Speaker arrays

[The Natural Philosopher]

That's an interesting issue: whether floating or fixed is better...

[The Natural Philosopher]

On 02/02/16 21:39, Johnny B Good wrote: I'm afraid TNP can

But this time, I am 97% sure I am correct in this.

[The Natural Philosopher]

Yes, I used one in the amp I posted a link to - it was noisy and if you used a slow clock rate it was audible..

At that time (mid 70s) real digital was bloody expensive. There were no A->D chips at sane prices, and RAM was discrete chips of not very high capacity and if you wanted DRAM you had top build your own refresher circuitry. And most CPUs only addressed 64K of it anyway.

If you wanted serious maths to do signal processing you bought a fiendishly expensive DSP chip.

[The Natural Philosopher]

Yep. With floating point processing there is no chance of 'overloading' anything and with recording done at reasonably high bit resolution - say

24 bit - there is n issues with needing to load a tape to saturation to get the dynamic range.

And as I realised when doing my proto code, there will never be a scratchy pot again. You bit DC on the controls and read the output into slow A->D converters

Digits are now infinitley preferable to analogue, which is in any case digital at is core. One elecrton two electron three electron four..

AND of course digital electronics is analogue electronics driven in strange ways

[The Natural Philosopher]

Proof by assertion again?

Again, you don't understand the subtleties of electromechanics.

What do you think the impedance of a *cabinet* and single *nominal* 8 ohm loudspeaker is, at:

- 10khz

- bass resonance (say 100hz)

I have plotted them. Have you?

A cabinet with a pair of loudspeakers in it has a rather different resonance from a cabinet with a single driver, and if they are in series then that's an even more complex situation.

You minimise the possibility of 'independent' cone motion within the cabinet for a pair of speakers, but that is exactly the case.

[Dave Plowman (News)]

The two things are different. But many digital artificial reverberation units may well include both.

You've missed out the obvious one. A tape delay.

Standard practice before digits arrived in pro studios was to feed the reverb plate from the output of a 3 head 1/4" tape recorder. Vary the speed of the recorder, and you vary the delay.

There were commercial devices to do much the same job - one being the Binson. Although they were built down to a price so rather too poor quality for pro use.

Of course a digital box makes all this easier. But given how much a decent EMT reverb plate fetches, not all agree they are better.

[Huge]

Err, no. If *you* annoy me, I killfile you.

[pamela]

Yes that's always an option. Although I hope we're still on talking terms.

[Huge]

Sure. At the moment.

[Johnny B Good]

Doesn't that 3% of uncertainty impell you to research other experts' findings such as the excellent article linked to by this:

Which Syd so obligingly located?

I have to say the author, Dick Pierce, puts, what seems so blindingly obvious to me, very eloquently indeed. If you haven't already done so, have a read and digest the argument he sets out. If you can see any flaw in this article, then you should at least be able explain where he has gone wrong.

[Johnny B Good]

Yes, very nicely put (particularly those last two statements). :-)

[Johnny B Good]

I wasn't sure about including it since it didn't exactly fall into the class of "compact electronic" methods (mind you, neither did the reverb plate).

Did they bother with equalisation correction for the effect of varying the speed? I would expect they'd keep deep bass content out of the picture so would only have to deal with the HF end of the EQ curve. Also, it being an add-in effect, the change of frequency response may have been regarded as simply a part of the overall effect (BICBW - it may have been an important consideration, I've never had the privilege of working in a recording studio).

I should imagine that part of that, aside from "The genuine EMT reverb plate" 'sound', would be due to a reluctance to consign an impressively sized piece of carefully crafted (and still functioning) kit to the scrap heap.

[Johnny B Good]

Jesus H Christ! There's no need to shout. We're none of us so deaf that we can't read what you have to say in the normal mix of upper and lower case text (even if it's mostly about "Teaching granny to suck eggs." without any form of punctuation other than elipses).

[The Natural Philosopher]

I read it and worked out where he had made a mistake yes.

I did.

[The Natural Philosopher]

I downloaded some 'free' reverb software and frankly their 'plate' reverb beats the heck out of anything I have ever heard in a studio recording back in the day.

They also had a 'spring' reverb which apart from not making a sound like the trumps of doom when you dropped the PC on the floor, did sound remarkably as bad as the two and three spring units I used to use.

The actual computations however to really simulate multiple reflections can get very huge, and so tricks are used to get away with less. Multiple paths and overall feedback and so on. Plus various filtering techniques. In the end every reverberator has a unique sound, and if a plate is what you like, stay with a plate.

[tabbypurr]

ISTR, and my memory's not 100% on this point, that current driving moving coil speakers, which zeroes damping, does more to eliminate bass peaks & troughs than any amount of damping.

NT

[Dave Plowman (News)]

I was talking about what was possible - not the size of the equipment needed.

At one time an edit suite was vast - you can do much the same on a PC these days. Etc.

Generally, you don't need the full audio bandwidth for reverb etc. And you'd generally remove the low frequencies anyway.

Vast numbers were. But those which survive can command high prices. I assume because the reverb they produce can't be replicated exactly in digits. Perhaps in the same way as you can't really add the sound of an LP or tape to a digital recording.

[tabbypurr]

I notice adding that 78 sound is also normally a fail, though it shouldn't be hard to do.

NT

[pamela]

I used to see those ellipses in some forum posts. Usually just two dots.

Is that from some Android text editor which turns repeated full stops into a space? If so, then why do the dots remain there in the posted text?

It looks messy.

[DerbyBorn]

Best explanation I have ever heard.

>