The principle is not sound (pun intended I guess...). The implementation is incredibly complicated and expensive, too.
Has a narrow band of frequencies reproduced by each driver, and thinks this will act like a Fourier Transform where there is one "bin" per frequency. Then mistakenly uses nude dipole drivers, which have a null on axis. Kind of a dumb implementation of a flawed concept...
But sure, they might sound great, just like a 4 inch fullranger in a huge horn might sound great, etc.
I will need to watch this video 3 or 4 more times before I begin to understand what is happening here (or maybe 30 to 40 times!) I wonder if @mcargill has seen these. They look like something that he might want to experiment with. In the video, they mentioned an earlier prototype version that looked really strange. Does anyone have a link to this early version?
Charlie, the prototype had 20 drivers. As I partially understand it, each driver would be playing a passband limited sine wave, convoluted into a sine wave via DSP from the original music waveform. Then the 20 band limited sine waves would sum back together again at the listeners ears to recreate the original waveform, since any complex waveform can be broken down into large number of sine waves. But this would require an almost infinite number of drivers to recreate the original waveform. How can they do this with only 20 drivers? They showed some graphs showing input verses output FR measurements to prove that it actually works. I would certainly like to hear these.
I'd be unhappy with the bass from a pair of 10s after paying for a 6-pack of 12s. Not that this is a budget solution with forty drivers & amp channels!
More fundamentally, you might be able to optimize the crossovers for a single point in space, but there must be phase errors at other locations. Using narrow passbands per driver won't automatically eliminate phase issues between the little mid on top & the drivers closer to the floor.
The explanation seemed like a long winded way of saying he made some 20-way speakers. The mounting was probably an interesting looking compromise to fit that many drivers in a small enough space to maintain some semblance of vertical alignment.
I definitely like the outside the box approach that this guy is going at. It's stuff like that that interests me. Sure the myriad of basic boxes with multiple drivers can and do sound great, but doing something different to get the same or next level performance is great for the hobby.
It's pretty much why I'll never build a regular box anymore. I'd rather have something that's a bit artistic AND sounds great. Dan's latest build is exactly that. Not everyone will find it attractive, but to those that do, it's that much better.
I hope the best for that guy and his creation. It's bad enough having to create around the fact of being in a hotel room, once out of that and in something a bit more conducive to audio, it'll work out even better. Then again, maybe all the drivers need to be rearranged🤷🏻♂️ That's the path we take to discover.
Comments
Wanted to hear them.
Although I'd like to hear them, how weird is that? Be fun to DIY something like that.
CoMplICated SoftWaRE
I don't understand the words he puts together. Maybe if I separated the words and ran them through a DSP and arranged them vertically.
The principle is not sound (pun intended I guess...). The implementation is incredibly complicated and expensive, too.
Has a narrow band of frequencies reproduced by each driver, and thinks this will act like a Fourier Transform where there is one "bin" per frequency. Then mistakenly uses nude dipole drivers, which have a null on axis. Kind of a dumb implementation of a flawed concept...
But sure, they might sound great, just like a 4 inch fullranger in a huge horn might sound great, etc.
I will need to watch this video 3 or 4 more times before I begin to understand what is happening here (or maybe 30 to 40 times!) I wonder if @mcargill has seen these. They look like something that he might want to experiment with. In the video, they mentioned an earlier prototype version that looked really strange. Does anyone have a link to this early version?
Charlie, the prototype had 20 drivers. As I partially understand it, each driver would be playing a passband limited sine wave, convoluted into a sine wave via DSP from the original music waveform. Then the 20 band limited sine waves would sum back together again at the listeners ears to recreate the original waveform, since any complex waveform can be broken down into large number of sine waves. But this would require an almost infinite number of drivers to recreate the original waveform. How can they do this with only 20 drivers? They showed some graphs showing input verses output FR measurements to prove that it actually works. I would certainly like to hear these.
I'd be unhappy with the bass from a pair of 10s after paying for a 6-pack of 12s. Not that this is a budget solution with forty drivers & amp channels!
More fundamentally, you might be able to optimize the crossovers for a single point in space, but there must be phase errors at other locations. Using narrow passbands per driver won't automatically eliminate phase issues between the little mid on top & the drivers closer to the floor.
I would think dispersion would be a mess with all those drivers stacked on top of each other.
In theory, the dispersion would be 360° or omni, which is good.
The stacked-nude-omni approach though is going to be rife with cancellations.
InDIYana Event Website
The explanation seemed like a long winded way of saying he made some 20-way speakers. The mounting was probably an interesting looking compromise to fit that many drivers in a small enough space to maintain some semblance of vertical alignment.
Where's the video of them actually listening to that heap?
Notice none of the planars on the side are mounted flush....
I definitely like the outside the box approach that this guy is going at. It's stuff like that that interests me. Sure the myriad of basic boxes with multiple drivers can and do sound great, but doing something different to get the same or next level performance is great for the hobby.
It's pretty much why I'll never build a regular box anymore. I'd rather have something that's a bit artistic AND sounds great. Dan's latest build is exactly that. Not everyone will find it attractive, but to those that do, it's that much better.
I hope the best for that guy and his creation. It's bad enough having to create around the fact of being in a hotel room, once out of that and in something a bit more conducive to audio, it'll work out even better. Then again, maybe all the drivers need to be rearranged🤷🏻♂️ That's the path we take to discover.
Hell, I have trouble getting one crossover point to sound right. I can't imagine dealing with 20!