thick bass and rough perceived character of the 650, as well as a zillion reviews circa 2014 saying that the HD 650 has a warm tilt. Here’s a Google AI output for “Is the HD 650 neutral” – it indeed replicates a zillion reviews from 10 years ago and parallels my direct experience too (highlights in original):
The Sennheiser HD 650 is widely regarded as a mostly neutral headphone with a slightly warm and musical sound signature. Its tuning emphasizes a natural and engaging midrange, making it an enjoyable headphone for long listening sessions. However, it is not perfectly “ruler flat” and exhibits some deviations from a strictly neutral reference sound.
The HD 650’s sound characteristics
Warmth and midrange focus: The HD 650 is known for its “laid-back” and “musical” sound signature. It has a slightly elevated mid-bass, which provides a sense of warmth and fullness. The midrange is a standout feature, praised for its accurate and lifelike reproduction of vocals and acoustic instruments.
Rolled-off bass and treble: Compared to a perfectly neutral target, the HD 650 has a noticeable roll-off in the deep sub-bass and the upper treble. The bass is present but lacks deep extension, which can make some bass-heavy music sound less impactful. The upper-treble roll-off contributes to the headphone’s smooth, non-fatiguing character and is sometimes referred to as the “Sennheiser veil”.
- Calibration for accuracy: For users who desire a more accurate and linear frequency response for critical mixing or mastering, the HD 650 is an excellent candidate for digital calibration software. Applying a custom EQ can flatten the frequency response, improving sub-bass extension and overall neutrality.
HD 650 vs. HD 600
For many years, the HD 650 has been compared to its sibling, the HD 600, which is considered the more analytically neutral of the two.
| Treble | More relaxed and gentle. | More forward and sometimes harsher. |
| Bass | A touch more mid-bass presence. | Leaner bass response. |
| Listening Style | Excellent for long, fatigue-free listening sessions. | Better suited for critical detail analysis. |
I again suggest starting the analysis from the logical beginning: philosophy and perception science. These scientific methods are often much more complex than electrical models and measurements, but certainly more relevant to human uses.
Lots of good information. There is definitely a divergence of opinion regarding this subject.
From the Sonarworks Blog:
“Sound
Sennheiser HD 650 are pretty much the most neutral headphones available on the market. However, their sub-bass response is lacking (frequencies below 100 Hz) and high-mids (around 2-5 kHz) need to be tweaked a bit. Otherwise, they are pretty much good to go straight out of the box.
Channel balance
You would expect from headphone manufacturers perfect matching between left and right drivers. Unfortunately, in the real world, this doesn’t happen too often – pretty much every other pair of headphones have channel inconsistencies even among the well established pro headphone brands. By contrast, Sennheiser is a true expert in offering headphones with matched left and right drivers so that soundstage stays consistent across the spectrum. With some minor exceptions, Sennheiser HD 650 has hardly any inconsistency between the two drivers. On a rare occasion, pairs have slight differences in bass and sub-bass region, but the magnitude is low enough to not cause audible issues with imaging.
The HD650 is the more relaxed sibling: warmer low mids, smoother treble, and a touch more bass body. The HD600, by contrast, has a leaner bass profile and a more forward presence region (~3–5 kHz), making it feel crisper and sometimes harsher on poor recordings. Think of the HD600 as a surgical monitor and the HD650 as the more musical long-term companion. For critical neutrality → HD600. For fatigue-free listening sessions → HD650.”
I appreciate you responses, even if I have a slightly different take on the subject. I just recently obtained the HD 650, and was amazed at it’s overall performance. I’ve gone through a lot of headphones over the past few years, and the HD 650 has a great overall balance, especially at it’s price. Many of the headphone I’ve tried cost more, but didn’t provide as much enjoyment as the HD 650.
I’ll look into the material you posted further. One thing is becoming clear: “The more one learns, the more one realizes there is much more to learn”
