Thanks - I’ll do some more listening when I get home as I’m traveling currently. I have to say though that the difference was not subtle at first listen so not certain why given your numbers.
I’d think the speaker membrane would be primarily vibrating at the loudest / highest excursion waveform then having that excursion be modulated +/- by the additional waveform needs?
I’ll slow down on the posts - sorry everyone but the square wave issue has also been discussed in context of audio elsewhere:
https://www.audiosciencereview.com/.../composition-of-a-square-wave-important.1921/
https://www.audiosciencereview.com/.../composition-of-a-square-wave-important.1921/
Yes, this is mixing. Modulating is something very different and I often see the two mixed up.
Digital theory says that you only need twice the frequency to represent the analogue equivalent. Despite any additional thoughts on the success of this, the point is that there is no mention of it needing to be different with a complex waveform.
As for the ability of the amp to slew between them, the slew rate has room to spare. Merely turning up the level on a sine wave calls for an increase in the slew rate without changing the bandwidth needed.
Agree re: the slew rate for amplitude, and I understand regarding the digital sampling needing to be 2x the frequency it’s trying to pass.
My point in all of this is I think the bandwidth requirements are substantially higher than we expect given most think 20khz is all that’s required as that’s what our ears can detect. Again my example of two sine waves representing two sound sources you’re trying to reproduce. If they are equal in frequency and separated by 90 degrees then you’d need twice the bandwidth to represent that signal than the base frequencies. Ie two 12khz waves separated by 90degrees would need 24khz signal bandwidth, but each source would be audible to us.
Also bandwidth is often cited as -3db bandwidth, so in order to faithfully reproduce you’d need to be in a linear area of the bandpass region, requiring much more headroom vs the quoted bandwidth.
My point in all of this is I think the bandwidth requirements are substantially higher than we expect given most think 20khz is all that’s required as that’s what our ears can detect. Again my example of two sine waves representing two sound sources you’re trying to reproduce. If they are equal in frequency and separated by 90 degrees then you’d need twice the bandwidth to represent that signal than the base frequencies. Ie two 12khz waves separated by 90degrees would need 24khz signal bandwidth, but each source would be audible to us.
Also bandwidth is often cited as -3db bandwidth, so in order to faithfully reproduce you’d need to be in a linear area of the bandpass region, requiring much more headroom vs the quoted bandwidth.
If you mix two sinewaves, one at 0 degrees and one at 90 degrees, you get a sinewave at 45 degrees.
Okay maybe my understanding of the resulting waves is flawed. But say if you mix the aforementioned sine wave plus a 5khz one plus a 15khz one plus a 10khz one, etc, I imagine you start getting something quite complex with multiple waves overlapping creating a fairly complex waveform to pass electrically.
There must be a reason for higher than 44khz digital sampling - many of the high def audio tracks I’m listening to are 96khz so could represent base frequencies of up 48khz in analog. Not sure why they would bother if 20khz is enough to pass the analog electrical signal faithfully.
There must be a reason for higher than 44khz digital sampling - many of the high def audio tracks I’m listening to are 96khz so could represent base frequencies of up 48khz in analog. Not sure why they would bother if 20khz is enough to pass the analog electrical signal faithfully.
This explains why:
https://usa.yamaha.com/products/con...ng_support/micro_tutorial/20170629/index.html
"In 1982, A/D and D/A convertors used analogue anti-aliasing filters to make things work. These filters had to be extremely steep to apply a brick wall slope between 20kHz and 22.2 kHz with difficult to hide artifacts, already starting the debate on using a higher sample rate. The issue was soon resolved by using oversampling and digital filter techniques.
Frequency Range.
Most broadcast and live systems today use a sampling frequency of 48kHz, this being slightly less demanding on the anti-aliasing filter. Using a higher sampling frequency of 96kHz extends the audio reproduction frequency range from 20kHz to to 40kHz. However, the human hearing range really ends at around 20kHz... the highest hair cell in the human ear's cochlea is tuned to about that frequency with a very narrow bandwidth and there are no hair cells listening to higher frequencies. So why do we need 96kHz?
Timing resolution.
The answer is timing resolution. For continuous signals, a 48kHz system is perfectly capable of reproducing time relationships with a very high accuracy. But when it comes to the start and end of audio signals, the timing resolution breaks down to about 21 microseconds - the reciprocal of the sampling rate. The challenge: the human hearing system is actually capable of detecting smaller time differences then 21 microseconds (known as dichotic difference). For example, experienced listeners, in ideal listening conditions, have been reported to be able to localise sounds with accuracy of less than five degrees. Since localisation is partially dependent on the arrival time difference between left and right ears, taking an average head size of 20cm sets the time difference between left and right for a sound coming from the side (90 degrees) at about half a millisecond. Five degrees then accounts for about six microseconds.
To be able to capture this resolution, a sampling rate of 192kHz would do the trick. However, most listening is done in less controlled circumstances such as a home living room, a hotel room, in a car or at a rock concert with a large PA system and thousands of fellow-listeners. In those cases, a 21 microseconds resolution may be more than enough - corresponding to 48kHz. Only for controlled listening situations - with perfect acoustics, perfect speakers and a single listener in the perfect sweet spot, is 96kHz worth going for."
Interesting, thanks! So in order to maintain the perceived spatial resolution due to timing differences and allow the edges of the waveforms to be onset/offset sufficiently you need a higher sampling rate for the waveform. So this all would need to be reliably passed in the analog realm too by the available analogue bandwidth. Is that a reasonable interpretation?
Sounds reasonable.
That said, doubling or quadrupling storage space for something that could only perceived by a select portion of listeners in a very controlled listening environment (like wearing headphones in a dead quiet room) listening to music that was recorded in a manner that would highlight spatial resolution less than 21 microseconds may not be a reasonable pursuit.
That said, doubling or quadrupling storage space for something that could only perceived by a select portion of listeners in a very controlled listening environment (like wearing headphones in a dead quiet room) listening to music that was recorded in a manner that would highlight spatial resolution less than 21 microseconds may not be a reasonable pursuit.
Yeah fair, unless maybe keeping the timing between different waves in the signal faithful in the signal chain also then results in maybe being able to discriminate individual instruments and their positions from each other too? If you lose that timing separation between onset of waveforms then they’d overlap more on the adjacent ones. Not sure if that would make it more difficult or easier for a speaker membrane to reproduce.
Considering the decay time of even the best tweeters extend into the millisecond range, differences of microseconds in the recording are not going to be in the range of perception.
Hey Joe. So somewhat recently, you were really intrigued by the sound of the NBAC with the Miflex when compared to the ULD. If one has the power to drive a NBAC Elsinore, does the NBAC seem to close the gap between the MFC and the ULD so long as you have the power to push it? Did you end up putting a Miflex cap in the ULD versions which pushed it ahead again? I have heard Ribbon drivers and do like them.
I am finally getting myself in gear to build the your NRX version... I have sourced the 1" material and have all the parts to get going. I had thought to myself, if I like the NRX quite a bit, what should be the next step. I also agree with you and others that a Satori version could be very interesting. I hope that you get a chance to explore that. The MW16P looks like a really nice driver.
I am finally getting myself in gear to build the your NRX version... I have sourced the 1" material and have all the parts to get going. I had thought to myself, if I like the NRX quite a bit, what should be the next step. I also agree with you and others that a Satori version could be very interesting. I hope that you get a chance to explore that. The MW16P looks like a really nice driver.
Another question. It looks like the purify uses a 178mm opening and 4mm deep recess. I can always make a 3mm shim for the recess. Would it be detrimental to make the openeing 7" (178mm) and install the SB17 drivers? This will make for a 3.5mm gap around the parameter of the SB17 instead of a 1.5mm gap. I am thinking of doing this to future proof the design a bit.
Mikerodrig27,
Maybe a removable black, silver or gold color "band" around the driver frame? Can make from MDF or something softer like PVC wood. Can be laser cut too.
Will be quite flimsy though...you will have to glue it to the cutout with an adhesive. Probably will break when you decide to install the Purifi drivers.
Maybe a removable black, silver or gold color "band" around the driver frame? Can make from MDF or something softer like PVC wood. Can be laser cut too.
Will be quite flimsy though...you will have to glue it to the cutout with an adhesive. Probably will break when you decide to install the Purifi drivers.
Last edited:
I had thought about that. Being that the difference is 2mm or .08", it would be hard to machine something like that. I do have veneer that is probably about that thick. I suppose I could always just stuff it in the gap or maybe use some kind of discreet putty that could be easily removed later on.
Yes, that could work. It's used to fill gaps around waveguide throats.