• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

ASR speaker (and electronics) review - is it really objective?

Sexter

New Member
Joined
Jul 22, 2021
Messages
1
Likes
2
Hello

I've read may excellent gear review on ASR, and as an electrical engineer, I have great respect for a scientific approach to audio review, but I'm afraid that not all reviews are truly objective and free from bias. Now I respect Amir and his work, and I know that the measurements results he presents are accurate. The problem is with his interpretation of the measurements results, especially when it comes to speaker measurements.

Let me just give an example. First, my gear is a simple home theater consists of Yamaha RX-2070 AVR, Focal Aria 906 and CC900 as front stage, Dali Spektor 2 as surrounds, and two sets of Dali Spektor 1 as front and back heights (effects). Together with dual SVS SB-3000, I'm overall very happy with this 5.2.4 system.

As the owner of both the Aria 906 and the Spektor 1, I was very excited to read both of these speakers reviews in on ASR:

https://www.audiosciencereview.com/forum/index.php?threads/focal-aria-906-speaker-review.14085/

https://www.audiosciencereview.com/...ali-spektor-1-review-bookshelf-speaker.25063/

But I was shocked from Amir's very negative interpretation of the Spektor 1 measurements results. I've have the Spektor 1 for 3 years, and I must say that O really like how they sound and perform, even when wall mounted. Yes, they're slightly on the low sensitivity side, but this is not an issue for an AVR receiver with automatic calibration. Yamaha YPAO had detected then as capable of 70Hz, but I had them crossed at 80Hz. The A2070 is quite powerful, and they can handle that power with ease w/o any audible distortion. Their SPL capability is more than adequate for Atmos speakers. Using an SPL and test signals, I was able to measure 98dBs, 4 meters away from them. Playing music as load as 97dB, I wasn't able to notice any distortion, even after I've connected them as main front speakers and re-run YPAO calibration.

Based on my memory, after listening to my favorite music for 2 hours using the Spektor 1, I can say with confidence that they're just as good as the Aria 906!
DALI.jpeg


In the recent Spektor 1 review, Amir presented the distortion test, and commented that at 96dB SPL, "the woofer is out of control", which is a misleading and false (subjective) comment. At 96dB from 1 meter way (I assume), the Spektor 1 is driven by 25W of power. For a speaker that designed to handle 100W of continues power, it should not have any problems handling 96dB from 1 meter! The low frequency <5% THD is very common in small speakers. This is called iron distortion of the crossover coils and/or voice coils due to magnetic saturation, but not due to over excursion of the drives, which will results in a very distorted sound. The measured distortion in the review is non audible, thus w/o including a disclaimer, these results are misleading.

For pure tones, the human hear cannot detect any distortion up to 30% THD, for low frequency tones (<300Hz). When it comes to music, no human can hear distortion up to 10% THD.

Have a look at this video, and judge for yourself:


Now, here are the distortion results graph from Warfedale Diamond 220, which Amir really liked and gave them a very positive review:

https://www.audiosciencereview.com/...ersus-frequency-audio-measurements-png.87985/

And here are the same distortion graphs for the Spektor 1, which Amir seemed to really hate:

https://www.audiosciencereview.com/...lf-speaker.25063/#lg=attachment142075&slide=0

They're almost identical (3.5% THD vs 4.5% THD), and at 86dB, the Spektor does much better! Still, Amir decided to interpret the results differently and add his negative subjective comment for the Spektor 1.

The bottom line, despite the negative review of the Spektor 1, I can assure anyone from my own experience and well trained ears, that these speakers sound amazing for their price! After repeating YPAO calibration if using a subwoofer, both the 1500$ Aria 906 and the 250$ Spektor 1, sound just as good and just as loud in two channel stereo.

It's unfortunate that many potential buyers that may be scared off by such biased to not very objective review and miss out these great speakers. I've also noticed that Amir also slaughtered the Yamaha RX-A1080 AVR. Now I do not own it, but before getting the A2070, I gave Denon X6400H a try for a couple of weeks, and while it sounded good for movies, it sucked for two channel music. This wasn't the case with my current Yamaha A2070, which sound great in both movies and music, even when using Cinema DSP to get all 5.2.4 speakers to work, including the Spektor 1.

Which brings me to my original question, are ASR reviews, especially speakers, really objective?

How come Cheap Chinese made DACs get excellent reviews, while they are very unreliable (my friend had his Topping D90SE failing on him after 6 months of use, he had it replaced and the second one came defective with background high pitch noise from the right channel).

Love to hear your thoughts.

Cheers.
 
Last edited:

beren777

Senior Member
Joined
Feb 25, 2020
Messages
339
Likes
589
"Subjectively though, I could not dislike it even without EQ. Its limits are farther than what I typically hear in small speakers. With EQ it becomes more pleasant but I personally could not warm up to them. I would save and buy a larger speaker with more bass. As noted, I am not a big fan of Dali Spektor 1 but you may interpret its performance differently. So while I won't recommend them, I don't have strong reasons to object to someone buying them either."

Seems like a fair disclaimer to me. Having looked at the two graphs you linked, one does look measurably worse than the other. Audible? I don't know. I find the reviews useful, but I consider only measurements to be objective, and even then only if performed to a consistent methodology.

I'm glad you're happy with your speakers.

Regarding DACs, I don't consider your second-hand anecdote sufficient to declare Topping "very unreliable." Either way, ASR doesn't measure warranties or warranty service. It measures technical performance. I've owned Topping and SMSL products with zero defects, but I wouldn't declare that their products are "superbly infallible."
 

Frgirard

Major Contributor
Joined
Apr 2, 2021
Messages
1,737
Likes
1,040
Hello

I'm new here. I've read may excellent gear review on ASR, and as an electrical engineer, I have great respect for a scientific approach to audio review, but I'm afraid that not all reviews are truly objective and free from bias. Now I respect Amir and his work, and I know that the measurements results he presents are accurate. The problem is with his interpretation of the measurements results, especially when it comes to speaker measurements.

Let me just give an example. First, my gear is a simple home theater consists of Yamaha RX-2070 AVR, Focal Aria 906 and CC900 as front stage, Dali Spektor 2 as surrounds, and two sets of Dali Spektor 1 as front and back heights (effects). Together with dual SVS SB-3000, I'm overall very happy with this 5.2.4 system.

As the owner of both the Aria 906 and the Spektor 1, I was very excited to read both of these speakers reviews in on ASR:

https://www.audiosciencereview.com/forum/index.php?threads/focal-aria-906-speaker-review.14085/

https://www.audiosciencereview.com/...ali-spektor-1-review-bookshelf-speaker.25063/

But I was shocked from Amir's very negative interpretation of the Spektor 1 measurements results. I've have the Spektor 1 for 3 years, and I must say that O really like how they sound and perform, even when wall mounted. Yes, they're slightly on the low sensitivity side, but this is not an issue for an AVR receiver with automatic calibration. Yamaha YPAO had detected then as capable of 70Hz, but I had them crossed at 80Hz. The A2070 is quite powerful, and they can handle that power with ease w/o any audible distortion. Their SPL capability is more than adequate for Atmos speakers. Using an SPL and test signals, I was able to measure 98dBs, 4 meters away from them. Playing music as load as 97dB, I wasn't able to notice any distortion, even after I've connected them as main front speakers and re-run YPAO calibration.

Based on my memory, after listening to my favorite music for 2 hours using the Spektor 1, I can say with confidence that they're just as good as the Aria 906!
View attachment 142941

In the recent Spektor 1 review, Amir presented the distortion test, and commented that at 96dB SPL, "the woofer is out of control", which is a misleading and false (subjective) comment. At 96dB from 1 meter way (I assume), the Spektor 1 is driven by 25W of power. For a speaker that designed to handle 100W of continues power, it should not have any problems handling 96dB from 1 meter! The low frequency <5% THD is very common in small speakers. This is called iron distortion of the crossover coils and/or voice coils due to magnetic saturation, but not due to over excursion of the drives, which will results in a very distorted sound. The measured distortion in the review is non audible, thus w/o including a disclaimer, these results are misleading.

For pure tones, the human hear cannot detect any distortion up to 30% THD, for low frequency tones (<300Hz). When it comes to music, no human can hear distortion up to 10% THD.

Have a look at this video, and judge for yourself:


Now, here are the distortion results graph from Warfedale Diamond 220, which Amir really liked and gave them a very positive review:

https://www.audiosciencereview.com/...ersus-frequency-audio-measurements-png.87985/

And here are the same distortion graphs for the Spektor 1, which Amir seemed to really hate:

https://www.audiosciencereview.com/...lf-speaker.25063/#lg=attachment142075&slide=0

They're almost identical (3.5% THD vs 4.5% THD), and at 86dB, the Spektor does much better! Still, Amir decided to interpret the results differently and add his negative subjective comment for the Spektor 1.

The bottom line, despite the negative review of the Spektor 1, I can assure anyone from my own experience and well trained ears, that these speakers sound amazing for their price! After repeating YPAO calibration if using a subwoofer, both the 1500$ Aria 906 and the 250$ Spektor 1, sound just as good and just as loud in two channel stereo.

It's unfortunate that many potential buyers that may be scared off by such biased to not very objective review and miss out these great speakers. I've also noticed that Amir also slaughtered the Yamaha RX-A1080 AVR. Now I do not own it, but before getting the A2070, I gave Denon X6400H a try for a couple of weeks, and while it sounded good for movies, it sucked for two channel music. This wasn't the case with my current Yamaha A2070, which sound great in both movies and music, even when using Cinema DSP to get all 5.2.4 speakers to work, including the Spektor 1.

Which brings me to my original question, are ASR reviews, especially speakers, really objective?

How come Cheap Chinese made DACs get excellent reviews, while they are very unreliable (my friend had his Topping D90SE failing on him after 6 months of use, he had it replaced and the second one came defective with background high pitch noise from the right channel).

Love to hear your thoughts.

Cheers.
The objective is what is measured.
The subjective is what is heard.
This is at you to find the balance between the two worlds.
 

pozz

Слава Україні
Forum Donor
Editor
Joined
May 21, 2019
Messages
4,036
Likes
6,825
For pure tones, the human hear cannot detect any distortion up to 30% THD, for low frequency tones (<300Hz). When it comes to music, no human can hear distortion up to 10% THD.
Your comments have no basis in psychoacoustics. You'll find numbers far, far lower than that for pure tone thresholds. Regardless, masking is always in play. If you're on the lookout for audible distortion it will generally be broadband and found in the higher frequencies.

ASR reviews aren't "objective" in some rigid sense. They provide you data and good commentary. The background for the interpretations and so forth is distributed around the forums. Keep in mind the tone for the reviews is largely conversational and dips in and out of technical discourse. It's not a journal paper submitted for review.

Small woofer breakup for low frequencies is a given. Remember that the coil assembly is part of the speaker's motor. Measured here, between the 86dB and 96dB outputs the THD went up very fast, meaning that certain motor limits have been reached. In terms of audibility? You have these comments in the listening portion:
  • Dali:
Good news here was dynamics. I could turn this little speaker up very high with no sign of heavy distortion. Two of them would do wonders to fill even a large space. So maybe there is something to that wood fiber!
  • Wharfedale:
The 220 could also play very loud. The only problem is that the port makes very obnoxious rattling sound that doesn't come out of the woofer. This only happens at very high volumes and you only hear it if you are close to the speaker.
Then you write this:
It's unfortunate that many potential buyers that may be scared off by such biased to not very objective review and miss out these great speakers.
Don't take this to heart: both the Dali and Wharfedale speakers rank ok to pretty good. They have flaws. Fact is what these flaws are and how they contribute to overall sound is not very well understood by the community at large and takes time to grapple with, even with the data given to us first hand. And even at the expert level there are grey areas which research has not fully covered. I think a pretty reasonable person can come to terms with that.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
The thing that you need to understand is that measurements focused reviews are primed to bring out the faults in a product. measurements make it very easy to see what deviates from the norm / ideal.

so when Amir sees a resonance, an anomaly or some sort of fuck up in the graphs it will taint the whole review, even when the speaker is good in all other aspects. People are also quite susceptible to that and their opinion of the product will be tainted as well.

And even when the speaker has no glaring faults, people look for Amir for impressions.

Look for example at the ELAC DBR-62 review, there is no glaring faults but nothing super impressive either. yet it's regarded by most here as a great bookshelf speaker.

on the other hand, KEF R3, which is possibly one of the best bookshelf speakers on the planet (8.2/10 Olive score with subwoofer without EQ) Amir thought was 'meh' so it's not really popular here.

it's important when assessing such technical reviews to know just enough about the topic so you would be able to seperate the reviewer's opinion and actual substance.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
This is misinformation by the way. with pure tones it's incredibly easy to hear distortion, even below 1% THD if you train yourself to

For pure tones, the human hear cannot detect any distortion up to 30% THD, for low frequency tones (<300Hz). When it comes to music, no human can hear distortion up to 10% THD.

Furthermore THD is a weird way of quantifying something. is it primarily second harmonic or are there 3rd, 4th and 5th harmonics too?

1% 5th order harmonic is INCREDIBLY AUDIBLE compared to 1% second harmonic.
 

AdamG

Proving your point makes it “Science”.
Moderator
Forum Donor
Joined
Jan 3, 2021
Messages
4,636
Likes
14,918
Location
Reality
Thread notice: This thread is temporarily closed while we investigate this supposed new user original account. At present and until this member responds to my PM, this new member is a former Banned member. I believe in second and even third chances. After all we are human and on occasion we let our emotions get the best of us. We understand this and are more than willing to review requests for account reinstatement. Creating a new account and pretending to be someone else is not the way. Send a PM to either myself or Amir @amirm and just ask. ;)
 
Last edited:

AdamG

Proving your point makes it “Science”.
Moderator
Forum Donor
Joined
Jan 3, 2021
Messages
4,636
Likes
14,918
Location
Reality

dfuller

Major Contributor
Joined
Apr 26, 2020
Messages
3,335
Likes
5,050
I don't think the D90SE has been released for that long has it ?
Certainly not, Amir's review was not long after it came out. The D90 though might make sense.
 

Kachda

Addicted to Fun and Learning
Forum Donor
Joined
May 31, 2020
Messages
903
Likes
1,578
Location
NY
on the other hand, KEF R3, which is possibly one of the best bookshelf speakers on the planet (8.2/10 Olive score with subwoofer without EQ) Amir thought was 'meh' so it's not really popular here.

The initial 'meh' impression was largely due to room modes which @amirm had not started correcting back then (R3 was one of the earliest speakers to go on the bench). He re-heard the speaker correcting for that when reviewing another (think Revel) speaker later when he realized room modes were causing problems, and his impression was a lot better.

Many people in the KEF R3 thread have asked for this correction to be included, but it has never been done. So anyone landing on that thread will come away with a bad impression, unless they get to page 20+ of the thread (which I doubt many bother with).
 

Laserjock

Major Contributor
Forum Donor
Joined
Feb 25, 2019
Messages
1,327
Likes
1,000
Location
Texas Coastal
The initial 'meh' impression was largely due to room modes which @amirm had not started correcting back then (R3 was one of the earliest speakers to go on the bench). He re-heard the speaker correcting for that when reviewing another (think Revel) speaker later when he realized room modes were causing problems, and his impression was a lot better.

Many people in the KEF R3 thread have asked for this correction to be included, but it has never been done. So anyone landing on that thread will come away with a bad impression, unless they get to page 20+ of the thread (which I doubt many bother with).
So the thing that hurt the Revel F208 scores, helped the KEF R3 ?
 

Kachda

Addicted to Fun and Learning
Forum Donor
Joined
May 31, 2020
Messages
903
Likes
1,578
Location
NY
Measurements/redone or whatever
Cant remember. The speaker that came right after the r3 on the nfs is the one that benefited from amir eqing the room mode
 

Sancus

Major Contributor
Forum Donor
Joined
Nov 30, 2018
Messages
2,923
Likes
7,616
Location
Canada
  • Like
Reactions: Ata

Laserjock

Major Contributor
Forum Donor
Joined
Feb 25, 2019
Messages
1,327
Likes
1,000
Location
Texas Coastal
Top Bottom