GR Research Speaker Upgrade Review (Sierra-2EX V2) [Video]

[Randolf]

At 28:18 he actually shows a measurement without smoothing applied.

He says:

Notice how jagged it is all across the top end. That's where the wavelength starts getting short, and the wavelengths will start reflecting of off things (?), and it causes little ripples all through there, and it's harder to read what's going on, when you're looking at all these little peaks, because what you really want is the average. You want to see what is the level ... I don't need to see every little wiggle, because you can't adjust for those. You can't go in and say "man there's one there, and one there..." You can't change that.

He seems to think that those ripples are caused by the speaker, but they are most likely caused by reflections from the mic stand and/or speaker stand.

It is probably not smoothed, but it is still 4ms time gated ("Unsmoothed ... Rectangular Start 2.85ms Stop 6.85ms FreqLO 250Hz"). So if properly done it at least should not contain room reflections but the usual tradeoff of limited frequency resolution in the lower mids due to time gate.

 

[Ageve]

It is probably not smoothed, but it is still 4ms time gated ("Unsmoothed ... Rectangular Start 2.85ms Stop 6.85ms FreqLO 250Hz"). So if properly done it at least should not contain room reflections but the usual tradeoff of limited frequency resolution in the lower mids due to time gate.

Yes, it's still gated, but he didn't apply smoothing on top of that, as he usually does.

The HF ripples are signs of small reflections, probably from the mic stand or speaker stand. Improper gating (too long window, capturing the first major reflection) would create ripples across the whole graph.

 

[Geert]

In his last video Danny explains how one of the main improvements he made is tame that resonance just above 1kHz, which according to him is far worse than the increased distortion that comes with his upgrade. I went back to the first upgrade video he did on this speaker because I seemed to remember something.

And here we have it, the stock spectral decay. I quote: "Spectral decay looks great", with Danny stressing 'great' and giving a 'thumbs up'.

This is the before and after spectral decay for the first version of his upgrade, where it looks like his upgrade made it worse (or his measurement reference isn't accurate). Altough according to him it didn't change that much and it was still fine.

Now over to his second video, where he comes back on the spectral decay. Now he shows this before (stock measurement) and after:

Now wait a minute ..., how come the resonance above 1kHz got so much worse in the 'before' measurement compared to the stock measurement he showed in the first video? The scale of the graphs didn't change, but looking at the low end of the graph it's almost like we're looking at different speakers. And now there is a problem of course; I quote "That ringing in the woofers response is a lot more audible than that 2,5% distortion that it might rise to". Although in the first video there was no problem?

Is that resonance that already dives below -25dB in less than 4ms audible and worse than 2% distortion, that's the real question of course. But I'm not sure if it makes sense to elaborate on that if the measurements are inconsistent.

@amirm, I know you don't like to pay much attention to the spectral decay and I understand why (the story above demonstrates it), but I would be interested to see the measurement you made just to understand what's going on with Danny's measurements. (*Edit: if Danny's measurement is correct, I guess that resonance comes from the bass reflex port).

 

[Chaos01]

I watched it. There is so much I can address in there. Should I do a video and do that?

I dont know, then He will do another 'answer video' and so on. But if you do so, stay fact based, no emotions like always and keep it short about 20min as possible. Or maybe say where he is right, where wrong and how the reality looks Like. And dont hang on the parts matter or not question to long, he believes what he is saying so much, that no study in the world can convince him.

I asked AI what its thinking about His Last Video and Gemini Put Out this:
"Disrespectful Behavior and Populism
The video is characterized by strong "us vs. them" rhetoric:
Degradation: Richie refers to his critics as "Flat Earthers". This is a populist method used to ridicule the scientific basis of the opposing side.
Allegations of Lying: He directly accuses Amir of slander and deliberate lying intended to damage his business.
Arrogance: Richie claims that Amir owns a $100,000 measurement system but does not know how to operate it or interpret the data."

I guess ASR members are the biggest group who watched his Last Video.
I really dont know, on one Side dont give him any Attention, on the other Hand His populist behavior and insults are so disrespectful... And He is Not wrong with every technical information but with some He is very wrong. So a answer hast to be on Point while Not wasting to much time for him, Not easy he knows exactly how to answer, Like a politics, He only Takes some of the criticsm, and where he was totaly wrong in His Last Video, he did Not talked about a second.

 

[SimpleTheater]

I added a reply to a reply to my comment above which is not visible (anymore) in the WebUI but my tool can download it, strange!

This means the YouTube channel has blocked you. You can see your comments, but no one else can, making it hard to know you’ve been blocked. Your tool is simply finding everything, hidden or not.

 

[SimpleTheater]

Let's do another one then, based on Danny's claim "I did have a full-size anechoic chamber when GR reserach first started. ...was like 23 feet Long.... we could take accurate measurements Achieving a reliable 40 Hz cutoff has rarely been attempted on a commercial scale due to space, cost, and engineering constraints, and is generally reserved for extreme research facilities.

So you’re saying Danny once had an extreme research facility, but is now storing old automobiles in it. Interesting.

 

[Geert]

So you’re saying Danny once had an extreme research facility, but is now storing old automobiles in it. Interesting.

Indeed. Back then the company was named ’GR Extreme Research”, but the research got to extreme.

 

[Nondemo01]

I watched it. There is so much I can address in there. Should I do a video and do that?

Amir, I understand wanting to respond point by point, especially if you believe Danny mischaracterized your test, your process, or your conclusions.

But at some point, I think this becomes less about the speaker, the measurements, or the upgrade, and more about two entrenched camps arguing past each other.

Audio is supposed to be a hobby. It is supposed to be enjoyable. Measurements matter. Listening matters. Design competence matters. Personal preference matters. But the objectivist-versus-subjectivist fight has become almost religious at this point.

Trying to convince hardcore subjectivists that measurements should matter more is like trying to convince religious people there is no God. Trying to convince hardcore objectivists that subjective enjoyment still has value is like trying to convince atheists there is one. Most people are not actually trying to be persuaded. They are defending a worldview.

That does not mean facts do not matter. It does not mean bad claims should go unanswered. But the endless point-by-point cage match rarely changes minds. It usually just hardens everyone’s position.

For most of us, the useful takeaway is pretty simple: measure what can be measured, listen for yourself, be honest about what you value, and buy what makes you enjoy music more.

At the end of the day, this is still supposed to be fun.

 

[RexrothPigeon]

Audio is supposed to be a hobby. It is supposed to be enjoyable. Measurements matter. Listening matters. Design competence matters. Personal preference matters. But the objectivist-versus-subjectivist fight has become almost religious at this point.

This isn't really that, though. Danny is making dozens of objective claims. They are mostly just false.

There is no objectivist-subjectivist problem here. If Danny said "my mods measure bad but I don't care, they sound good," that wouldn't be objectionable. But he's not saying that. He's making objectively false claims about the measurement regimen, ASR, Amir, Klippel NFS, his own measurement regimen, etc.

 

[RexrothPigeon]

this is still supposed to be fun.

People say this here a lot and it's never anything more than projection and assumption. You have no idea what is fun for other people.

 

[Chrispy]

Just makes me wonder how less credible Danny can make himself?

 

[PristineSound]

I remember well when Danny used to claim he had an anechoic chamber. This was a long time ago, perhaps back in the early 2000's.

Not long after our CBM-170's were measured in the anechoic chamber at Canada's NRC, I had a disagreement with Danny's old boss, Mark Schifter, about the "anechoic chamber" claim. I told him it was more of a treated room than an actual chamber. It was far too small and did not have the requisite suspended mesh flooring over thick fiberglass wedges to absorb floor bounce.

Danny's "wedges" were only about 4 inches deep, maybe 6" at most and were open cell acoustic foam. At such limited thickness and the wrong material, it would, at most be "anechoic" down to, at best 500-600Hz - assuming the floor was covered in wedges and the walls were thick concrete.

His claim of anechoic to 40Hz is absurd, he either truly doesn't understand or is just outright lying. To be anechoic to 40Hz would require specialized fiberglass wedges (not acoustic foam) that would need to be at least 4 FEET in depth. I believe there are only a few chambers in the world that are anechoic to this low of a frequency. Even the widely used chamber at the NRC is generally only good down to ~ 80Hz.

Here is a pic of GR's so called anechoic chamber from their website back then...



View attachment 531742

I think we all can fairly confidently agreed that neither Danny nor his previous boss, ever had a legit anechoic chamber from that picture.

Danny also talked about how he recently spent like $30k on wedges in his latest video response, which probably implies that the pictured treated room no longer exist and he is trying to DIY one himself.

To build an anechoic chamber is not just the specialized wedges with a suspended wire mesh floor, it needs to be completely sound proof. It's virtually impossible to DIY, I think we can expect Danny to brandish his soon to be Temu anechoic chamber in his next video.

 

[AscendDF]

In his last video Danny explains how one of the main improvements he made is tame that resonance just above 1kHz, which according to him is far worse than the increased distortion that comes with his upgrade. I went back to the first upgrade video he did on this speaker because I seemed to remember something.

And here we have it, the stock spectral decay. I quote: "Spectral decay looks great", with Danny stressing 'great' and giving a 'thumbs up'.

View attachment 531810

This is the before and after spectral decay for the first version of his upgrade, where it looks like his upgrade made it worse (or his measurement reference isn't accurate). Altough according to him it didn't change that much and it was still fine.
View attachment 531813

Now over to his second video, where he comes back on the spectral decay. Now he shows this before (stock measurement) and after:

View attachment 531814

Now wait a minute ..., how come the resonance above 1kHz got so much worse in the 'before' measurement compared to the stock measurement he showed in the first video? The scale of the graphs didn't change, but looking at the low end of the graph it's almost like we're looking at different speakers. And now there is a problem of course; I quote "That ringing in the woofers response is a lot more audible than that 2,5% distortion that it might rise to". Although in the first video there was no problem?

Is that resonance that already dives below -25dB in less than 4ms audible and worse than 2% distortion, that's the real question of course. But I'm not sure if it makes sense to elaborate on that if the measurements are inconsistent.

@amirm, I know you don't like to pay much attention to the spectral decay and I understand why (the story above demonstrates it), but I would be interested to see the measurement you made just to understand what's going on with Danny's measurements. (*Edit: if Danny's measurement is correct, I guess that resonance comes from the bass reflex port).

Thank you! I have not watched the last few videos, but seeing this makes me suspicious and requires deeper investigation. Not only do Danny's CSD measurements of the same speaker vary dramatically (as do his comments from one video to another) but the frequency response varies considerably. The first "slice" (the top most) is the frequency response of the speaker. With his CSD plot dated 12-3-2025, you can see a high frequency roll off starting at about 10kHz (which was why he unnecessarily boosted the highs in his 1st modification). In his 2nd CSD measurement, dated 03-11-2026, the HF roll off is magically gone.

In addition, in his first CSD measurement, there appears to be a gradual roll off starting at ~1 kHz and downward, while in his 2nd measurement, there is a pronounced dip at ~800Hz, and a scoop at about 400Hz. His measurements don't seem to be repeatable, that or something was modified between between his 1st and 2nd measurements. And, if you look at the dates, his 1st measurement was back on 12-03-2025 and his 2nd measurement was on 03-11-2026. Am I supposed to believe that he had a supposed customer's speakers in his shop for nearly 4 months?

I wish this customer would speak up, I have done everything possible to find out who this customer might be, with zero luck. I would love to measure this pair on the NFS.

 

[Ropeburn]

Thank you! I have not watched the last few videos, but seeing this makes me suspicious and requires deeper investigation. Not only do Danny's CSD measurements of the same speaker vary dramatically (as do his comments from one video to another) but the frequency response varies considerably. The first "slice" (the top most) is the frequency response of the speaker. With his CSD plot dated 12-3-2025, you can see a high frequency roll off starting at about 10kHz (which was why he unnecessarily boosted the highs in his 1st modification). In his 2nd CSD measurement, dated 03-11-2026, the HF roll off is magically gone.

In addition, in his first CSD measurement, there appears to be a gradual roll off starting at ~1 kHz and downward, while in his 2nd measurement, there is a pronounced dip at ~800Hz, and a scoop at about 400Hz. His measurements don't seem to be repeatable, that or something was modified between between his 1st and 2nd measurements. And, if you look at the dates, his 1st measurement was back on 12-03-2025 and his 2nd measurement was on 03-11-2026. Am I supposed to believe that he had a supposed customer's speakers in his shop for nearly 4 months?

I wish this customer would speak up, I have done everything possible to find out who this customer might be, with zero luck. I would love to measure this pair on the NFS.

Lack of consistency and repeatability is a problem indeed and makes measurements useless. Inconsistency in many regards is perhaps his biggest weakness.

Oh, and his customers DO speak up. It's just that he's notorious for deleting comments he doesn't like.

 

[VQR]

He did no such thing. He *claims* AI said it pulled such quotes out of ASR. But he neither shows the prompt he used, or the response.

I repeated his experiment, with exact question he said he asked: "audible effects of electrical shift of phase in speaker's impedance"

This is the answer Gemini gave:

View attachment 531654

Notice how the reference is NOT to ASR but Parts Express. If you click on that link, it will take you to a short, random thread about this topic with hardly any of the above discussed.

If you highlight the link then it will give you one reference to ASR among three others. That link is to a post by Don:

Variable impedance - what's the audible impact

Stupid question time: Is it true or not that load dependence of amplifiers impacts frequency response only at higher frequencies? I ask because the subject of this thread is "what's the audible impact" of "variable speaker impedance." And as we know from other threads, a major bone of...

www.audiosciencereview.com

He was answering this question: "Stupid question time: Is it true or not that load dependence of amplifiers impacts frequency response only at higher frequencies?"

Don's answer starts with:


This is precisely what I said in my video. Which is, this is only an issue if your amplifier has high output impedance, i.e. a tube amp. Solid state amplifiers have very low output impedance with essentially no change with frequency. Exception is class D amps that don't have post filter feedback.

So nothing in what AI generated is backed up by what is posted on ASR in the way Danny reading it. Indeed, I asked AI about this:
View attachment 531656

I have more on this but I finished my AI inquiry by giving Google Danny's video and asking this question in context:

View attachment 531657

Pretty damning I would say.

Finally, and this is really stupid but I see you are falling for it as well, ASR is not me. Ton of topics are discussed and naysayers come here argue points like what Danny says. Summarizing ASR as a forum does not at mean that what I state is wrong.

If you do decided to make a follow-up to Danny's response, you definitely should bring up his use of AI. AI is like Wikipedia: it can have valid information, but when push comes to shoves you must double-check its source. I see more and more people treat AI like it's infallible and purely objective, instead of a utility than can be innaccurate or hallucinate.

I wouldn't spend more than maybe 20 to 30 minutes in a video to debunk him. You can breeze through most everything succintly. Danny's videos can be played back at 1.5x speed yet still feel tedious as all heck.

 

[Ropeburn]

If you do decided to make a follow-up to Danny's response, you definitely should bring up his use of AI. AI is like Wikipedia: it can have valid information, but when push comes to shoves you must double-check its source. I see more and more people treat AI like it's infallible and purely objective, instead of a utility than can be innaccurate or hallucinate.

I wouldn't spend more than maybe 20 to 30 minutes in a video to debunk him. You can breeze through most everything succintly. Danny's videos can be played back at 1.5x speed yet still feel tedious as all heck.

Bad comparison. Wikipedia is being constantly peer-reviewed by the public, which is why (as studies found) it isn't any less accurate than the average encyclopedia.

None of this happens with AI. You have to check it every time anew by yourself.

 

[Randolf]

Since Danny continues to publish false claims on the effect of time gating and smooting on the measurement resultion and quite a lot people replied to my comment in his video, I decided to become creative and cretaed my very firts short Youtube video on this topic:

The goal is to fight against Danny`s disinformation on this topic in a format many people with and without deep knowledge on this topic can understand. I hope it makes the world a tiny little bit better and less disinformed. Sorry to bother you with advertisment of my lastet block buster

 

[Karmacoma]

Since Danny continues to publish false claims on the effect of time gating and smooting on the measurement resultion and quite a lot people replied to my comment in his video, I decided to become creative and cretaed my very firts short Youtube video on this topic:

The goal is to fight against Danny`s disinformation on this topic in a format many people with and without deep knowledge on this topic can understand. I hope it makes the world a tiny little bit better and less disinformed. Sorry to bother you with advertisment of my lastet block buster

Good job!

 

[DanielT]

Frequency response already tells us there are resonances. It is all those little peaks. My first graph shows them lingering. But if you raise the noise floor, or reduce resolution, they go away.

So Danny claims in his latest Youtube video response that he created a dip. A dip you showed:

GR Research Speaker Upgrade Review (Sierra-2EX V2)

This is a detailed review and measurements of a speaker modified by GR Research and compared to factory performance. The modification kit costs US $510 that you have to assemble: I know, not much to look at. :) But the left one was upgraded by the Ascend founder/designer Dave to include...

www.audiosciencereview.com

...to balance out, compensate for ringing he sees in the spectral decay. But how come there is still a dip in FR in Danny's modification?

Have I missed something? Or am I thinking in the wrong way now in connection with my FR - spectral decay question?

 

[VQR]

Bad comparison. Wikipedia is being constantly peer-reviewed by the public, which is why (as studies found) it isn't any less accurate than the average encyclopedia.

None of this happens with AI. You have to check it every time anew by yourself.

I stand corrected. Even after many years, I still have my old grade teacher in my head, insisting the class never use Wikipedia since "anyone can edit it!"

All the way back in the halcyon days, when teachers worried only about students citing and copy/pasting from Wikipedia, before the advent of LLM's to write papers for little Timmy... those are so long gone. Now we have to worry about crazy ladies falling for their psychiatrist since their AI buddy corroborated her delusions, or a charlatan like Danny hiding behind AI as a shield for his chickanery.