The making of: The Two Towers (a 25 driver Full Range line array)

Just to clarify things:
The flac file in the link is a studio take not a live session.
IMHO here you can imagine what it looked like when the musicians were actually in the studio.
I am not that fond of takes where acoustic instruments are recorded very lifelike, but placed in the soundscape where they cannot be .
But again, maybe it tells more about my poor brain...:cool:
https://www.dropbox.com/sh/jaklk983eqbby5i/AADwxtyxCEyhjnFdzlJwJrNPa?dl=0
 
Perceval.
Still I am a little surprised that you cannot find a Peter Gabriel with depth of field?

Try listen to Apr_s Moi from Scratch My Back just as an example.

Peter Gabriels acoustic is of course artificial so you cannot compare the two..
Koldby

No, I meant all my CDs of Peter are badly mastered and lifeless... I think I've been unlucky with the batch of commercial CDs I got, and I should try to find some high resolution tracks on the net.
By comparison, his DVD "Secret World" sounds great.

I will have a listen to Après Moi... I've not connected with his latest work so much... I'm more a 1-2-3-4 albums kind of fan...
 
Last edited:
Arrays and near field

I heard a rule of thumb that you need to sit at least 2x height of towers away from them. Otherwise the comb filtering will be overwhelming.

Hi. Do you mean that 2m is not nearfield, or that the towers are not ideally suitable for 2m listening?

Well after measuring the distance from my listening position to the midpoint, I would say 2 m is a little close depending of course on the distance between the speakers, but if the distance between also is 2 m I would say too close to the speakers.

Wow, I am sitting way too close to my arrays then. I am definitely in the Near field. But it is the near field sound I like: big headphones with big soundstage. This is one of the advantages of line arrays James Griffin mentions in his white paper:

http://www.audioroundtable.com/misc/nflawp.pdf

Whether or not You are in the near field is based on distance and frequency. The higher the frequency, the longer the distance You are in near field. The math for this is on page 7-8. James explains it better then I can.

Comb filtering does roll off the highs at closer listening distances, but I EQ for this and Ronald has gone a lot farther with other processing with his arrays. If I recall, his listening distance is about 2 meters, but my memory could have this wrong, I will let him chime in on this.

I will admit, one short coming of my NSB Array are the highs. Part of it is due to the highs of the drivers themselves. the highs of the NSB are pretty rough and it is something I can not fix with DSP. The Vifa does have smoother highs. Maybe the combing is making the problem worse with my current arrays, but I can not say for sure. The answer to this will be clearer once I build my Vifa array. This is also why I am interested in Peter's array, because he is using a line of planers for his highs which in theory has very little combing and combine as a true line source. (see the picture on page 15 of the white paper linked above.) It would be cool to compare the two designs, given the compromises of both.

dublin78, I would encourage You read James' paper. There seems to be mixed information on this topic.
 
Last edited:
Near field vs Far field

Another thing to bear in mind about listening distances with arrays:

In the Near field, SPL drops 3dB per doubling of distance.

In the Far field, SPL drops 6dB per doubling of distance.

As the high frequencies stay in the near field at longer distances, it is no wonder we can hear the highs "cut thru" better at longer listening distances. In essence, the low frequencies "roll-off" quicker then the highs at longer distances.

Line arrays do project well in large spaces, yet the benefits of near field listening in a smaller room is very special indeed. Some close miced recordings "bloom", which I do not mind, as they are artificial anyway. But movies and acoustical recordings have a very "true to life" scale to them.
 
Wesayso
Inspired of your flac upload, I got the Anne Bisson album and I have a couple of comments after listening to that.

As maybe will be known I also have Line Arrays with the same Vifa drivers as Wesayso.

These arrays are very good @ placing instruments and recreate the acoustic.
Now here comes my concern when I listen to a album like Anne Bisson Blue Mind. I know it has high ranks as an Audiophile album, but I have strong reservations concerning recordings like this.
I am very used to hearing well recorded classical/jazz/folk acoustic music and also rock with use of electric instruments (Peter Gabriel Pink Floyd aso.) and I can certainly appreciate the fact that an electrical guitar or synthesizer can be placed wherever the producer/artist wants it to be...But..

I start listening to Anne Bisson and her voice is very clear, very warm and well recorded. I must be very close to her with all these details and she must be facing me. Then the piano comes in. I know she should be playing it, but how can she do that. The piano (flygel it must be form the sound of it)
is clearly further away...
So we are sitting 2 meters or less from Anne , facing us, and she is playing a piano that sounds as if is 5 meters away .
And the comes the weirdest thing : A drummer joins the session! But where is he?
Sounds like the distance is about the same as the Flugel.. well that's ok but it must be behind the piano or Anne couldn't play on it. WOW this drummer guy must have REALLY long arms as one of the cymbals are to the left of the (pretty big) piano and another is placed to the right of the piano.
Could there be two drummers? No not just two because on cymbal pops up just in the middle of the piano also. I dont even want to start placing the bass player in all this mess.

Everything is well recorded and believable as separate instruments, but when it is obvious that it is a recording of acoustical instruments (and voice) in an acoustical environment, my brain cannot accept the result very easily.

Maybe this just tells a very sad story about my brain:eek:
Koldby

Hahaha, I did not pick that song for it's imaging properties but for the central voice part, with even a spoken word from Anne.

I've used September in Montreal as a test track and that one has some fun features about it in how it sounds. The piano is on the right side but you can actually follow the notes all the way to the (phantom) center. But in order for Anne to sit there, playing that piano and sing to us she would have to pull of a scene from the Exorcist.

That song also has separate drum tracks with cymbal rides way to the sides.
But the piano is "in a room". The first drum coming in from the left has a nice "reflection" going on, on the right "wall". But the rest of the drum kit, being more in the back of the room does not seem to encounter that wall at all. So yeah, it's a mixed bag in imaging.

I've used it for Anne's voice, the piano on the right, overall room size (as far as I could tell) and the low bass parts are interesting and quite direct. I've used that bass to judge timing of my arrays, as I did experience that particular bass part as "off" when I used complete linear phase filters.
I realised I needed to back off a bit and let my arrays follow minimum phase.
After that the timing locked pretty good. This doesn't mean linear phase would be dead wrong, just that if I implement that in a real room I might have some parts a bit too early.

It seems very little processing is done on the separate streams. But they do not match up in imaging. That's why I listen to parts of it and use that as a reference to judge other DSP setting with. Knowing what each separate instrument is doing. Plus the voice is very clear (but not challenging).

I have more recordings I use for partial features, but also some that do display a recording as if it were a real life band. I can listen to, and enjoy both (as long as tonality is right).

It doesn't always have to be strict in placement for me. For instance, my avatar makes it clear I do like Van Halen. Yet I'm not impressed by Ted Templeman's choices in imaging on the first 3 albums. It gets boring to always have the guitar far left, singer dead center and usually the same effects in reverb. Drums+bass in the middle (always) except a few more entertaining recordings like some of the more free and experimental guitar parts and "Everybody wants some" from "Women and Children First". But even that goes back to the "standard" just past one minute.
If you add to that the CD versions all have way too little low end on them compared to how I remember the songs on vinyl from my youth... except the DCC version of VH1, mastered by Steve Hofmann. So sad he didn't do all others too.

I guess we can't have it all :p. But at least the songs are still part of the soundtrack of my life, and as such I can enjoy them.
 
Last edited:
Comb filtering does roll off the highs at closer listening distances, but I EQ for this and Ronald has gone a lot farther with other processing with his arrays. If I recall, his listening distance is about 2 meters, but my memory could have this wrong, I will let him chime in on this.

I'm at 2.8 meter distance from each array (mic measurement position). The arrays are 2.8 meter apart. The actual listening position varies if I sit upright on the edge of the couch (mic position) or have a more relax position.

In critical mode I'm sitting upright where the mic is, leaning back (meaning closer to back wall) I do get a bit more bass, just a hair.

All the stuff like "2x height of towers away" does not really fly with floor to ceiling full range speakers i.m.h.o.

I'm in the near field for my listening spot, or at least I consider it like that. Especially if we apply the theory of "mirror images of floor and ceiling" like Don Keele.

Don't judge the the center to center spacing of the drivers only without also thinking a bit further. In this case the driver to ear distance is at least as important with all drivers playing the same material. I do not think a pure cylindrical wave front exists in these arrays as in the pure theoretical versions. Yet the drop of with increasing distance comes quite close to following the 3 dB/double distance rule of thumb.

Look at ra7's 3 ft, 6 ft and 12 ft measurements to see how 3 ft is way to close and the response gets better on the top end by increasing the listening distance to the arrays. There's already a world of difference between 3 ft and 6 ft.
He also proved moving up and down at the listening spot has little to no ill effects. This one is impressive though, comparing 6 ft to 12 ft and having the two only vary a few dB here and there (these are raw measurements):
525561d1453010905-corner-floor-ceiling-line-array-using-vifa-tc9-6ft_12ft_response.jpg

(image and measurement from ra7's corner arrays)

That image shows there's merit to using full range floor to ceiling line arrays at a relatively short listening distance (6 ft or 1.8 meter). Getting further away is moving the comb effect further up in frequency.
 
Hi Folks
I thought it wise to stop sidetracking this great line array thread with DSP stuff, so I have started a new thread "Off The Shelf DSP for an Old School System"

I do hope you get a few of the readers that followed this discussion here with you over to this new thread.
Here's a link: http://www.diyaudio.com/forums/digital-line-level/298154-off-shelf-dsp-old-school-system.html

I already see a few familiar names popping up. I do hope you're committing to making some measurements, I'd sure like to see what your setup does.
 
"In essence, the low frequencies "roll-off" quicker then the highs at longer distances."

This statement is reversed, high frequencies drop much faster with distance. In-room not a real issue but this is always something you have to deal with with outdoor live events.

I'm thinking that sentence was getting at the near field/far field transition of line arrays in particular. Arrays should be extremely long to truly act like a line source at low frequencies. A floor to ceiling line array can come close. But a PA array would never get that, hanging in free air.

In a room though, you might as well consider a floor to ceiling array a near field device. Or you're living space is much bigger than most of us have :D.

We need to look at floor to ceiling arrays as different animals than shorter arrays. The question I get most: couldn't you use half the length? I wouldn't even think about it.
 
Upper opppp n ghfhxkkzjxh


Sent from my iPhone using Tapatalk

I guess this means you're still reading :D.

Apologies, I really enjoy reading this thread because of the attention to detail and the quest to find the measurement techniques to tease out why certain things have a disproportionate impact on the sound we hear....

As I'm reading this on my phone surrounded by little people, sometimes the phone gets mashed and apparently it sends these blank messages - I don't know how to delete them, so please take them as a sign of interest and approval!
 
I'm thinking that sentence was getting at the near field/far field transition of line arrays in particular. Arrays should be extremely long to truly act like a line source at low frequencies. A floor to ceiling line array can come close.

Thanks, Yes, I was speaking in terms of near/far field transition in floor to ceiling Line arrays. :)

My main listening seat is 7 feet or 2.1 metres from the arrays and my arrays are 6 feet, 6 inches or 2 metres apart. So I am very close to the minimum based on RA's measurement. I do sit closer to the arrays when we play games, the highs are rolled off, I am sure partly due the narrowing dispersion pattern of the highs. (this is not a critical listening spot). This is why I am curious about a two-way design with planar tweeters. Maybe that would solve the listening distance issue, of course bringing in a new issue: a cross-over point in the critical mid-high region. :eek: But we have FIR cross-overs now. Lots of experimenting to do. :D

I really like what floor to ceiling line arrays do, and I do not want to see anyone dismiss the concept without being fully informed. ;)
 
As my son has a holiday week we've been watching movies every evening. The whole Pirates of the Caribbean sequence.

I have to admit running into trouble at the bottom end if I don't roll of the bass a bit below 40 Hz. I'm guessing it isn't healthy to ask my amp to play this nor will it be a good thing for the drivers.

Some day I'll have to try a more powerful amp combo. I just can't make up my mind what "a great amp" really is. Every guy you ask about this has a different opinion about it.
For a while I have been following the "First One" thread. While that started out as a science story lately it has been moving into exotic caps and wires.... without defending the choices behind it. Ah well...

As said I'm still using my old (but well serviced) Pioneer A-757 Mark II and I'd just like to get a similar, but more powerful amplifier to try. This amplifier has served me well, measurements don't show any strange behaviour, I'd just like to have more power on tab.
This one is about 100 watt into 8 ohm. I'd like to try at least 250 watt (into 8 ohm).

Whatever I end up choosing probably needs me to save up for a while. If I can even make an educated guess of what a good amplifier really means.

That doesn't mean I don't think it's important. I just can't understand why we cannot agree on what is good and what is bad. We are so gullible on all things audio (me included). Yet I've been able to prove some things to myself.

When I bought my Pioneer amp new, long ago (I think 1989/1990) I had saved up to get it, being young and all. The thing weight 20 Kg, and all of it was used to produce sound, no bells no whistles. Internally it had a construction to keep path lengths short, you could see it was a concept. The amplifier Pioneer came out with a year later was about 3 Kg. No more extra beefy power supplies (two in the old model that caused it to weigh 20 Kg).

Somehow that old model made sense to me. The new flimsy style never won me over. I'm also convinced Sd area in speaker systems matter. I just need to add some Sd area to also be able to play the below 40 Hz notes needed in movies. In music I've never encountered this and I did try a lot of different genres.

I'll stop rambling... I do miss some open discussion about this though. Without all the subjectiveness. To use a metaphor: I'm not impressed by a Ferrari, never have been. That's why I actually drive a Porsche 911. Something that performs and isn't all about that "look at me, I got money". You CAN drive a Porsche daily, the Ferrari? Not so sure about that. Not that a Porsche comes cheap, but I can actually see and understand why.
(not talking about the new models though, mine is from 1982 and everything after 1998 isn't of any interest to me. I'd rather have a '73 RS. (not that I ever could afford that one)
 
Last edited:
I have been working with a NuPrime MCH-K38 for about a month now... that thing does not run out of steam, plus... it has 8 channels.

It has had glorious reviews so far, and it does sound like they are deserved.

With 8 channels, you can use all of them in stereo, so plenty for your extra ambient speakers and a couple of subs for the future... but each pair is also bridgeable to give a more meaty sound.

Quite the beast. It looks great, weighs more than you would think. So far... I like it.

Check it out.
 
If I can even make an educated guess of what a good amplifier really means.

I am pretty sure you can do that :), but you might as well ask xrk971 if you're open to DIY - he's done quite a few builds in the past few months and done some comparisons too.

Have you tried reaching out to any fellow enthusiasts around your place? Especially regarding something like the NCore which you had mentioned earlier.
 
As my son has a holiday week we've been watching movies every evening. The whole Pirates of the Caribbean sequence.

I have to admit running into trouble at the bottom end if I don't roll of the bass a bit below 40 Hz. I'm guessing it isn't healthy to ask my amp to play this nor will it be a good thing for the drivers.

Most amps dont have trouble delivering their full rated power down to 20hz. While a bigger amp may stress less, that wouldn't protect your drivers. Seems like a subwoofer might be your best solution.
 
I have to admit running into trouble at the bottom end if I don't roll of the bass a bit below 40 Hz.

If you want to cut some low end, try this:

Load your corrected response into REW and (using the EQ) set the LF cutoff to 20hz. Now, tune a PEQ to 20hz with a Q of .71 and -3db gain. The bottom end will now be non-resonant, and at medium to high SPL you should still have some nice presence at 20hz (now the -6db point). I do this at 35hz for my system and run this "low end target filter" in a separate convolver. If you can only use one convovler, then of course you'll have to convolve the REW filter with the DRC filter.