Soundtheory GULLFOSS computional auditory perception EQ

[vitocorleone123]

I can't spend $200 on something that I, and apparently everyone else, have no idea what the plugin is actually doing.

Budget aside, do you like what it does? Does it make stuff sound better? What do your ears say, regardless of what your thoughts are?

Can you achieve the same result with one or more different tools and, if so, do it with an efficiency that doesn't detract from your approach? If so, then maybe it's not worth duplicating by getting Gullfoss.

If not, then it's more a matter of what someone is willing to spend for something that

1) can make things sound better (esp. at the last step(s)... though I find it can help with a vocal track as well, among others)

2) is pretty fast and easy to dial in and move on to the next thing

It's a personal choice. As a hobbyist, buying Gullfoss at $99 was an easy choice to make. But that's me, and that certainly doesn't make it the best choice for anyone else.

(Side note: I don't know how any of my plugins do what they do, other than what they display at the UI level... and I personally don't care as long as it sounds good and helps me achieve goals)

[plexuss]

I’ve never seen anyone make so many completely incorrect and unproven assumptions in a single thread. It’s clear you have an agenda against Sound Theory.

I see no clarity here. I also see no evidence that Mathematics has an agenda against Sound Theory. In fact I think v1o's assertion is irrational and should be retracted unless they have proof to back up their claim.

All his assertions are unproven and based on him trying to disprove Sound Theory’s auditory perception mathematical model as nothing more than pink noise shaping without facts or evidence apart from some truly shoddy testing. Basically he’s made up a fake conspiracy in his head. Sound Theory have gone on the record several times and explained how their system works as best they can without giving away too much proprietary information. There is nothing to be gained from them being dishonest and lying about how Gullfoss actually works. If Gullfoss didn’t work the way they say it does it would have been picked up already.

Life must be tough for you.

Mathematics analysis was valid and interesting but of limited scope. Not enough to draw conclusions about what Gullfoss does internally.

[v1o]

I’ve never seen anyone make so many completely incorrect and unproven assumptions in a single thread. It’s clear you have an agenda against Sound Theory.

I see no clarity here. I also see no evidence that Mathematics has an agenda against Sound Theory. In fact I think v1o's assertion is irrational and should be retracted unless they have proof to back up their claim.

All his assertions are unproven and based on him trying to disprove Sound Theory’s auditory perception mathematical model as nothing more than pink noise shaping without facts or evidence apart from some truly shoddy testing. Basically he’s made up a fake conspiracy in his head. Sound Theory have gone on the record several times and explained how their system works as best they can without giving away too much proprietary information. There is nothing to be gained from them being dishonest and lying about how Gullfoss actually works. If Gullfoss didn’t work the way they say it does it would have been picked up already.

Life must be tough for you.

Mathematics analysis was valid and interesting but of limited scope. Not enough to draw conclusions about what Gullfoss does internally.

Its very shoddy and nothing meaningful can be concluded from it.

Why don't you believe Sound Theory?

[jbraner]

It sounds good and it *works*.
So, who cares *how* it works?

[Mathematics]

I’ve never seen anyone make so many completely incorrect and unproven assumptions in a single thread. It’s clear you have an agenda against Sound Theory.

I see no clarity here. I also see no evidence that Mathematics has an agenda against Sound Theory. In fact I think v1o's assertion is irrational and should be retracted unless they have proof to back up their claim.

All his assertions are unproven and based on him trying to disprove Sound Theory’s auditory perception mathematical model as nothing more than pink noise shaping without facts or evidence apart from some truly shoddy testing. Basically he’s made up a fake conspiracy in his head. Sound Theory have gone on the record several times and explained how their system works as best they can without giving away too much proprietary information. There is nothing to be gained from them being dishonest and lying about how Gullfoss actually works. If Gullfoss didn’t work the way they say it does it would have been picked up already.

Life must be tough for you.

Mathematics analysis was valid and interesting but of limited scope. Not enough to draw conclusions about what Gullfoss does internally.

Its very shoddy and nothing meaningful can be concluded from it.

Why don't you believe Sound Theory?

Hey guy. I can understand why you think I was taking a stab and denigrating SoundTheory's Gullfoss to a cheap gimmicky Pink Noise Matching EQ, but if you read carefully, my original post asked, "How do you guys use it most effectively?"

In fact, this is the point where I started in on this thread:
viewtopic.php?p=7470030#p7470030

I never saw your name pop up with any significant help or advice, but you're here casting the first stone at me when technically, I was only posting what I did on my own. I only decided to post my tests because I couldn't think of a better way at the time, to create a variable input that covers every single frequency, and to get constructive feedback. Yes, the tests are inconclusive because I didn't account for many things because I just don't have the time. I don't do this full time.

There's no need to start calling names or trying to make me look stupid. Something I once learned from an ex-convicted felon while studying martial arts in my early 20's was this..."it's better to make a soldier than break a soldier." You seem like an otherwise intelligent individual. Why not take his advice now, and ADD or suggest a better test, so that we can all benefit from spending our time reading about results rather then me teaching you a lesson in life.

That said, I never had an agenda. I am not supporting Sound Theory nor am I against them. Overall, it appears as if though Gullfoss uses Pink Noise as a reference but like I mentioned a few posts ago, I do think there is more going on under the hood. Things like: time intervals, q factor, inertial thresholding, F-M curves...technical stuff that you probably are much more qualified to talk about than the rest of us peasants.

One last thing. I didn't want to mention this before, but I specifically asked NOT to be quoted anymore as if I was some kind of authority on this...this means coming in late conversation, reading a couple posts and taking what I post as something you should take religiously.

I think Plexuss said it best, "plugin code can be doing all kinds of non-linear, conditional and sometimes unpredictable things." I contacted Sound Theory with 6 emails and links to this discussion and got a reply back that Andreas will answer my questions. That is, technically, where we're at.

Again, as a reminder, take a convicted felon's advice and quit trying to break soldiers. Evolve.

[Mathematics]

Gullfoss does need to improve that user manual though. I notice that there is a tilt about 1200Hz. So, if you look at the frequency spectrum of different genres, they all have distinct characteristics. EDM is the more pink noise-ish looking than rock or classical. So, I notice that it still boosts or cuts by a percentage of the maximum limits of gain and attenuation.

I am working on a piano based song with choirs and a sub line. Ambient-ish. It dramatically wants to add air where air wasn't technically mixed in. Meaning, for some parts, there isn't an instrument for that band. When something does come it, it sharply reduces the gain for that section but it sounds great. Once that section is over, it brings back up the tail of the frequency which sounds odd.

That said, you have to choose the type of song wisely. Again, it really adds smoothness to complete mixes with a busy spectral signature. However, I'm finding that it works really well as a "mix checker".

I read, a post from VI Control, Beat Kauffman was giving Gullfoss a try and he had decent things to say about it, and mentioned a few shortcomings. Nothing too different than what has been discussed already. However, there was one person that talked about the algorithms used in Ozone 8, and he said that some of the predictions it gives, is based on a similar pink noise reference. Now, what I said a moment ago, different frequency spectrums have distinct characteristics but they all follow that 'somewhat 3db slope' as an upper bound limit. This is true with all music, with the exception of genre specific bass heavy tracks like EDM and Hip Hop. The Sub doesn't follow the same rule. If you've ever mixed LoFi tracks or look at any popular Hip Hop tracks, you see that the sub is significantly disproportionate in amplitude to the rest of the track. So, perhaps Ozone 8 uses a balance between the two, by offering an overall shape and just taming the in between frequencies. I'll be honest, Ozone 8 usually, like 99% of the time, gives me a darn good approximation of the final release that I achieve on my own. Usually, my self-mastered mixes always need about a +/-2db gain/red at the cutoffs and +/-1db at certain notches frequencies around 400 and 2k...something like that, I don't remember.

Anyway, I can see Gullfoss evolving toward something like that. In fact, it's great at suggesting. THAT is how I think I would use it effectively in my production. Like a suggestion of what to boost and cut for smoothness. I probably WOULD NOT want to take that advice all the time because that makes everything sound like everything...if that makes sense. Like, nothing guiding you. It's like looking at a blurred out picture of a person's face. You're never drawn in to...say, their eyes, or their smile. It seems like Gullfoss is like the Gaussian Blur for audio tracks. Use with care, or use a layer mask to white out where the detail needs to be.

Anyway, I feel like they could create another plugin called "Pseudo GEars" Pseudo Golden Ears, where it

[atell]

I contacted Sound Theory with 6 emails and links to this discussion and got a reply back that Andreas will answer my questions. That is, technically, where we're at.

I'm surprised that you did not mention the verbose response I have sent to you and that has been sitting in your inbox for more than 10 hours at the time you wrote this. It seems you have no interest in sharing it, so I take the freedom to post it here for everyone to see it.

Hi Ron,

thank you for coming to us for clarification. We've already answered most of these questions, but I can understand that this information is hard to find and spread out over the internet.

First, it is important to understand that Gullfoss does not use any kind of internal reference curve for the spectral power density of a signal. In fact, such an approach would be flawed in many ways and Gullfoss would not be able to operate as it does with it. Determining if a signal is distributed like pink noise would require to integrate over some representative time interval. Any kind of such integration would necessarily result in a slowed down response to the input signal. It would also introduce a more or less arbitrary integration time and there is no reason to believe that matching to such a short-time power density would result in any kind of musically meaningful output. Gullfoss, on the other hand, reacts instantly to transients and at the same time produces a stable response for longer-term features. And it sounds good, too.

Our approach is much more fundamental. We have developed a model of information transport in the human brain. This model is based on first principles and physical constraints on the operation of the auditory system, not on psychoacoustic data. The model predicts which information is available to the brain at that time. This is closely related to the phenomenon of psychoacoustic masking, but our predictions go beyond the usual understanding of masking and are vastly more general.

Gullfoss uses the predicted information and tries to change the signal in a way that maximises the information perceived by the brain. This improves clarity, detail and reveals more spatial cues. In order to do that, we offer the two strategies "Tame" and "Recover", which basically use different heuristics to maximise the information. Tame reduces the dominance of the signal components that suck up all the information, and because of the sub-additivity of information measures reduce the total available information, while "Recover" tries to lift information up from the previously inaudible background.

Neither of these strategies works with a global reference. They both use local cues, only. There is also no built-in limit for the maximum change. Any limit that you observe emerges from the local context and properties of the perception model.

Because the decisions are made locally and allow for a certain range of changes to fully reveal the information, there is a certain freedom of a global distribution that can be controlled with brighten. So the brighten control, in fact, has some minor influence on the global power balance between frequencies if either Tame or Recover is active.

The changes of the signal also are fully determined by the perception model, and they happen in a way that respects the information grouping in the brain. That means the equaliser works in a way that avoids artefacts that can occur because originally grouped information is torn apart or distorted. The equalisation also happens in a special perception space that allows us to apply these sorts of constraints. Only after determining the equalisation change in this space, it is mapped back to a normal equalisation change that then affects the signal in exactly the way required to affect the perception as intended. The perception space used for equalisation comes with the nice feature of being loudness neutral. That means the perceived instantaneous loudness of the output matches that of the input exactly (- if you listen at the same volume as the internal perception model, see below.)

This is why changing the sharpness, or "q-factor" of the equaliser response would break the design of Gullfoss and result in less-than-optimal signal modifications, according to our model.

The perception model listens to the input signal like a human listener at a certain sound pressure level. Per default, it is calibrated to perceive a full-scale sine at 1kHz at a sound pressure level of 67dBSPL. This calibrated can be affected by the boost parameter. Positive boost settings reduce the listening level, negative ones increase it. That's why boost is measured in dB(SPL). The cutoff floor for Gullfoss' processing is determined by the absolute threshold of hearing as implemented by the model. But long before you hit that absolute threshold, the low volume characteristics of perception will change enough to slowly saturate the EQ changes. So there is no hard threshold.

Allowing the user to set Tame and Recover as a function of frequency is problematic. Both strategies only work correctly if they are applied equally to the entire local context. We already have the range limiters, which allow you to focus in on or fully exclude a certain frequency range. This feature is already a compromise because near the transition the unmasking does not work perfectly. So you have to be careful where to set the limiters. Giving the user the ability to freely choose the Tame and Recover parameters would likely encourage to produce transitions that would break Gullfoss' internal assumptions and reduce its ability to fix masking issues. It would also make the user interface more complicated and has the potential to confuse less technically inclined users. So we have decided against it. There is a chance that it will come in a future version that is geared towards power users, but certainly with limitations regarding the transitions.

So, with these explanations, I think you will understand that running a stationary ergodic noise process through Gullfoss is merely a special case. And in this special case, the optimal information distribution does result in something that is close to pink noise but does not fit exactly. If you change Boost and Brighten, you will see that there are degrees of freedom that will make it deviate even more from pink noise. But already making the input signal non-ergodic while keeping it pink will reveal that power density statistics are not the foundation of Gullfoss' processing. Try it with a Poisson process for example and observe how the response changes if you modify the expectation value of the lag time.

Finally, we do believe that A/B and Undo are services that should be provided by the hosting DAW, and in fact, most of them do. Implementing these in a plugin causes all sorts of problems with parameter automation writing and it diversifies the user interface of common operations. It is also much simpler to implement on the host side and is specified as such in most plugin standards.

Allow me also to point you to two resources that could be of particular interest to you:

https://www.prosoundnetwork.com/pro-sou ... -tells-all

https://www.gearslutz.com/board/showpos ... tcount=601

I hope you understand now that Gullfoss is not your average audio processor.

Cheers!
---
Andreas Tell
Soundtheory

[atell]

That said, I never had an agenda. I am not supporting Sound Theory nor am I against them. Overall, it appears as if though Gullfoss uses Pink Noise as a reference but like I mentioned a few posts ago, I do think there is more going on under the hood. Things like: time intervals, q factor, inertial thresholding, F-M curves...technical stuff that you probably are much more qualified to talk about than the rest of us peasants.

May I remind you of a few things you wrote earlier?

They market it to be something more than what it is. It's just a dynamic matching eq with pink noise as a reference. That's it.

Personally, I, don't need to run tons of audio through it because it's clear to me what Gullfoss is doing.

Guys, run with the idea of playing music through it. I'm done testing this thing. I already have my conclusion.

That's not really in agreement with how you are presenting your position now, is it? At the very least you will have to live with the accusation of following bad science practice.

As a trained scientist, which you say you are, you should be aware that the essence of the scientific method lies in falsification. If you come up with a hypothesis as a scientist, you will try to find evidence to falsify your own hypothesis.

Trying to only find supporting facts plays into your confirmation bias. As far as I can see, you did not attempt to falsify your own initial hypothesis, that Gullfoss uses a pink noise reference and is just a matching EQ. In fact, you did not even consider the possibility that it was anything else, by directly jumping to the conclusion you had set your mind on by just collecting supporting evidence.

This is very bad science, because to the non-scientifically trained it suggests that you used the scientific method when in fact you abused it to prove your point. Haven't you learned this during your science education? Or are you deliberately ignoring what you have learned?

Andreas

[Mathematics]

Andreas, thanks for posting this here. By the way, I had over 100 messages before yours in my personal email, I see it now.

[Ah_Dziz]

Anybody know when the windows version goes to full price? It’s probably right up there on the site, but I seem to be really good at missing that sort of thing.

[Mathematics]

As a trained scientist, which you say you are, you should be aware that the essence of the scientific method lies in falsification. If you come up with a hypothesis as a scientist, you will try to find evidence to falsify your own hypothesis.

Trying to only find supporting facts plays into your confirmation bias. As far as I can see, you did not attempt to falsify your own initial hypothesis, that Gullfoss uses a pink noise reference and is just a matching EQ. In fact, you did not even consider the possibility that it was anything else, by directly jumping to the conclusion you had set your mind on by just collecting supporting evidence.

This is very bad science, because to the non-scientifically trained it suggests that you used the scientific method when in fact you abused it to prove your point. Haven't you learned this during your science education? Or are you deliberately ignoring what you have learned?

Andreas

Again, like I mentioned earlier...I don't make claims as an authority on this theory, plugin, or software engineering. I'm getting pretty tired of repeating that.

I'm sure when you studied waves and oscillation theory, you can't deny that a basic test could have been the white and pink noise. That's obvious. It's as basic as using lambda for solving a differential equation. I wasn't trying to prove or disprove anything, I wanted to see what the "characteristic function" looked like after.

It's like you guys have to attack someone with some sort of common sense. In your case, you're job is much more difficult, because you have to explain why Gullfoss looks like it produces the same result but by a different means based on subjective phenomenon that you have coded the algorithm for. You chose wisely/unwisely to argue with someone that has more than a degree from YouTube; epecially, after I had just mentioned to someone about this very thing. Lol. You guys. Even the educated can take the advice from a convicted felon.

Going back to the beginning of my original question about the best use of Gullfoss, everything after that is a basic observation. Then I got the similar attacks from people about how the tests were flawed. I never argued that they were conclusive but they were somewhat good enough to at least show the direction of what happens to input. That was good enough for me and it still is, regardless of how Gullfoss is coded.

As a programmer, you claim that you have the algorithm that can create what the brain wants to hear. Let's let that sink in. You can fool most everyone here that Gullfoss is much more but you can only code something based on some quantifiable result. This means, you must be using some kind of reference curve for the minimal amount of calculations it's doing, i.e, clock cycles. Ok, so let's settle on that it's NOT a pink noise reference curve since you claim there's something more fundamental. I like that idea and I can see that. The test show that it's not exactly a reference curve...only close. It is though, some kind of threshold for attenuation, but not a curve. That's great. In fact, that's something purposeful. So 1, 2 ,3 ,4 don't show a slope of 1. Got it.

This brings up something important for our audience. One thing most guys don't know here, is the magnitude of computation. For instance, if I were to code the algorithm for a tube emulation plugin that resembles the magintude of what Gullfoss is achieving, I would need a minimum of 24cores to process a sound that was nearly imperceivably different than the actual hardware. This is why UAD plugins sell those co-processors. Anway, Gullfoss must be cutting some serious corners since it runs as light as a simple compressor emulation. Ok, so good coding? I'm sure. I would expect nothing less.

However, if what you want people to believe Gullfoss is, some kind of algorithm based on the neurotransmission of the human brain...it would would probably need double the amount of cores, but it isn't. Tangent time...

The point I'm trying to make, is that you are trying to say that I'm not using the scientific method as it was intended. Of course not. What's the point? Your website, user manual, and all that you have said, doesn't give enough information to draw a hypothesis in the first place. Ridiculous. I think you're trying to win the hearts of the believers in that it is not what it looks like. Not professional Andreas.

I mean, unless you studied brain neurology and Gullfoss requires an organic based processor, it boils down to a mathematical algorithm with an input and an output. I tested out Gullfoss on a the human voice. It smears the audio. No problem, I just know that I probably wouldn't use heavily on a vocal track.

I tested it out on complete tracks, individual instruments, and your algorithm, every time, always tries to adjust to a pink noise curve, or close to it. Look, I'm not trying to make Gullfoss seem less effective than it is, but whatever algorithm it generates its output from, an incoming input has to be any kind of noise other than a single tone.

The results speak for themselves. I never spoke badly of the plugin. In fact, the results I posted only show something logically good, even to the simple hobbyist. Thanks for posting your thoughts though. I'm sure this will help people make a definitive decision.

Please, don't take my retort negatively. I just think as a professional, you should resort to teaching for understanding rather than take results as a personal attack. Also, people don't care about taking sides. Like me, we're working toward the next thing. Tools come and go. People don't care how things work, so long as they do as intended.

...this all came about because you didn't provide enough information about your plugin, how it works, and "how to effectively use it". That was my exact line that started this. Unbelievable. It took me sending you 6 emails, start an online debate about a test that was nothing more than a test with results and that wasn't aimed at proving anything in the first place; no agenda either as someone claimed. Technically, you just brought all this arguing between you and I upon yourself for your inadequacy to provide sufficient information for something that is "like nothing else."

If you read back at my comments...I have great things to say about it.

My suggestions:
- Make a better user manual and update your FAQ to incorporate the questions I asked.
- Add more control.

[Mathematics]

So, let's try this again. How do you guys effectively use this?

[v1o]

As a trained scientist, which you say you are, you should be aware that the essence of the scientific method lies in falsification. If you come up with a hypothesis as a scientist, you will try to find evidence to falsify your own hypothesis.

Trying to only find supporting facts plays into your confirmation bias. As far as I can see, you did not attempt to falsify your own initial hypothesis, that Gullfoss uses a pink noise reference and is just a matching EQ. In fact, you did not even consider the possibility that it was anything else, by directly jumping to the conclusion you had set your mind on by just collecting supporting evidence.

This is very bad science, because to the non-scientifically trained it suggests that you used the scientific method when in fact you abused it to prove your point. Haven't you learned this during your science education? Or are you deliberately ignoring what you have learned?

Andreas

Again, like I mentioned earlier...I don't make claims as an authority on this theory, plugin, or software engineering. I'm getting pretty tired of repeating that.

I'm sure when you studied waves and oscillation theory, you can't deny that a basic test could have been the white and pink noise. That's obvious. It's as basic as using lambda for solving a differential equation. I wasn't trying to prove or disprove anything, I wanted to see what the "characteristic function" looked like after.

It's like you guys have to attack someone with some sort of common sense. In your case, you're job is much more difficult, because you have to explain why Gullfoss looks like it produces the same result but by a different means based on subjective phenomenon that you have coded the algorithm for. You chose wisely/unwisely to argue with someone that has more than a degree from YouTube; epecially, after I had just mentioned to someone about this very thing. Lol. You guys. Even the educated can take the advice from a convicted felon.

Going back to the beginning of my original question about the best use of Gullfoss, everything after that is a basic observation. Then I got the similar attacks from people about how the tests were flawed. I never argued that they were conclusive but they were somewhat good enough to at least show the direction of what happens to input. That was good enough for me and it still is, regardless of how Gullfoss is coded.

As a programmer, you claim that you have the algorithm that can create what the brain wants to hear. Let's let that sink in. You can fool most everyone here that Gullfoss is much more but you can only code something based on some quantifiable result. This means, you must be using some kind of reference curve for the minimal amount of calculations it's doing, i.e, clock cycles. Ok, so let's settle on that it's NOT a pink noise reference curve since you claim there's something more fundamental. I like that idea and I can see that. The test show that it's not exactly a reference curve...only close. It is though, some kind of threshold for attenuation, but not a curve. That's great. In fact, that's something purposeful. So 1, 2 ,3 ,4 don't show a slope of 1. Got it.

This brings up something important for our audience. One thing most guys don't know here, is the magnitude of computation. For instance, if I were to code the algorithm for a tube emulation plugin that resembles the magintude of what Gullfoss is achieving, I would need a minimum of 24cores to process a sound that was nearly imperceivably different than the actual hardware. This is why UAD plugins sell those co-processors. Anway, Gullfoss must be cutting some serious corners since it runs as light as a simple compressor emulation. Ok, so good coding? I'm sure. I would expect nothing less.

However, if what you want people to believe Gullfoss is, some kind of algorithm based on the neurotransmission of the human brain...it would would probably need double the amount of cores, but it isn't. Tangent time...

The point I'm trying to make, is that you are trying to say that I'm not using the scientific method as it was intended. Of course not. What's the point? Your website, user manual, and all that you have said, doesn't give enough information to draw a hypothesis in the first place. Ridiculous. I think you're trying to win the hearts of the believers in that it is not what it looks like. Not professional Andreas.

I mean, unless you studied brain neurology and Gullfoss requires an organic based processor, it boils down to a mathematical algorithm with an input and an output. I tested out Gullfoss on a the human voice. It smears the audio. No problem, I just know that I probably wouldn't use heavily on a vocal track.

I tested it out on complete tracks, individual instruments, and your algorithm, every time, always tries to adjust to a pink noise curve, or close to it. Look, I'm not trying to make Gullfoss seem less effective than it is, but whatever algorithm it generates its output from, an incoming input has to be any kind of noise other than a single tone.

The results speak for themselves. I never spoke badly of the plugin. In fact, the results I posted only show something logically good, even to the simple hobbyist. Thanks for posting your thoughts though. I'm sure this will help people make a definitive decision.

Please, don't take my retort negatively. I just think as a professional, you should resort to teaching for understanding rather than take results as a personal attack. Also, people don't care about taking sides. Like me, we're working toward the next thing. Tools come and go. People don't care how things work, so long as they do as intended.

...this all came about because you didn't provide enough information about your plugin, how it works, and "how to effectively use it". That was my exact line that started this. Unbelievable. It took me sending you 6 emails, start an online debate about a test that was nothing more than a test with results and that wasn't aimed at proving anything in the first place; no agenda either as someone claimed. Technically, you just brought all this arguing between you and I upon yourself for your inadequacy to provide sufficient information for something that is "like nothing else."

If you read back at my comments...I have great things to say about it.

My suggestions:
- Make a better user manual and update your FAQ to incorporate the questions I asked.
- Add more control.

Right I’m not trying to break any soldier. I’m only being honest by stating the obvious, that is you don’t know what you’re talking about and your tests are invalid. And rather than owning up to your errors you only continue to push through with your assertions even though they’ve been debunked by Sound Theory.

[Mathematics]

They're not completely invalid though. To be honest, I actually got more insight on how to use Gullfoss because of the test. The results show one thing. I think we all agree it certainly looks like what it looks like. Ok, we now know the result is due to something "more fundamental". Instead of trying to talk crap and point fingers, you should try to help bring understanding. You've done nothing for this thread.

As far as trying to claim that I don't know what I'm talking about. If it looks like a duck and quacks like a duck, but it isn't a duck, how can you blame me for thinking it is a duck? You...make no sense. I think you're the one that doesn't know what you're talking about.

Seriously, this is why I asked to quit quoting me. Again, you're so caught up thinking that I care about these simple tests. Of course they don't explain whatever is going on.

What's funny is that I'm sure people will say that it looks like it's eq matching to a pink noise and not that it's eq'ing to what the human ear wants to hear. What the heck does that mean? I bet you have a better idea on how to use it now seeing the results I posted. You're welcome.

[v1o]

They're not completely invalid though. To be honest, I actually got more insight on how to use Gullfoss because of the test. The results show one thing. I think we all agree it certainly looks like what it looks like. Ok, we now know the result is due to something "more fundamental". Instead of trying to talk crap and point fingers, you should try to help bring understanding. You've done nothing for this thread.

Everything you said is invalid and quite misleading. Especially to people who might stumble upon this thread and read your posts. For instance below some of the things you said:

"Gullfoss isn't as intelligent as they say"

"They market it to be something more than what it is. It's just a dynamic matching eq with pink noise as a reference. That's it."

"it's also kind of gimmicky when you really think about it. I mean...who really eq matches to pink noise. Unless you're doing ambient or ethereal...no one. "

"I can see why they put high and low pass filters. They probably couldn't figure out how to keep the dynamic eq from eq'ing dead air or eq'ing irrelevant parts."

All that is inaccurate and stated by you as if it's fact.

As far as trying to claim that I don't know what I'm talking about. If it looks like a duck and quacks like a duck, but it isn't a duck, how can you blame me for thinking it is a duck? You...make no sense. I think you're the one that doesn't know what you're talking about.

Nonsense

Seriously, this is why I asked to quit quoting me. Again, you're so caught up thinking that I care about these simple tests. Of course they don't explain whatever is going on.

Nonsense

What's funny is that I'm sure people will say that it looks like it's eq matching to a pink noise and not that it's eq'ing to what the human ear wants to hear. What the heck does that mean? I bet you have a better idea on how to use it now seeing the results I posted. You're welcome.

Actually no in actual usage it does not look or sound like it's matching a pink noise reference. And I have a good idea on how to use it based on my own testing and experience.