Monaural Speaker Recordings - Blind Testing Six Speakers for Fun

[GXAlan]

I’m pretty sure JakeK meant that the recorded signal may have been driven above digital zero and caused clipping, not that the amp used for the speakers or the microphone was clipping.

To make sure no clipping occurs while recording you need to leave enough headroom, the peaks should never reach higher than say -3/-6 dB. You must turn the gain knob down on the audio interface until you reach that level of headroom.

Not sure what is meant, which means there may be an error in my technique.

When recording with audacity, the recording VU meter had *peaks* around -20 dB. The UMIK-2 was set at 100% since I assumed I had plenty of headroom.

Setups
SACD player into integrated Class D amp, volume set with int amp
Digital into active speaker, volume attenuation done on the speaker
HDMI PCM into AV separates with Class AB amp, volume set by volume knob of AV processor
USB drive into lifestyle product, volume set on lifestyle product

 

[goat76]

Not sure what is meant, which means there may be an error in my technique.

When recording with audacity, the recording VU meter had *peaks* around -20 dB. The UMIK-2 was set at 100% since I assumed I had plenty of headroom.

Setups
SACD player into integrated Class D amp, volume set with int amp
Digital into active speaker, volume attenuation done on the speaker
HDMI PCM into AV separates with Class AB amp, volume set by volume knob of AV processor
USB drive into lifestyle product, volume set on lifestyle product

Okay, I saw that now when looking at you recorded files that they were way down at -16 dB.

 

[JakeK]

No way it’s the amps, but it could be the mic. The UMIK-2 didn’t clip electrically, and MiniDSP claims UMIK-2 handles “125 dB” but maybe for something that isn’t a frequency sweep, it starts to distort.

Depending on the setup,
1) Bridged NC500, 625W at 1% into 8 ohms
2) Active speaker with 250W class D x 2 (biamplified), no THD rating
3) Class AB amp, 177W and 250W into 8 and 4 ohms at 0.1%, CEA burst of 225 and 396W (1%)
4) Lifestyle product, at risk for clipping (but unlikely once I tell you). Partially measured since it is a Class D with switching voltage Class G/H style. 65 W into main drivers at 0.30% at 4 ohms, and 130W at 0.08% into 2 ohms. Woofers are 2 ohms and power is advertised as 160W and peak 320W there.

Maybe it is the mic at fault as it's made to measure levels instead of making accurate recordings? All I know is it sounded quite badly distorted to me. Can you hear the harshness in person? If the volume was quite high it could of course be distortion in the speaker.

 

[tmtomh]

Very interesting - thanks @GXAlan ! And thanks @IAtaman for the level-matched versions, which I used.

I think there are likely multiple dependent variables here - not least of which is the two different rooms, and even more importantly, different listening positions. I mean, if you've ever used a UMIK mic to take multiple room measurements, you know that you can move 6 inches and the measurements can look radically different.

That said, I tend to agree with what seems to be the rough consensus: A and B sound the worst to me; C is an improvement but there's some kind of noise or distortion; D is smoother than A though C but somewhat muffled. E is more open sounding but also has some harshness; F is sort of a decent medium between D and E. So I guess I'd choose F and then 2nd choice would be a toss-up between D and E depending on my mood (and how long I had to listen - couldn't do E for any extended period.)

But my overwhelming feeling when I heard A and B was not, "those speakers sound bad" or even "those speakers sound bad in that room." My overwhelming perception was "these are poor recordings." No offense at all to Alan - this is interesting and fun, and it clearly took a good deal of effort. It's just that my brain was listening to recordings - I couldn't separate out what I thought I might be hearing from the speakers themselves.

 

[GXAlan]

No offense at all to Alan - this is interesting and fun, and it clearly took a good deal of effort.

None taken. I agree these recordings are “not ideal”

i think the area where subjectivists and objectivists could agree in is if we could record a speaker well, it would be an interesting way to compare “systems”.

When I have previous recorded events using a Sony PCM-M50 it was fine. I have heard great PCM-D1 recordings too.

Using the UMIK-2 is bad, but it also is interesting to see how we use this for our room correction. I am not sure where the issues lie, bht

1) it’s interesting to see that for speakers with very different sensitivities and frequency responses, how do you level match? These were all level matched with pink noise and yet it’s clearly different. It makes sense — pink noise is going to capture reflections, on axis sound, and these speakers are *very* different.

Room info

Spoiler

2) A and C are in the same large room. B, E, F are in the same smaller room. D is a third environment.

Since you voted and commented, I will PM you with the identities.

 

[GXAlan]

Maybe it is the mic at fault as it's made to measure levels instead of making accurate recordings? All I know is it sounded quite badly distorted to me. Can you hear the harshness in person? If the volume was quite high it could of course be distortion in the speaker.

I don’t hear that kind of distortion. Yes, I agree. Yet we use the REW sweep to calculate distortion using a UMIK-1 or UMIK-2.

Since most of you have a UMIK-1/2 at home anyway, and Audacity 2.2.2 is free, we can give it a try. You need the older version of Audacity for the correction spreadsheet here:

Audio recording through calibrated UMIK-1?

I would like to do audio recordings with my UMIK-1, but I need software which would accept the calibration file to be included to allow for the most linear frequency response. I can record through the UMIK-1 without any calibration, but the result is not really convincing as it doesn't sound...

www.minidsp.com

 

[rationaltime]

I listened to the youtube version only. Frankly I find this performance suboptimal
for comparing speakers, instruments, or musicians. While demonstrating the
proficiency of the muscians, the tempo is so fast that successive notes cut off the
timbre and harmonics of the instruments. We judge instruments by their
characteristic harmonics and damping. Good violinists are able to extract good
tone from a violin. There is no chance to hear that in this piece. Maybe you can
if you are used to listening to this kind of music.

I would not use this performance to select speakers. That is my opinion.

 

[GXAlan]

I listened to the youtube version only. Frankly I find this performance suboptimal
for comparing speakers, instruments, or musicians. While demonstrating the
proficiency of the muscians, the tempo is so fast that successive notes cut off the
timbre and harmonics of the instruments. We judge instruments by their
characteristic harmonics and damping. Good violinists are able to extract good
tone from a violin. There is no chance to hear that in this piece. Maybe you can
if you are used to listening to this kind of music.

I would not use this performance to select speakers. That is my opinion.

That is a good point about the speed blurring the harmonics.

On the one hand, we have the strict objectivist perspective that the best speaker is ruthlessnessly transparent. On the other, we have subjectivist perspective that some speakers are too accurate and you may want a “forgiving speaker.” Based upon your description, since you cannot hear damping or the harmonics due to the tempo, maybe this is a “forgiving recording”

Second, maybe someone likes fast tempo’d classical music. The performance on this type of music is important. Maybe full range bass extension is more important than accurate treble/harmonics?

Even though this may not reflect the “best” speaker (nor the best recordings), did you have a preference among the six?

 

[Blumlein 88]

It comes back to what is your purpose in recording? If it is to do a youtube, where people can get an idea how a speaker sounds, recording at the LP won't do that. Even purist stereo recordings have the microphones placed closer to the musicians than where the audience can sit. This is to get a sound more like the front rows of an audience hears.

If the idea is to do a youtube so people can get an idea of the sound of the speaker in that room, you still need to record closer than the LP, but not as close as you can get to the speaker. Any youtube recording from the LP is ill-conceived as it will not be like what you hear at the LP, nor at all like the speaker either.

A bit of an exception will be binaural recordings though they too might work best a bit closer than the LP normally used. Here is a highly detailed example of that by mitchco who does at times post here.

KEF LS50 (David) Versus JBL 4722 Cinema (Goliath) Speaker Comparison with Binaural Recordings

Here are two extreme opposite loudspeakers when it comes to sensitivity, size, directivity, topology and looks. Does the big speaker sound perceptually larger than the smaller speaker? Do the speakers sound the same if eq’d the same and level matched? Does the LS50 sound perceptually bigger when ...

audiophilestyle.com

 

[JakeK]

I don’t hear that kind of distortion. Yes, I agree. Yet we use the REW sweep to calculate distortion using a UMIK-1 or UMIK-2.

Since most of you have a UMIK-1/2 at home anyway, and Audacity 2.2.2 is free, we can give it a try. You need the older version of Audacity for the correction spreadsheet here:

Audio recording through calibrated UMIK-1?

I would like to do audio recordings with my UMIK-1, but I need software which would accept the calibration file to be included to allow for the most linear frequency response. I can record through the UMIK-1 without any calibration, but the result is not really convincing as it doesn't sound...

www.minidsp.com

Not all recordings had the same harshness to me so it can't be the mic if the same one was used for all? It'll be interesting to see what the speakers were!

 

[GXAlan]

Not all recordings had the same harshness to me so it can't be the mic if the same one was used for all? It'll be interesting to see what the speakers were!

Same mic. Same laptop. Same volume calibration.

If you have voted, made a comment and promise not to reveal answers publicly before 50 votes, I can PM you the identities.

 

[Thomas_A]

I have voted. I think though these kind of tests must be done in the same environment and with binaural recordings.

 

[GXAlan]

I have voted. I think though these kind of tests must be done in the same environment and with binaural recordings.

The different environment is because you are listening to different systems and environments.

I agree with binaural recording — do you have recommendations? Ambisonic mic that converts to binaural? MiniDSP HATS as a mic? Amazon random headphones you wear on your ears and use a basic laptop mic-in?

I prefer something with an internal ADC or a mix with a battery for a line level output

 

[Thomas_A]

The different environment is because you are listening to different systems and environments.

I agree with binaural recording — do you have recommendations? Ambisonic mic that converts to binaural? MiniDSP HATS as a mic? Amazon random headphones you wear on your ears and use a basic laptop mic-in?

I prefer something with an internal ADC or a mix with a battery for a line level output

I used in-ears. I think it works ok. Main problem is to eq the headphones to binaural curve which means bass correction. One example:

Danny's ASR "Will you ACCEPT my Challenge?!"

What it supposed to prove? I'm the one missing something here? I mean even in the audiophiles realm it's extremely bizarre and pointless.

www.audiosciencereview.com

 

[Blumlein 88]

The different environment is because you are listening to different systems and environments.

I agree with binaural recording — do you have recommendations? Ambisonic mic that converts to binaural? MiniDSP HATS as a mic? Amazon random headphones you wear on your ears and use a basic laptop mic-in?

I prefer something with an internal ADC or a mix with a battery for a line level output

KEF LS50 (David) Versus JBL 4722 Cinema (Goliath) Speaker Comparison with Binaural Recordings

Here are two extreme opposite loudspeakers when it comes to sensitivity, size, directivity, topology and looks. Does the big speaker sound perceptually larger than the smaller speaker? Do the speakers sound the same if eq’d the same and level matched? Does the LS50 sound perceptually bigger when ...

audiophilestyle.com

Read this including the in ear mics he uses. It covers details of room eq, but embedded in it is a wealth of information about binaural recording for your purposes.

 

[dasdoing]

Bose 901 Series VI

Spoiler

wow. that explains my "around the corner" impression of A. lol

 

[badspeakerdesigner]

Aight. DT770 Headphones, didn't read anyones spoilers.

Spoiler: my thoughts

E, A, and F are my preferred with F beating E and A a bit but I cant only vote for two.

D sounds like controlled dispersion or different room, stands out as being much less reflective, sounds like it would be much better in the room vs this mic. Comes off nasally though.

F seems to most balanced. The section I ended up focusing on was around 7-8 seconds in, those lower piano notes, it renders them the best. F is the only one that sounds right to me at that section

E is good, not much complaint there, maybe a tad bright.

C is ok but bright (is this da focal?)

B is just too dark.

A pretty good. It does the intro the best IMO. I had to change my vote after I kept listening and started to like A.

 

[dasdoing]

live

 

[GXAlan]

I will post identities next week. There hasn’t been much in the way of new votes, so I will give it a full month and then delete the recordings.

 

[GXAlan]

Answer Key
A - JBL 708P in family room
B - Bose 901 ‘68 in office
C - Bose 901 Series VI in family room.
D - BMW 640i GT, driver’s seat
E - Onkyo D-TK10 in office
F - JBL XPL90 in office