TOPIC:

Mini digi - mini dsp - third party amp?

Simple enough?

Lorienblack wrote:

All "solutions" are compromises of a sort and the decisions get harder when attempting to put together an active speaker. The dsp is fine until people want to feed in a line level analouge, then the signal is digitized and rap up through a DAC again. For people with high end components, that's like doing your washing in dirty water.

Well, I disagree, really, because decisions can be much easier, as many driver problems either go away or are easily avoided. Yes, you can argue that there is SOME distortion created by digitizing something, but it is so small that they have run tests showing that it takes dozens of such conversion in a row to even begin to become measurable, let alone audible. Whereas the distortion in any driver is readily measurable and audible. So, yes, it is like wasthing a muddy pair of jeans in water that has one or two grains of sand at the bottom. Even if those grains of sand fix themselves to the jeans, the net effect is an apparently clean pair of jeans.

The MA7's lack nothing in the HF, sit down in front of a pair one day, maybe you'll be the next old fart banging on about the magic of FR

Maybe, but probably not. The problem is, it is almost impossible to coax more than 2-3 good octaves from any driver, but there are 10 octaves in play. Most, let's say, 5" drivers will begin to have obviously measurable motor distortion below about 500Hz. You can beef up the motor and lengthen the travel but that usually means sacrifices above 500Hz in linearity. And then above about 2000Hz, you end up with narrowing dispersion, measurable aritifacts from the beaming of the driver and cone resonances. If you make the driver super rigid, it will still ring by about 5000Hz-8000Hz, if you make it softer and more pliable like a bending wave transducer, you may increase disperson, but also inband issues at the same time. Many companies claim extended bandwidth for FR drivers, but I've never seen them hit those specs in real life. Most fall off or go completely non lineary by 10kHz. Bose claims 17kHz, but I've never seen it measure over 10kHz. And even if they can do it, they can only do it in one spot. That's just where we are when it comes to physics and driver design.

But I also accept that I'm a FR fan and generally find the compromises painless. The modern sound of most high end gear sounds painfully artificial to my ear, despite the XO, I would be happier sat in front of some 40yo celestion ditton 44's.

I would just encourage more experimentation outside your comfort zone. In my listening, the most natural sounding, most effortless, most musically enjoyable speakers have all been digital active drivers with very rigid drivers and steep crossovers that substantially lower motor distortion, cone resonance and intermodulation distortions between drivers. Even when the signal must be converted to digital and back. Digital can't remove distortion from a full range driver, but it can confine a driver to only the areas it does well and effortlessly pass off the rest to other drivers that are specialists in their own FR.

Many of the current designs - horns, transmission lines, full range, concentric drivers, etc, etc, all address fundamental flaws of non-digital designs. But all bring their own problems. Horns may increase dynamic range, but bring all kinds of other distortions. Full range may eliminate the sound of a passive crossover, but have serious limitations at both frequency extremes Concentric drivers may create a point source, but you're still horn loading the tweeter with a moving horn.

DSP crossovers and impulse response correction is the closest thing to musical panacea as we've gotten handed to us since cones and domes were first created. And what a wonderful thing we can add it for such little money now.

I have no beef with any of the above your quite right.

I like the sound of full range, more specifically the MA drivers, the LF can be overcome to a large extent by great cabinet design (see pensil series) and the HF is much less of an issue when you have turned 40 and can't accurately above 13,000khz and fail to hear above 15,000khz anyway. There is a quality to FR that can't be reproduced by multiway, I guess that's where personal choice carries more weight than response curves. I have yet to hear an affordable setup that makes me melt into the music so rapidly, to simply lose the will to be analytical.

Back on topic - there's no shortage of people reporting that the mini dsp, for all it's wonder, degrades SQ. That was my concern. People will want to run analouge into my speakers and I sort of need that to be done without detriment.

What route would you suggest? I value your opinion.

Would mini digi, mini dsp, third party amp be a viable cost effective solution?

Yeah, I think that is the least expensive way to go and should be fine for what you want to do. That would also give you the capacity to experiment with subs or tweeters if you want.

The higher end way to go would be an OpenDRC driving a DAC and amp or an amp with a DAC.

FWIW, TruAudio just started shipping its new 4-channel amp with optical input for about $400 I think. It has some interesting flexibility options, but you could biamp your speakers and run digital in from an Open DRC. Not sure about the sound quality, it's brand new.

www.truaudio.com/dealer_resources/oneshe..._sheet_(amp-440).pdf

But it's interesting because apparently, the volume of all 4 amps go up and down together, if both are selected to "variable". So it could have some interesting uses with MiniDSP. Going to build a new Near 10M clone with DSP crossover, probably with this amp, for the living room.

Thanks for that - I'll look into the amp with interest. And if I go down the mini dsp route, I'll post my finding here for those who might be interested.

Too bad everyone's so spread out, it would be great to have a MiniDSP project summit.

I have about 4-6 designs in my head that I intend to build, just waiting cash to fall from the sky.....

If cash starts falling from the sky, I want you buggers as far away as possible

Um, by the time you get that amp, a miniDSP and miniDIGI it would cost more than a pair of PWR-ice125 (and you get a lot less).

Ummm...nooooo....not correct. That system is about $530, or could be the nanodigi for about $535, versus $550 for two Power ICE modules

And you get things like a cabinet for the amp, remote control, a preamp, switching, etc. You do get more power with the Power ICE modules but it's not going to be terribly useful driving a couple of small drivers. You also get more flexibility because you can swap amps or move on to another project and he doesn't have to cut open his speakers or build amp cabinets.

Hah, prices have gone down since I last looked But don't forget the chassis and power supply for the miniDSP+DIGI

After reading this thread I'm thinking I'd like to try using the PWR-ICEs with my Alpair 7.3 and a larger woofer underneath. No time though / too many projects

It's a shame there isn't a Power ICE with 3 or 4 80W amps instead.

In any case, looking at the measurements and waterfall of the Alpair 7s, they are pretty decent up through about 3kHZ, but then start to roll off in dispersion as would be expected, but also have some off axis suck outs above that, which are probably cone reflections if I had to guess, and have a nasty 10kHz resonance peak. 'Tis why I'm a fan of tweeters. Also because I'm a drummer and really picky about cymbal reproduction, which soft domes and full range speakers just can't do properly.

Going to be doing some stuff with aluminum midranges and titanium cone tweeters.

Hi John, are you looking at Bob Brines' measurements? I'm curious to know if that is the Gen 3 driver, but at any rate a. I'll get around to measuring mine eventually and b. unlikely they will be as clean as a well-designed multi-way system. Still, there is this interesting question as to exactly how the ear interprets an acoustic signal, as what you hear doesn't sound nearly as bad as what a measurement may look like. (And vice versa, actually).

John wrote:

It's a shame there isn't a Power ICE with 3 or 4 80W amps instead.

One with a 50W ICEpower and a 125W would give 50+50+125+125 or 50+50+250 (lower module bridged) which would work in some pretty nice applications (I think). However it would of course cost a fair bit more than the PWR-ICE125.

Good point, if response curves and graphs were the holy grail, the Quad ESL57's would never have made it past the drawing board.

Yeah, our ear's ability to smooth out the obvious garbage that is the sound of a speaker is pretty amazing. But that's mainly fine, sharp peaks and dips that we ignore, not huge sweeping errors or cone resonances. The cool thing is that DSP makes it easy to drop in a tweeter or woofer and see how it works.

Still, we're now at a point that we can have speakers that measure well and sound good at the same time.