Great article on "Analogue Warmth"

[microstrip]

(...)
microstrip..

some times the music is ment to disturb in such cases microstrip(you just take your socks off or what? if so i aint booking you for my 40th regardless of the special rate you offered me in your pm). that what i want. not some dumbed down soft sounding bollocks. what ever you would call that is what i ment.

Would you mind translating it in english?

 

[microstrip]

As I say, I just want 'neutral' so there's not much to say. I am sure they measure excellently (I wouldn't expect anything else), but if I were in his position I, too, would seek to cast unprovable aspersions on my competitors whose products measured just as well but were cheaper.

The question is how we assert "neutral". Listening tests, null tests, some specific measurements, asking the Oracle?
90% of manufacturers claim they are the "true" neutral. Unfortunately neutral is very often associated with a boring sound.

In the good old days the Quad setup (34 or 44 with the 405) was a benchmark for neutrality. Although I still keep them, neutrality has really evolved a lot.

 

[Purite Audio]

The question is how we assert "neutral". Listening tests, null tests, some specific measurements, asking the Oracle?
90% of manufacturers claim they are the "true" neutral. Unfortunately neutral is very often associated with a boring sound.

In the good old days the Quad setup (34 or 44 with the 405) was a benchmark for neutrality. Although I still keep them, neutrality has really evolved a lot.

Neutral would be defined by the loudspeakers measurements, flat on axis response, and smooth and even off axis.
I can imagine a fine measuring loudspeaker in a well treated room initially might well sound a little boring, but IME once you are used to a neutral presentation everything else sounds wrong.
Keith.

 

[Phelonious Ponk]

Groucho, referring to modeling guitar amps:

Yes, and it assumes there is a neutral amplifier at the end of it. But if we believe that the 'magic' is the interaction of the amplifier with the load itself, interaction that varies dynamically with coil temperature and therefore ambient temperature, or maybe the room-dependent microphony of valves (that is real and measurable, but not so easily simulated), it can't, in fact, be modelled and simulated to the satisfaction of audiophiles.

With guitar amps, it has much to do with the amps interaction wit the source, the electric guitar. But that's a long story and WAY off topic.

But this is all starting from the assumption that it is self-evident that non-neutral amplifiers do, in fact, improve the sound. My 'philosophy' is the other way round: I want a neutral, powerful amplifier that remains demonstrably neutral with any load I give it. I can then forget about it. I want a speaker that is as neutral as I can make it. Naturally this involves active crossover, three-way, driver correction, prodigious bass capability etc. I can then forget about it. I want a source that is as neutral as I can make it. Digital fits the bill. If this combination sounds wrong to me, I will try to find out why and fix it - maybe I just need to get used to it. Only after I have exhausted all that, will I cast around for a suitable virtual DSP fuzz box to put in the way. But have you ever heard a system that does genuinely attempt to be neutral all the way? (stressing the need to be truly neutral, including the bass - it is not 'compact' - including the phase correction). You don't need to get used to it, and you don't need the fuzz box!

Agreed on all counts, though I have a strong preference for the other kind of three-way -- two-way with subs.

Tim

 

[Groucho]

I have a strong preference for the other kind of three-way -- two-way with subs.

Actually my speakers are two-way boxes on top of large woofer enclosures. Not sure whether you'd call them subs however, because I am crossing them over at a slightly higher frequency than people generally define a "sub-woofer".

 

[Atmasphere]

It would be interesting to see if the technology I've been talking about will have the ability to exactly emulate the MA-3. Do you think this is possible with advanced DSP combined with SOTA class D power?

In a word, no.

The problem you have there is that the MA-3 does not make much in the way of distortions to which the ear is most sensitive. The technology you are referring to still has a ways to go in that regard!

Would it be possible in the future where we all have flying cars? Possibly. Its certainly not on the horizon right now!

We already have that with vinyl and tube gear The sound from the mic feeds is awful. This is why we need the DSP, to enhance the horrible reality with tonal density and timbral richness.

Apparently you have not heard a decent mic feed or you are referring to something that is not obvious from this post. A good mic feed is so good that it will fool even the most jaundiced audiophile into thinking its real! The problem comes in with the recording media- that sense of being absolutely real is gone once its been recorded. You certainly don't need to enhance a good mic feed in any way. What you have to work with is what happens to the signal downstream, but IMO/IME of running both digital and analog systems in a recording studio is the thing you really want to keep away from any recording that you want to sound good is DSP, unless by 'good' you mean 'suck the life out of it'. I'm not trying to be inflammatory BTW; this is very real. Spend some time in a decent studio with decent mics and you will see what I mean right away!

 

[Groucho]

The question is how we assert "neutral". Listening tests, null tests, some specific measurements, asking the Oracle?
90% of manufacturers claim they are the "true" neutral. Unfortunately neutral is very often associated with a boring sound.

In the good old days the Quad setup (34 or 44 with the 405) was a benchmark for neutrality. Although I still keep them, neutrality has really evolved a lot.

I own one of the Quad current dumping amps, too, but I don't use it since I went 'active'. Active speakers present such an easy, efficient load that my presumption is that any characteristic sound that may, or may not, have been heard will not apply anyway. If amplifiers have any sound at all (they shouldn't), I am diluting that characteristic three ways. My bass will not intermodulate with the mid etc. The only way I could get the conventional "analogue warmth" would be to pre-process the audio with a DSP simulation of intermodulation distortion - or whatever we think is the nicest form of analogue warmth. As I said before, a truly neutral system (or as close I can get) sounds perfectly warm to me anyway - or cool if the recording is cool.

Edit: By "cool" I don't mean "edgy" or "sharp" or "harsh" - which is what I think you get with passive, non-DSP speakers possibly because of the phase distortion, or with speakers that chop the bass off.

 

[Phelonious Ponk]

Actually my speakers are two-way boxes on top of large woofer enclosures. Not sure whether you'd call them subs however, because I am crossing them over at a slightly higher frequency than people generally define a "sub-woofer".

I'm sure that can work very well. I actually prefer smaller subs, crossed over higher. I've assumed that's because I don't care about home theater, organ music...or even very much orchestral, but I really like it when a kick drum and a standup bass get close. There are a lot of subs I haven't heard, though, FWIW.

Tim

 

[Blizzard]

If DSP proponents had something to suggest as effective as the electronic chain to please audiophile preferences it would be very well accepted. BTW, Nelson Pass answered your main question in his manual for the XS series:


If you are concerned that your power amplifier (or anything else
for that matter) is as objectively and technically accurate as possible,
that is a perfectly legitimate criterion. You will certainly find many
products in the marketplace that excel at conventional objective
performance, and most of them are much cheaper.

Our real customers care most about the experience they get when
they sit down to listen to their music. We create amplifiers that we
like to listen to, on the assumption that we share similar taste
We want our products to invite you to listen. We want you to enjoy
the experience so much that you go through your entire record
collection - again and again. This, by the way, is a very strong
indicator.
A simple survey of really successful audio amplifiers shows
that objective performance numbers by themselves are not that
important.

That Would Be You.

If you are concerned that your power amplifier (or anything else
for that matter) is as objectively and technically accurate as possible,
that is a perfectly legitimate criterion. You will certainly find many
products in the marketplace that excel at conventional objective
performance, and most of them are much cheaper.

Our real customers care most about the experience they get when
they sit down to listen to their music. We create amplifiers that we
like to listen to, on the assumption that we share similar taste.
We want our products to invite you to listen. We want you to enjoy
the experience so much that you go through your entire record
collection - again and again. This, by the way, is a very strong
indicator.
A simple survey of really successful audio amplifiers shows
that objective performance numbers by themselves are not that
important.


I marked in bold what I find basic concepts behind high-end - creating a sound reproduction that is systematically enjoyable for the consumer. But also acknowledging that excellent stereo reproduction involves the listener and needs his collaboration in such a strong way that it is not possible to create unique products that are the preferred by everyone.

Unfortunately that model is soon to be outdated with SOTA DSP that can create any sound profile one desires. Now the best gear will be the most transparent, efficient and reliable. The "artists" will be building DSP profiles rather than hardware based effects boxes.

 

[Groucho]

I'm sure that can work very well. I actually prefer smaller subs, crossed over higher. I've assumed that's because I don't care about home theater, organ music...or even very much orchestral, but I really like it when a kick drum and a standup bass get close. There are a lot of subs I haven't heard, though, FWIW.

Have you auditioned many sealed subs (as opposed to ported)? It's my guess, for what it's worth, that with a larger sealed sub-woofer you may get closer to what you want. Ported speakers are, by definition, smearing the bass and giving the misleading impression, perhaps, that larger subs are "slow". The sealed version is less 'flabby'. Sealed speakers roll-off at a shallower rate, and the bass is controlled throughout. The ability to go deeper with more precision is also a contributor to "warmth", I would say.

(apologies if you have already tried that, though)

 

[Blizzard]

In a word, no.

The problem you have there is that the MA-3 does not make much in the way of distortions to which the ear is most sensitive. The technology you are referring to still has a ways to go in that regard!

Would it be possible in the future where we all have flying cars? Possibly. Its certainly not on the horizon right now!



Apparently you have not heard a decent mic feed or you are referring to something that is not obvious from this post. A good mic feed is so good that it will fool even the most jaundiced audiophile into thinking its real! The problem comes in with the recording media- that sense of being absolutely real is gone once its been recorded. You certainly don't need to enhance a good mic feed in any way. What you have to work with is what happens to the signal downstream, but IMO/IME of running both digital and analog systems in a recording studio is the thing you really want to keep away from any recording that you want to sound good is DSP, unless by 'good' you mean 'suck the life out of it'. I'm not trying to be inflammatory BTW; this is very real. Spend some time in a decent studio with decent mics and you will see what I mean right away!

I suppose you would have to test drive this technology to know if it has a ways to go still. It's such high resolution it picks up everything. Even what we don't know how to measure yet. Much the same as if you connected a merging HAPI/Pyramix up to the outputs and made a DSD 256 copy.

Think of it as an ultra highend version of this, put for the playback chain, rather than the studio. Also DSD compatible up to 1024.

http://www.uaudio.com/uad-plug-ins.html



I was being scarcestic about the mic feeds. As some folks think we must add coloration after the mic feeds in order to obtain "audiophile grade" tonal density and timbral richness.

 

[Groucho]

Unfortunately that model is soon to be outdated with SOTA DSP that can create any sound profile one desires. Now the best gear will be the most transparent, efficient and reliable. The "artists" will be building DSP profiles rather than hardware based effects boxes.

How will you establish the existing amp's characteristics? (apologies if you described that earlier and I missed it). Does it involve a test signal, test load and measurements? If so, what will your test signal and load be?

 

[Blizzard]

How will you establish the existing amp's characteristics? (apologies if you described that earlier and I missed it). Does it involve a test signal, test load and measurements? If so, what will your test signal and load be?

For DACs and preamps you simply connect to the analog outs, and capture a sweep. You either use a connected DAC playing back a special sweep, or you must connect the analog outs of the measurement gear to the inputs if it's a stand alone pre.

For amps, if you really wanted you could load the amps to a level where they produce extra distortion, but most amps preform best at very low load. and long as the sweep is captured within a the range that the amp isn't straining, it would be fine.

 

[microstrip]

I own one of the Quad current dumping amps, too, but I don't use it since I went 'active'. Active speakers present such an easy, efficient load that my presumption is that any characteristic sound that may, or may not, have been heard will not apply anyway. If amplifiers have any sound at all (they shouldn't), I am diluting that characteristic three ways. My bass will not intermodulate with the mid etc. The only way I could get the conventional "analogue warmth" would be to pre-process the audio with a DSP simulation of intermodulation distortion - or whatever we think is the nicest form of analogue warmth. As I said before, a truly neutral system (or as close I can get) sounds perfectly warm to me anyway - or cool if the recording is cool.

Edit: By "cool" I don't mean "edgy" or "sharp" or "harsh" - which is what I think you get with passive, non-DSP speakers possibly because of the phase distortion, or with speakers that chop the bass off.

OK, I see now your definition of a neutral amplifier is mostly anything decent that should not play more than a few octaves, if there are no passive components in series with the speaker coil. But I could not understand the concept of "diluting" characteristics three ways.

BTW, Philip Newell has expressed a similar idea about speaker cables in HiFi Critic- no speaker cable should need to pass ten octaves.

 

[Atmasphere]

I suppose you would have to test drive this technology to know if it has a ways to go still. It's such high resolution it picks up everything. Even what we don't know how to measure yet. Much the same as if you connected a merging HAPI/Pyramix up to the outputs and made a DSD 256 copy.

Think of it as an ultra highend version of this, put for the playback chain, rather than the studio. Also DSD compatible up to 1024.

http://www.uaudio.com/uad-plug-ins.html

I don't doubt that it picks up stuff. That's not what I was talking about. What I was talking about is that the current crop of DSP and class D amps (which are analog, BTW) have their own artifacts that will not allow them to sound like a set of MA-3s. They are a long ways from that!

In studying the class D technology over the last 12 years or so, its pretty obvious we have something to bring to that table. I'm of the opinion that class D has arrived as a technology but it is by no means mature.

 

[Groucho]

For DACs and preamps you simply connect to the analog outs, and capture a sweep. You either use a connected DAC playing back a special sweep, or you must connect the analog outs of the measurement gear to the inputs if it's a stand alone pre.

For amps, if you really wanted you could load the amps to a level where they produce extra distortion, but most amps preform best at very low load. and long as the sweep is captured within a the range that the amp isn't straining, it would be fine.

OK, so the result is a 'look-up table'..? Input sample is an index into a table that gives you the value for the output? And then this relies on an ultra-neutral (class D) amp to drive the actual speaker..?

 

[Blizzard]

I don't doubt that it picks up stuff. That's not what I was talking about. What I was talking about is that the current crop of DSP and class D amps (which are analog, BTW) have their own artifacts that will not allow them to sound like a set of MA-3s. They are a long ways from that!

In studying the class D technology over the last 12 years or so, its pretty obvious we have something to bring to that table. I'm of the opinion that class D has arrived as a technology but it is by no means mature.

Okay, so the issue is on the amp end. Must be a very well built class D amp then. Have you tested the new Hypex NC500 units yet?

 

[Blizzard]

OK, so the result is a 'look-up table'..? Input sample is an index into a table that gives you the value for the output? And then this relies on an ultra-neutral (class D) amp to drive the actual speaker..?

You won't have to worry about how the measurement was taken. As an end user you just have to select the sound profile you want to listen to and that's it. Super simple.

What would even be cooler is if I could project a 3D hologram of the actual gear on a platform that sits on the audio rack!

The profile database on the website will be similar to this:

http://www.uaudio.com/store.html

But will have a custom area for custom profiles users made by blending other profiles together.

 

[Groucho]

But I could not understand the concept of "diluting" characteristics three ways.

I am thinking of it this way:

If I were to feed a signal into an amp with a gradual clipping characteristic and then into a passive speaker I would get a certain result.

If I were, instead, to filter the signal into three bands and then feed those into three amps with the same transfer function, then sum the result (electrically or, in the case of active speakers, acoustically) the result is different. I am no longer getting the intermodulation between bass and mid, for example. Also, assuming the same amps in both cases, I am not driving the amps into the same degree of clipping.

Would a certain audiophile description for that amp in passive speaker mode ("feisty", "dark", "warm" etc.) still apply after we had split the signal three ways and passed it through three of them in parallel? I imagine that the result has got to be more 'neutral' as a result.

 

[Atmasphere]

Okay, so the issue is on the amp end. Must be a very well built class D amp then. Have you tested the new Hypex NC500 units yet?

Its also on the DSP end. If you want to know an effective method of messing up an otherwise excellent digital recording, simply apply DSP and watch it go down the drain.