What distance is nearfield?
-
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!
Sigberg Audio Manta Cardioid active speakers: Full measurements (Spinorama)
- Thread starter sigbergaudio
- Start date
OP
- Thread Starter
- #42
Yes, to be even more precise, the goal isn't "to keep the frequency response as flat as possible", the goal is balanced, natural in-room sound.
Flat / smooth anechoic on-axis response is a good starting point towards that goal, but may not always be the right answer. Another explicit deviation in our speakers is a slightly increased midbass. This too looks "wrong" in an anechoic chamber, but translates to a balanced sound in most rooms.
OP
- Thread Starter
- #43
This specific measurement was made at ~40cm.
Sign me confused … I thought perfectly flat FR was always the target, but yet now we say only sometimes? It depends? Either perfectly flat FR is the ultimate target for what we here on ASR consider a “correct” design, or it isn’t. Which is it?
Not trying to be confrontational, I truly am struggling a bit with this seeming inconsistency vs what ASR is all about. Certainly the crowd here has lambasted other designs for less.
And I should add I REALLY appreciate and respect your transparency here @sigbergaudio, truly above and beyond and very admirable indeed.
This has been discussed several threads already. It is clear that listening to a single speaker in the zero degree axis should have flat frequency response. Also when speakers are setup up in multichannel mode with center speaker + R/L speakers. Entering stereo, you have a flawed system with errors. There are "tricks" to make the center phantom image a bit more like a center speaker and is related mostly to the errors in the 1-5 kHz range. This should be taken care both on-axis but especially off-axis (early reflections). Early reflections (side reflections) should *never* have a dip in the 1-2 kHz range, rather a small boost compared to the 3-4 kHz range. I've done many years listening and found the same thing. Making sure that there is a bit more energy 1-2 kHz (now read this is most important off-axis) and a small dip 3-4 kHz sound smooth and natural IMO.
ernestcarl
Major Contributor
I am aware, but I have heard many designs that are flat and still have very good on/off axis. My comments were not a “challenge” rather I genuinely am trying to better understand why this trade off was made here when it wasn’t necessary in other designs.
Seems to be the case with the Mantas off-axis
OP
- Thread Starter
- #49
That is a real problem with the increased focus on measurements (there are of course many good things as well). To understand a full set of CEA2034 measurements and what it actually translates to with regards to in-room sound is pretty complicated. I would go as far as to say that most people in here should probably focus on the estimated in-room response and the listening window graphs, and try not to judge the speakers on the rest of the graphs, because they're not in a position to understand what they mean.
I've had loooooong discussions on this topic, and the two sentence summary of my stance on this is really the same I mentioned above: Flat anechoic response can never be the end goal. The end goal is that the speaker sounds right and balanced to the consumer, in as many listening situations as possible.
It's generally accepted that a speaker should be flat anechoically, and that is the only correct response. This makes logical sense, we want all of our other components like amps and dacs etc to be perfectly flat. Why not the speakers? It would be very neat and tidy if this was the case, but in reality it's a bit more complicated than that. The design and directivity of the speakers affect what works.
And personally I think especially the midbass is simply wrong in way too many speakers. Due to our listening environments and how we typically place our speakers, a flat anechoic response will result in too little midbass - SBIR effects (reflections / cancellations from the wall) eats up that range, and it requires a bit of extra level to compensate. Kef is one of the few measurement focused manufacturers I know of who does this at least on some models, and have confirmed that they do this intentionally because it simply sounds more accurate.
...Then you have other manufacturers that you will find on the top of the Spinorama ranking that are very focused on a flat on-axis response. You say they didn't need to make this trade off. An alternative interpretation is that they have chosen to focus harder on on-axis response than off-axis response. Every manufacturer have their own design goals and their own understanding of what is "right" or works best.
Personally I disagree with for instance how Genelec tune their speakers and their room correction system. The engineering is fantastic and they're an amazing company. But to my ears, the in-room result is usually too bright. Right or wrong, that's not how I want our speakers to sound, and I don't think it sounds right.
Even worse, many expensive, high-end speakers have elevated highs and inconsistent midrange that sounds harsh and anything but balanced in-room. We've ended up with this hunt for "resolution" and "revealing" speakers. Real instruments (for the most part) don't sound bright and hard.
Finally, and interestingly, this is not a perfect and easily explained science. Our own SBS.1 speakers have cleaner measurements and a higher preference score. But I don't think you will find anyone who wouldn't prefer the Mantas based on a listening test. From another perspective, they sound remarkably similar in-room despite clear differences in the measurements.
Thank you, it's a privilege to have access to interested and knowledgable consumers and be able to share and discuss our designs and the pilosophy behind them! Part of the reason for posting the Manta graphs was also exactly in the hope of getting this somewhat more nuanced discussion around measurements, because it is an interesting one.
Sokel
Master Contributor
- Joined
- Sep 8, 2021
- Messages
- 7,154
- Likes
- 7,366
Genelecs-only preference,80xx series by far for me and I'm not the only one.
Perfectly sloped in room,the most natural Genelecs to me.
(now eat me alive
)
OP
- Thread Starter
- #51
Genelecs-only preference,80xx series by far for me and I'm not the only one.
Perfectly sloped in room,the most natural Genelecs to me.
View attachment 323205View attachment 323206
(now eat me alive
Several models have good, natural in-room slopes. But at least with their earlier GLM room correction software, the software "fixed it" and made it too flat. Luckily with newer versions it is more conservative, and you can also adjust what it does.
Sokel
Master Contributor
- Joined
- Sep 8, 2021
- Messages
- 7,154
- Likes
- 7,366
It's raw performance's fall before 2Khz and the further dip until 6Khz (early reflections) is ingenious for people with my condition (round headache if that area is pronounced).Several models have good, natural in-room slopes. But at least with their earlier GLM room correction software, the software "fixed it" and made it too flat. Luckily with newer versions it is more conservative, and you can also adjust what it does.
To adjust them to taste after that (and RC) is a plus.
OP
- Thread Starter
- #53
I think you've hit on a very key point here, @sigbergaudio .
It is too easy to lose sight of this as the most important part of the equation. I am all for seeing the measurements, and frankly feel the more the better! Yet in that, one still has to account for the fact that a perfectly flat Speaker FR doesn't necessarily guarantee the best SQ. It is one thing to strive towards excellence in engineering while neglecting the usability of the finished product. The best designs strike a balance. This understanding extends to all things.
- Joined
- Jan 15, 2020
- Messages
- 7,026
- Likes
- 17,582
Research has shown that most listeners prefer direct flat sound combined with smooth radiation and directivity. In real life the second usually is very hard to achieve so some compromise in the direct sound is done to avoid for example sound power peaks in some regions due to the non-perfect directivity.That is a real problem with the increased focus on measurements (there are of course many good things as well). To understand a full set of CEA2034 measurements and what it actually translates to with regards to in-room sound is pretty complicated. I would go as far as to say that most people in here should probably focus on the estimated in-room response and the listening window graphs, and try not to judge the speakers on the rest of the graphs, because they're not in a position to understand what they mean.
I've had loooooong discussions on this topic, and the two sentence summary of my stance on this is really the same I mentioned above: Flat anechoic response can never be the end goal. The end goal is that the speaker sounds right and balanced to the consumer, in as many listening situations as possible.
It's generally accepted that a speaker should be flat anechoically, and that is the only correct response. This makes logical sense, we want all of our other components like amps and dacs etc to be perfectly flat. Why not the speakers? It would be very neat and tidy if this was the case, but in reality it's a bit more complicated than that. The design and directivity of the speakers affect what works.
And personally I think especially the midbass is simply wrong in way too many speakers. Due to our listening environments and how we typically place our speakers, a flat anechoic response will result in too little midbass - SBIR effects (reflections / cancellations from the wall) eats up that range, and it requires a bit of extra level to compensate. Kef is one of the few measurement focused manufacturers I know of who does this at least on some models, and have confirmed that they do this intentionally because it simply sounds more accurate.
...Then you have other manufacturers that you will find on the top of the Spinorama ranking that are very focused on a flat on-axis response. You say they didn't need to make this trade off. An alternative interpretation is that they have chosen to focus harder on on-axis response than off-axis response. Every manufacturer have their own design goals and their own understanding of what is "right" or works best.
Personally I disagree with for instance how Genelec tune their speakers and their room correction system. The engineering is fantastic and they're an amazing company. But to my ears, the in-room result is usually too bright. Right or wrong, that's not how I want our speakers to sound, and I don't think it sounds right.
Even worse, many expensive, high-end speakers have elevated highs and inconsistent midrange that sounds harsh and anything but balanced in-room. We've ended up with this hunt for "resolution" and "revealing" speakers. Real instruments (for the most part) don't sound bright and hard.
Finally, and interestingly, this is not a perfect and easily explained science. Our own SBS.1 speakers have cleaner measurements and a higher preference score. But I don't think you will find anyone who wouldn't prefer the Mantas based on a listening test. From another perspective, they sound remarkably similar in-room despite clear differences in the measurements.
Thank you, it's a privilege to have access to interested and knowledgable consumers and be able to share and discuss our designs and the pilosophy behind them! Part of the reason for posting the Manta graphs was also exactly in the hope of getting this somewhat more nuanced discussion around measurements, because it is an interesting one.
Below are some quotes from Toole about this topic as they are the best descriptors of the ideal case and its restrictions:
"Almost 50 years of double-blind listening tests have shown persuasively that listeners like loudspeakers with flat, smooth, anechoic on-axis and listening-window frequency responses. Those with smoothly changing or relatively constant directivity do best.
...
in double blind tests in normally reflective rooms (different ones over the years) listeners give the highest ratings to loudspeakers that measure essentially flat and smooth on axis, and at least smooth off axis in an anechoic chamber or functional equivalent. What they are recognizing and responding favorably to is the absence of resonances - i.e. neutrality.
...
Achieving a desirable flat on-axis sound using passive or active networks can result in non-flat off-axis behavior because transducers have frequency-dependent directivity. In a room the result is that even with flat direct sound, the early reflected and later reflected sounds may exhibit emphasis over a range of frequencies that could forgivably be interpreted as a low-Q resonance."
Then also there are the issues of individual room acoustics (although humans tend to hear to a big part tonally through them), individual hearing problems (which I think is the main reason many high end loudspeakers have boosted treble) and unfortunately strongly variating quality of recordings (see https://seanolive.blogspot.com/2009/10/audios-circle-of-confusion.html ) Due to all of these I personally hope that studio monitors will be all standardised in terms of flatness and directivity but for home hifi loudspeakers I just recommend buyers to get what they enjoy most, although my own limited experience confirms that most people prefer loudspeakers which measure as well and for poor recordings and individual taste still EQ is more expedient to use like Toole also recommends.
OP
- Thread Starter
- #56
Yes, I am familiar with the paper.
Colonel7
Addicted to Fun and Learning
Yes. In my experience they should sound smooth in-room in stereo. I have not heard them but my guess is you can crank them up to high volumes and they will still sound smooth and natural.
I basically agree with Toole but as he also say ”stereo is a flawed system so use multichannel instead ( where the center channel takes care of the flaw)”. One have to define the end goal. Is it mono , stereo or multichannel we are optimizing our speakers for? That said any ”corrections” may be within +/- 1.5 dB which is anyhow quite linear to start with.
OP
- Thread Starter
- #60
Just to put this dip (and the overall response) a bit into perspective as well with regards to linearity:
At 20 degrees, which is closer to a typical listening angle than on-axis, the difference between the peaks at 1khz and 1.8khz vs the dip at 1.3khz, is a total of 2.4db. So the response across the dip is +/-1.2dB.
Overall frequency response deviation as measured by spinorama.org (300hz-5khz) is +/-2.5dB on-axis.
Overall frequency response deviation as measured by spionorama.org (300hz-5khz) is +/-1.9dB at 20 degrees.
At 20 degrees, which is closer to a typical listening angle than on-axis, the difference between the peaks at 1khz and 1.8khz vs the dip at 1.3khz, is a total of 2.4db. So the response across the dip is +/-1.2dB.
Overall frequency response deviation as measured by spinorama.org (300hz-5khz) is +/-2.5dB on-axis.
Overall frequency response deviation as measured by spionorama.org (300hz-5khz) is +/-1.9dB at 20 degrees.
Similar threads
