Interview with Dynaudio chief engineer Stephen Entwistle

[abdo123]

You have misunderstood me there. Below the critical distance, the direct/free field is dominant. In an ideal free field there would be no reflections and the sound character of the loudspeaker would be determined solely by the listening axis - e.g. the on-axis FR when the loudspeakers are directed towards the listener.

In this case (ideal free field) the directivity of the LS would not matter at all, only the FR on which the listening is done would play a role.

In an ideal diffuse field, the sound source direction could no longer be determined, since all reflections would be equivalent. Since this can never be realized in a normal listening room, there will always be a mixture of direct and diffuse field at the listening position, with the diffuse field being dominant over wide frequency ranges (see also the calculation of the critical distance).

So you need a measure that describes the behavior and the sound character of a loudspeaker at the listening position including the room reflections.
This is where "sound power" SP and "predicted in-room amplitude response" PIR come into play. In the calculation of these two curves the radiation outside the on-axis FR of the speaker is included.

Historically, the SP curve has been used in loudspeaker design for a very long time - I come from the DIY field and there the SP (calculated from measurements +-90° horizontally and, not always, +-90° vertically), has been used for over 15 years.
The PIR reproduces the behavior of the LS at the listening position even better. Whereby the weighted average of 44% SP also flows into the PIR (see CTA-2034a definitions).

Why is SP and PIR important?
It helps to explain why, for example, Amir found the Wilson-Audio tunetot tonally not bad and with a little EQ even good.
View attachment 226575 View attachment 226576
Especially with the PIR, you can see that the LS does not show sudden or wide humps (when we ignore the ugly bass hump).
SP/PIR can help explain why even poorly designed speakers can still sound good in the listening room if they are tuned properly with SP or PIR in mind.

The better a loudspeaker is designed, the more the on-axis, listening window LW, SP and PIR curves converge:
View attachment 226579

For a loudspeaker that is nearly SOTA, the various curves are nearly parallel (ignore FR >15kHz) because the radiation is so uniform that the off-axis FRs hardly change. You can then focus on small details, for example, and use different XO designs to strongly influence the vertical radiation around the crossover frequency. Same LS with two different XO:
View attachment 226580 View attachment 226581
In both cases on-axis FR (black), LW (green), SP (blue) and PIR (orange) >700Hz are nearly parallel (DI is red).

However, since very few loudspeakers are SOTA, it is important to look at SP/PIR, or to consider these curves when designing and tuning a loudspeaker, and to value a smooth SP/PIR response more highly than a flat on-axis frequency response - at least that is what Stephen Entwistle probably wanted to say, and with which I strongly agree (I am more skeptical about other statements in the interview).

It is important that a uniform SP/PIR response is present regardless of wide or narrow directivity.

With a narrow directivity it is usually easier to achieve a uniform SP/PIR response and the curve shows a steeper drop.
Therefore horn loudspeakers can often be corrected quite well via EQ (if the directivity is even, which the DI shows us). Example without and with EQ:

View attachment 226588 View attachment 226589
Source

That was another annoyingly long post, but it try to show that not every LS with a (slightly) wavy FR has to sound bad.

Conversely, you can say that a classic 2-way LS (6.5'' woofer and 1'' tweeter) in a normal cabinet with classic XO around 2kHz with an optimally flat on-axis FR will not sound optimal in a normal listening room (too aggressive at high SPL).

One can't help but wonder where that picture perfect VituixCAD project is coming from.

 

[tuga]

Not really. Sighted listening is full of bias. Also, our auditory memory is pretty bad. This has been shown over and over again. Toole, "Sound reproduction" for a primer. Or this example from Olive: https://seanolive.blogspot.com/2009/04/dishonesty-of-sighted-audio-product.html

If you want to compare two different sound fields in an objective manner you have to design a blind listening test that controls all possible bias. That's pretty hard to do. Probably involves doing binaural recordings.
But don't throw out the baby with the bathwater and expect that what you're proposing above would be worthwhile. I mean it is what audiophiles are doing and have been doing since the beginning of audio reproduction. It's just human to believe our individual perception would let us experience the world in an objective way. Hardly useful though if one wants to learn something objective about the real world. That's why we haven't made as much progress as we could have in the last 100 years or so in the field of audio reproduction. More or less it's all just been driven by business interest and listener confusion.

No, it has been scientifically proven over and over again how cognitive bias renders sighted listening tests, or any sensory testing, rather useless – see my avatar. Just because you're not aware of the research that has been done doesn't mean it wouldn't exist. Your following statement shows the full extend of confusion which is common amongst "golden ear" audiophiles:

Speakers are sound reproduction devices. They are not part or should not be part of a recording and the process of making artistic decisions. You're confusing the pizza oven with the pizza.

You’ve been brainwashed.

Also Toole has admitted here the limitations of Harman’s listening test methodology. I’ve discussed this to death. Look it up.

 

[Thomas_A]

On one hand, a narrow directivity means hearing less of the room and more of the recording... on the other hand, the reflections can create a wider soundstage that can be pleasant.
It's a hard life

It is. If my model would be that the band is in a room behind the speaker wall, and the wall is acoustically invisible, how narrow would the dispersion be from the band?

 

[markus]

You’ve been brainwashed.

Also Toole has admitted here the limitations of Harman’s listening test methodology. I’ve discussed this to death. Look it up.

You're missing the point. I wasn't specifically talking about "Harman’s listening test methodology" but about how cognitive bias affects our perception and thinking, and how hard it is to design a proper test if you're interested in objective data. If you don't like Toole try Kahneman. Plenty of literature, also from other scientific fields. Sometimes test results can be invalid or misleading even when no sensory testing was involved whatsoever but bias affected the test design itself.
In any case, sighted listening testing is really just a massive waste of time. Audiophiles seem to like this process though, it's their hobby, and that's okay. Just don't make the mistake to believe you would learn something objective by doing so.

 

[ctrl]

One can't help but wonder where that picture perfect VituixCAD project is coming from.

The example projects without references are my personal projects - if I ever find the time, there will be more here in the forum at some point.

 

[lowkeyoperations]

That's why we haven't made as much progress as we could have in the last 100 years or so in the field of audio reproduction. More or less it's all just been driven by business interest and listener confusion.

Given the general obsession with measurements, how about you give us the FR of the first AM broadcast 100 years ago?

I think you will find the difference between that, and almost any half reasonable hifi system of today is extreme. :lol:

Your following statement shows the full extend of confusion which is common amongst "golden ear" audiophiles:

Your point is about double blind tests, not the speed at which the switching is done. The point earlier was listening tests must switch back and forth quickly to be relevant. There’s no study of that.

Speakers are sound reproduction devices. They are not part or should not be part of a recording and the process of making artistic decisions. You're confusing the pizza oven with the pizza.

Not sure how that is relevant to this discussion.

But if that’s what you think, do tell me the preferred way electric guitars are recorded in a studio. Speakers are part of recording and are part of the process of making artistic decisions.

Speakers are part of the recording. Microphones are part of the recording.
Preamps are part of the recording.
Even the recording medium is part of the recording!

Not sure how that’s relevant to pizza.

 

[abdo123]

Your point is about double blind tests, not the speed at which the switching is done. The point earlier was listening tests must switch back and forth quickly to be relevant. There’s no study of that.

There are multiple studies that show that auditory memory in humans is very ******, remarkably ******.

Achilles’ Ear? Inferior Human Short-Term and Recognition Memory in the Auditory Modality

Studies of the memory capabilities of nonhuman primates have consistently revealed a relative weakness for auditory compared to visual or tactile stimuli: extensive training is required to learn auditory memory tasks, and subjects are only capable of retaining acoustic information for a brief...

journals.plos.org

 

[Thomas_A]

What is the ideal directivity of a speaker in comparison to the recorded source vs. the event itself? What should a speaker really reproduce?

I wonder...

 

[lowkeyoperations]

Do you really think I can’t remember whether my speakers in my lounge room are more enjoyable than my office speakers?

Every day I need to measure them just to check?

Audio science at its best.
Don’t listen to your speakers.
Just measure them.

LOL

Do you know how the Dynaudio listening tests were undertaken? What were they listening for? For how long?

Trying suggest that listening to speakers during their development is worthless without that knowledge seems presumptive.

 

[MattHooper]

Hi markus, just to be clear, I'm wondering if you agree with the following:

Not really. Sighted listening is full of bias.

Not just sighted listening. Virtually everything we perceive and think about can be infected by bias.

But that doesn't mean everything we conclude without the use of scientific controls is useless, right? That would be truly throwing the baby out with the bathwater.

Take optical illusions, for an example. Like the classic checkerboard pattern where a "shadow" cast on the board fools our perceptual system in to thinking two squares of the same luminance are different. That clearly shows our perception can be unreliable. But it would be an incautious (and unscientific) leap from that to say "therefore our sight is wholly unreliable." Right?

Because that would not explain the fact we use our sight to successfully navigate the world, for the most part. It wouldn't explain all the instances in which it is clearly reliable (e.g. we find our way out our front door every day, how you manage to drive a car, etc).

So if you are thinking scientifically, you want to be cautious not only about the results you are accepting (ensuring good methodology) but you ALSO want to exhibit caution about the conclusions you can draw from experimental data. I'm sure you agree?

Likewise with our perception of sound. Certainly it can be influenced by biases. But does that make all inferences we draw about "what we hear" to be unreasonable, if they are not done under scientific controls? That would be an unreasonable conclusion, right?

So: Yes, if you want a scientific level of certainty you will want to use scientific controls. No question.

But let's be careful about leveraging this to dismiss any inferences drawn without scientific controls to be "useless" or a "waste of time."
That would be throwing the baby out with the bathwater. Just as with any sense perception or empirical inference we make, there is going to be a large area of fuzziness in between "wholly unreliable" and "very reliable" in which pragmatic concerns will mean it's reasonable to draw tentative conclusions. (As I've said many times on this subject: in my business, post production sound, we operate without scientific controls and yet manage to get our job done day in and day out).

Also, our auditory memory is pretty bad. This has been shown over and over again. Toole, "Sound reproduction" for a primer. Or this example from Olive: https://seanolive.blogspot.com/2009/04/dishonesty-of-sighted-audio-product.html

Just to be clear: the link doesn't establish our audio memory is bad; it establishes the influence of sighted bias on our perception. If our audio memory was that bad, the test wouldn't have been able to establish that very conclusion (you'd have to throw out the blinded conditions as telling you anything about what things 'really sound like' as well). Blind tests can actually show your auditory discrimination is quite robust. (That's why the results of the Harman Kardon and other blind tests are useful).

In any case, sighted listening testing is really just a massive waste of time.

To be clear, though: Sighted listening is a "waste of time" IF you are looking for scientifically reliable results.

Which is a sort of tautology: If you aren't doing science; you aren't doing science.

But this does not entail they are therefore inferences drawn without scientific controls is a "massive waste of time" in the sense of being utterly useless or unreasonable.

When cooking if we feel the recipe needs to taste a bit more salty, then we add a few more pinches of salt and now it tastes a bit more salty...bias is always possible. Maybe we think it tastes more salty simply because we know we just added more salt. On the other hand, we know that increasing the salt content CAN indeed make something taste more salty. And since we normally don't have the ability to do scientific blind test controls with our cooking, we can move along pragmatically: yeah, I just added salt, it's reasonable to infer I detected that change.

Likewise with speakers that measure quite (audibly) differently. If someone reports those differences, and attempts to describe them...yes of course bias is always looming as influencing the scenario. On the other hand, speakers DO sound different and it's entirely plausible someone heard those actual differences even under sighted listening. You can't actually determine otherwise, unless you'd done blind tests on the specific models. (And, note: even in the examples given for how sighted listening influences preference scores, there is actually some continuity maintained between the sighted and unsighted listening - e.g. some speakers consistently rated higher than the others in both the sighted and blinded results).

Cheers.

 

[abdo123]

Do you really think I can’t remember whether my speakers in my lounge room are more enjoyable than my office speakers?

Every day I need to measure them just to check?

Audio science at its best.
Don’t listen to your speakers.
Just measure them.

LOL

Do you know how the Dynaudio listening tests were undertaken? What were they listening for? For how long?

Trying suggest that listening to speakers during their development is worthless without that knowledge seems presumptive.

You're going to have to invest in this discussion a little bit by reading the paper i shared before you warrant an actual response.

 

[abdo123]

Hi markus, just to be clear, I'm wondering if you agree with the following:



Not just sighted listening. Virtually everything we perceive and think about can be infected by bias.

But that doesn't mean everything we conclude without the use of scientific controls is useless, right? That would be truly throwing the baby out with the bathwater.

Take optical illusions, for an example. Like the classic checkerboard pattern where a "shadow" cast on the board fools our perceptual system in to thinking two squares of the same luminance are different. That clearly shows our perception can be unreliable. But it would be an incautious (and unscientific) leap from that to say "therefore our sight is wholly unreliable." Right?

Because that would not explain the fact we use our sight to successfully navigate the world, for the most part. It wouldn't explain all the instances in which it is clearly reliable (e.g. we find our way out our front door every day, how you manage to drive a car, etc).

So if you are thinking scientifically, you want to be cautious not only about the results you are accepting (ensuring good methodology) but you ALSO want to exhibit caution about the conclusions you can draw from experimental data. I'm sure you agree?

Likewise with our perception of sound. Certainly it can be influenced by biases. But does that make all inferences we draw about "what we hear" to be unreasonable, if they are not done under scientific controls? That would be an unreasonable conclusion, right?

So: Yes, if you want a scientific level of certainty you will want to use scientific controls. No question.

But let's be careful about leveraging this to dismiss any inferences drawn without scientific controls to be "useless" or a "waste of time."
That would be throwing the baby out with the bathwater. Just as with any sense perception or empirical inference we make, there is going to be a large area of fuzziness in between "wholly unreliable" and "very reliable" in which pragmatic concerns will mean it's reasonable to draw tentative conclusions. (As I've said many times on this subject: in my business, post production sound, we operate without scientific controls and yet manage to get our job done day in and day out).



Just to be clear: the link doesn't establish our audio memory is bad; it establishes the influence of sighted bias on our perception. If our audio memory was that bad, the test wouldn't have been able to establish that very conclusion (you'd have to throw out the blinded conditions as telling you anything about what things 'really sound like' as well). Blind tests can actually show your auditory discrimination is quite robust. (That's why the results of the Harman Kardon and other blind tests are useful).



To be clear, though: Sighted listening is a "waste of time" IF you are looking for scientifically reliable results.

Which is a sort of tautology: If you aren't doing science; you aren't doing science.

But this does not entail they are therefore inferences drawn without scientific controls is a "massive waste of time" in the sense of being utterly useless or unreasonable.

When cooking if we feel the recipe needs to taste a bit more salty, then we add a few more pinches of salt and now it tastes a bit more salty...bias is always possible. Maybe we think it tastes more salty simply because we know we just added more salt. On the other hand, we know that increasing the salt content CAN indeed make something taste more salty. And since we normally don't have the ability to do scientific blind test controls with our cooking, we can move along pragmatically: yeah, I just added salt, it's reasonable to infer I detected that change.

Likewise with speakers that measure quite (audibly) differently. If someone reports those differences, and attempts to describe them...yes of course bias is always looming as influencing the scenario. On the other hand, speakers DO sound different and it's entirely plausible someone heard those actual differences even under sighted listening. You can't actually determine otherwise, unless you'd done blind tests on the specific models. (And, note: even in the examples given for how sighted listening influences preference scores, there is actually some continuity maintained between the sighted and unsighted listening - e.g. some speakers consistently rated higher than the others in both the sighted and blinded results).

Cheers.

Our auditory memory IS bad though, we’re talking seconds like a gold fish (although that’s not really scientific either)

The study i shared earlier that accuracy of answers dropped by like 30% after only 15 seconds have passed.

 

[MattHooper]

Our auditory memory IS bad though, we’re talking seconds like a gold fish (although that’s not really scientific either)

The study i shared earlier that accuracy of answers dropped by like 30% after only 15 seconds have passed.

Again...you have to look at precisely the conditions of a test and be careful what you can infer. If you infer from those particular test conditions that "our auditory memory IS bad" what exactly is that claim? If it is that "bad" it would be useless, but it isn't useless...right?
That's why we evolved hearing and memory. It's why you don't wonder if the voice on the phone is your mom or Donald Trump, etc.
Whenever you are trying to draw conclusions they have to make sense within context, which includes the context that there are ways in which our auditory memory clearly IS reliable.

As to the specific tests you cited and their implications, note that the main thrust is the unreliability of auditory memory RELATIVE to visual and tactile memory. From the link

"In general, we observed that retention was inferior for acoustic stimuli compared to visual and tactile stimuli, whereas retention for visual and tactile stimuli was approximately equal."

"inferior"

It doesn't produce the result "our auditory memory is pure sh*t." Is auditory memory of lower accuracy over time? Yes. Is it therefore utterly inaccurate? No.

For instance, look at the results for Experiment 2, investigating the "real world applicability" of their findings (where sound stimuli was similar to "every day life").

From the graphs, while the visual/tactile memory accuracy was highest, the "same day recognition" for auditory memory was
88.61 %!

The next day retention was 82.85% !

The next week retention accuracy was still 76.25%

So the auditory memory is still performing well above mere chance, right? It's hardly "bad" or useless. (Which, again, is why it evolved).

Further, note that the authors say that even on the "week later" accuracy results: "However, although accuracy was lower in the auditory block than in the visual block (79.86%), this difference was not significant."

And:

"Although accuracy predictably decreased with increasing time between the study and recognition phases, as indicated by mean accuracy scores, the magnitude of the deficit in auditory recognition compared to visual and tactile recognition diminished at the longer delays. This outcome contradicted our a priori expectation that, since auditory recognition accuracy was relatively poor after a short delay period, this difference would become more pronounced with time."

(Note again the caveat "relatively")

So, yes, auditory memory is not perfect and all perception accuracy diminishes over time. But the study above indicates that auditory memory IS fairly accurate and that at longer time periods the difference between visual and auditory accuracy (at least up to a week) starts to become "not significant."

All this is to say we need to be careful about the strength and scope of claims we make based on specific test situations. When scientists are at their most scientific, they are acutely aware of such cautions.

 

[abdo123]

Again...you have to look at precisely the conditions of a test and be careful what you can infer. If you infer from those particular test conditions that "our auditory memory IS bad" what exactly is that claim? If it is that "bad" it would be useless, but it isn't useless...right?
That's why we evolved hearing and memory. It's why you don't wonder if the voice on the phone is your mom or Donald Trump, etc.
Whenever you are trying to draw conclusions they have to make sense within context, which includes the context that there are ways in which our auditory memory clearly IS reliable.

As to the specific tests you cited and their implications, note that the main thrust is the unreliability of auditory memory RELATIVE to visual and tactile memory. From the link

"In general, we observed that retention was inferior for acoustic stimuli compared to visual and tactile stimuli, whereas retention for visual and tactile stimuli was approximately equal."

"inferior"

It doesn't produce the result "our auditory memory is pure sh*t." Is auditory memory of lower accuracy over time? Yes. Is it therefore utterly inaccurate? No.

For instance, look at the results for Experiment 2, investigating the "real world applicability" of their findings (where sound stimuli was similar to "every day life").

From the graphs, while the visual/tactile memory accuracy was highest, the "same day recognition" for auditory memory was
88.61 %!

The next day retention was 82.85% !

The next week retention accuracy was still 76.25%

So the auditory memory is still performing well above mere chance, right? It's hardly "bad" or useless. (Which, again, is why it evolved).

Further, note that the authors say that even on the "week later" accuracy results: "However, although accuracy was lower in the auditory block than in the visual block (79.86%), this difference was not significant."

And:

"Although accuracy predictably decreased with increasing time between the study and recognition phases, as indicated by mean accuracy scores, the magnitude of the deficit in auditory recognition compared to visual and tactile recognition diminished at the longer delays. This outcome contradicted our a priori expectation that, since auditory recognition accuracy was relatively poor after a short delay period, this difference would become more pronounced with time."

(Note again the caveat "relatively")

So, yes, auditory memory is not perfect and all perception accuracy diminishes over time. But the study above indicates that auditory memory IS fairly accurate and that at longer time periods the difference between visual and auditory accuracy (at least up to a week) starts to become "not significant."

All this is to say we need to be careful about the strength and scope of claims we make based on specific test situations. When scientists are at their most scientific, they are acutely aware of such cautions.

You have to keep in mind that people will always have a 50% chance to give the correct answer, so when the accuracy is already at 60% at 15 seconds that means it's almost practically inaccurate.

Usually when I do intense ABX stuff i need to switch within a second or two. I can't even think of making small scale tuning adjustments to something if i'm just listening weeks on end semi-attentively.

 

[lherrm]

You have to keep in mind that people will always have a 50% chance to give the correct answer, so when the accuracy is already at 60% at 15 seconds that means it's almost practically inaccurate.

Usually when I do intense ABX stuff i need to switch within a second or two. I can't even think of making small scale tuning adjustments to something if i'm just listening weeks on end semi-attentively.

Doesn't it all depend on the scale of the variations ?
As @MattHooper said, you can easily distinguish between your relative's voices over the phone without any hesitation even weeks after not having listened to them.

 

[MattHooper]

You have to keep in mind that people will always have a 50% chance to give the correct answer, so when the accuracy is already at 60% at 15 seconds that means it's almost practically inaccurate.

Refer to the accuracy results I cited from your link. In "real world sounds" experiment #2, "Same day recognition" for auditory memory
was 88.61 %

That's already usefully accurate. And it remained above 75% a week later.

Does this mean that the memory was totally reliable? No. Does it mean that drawing an inference from that memory was more likely false than true? No. In practical terms, it's still useful.


And speaking of practical:

Usually when I do intense ABX stuff i need to switch within a second or two. I can't even think of making small scale tuning adjustments to something if i'm just listening weeks on end semi-attentively.

In my job as a sound editor I'm adjusting sound levels, often in minute amounts, all day long. Literally, constantly. I'm also making judgements about often subtle audible differences in character. Am I practicing scientific controls? No. Are my efforts therefore rendered useless or utterly unreliable? Of course not.

(In fact, the very study you cited supports this)

 

[abdo123]

Doesn't it all depend on the scale of the variations ?
As @MattHooper said, you can easily distinguish between your relative's voices over the phone without any hesitation even weeks after not having listened to them.

Yeah ofcourse. But the whole discussion was about whether the data from short interval changes is more valuable than long interval changes. Which evidence suggests is indeed more valuable.

 

[abdo123]

Refer to the accuracy results I cited from your link. In "real world sounds" experiment #2, "Same day recognition" for auditory memory
was 88.61 %

That's already usefully accurate. And it remained above 75% a week later.

Does this mean that the memory was totally reliable? No. Does it mean that drawing an inference from that memory was more likely false than true? No. In practical terms, it's still useful.


And speaking of practical:



In my job as a sound editor I'm adjusting sound levels, often in minute amounts, all day long. Literally, constantly. I'm also making judgements about often subtle audible differences in character. Am I practicing scientific controls? No. Are my efforts therefore rendered useless or utterly unreliable? Of course not.

(In fact, the very study you cited supports this)

And you’re able to judge whether a change is necessary or not because the change is instant.

If you had to make adjustments that you can only control their consequences weeks later you will not be as efficient. If at all.

 

[MattHooper]

And you’re able to judge whether a change is necessary or not because the change is instant.

If you had to make adjustments that you can only control their consequences weeks later you will not be as efficient. If at all.

That will depend on what is being asked of the memory.

Generally speaking: The more minute the audibility, the worse for acoustic memory. The more distinct the changes, the better.

I don't just rely on my immediate seconds-long short term memory (e.g. for changing volume or EQ on a sound). I also use my memory of sounds in my library, or that I have recorded, from yesterday, last week, last month, even years ago.

Just a few days ago I was creating a sound. I felt it needed a deep rumbling tone with a certain texture. I remembered creating just such a tone for a movie several years ago. I found it...it was just as I remembered and worked perfectly. People in my line of work do this all the time.

I'm constantly using my memory to recall sounds I've used before, or heard in my library, that will suit a scene I'm working on. For instance if I need a "dog barking" I have good recall of lots of the dog barks in my library, which allow me to find the one I want quickly. (One thing I often do before working on a movie is go through a relevant library and audition lots of the sounds. I can then often remember, often during months of work, a sound or recording I'd heard earlier that will suit my needs and go grab it).

If acoustic memory were truly "awful" or "bad," this wouldn't be possible.

This is why I keep emphasizing that you have to be cautious about what can be extrapolated, even from specific careful experiments. Those experiments don't show acoustic memory isn't useful.

 

[lowkeyoperations]

You're going to have to invest in this discussion a little bit by reading the paper i shared before you warrant an actual response.

It’s pretty obvious from the posts above that the paper shows different results and conclusions than your claims.

88.61% memory retention is not “60% which is pretty much 50% which is there completely cr#p”.

Simply linking papers doesn’t make you scientific. Especially when you ignore and misrepresent the contents to come up with the answer you already wanted.