Here's an update on what was achieved this week so far;
(I will never be able to inject the enthusiasm I had in my lost post because then right after the fact)
Monday I started thinking more towards the ringing of filters, knowing that passive loudspeaker filters/XOvers ring just the same (post, IIR). So, went back to the DSP module.
But actual reason for that was to easily apply a needed notch because a self resonance frequency of the baffle. And notice : in this version of the Orelino the drivers are mounted in the baffle, which is NOT the idea for the final version (drivers will shine-through only, like the Big Orelo).
Again the bass and mid-high where aligned by means of phase (and not impulse), but, with my now "known" technique how to do that, I also applied the time alignment (since I was in DSP mode anyway);
I had slept a night on how to improve the alignment technique as a whole, and thought that very precise alignment would be possible now.
But my few cm of headroom on shifting the horn was sufficient ?
Haha, yes. With normal alignment maybe. But not with what I had dreamt about.
Of course I'll keep the means to myself, but I am now capable of creating a 100% XOver. Thus, prove that it's correct. I don't think anyone can do that.
Example from yesterday (and now in reverse mode making this consistent with my earlier posts about "this" subject) :
I wanted to lower the low end of the mid-high horn by 1dB (remember, that's passively filtered anyway), and as I told in that earlier post, this is not allowed without changing the XOver. My special means shows that in less than half of a second, and getting it right again takes maybe 10 minutes. This is about the slopes on both sides (24dB/Octave in this case), also knowing that when only one side is lowered in output (like my 1dB) the slopes won't match (for the alignment) anymore either. This too can just be checked in a second as well. So, change that when needed and together with the phase alignment with as means the time alignment and all is OK again.
Remember, the big change compared to the first post in this topic is about the now
precise alignment. This is the most different from running a sweep or all-frequency noise normally used. You see nothing. Well, relatively. So actually we hope to get close, no, think we do, because the severe wobbling plot "now" looks the best (through the software used). But of course there is a reason to use wideband drivers because they don't need a devistating XOver ...
Haha, listen to this ...
In that first post I talked about the "free sound of open baffle". Oh, is that so ? Well, apparently I didn't know sh*t, but I'm pretty sure I am no different than anyone elsewhere. And no, those with a single (wideband) driver also don't know, because it's all about the bass.
Oh, didn't I tell you ?
The Orelino now goes straight to 18Hz. (+/- 1dB in the region up to 100Hz).
17Hz is feasable too, but I just stopped (hey, the illegal boosting
)
100dBSPL continuous is still a piece of cake (3-4mm of excursion).
The soundFirst of all : No standing waves anywhere no matter how low it goes (there's my hobby horse again). Orelino's are 65cm (2ft) from the back wall.
Next, no coloration anywhere. And on this subject, heck, what difficult that is. So, this is how I had to go back to the DSP, because any slidest resonance in something and you're gone. Also notice that this can't be checked with normal loudspaker measurement means because they measure the frequency involved (like e.g. 120Hz at some stage) and not the resonating frequencies which can spring from it, like 60Hz. So, play 120Hz and perceive 60Hz ? haha, may sound as a gag, but that really colors. So what I did was hooking up an FFT analyser and watched all music playing for hours and hours (and more). Only then you learn where issues are and that what you hear must be fake (because too often occurring as seen through the analyser = ... not neutral.
When this "100% filter" was first done on Monday, I could not believe what I heard. So, listened to the speaker for more than a week already, was the most enthusiast already, but THIS ? How to explain it to you. I think I can do that in a kind of technical way :
Envision this slow blues electric bass. The guy (with hat or not) plays his "chords". Well, not real chords, but the melody. This goes from the one hey to the other, right ?
WRONG. Not anymore !
It now seems that no bass player plays like that. And maybe I also found the reason why the bass player at that concert always sounds better at that concert than at home through the "stereo". And I hinted to this long ago : we should have a separate speaker for that bass player in our set. The why was beyond me, but it felt like that.
Not so anymore;
What happens is that any XOver normally "jumps" from the one key (frequency) to the other. Now think of the player moving the string between the frets (also without frets being there of course). So, the string is stretched and the frequency goes up somewhat. Also, such a string sound is always more squary (at least far from a sine) which means that higher frequencies are involved as well (harmonics). This by itself implies that wherever the XOver is (285Hz in our case) you will be playing right *through* that. Mind the through ...
When the frequency graduately changes, they will cancel out "step by step" and the net effect is that you don't perceive the graduate change at all. Well, that is my not-so-scientific reasoning for it. And, instead you just hear the note played die out until the next note is hit (on the string). Not bad when you don't know about the difference (or "reality").
Next thing to tell is that now all basses snorr (my Dutch expression about snorring cats). "Yea, but my basses snorr too" I hear you say, and well, mine did too. But only the lower key ones. And again long ago I told about my waiting (for SQ improvement) until cello's doing that, which in my vision was possible. Ha !
I can now hear that the "snorring" is actually implied by higher frequencies (think about the squares again) and not by low frequency sines at all. So, when low enough (think 30Hz) a pure sine will exhibit similar, but hardly any bass plays that low, so it is not about that in the first place. It again is about how the Xover destroys, and actually makes anything of it, which is vaguening. Also, even a 24dB XOver is rather wide (spreads something like 260Hz around my 285Hz mid point) and a lot can get lost there. Think about a 80Hz more square sound, which has it's third (needed for "square") harmonic at 320Hz. All right in the XOver area ... (and 80Hz plays easily through a bass instrument).
Next thing which clearly occurs - and what I also never perceived before - is how the bass player slowly releases his (chord playing) finger from the string. So, while I talked about dying out of the sound because of the XOver while graduately changing frequency, the other way around happens the same : there is mucho variation in how the player releases the string. This by itself is "enabled" by again those harmonics (higher frequencies) now playing (the snorring actually) which are so easy to hear dying out when they do. I mean, when not there, how to hear them die out, right ?
So there's a whole freaking new dimension added.
And it is seriously freaking, because you won't recognize your favorites anymore.
Sub lowTalking about freaking ... what about frightening;
So, I use(d !) two subwoofers. Can do 12Hz blabla.
Oh, is that so.
Well, no, that is not so. They create low sound. They can indeed create that blush of air if you want. Can shake the room or even your trousers. But they do not play music.
I've had similar stories about my ambient favorites and how they are hard to recognize after a SQ change. But this ?
My "Demo Gallery" is full with that stuff. NOTHING can be recognized anymore. I now see that all this sh*t is full of sub-low I did not know about. I now see that all that stuff seemed to have its "melody" in the higher regions, while it's actually the roar. It is totally unbelievable.
Been to a good movie with great surrounding sound lately ? this does just that;
It is not about adding a blush of air anymore. It is about sheer 3D (??) left/right pressuring the room with all kind of events which makes you think you are in the middle of a nasty thing happening soon. Can I call it orgasmic ?
And still this is open baffle (based). Our Bert applied some nice efficiency principles (meant to supercede the Big Orelo), and they work better than the man expected (and heard) himself. See you tonight Bert !
Oh, finally finished with undoubtedly again too long post, I suddenly recall how I started out the accidentally deleted one :
My (woman) partner in crime's first remark was that the electric guitars sounds so beautiful. An electric guitar sounding beautiful eh ? Women ...
Yesterday she repeated that and added to it that all woman voices now can go easily. Hmm ...
But it is true, there is a purity unheard in it all.
And on the woman voice ... Take Emma Shaplin. Ever back I used a specific track of her for testing. Played it often. And yesterday there was a completely profound synthesizer playing along ? Never heard that. But this is how albums completely change ...
Ok, done. Best speaker ever I dare say, although it is still a proto. Right after the X-fi show it will reach its final stage (within a week) and it will only get better of that.
What you didn't know is that I have been doing this for myself, being the first on the order list (well, for a different but similar version which is a
). I guess we'll reveal that at the Show. Ah, right. Ok, I did it for the Show.
No ... it was for Bert.
For both of us ?
For you ?
Peter
PS: If you see how a speaker like this can be changed from one day to the other and from "best" to "outragious", you should also be able to see how we (or customers) can tweak for the better in the DSP department to next share the configuration files amongst all. This is also how it's crucial to NOT apply notches (or boosts) for room correction, which I did not apply anywhere (and which I claim not to be needed). So, applying a new "filter" is a matter of 5 seconds (plugging in the USB cable for it takes the most of that time).
