Millennia HV-3 vs. Forssell SMP-2: AB samples

[ISedlacek]

Originally Posted by DaveD

Ivo, are you going to use Mk2 for the choir?

Most probably MK21 NOS main, MK2 ambience, MK4V (or Sennheiser 8040) spots for piano and/or soloists

[Marlan]

Ivo,

First off, nice sound on everything. My question is this: After a simple listen of the website audio file as played through decent computer speakers, it seems as if the overall amplitude of the monochord recordings are not evenly matched. Is this the case? The Forsell example sounds LOUDER.

I am not trying to start trouble in any way mind you but I think this has probably been raised by other people on tests such as this (not just yours) so I'm probably not alone in stating that if these preamp outputs were not accurately calibrated by scope and meter with tones, then I feel one cannot accurately judge the validity of such statements as more natural, open, or statements like this on a test such as this.

On the other hand, if you are making a rough test to see if you can discern extremely obvious and very different dynamic differences in timbre and color then that is something else entirely.

The reason for my post is that I feel it is irresponsible for people to be making huge disclaimers about gear in an open forum like this if they have not made accurate comparrisons themselves with the equipment in question.

Regardless, the playing is great and the recordings sound beautiful. Bravo.

[Stefanizzi]

Originally Posted by Marlan

Ivo,

First off, nice sound on everything. My question is this: After a simple listen of the website audio file as played through decent computer speakers, it seems as if the overall amplitude of the monochord recordings are not evenly matched. Is this the case? The Forsell example sounds LOUDER.

I am not trying to start trouble in any way mind you but I think this has probably been raised by other people on tests such as this (not just yours) so I'm probably not alone in stating that if these preamp outputs were not accurately calibrated by scope and meter with tones, then I feel one cannot accurately judge the validity of such statements as more natural, open, or statements like this on a test such as this.

On the other hand, if you are making a rough test to see if you can discern extremely obvious and very different dynamic differences in timbre and color then that is something else entirely.

The reason for my post is that I feel it is irresponsible for people to be making huge disclaimers about gear in an open forum like this if they have not made accurate comparrisons themselves with the equipment in question.

Regardless, the playing is great and the recordings sound beautiful. Bravo.

Very few GS's members, maybe one in a hundred, deserve deep attention.

My kind respect Mr. Marlan, you are one of them.

[ISedlacek]

Originally Posted by Marlan

Ivo,

First off, nice sound on everything. My question is this: After a simple listen of the website audio file as played through decent computer speakers, it seems as if the overall amplitude of the monochord recordings are not evenly matched. Is this the case? The Forsell example sounds LOUDER.

I am not trying to start trouble in any way mind you but I think this has probably been raised by other people on tests such as this (not just yours) so I'm probably not alone in stating that if these preamp outputs were not accurately calibrated by scope and meter with tones, then I feel one cannot accurately judge the validity of such statements as more natural, open, or statements like this on a test such as this.

On the other hand, if you are making a rough test to see if you can discern extremely obvious and very different dynamic differences in timbre and color then that is something else entirely.

The reason for my post is that I feel it is irresponsible for people to be making huge disclaimers about gear in an open forum like this if they have not made accurate comparrisons themselves with the equipment in question.

Regardless, the playing is great and the recordings sound beautiful. Bravo.

Marlan,

you can evenly match two identical tracks but not possibly two separate, slightly different improvised performances. It has nothing to do with calibrating the preamps outputs etc. What may be slightly "louder" at one particular spot, may be opposite at another spot. A human being is playing, not a synth or a machine. The overal average level is however approximately the same.

This is not a chemical analysis. The overal sound characteristics (and the difference) is usually obvious anyway, regardless some natural minor dynamic differences (in both directions).

My basic attitude is: if you can trace some sound difference only under very laboratory conditions and with a microscope, it is not worth consideration for a practical studio life. Only the differences which are continuous, obvious at any moment have some practical value.

The source of my subjective impressions that I sometimes may share is not one sample that I listen through the internet - but rather days, weeks and months of working with the mentioned devices in a practical life ... And I hope, I can share my impressions in the same way as everybody else can. But I am the source of the impression, rather than some short, limited, purely illustrative sound example, that with an extra effort I post (not for my own benefit).

For sure, if my intention would be to RECORD very "scientific", analytical comparisons, I will be able to that, but it was not my intention so far and I don´t think it will be any time soon .. I know how it sounds for myself and it is enough for me. If somebody asks, I may share ... that´s all. For a detailed type of comparisons I would refer to Lynn Fuston and his comparison CDs - and you can ask him how much time and effort it required.

Yet, I personally don´t need to create a comparison CD to be able to easily hear basic sound difference between various types of gears ...

Originally Posted by Stefanizzi

Very few GS's members, maybe one in a hundred, deserve deep attention.

My kind respect Mr. Marlan, you are one of them.

Mr. Stefanizzi,

in the thread about DAD AX-24 samples recorded in various sample rates, you showed very bright insight and admirable ears and made very clear and authoritative quality scale from these samples. I am not sure whether you noticed that these samples (although in this case it was just one source, one performance) actually quite differed in the dynamic level. See the beginning:

[MichaelPatrick]

Ivo, this is a bit off subject so you please ignore if you like...

I'm wondering if you'd prefer DACS over Millennia for acoustic instrument recording? I know you had both of these before settling on Forssell SMP-2.

[ISedlacek]

Originally Posted by MichaelPatrick

Ivo, this is a bit off subject so you please ignore if you like...

I'm wondering if you'd prefer DACS over Millennia for acoustic instrument recording? I know you had both of these before settling on Forssell SMP-2.

It was very long time ago these two met here for a while ... but I know I prefered Millennia ... sounded a bit more "decent" so to say ... DACS sounded somehow ... more "hollow" ...(?)

[bcgood]

the listening sessions' mic pre graph.

[Stefanizzi]

Originally Posted by ISedlacek

[...] Mr. Stefanizzi,

in the thread about DAD AX-24 samples recorded in various sample rates, you showed very bright insight and admirable ears and made very clear and authoritative quality scale from these samples. I am not sure whether you noticed that these samples (although in this case it was just one source, one performance) actually quite differed in the dynamic level. See the beginning:

Mr. Sedlacek,

I appreciate your attempt to find tangible arguments to support your convictions or suppositions. Nevertheless I have to conclude that, right with such argumentation, you ended to cut your own throath:

a) Tests you mention here are deeply different in nature. The objections raised by Mr. Barry relate, if I am not wrong, to a renowned phenomenon like that investigated by Fletcher & Munson in the thirties: a study on perceptual differencies in timbre, at different levels of volume; and while everyone agree that a louder program is perceived as clearer, more definite, more convincing and authoritative, with more bass and so on, you can easily verify that the files I liked more in downsampling-test were not necessary the louder ones.

b) A problem of method: even if that were the case (supposing someone would prefer the louder ones), the differences in amplitude between the downsampling files of DAD AX-24 and the ones you posted in this thread, are of several order of magnitude, therefore no doubt or insinuation is legitimate here. In downsampling test we deal with very small amplitude variation of a single performance; in your test we deal with (among other even more dramatic variations) greater amplitude variation of two different performances. One can’t muddle apples with oranges.

c) In spite of everything, test like yours are interesting. I believe nobody knows the acoustic peculiarity of your studio, or that of your own instrument better than you… here you are right. But this statement doesn’t necessarily invalidate the legitimacy of your opponent’s assertions. If you couldn’t split the signal and were forced to play twice the same musical piece, you should calculate a margin of error, and above all you should expect some risk of debate, or even criticism, from the outside.

d) I hope you didn’t know which preamplifier was in function. I myself did some not-blind tests in the past. From a scientific point of view, well… very ineffectual tests. If the player and the engineer (sometimes producer and studio owner too?) coincide, the reliability in the judgement is very weak. More and more weak if the engineer has at his disposal for a couple of day a new component supposed to be better than the one he owns. Little nuances lie in ambush… and act unconscious. After all, even if you are extremely concentrated and try to play the same piece exactly twice (like a cyborg), the differencies between two performances are likely greater than that between two hi-end preamplifiers. If you listen to the piano samples in DAD website you’ll understand what I mean (moreover I’m pretty sure the pianist acted in that case as pure performer, unaware of technical matters).

Nobody would measure an ant with a pocket rule.
But I’m pretty sure you took precautions against this common, silly accident.
I hope I could help.


Thanks for giving us the opportunity to hear the beautiful, seductive voice of an exotic instrument, beautifully played (ok, uneven… but who cares?) by a passionate human being.

Regards

[ISedlacek]

Thank you for your elaborate ideas. I would just again say

1) I did not offer, intend and announce any "laboratory" scientific test here. If I did, your objections would be perfectly legitimate. I do not issue a kind of "certificate" by sharing my impressions about various type of gears. It is just within a common chat on our favourite topics.

2) Various types of gear usually do not sound very much alike (they are even not intended to). For noticing that, you do not need to align identical tracks by 0,001 dB. You can simply hear it all the time if you want. If you have Behringer C-1 mic and U47 - do you think you really need a laboratory to hear and describe the difference in character ? If you have Telefunken V76 and GML preamps - will you not hear the basic sound differences whatever you play ? Etc. Please, let us not unrealistically overestimate this kind of pedantic "pseudo-scientific" approach and doubt simple hearing evaluation and descriptions based on a practical use - everybody can hear that. If not, the differences are meaningless.

When I say that for example V76 sounds a lot more coloured, more wooly, with emphasised low mids etc. comparing to say Millennia - would not that be a simple truth and would not be doubting it and directing to scientific laboratory tests a bit funny ?

[MichaelPatrick]

Marlan and Massimo,

Ivo doesn't need me to defend him. Here are my thoughts about his method...

Ivo has always been candid about his method. He admits to NOT using scientific controls in his experiments. Wanting to prove something conclusively to others is just not his purpose.

His opinions have value because they are grounded in consistent aesthetic goals and years of thoughtful listening, and his clips are plucked from a long stream of experiments guided by those goals. He observes sonic qualities that not everyone may care much about, and the clips illustrate what he hears.

I don't have enough time to demo all the great gear he's used in production, yet because we have similar aesthetic goals Ivo's observations have helped me make better gear choices for acoustic music in live spaces. Fred Forssell's SMP-2 is a case-in-point. I bought it on his opinion and enjoy it for many of the reasons he does.

Running Ivo's experiments with more controls may help you hear what he hears, but you'll miss the point entirely if you don't grasp his aesthetic. My question for you is not about method, it is this: What sonic qualities that Ivo seeks are being, or may be, misrepresented by his clips and opinions?

Perhaps you understand his aesthetic 100% and you're just passionate about experimental orthodoxy. Then you should tell us why his opinions matter enough to be proven right or wrong. This could perhaps help motivate someone (you?) to reproduce the experiments with better controls, and we'll all benefit from the findings.

Ivo's opinions are founded --or not-- on HIS methods, and he tells us what they are. I don't think he needs to produce scientific proofs to justify clips and opinions. Rather, I think you should demonstrate the validity of Ivo's findings yourself, or lack thereof, if scientific rigor matters enough to you to make it an issue or you really believe others are being mislead.

[vitreouswindows]

How did the comparison work with the Horsch for vocals? Impressions?

[d_fu]

Massimo,

Originally Posted by Stefanizzi

Mr. Sedlacek,

As on many other internet fora and newsgroups, it is common practice here to address each other with first names or the forum nickname. Without implying that you have any such intention, adressing someone as "Mr. so-and-so", esp. when in connection with criticism, may be interpreted as a sign of contempt. Looking through earlier postings of yours, you seem to vary the way you address people with no particular reason or intent, therefore I don't think you meant to indicate contempt here. I still would like to point out that in order to make it clear that this is a discussion based on technical matters alone, and to avoid confusion, you might want to do as everybody else does here. Or at least stick to a uniform way of addressing people, if you dislike calling everybody by their first name.

There's not much to add to Michael's fine posting, just my 2c.

b) in your test we deal with (among other even more dramatic variations) greater amplitude variation of two different performances. One can’t muddle apples with oranges.

I don't have the feeling that the difference is dramatic - also, anyone is free to load these files into the DAW of his choice to adjust levels. The performance will still be different, but the playing is uniform enough to allow hearing a certain difference. Also, it is my impression that the differences in question don't relate to loudness much. Esp. on the monochord samples, the Forssell seems to create a wider space - which almost sounds like something's been added to the sound, because it's hard to believe the HV-3 could "take away" as much...

That said, I agree in principle that identical performances would be preferrable, and I believe I've also said so here on earlier occasions. But Ivo is free to conduct his own tests and share them. He's always been clear about the fact that they are not entirely "scientific".

c) If you couldn’t split the signal and were forced to play twice the same musical piece, you should calculate a margin of error, and above all you should expect some risk of debate, or even criticism, from the outside.

As for the risk of debate, I believe Ivo is well prepared... As for margin of error, the specific differences Ivo detects are not too vulnerable to differences of texture and volume caused by subtle differences in playing, IMHO.


Daniel

[ISedlacek]

I almost feel like being presented to the court

Anyway, I do nothing more here than sharing about the things I tried and telling how I liked them and why ... That´s all ...

Originally Posted by d_fu

Esp. on the monochord samples, the Forssell seems to create a wider space - which almost sounds like something's been added to the sound, because it's hard to believe the HV-3 could "take away" as much...

Well, that has been my point and surprise most of the time ... To my ears, Forssell does not add anything. Through it, it sounds as it really sounds when I play and hear it, especially the space and depth around (or very close to it). HV-3 takes away something ... (not only in case of monochord). I am not much concerned about the actual brands so much. To be on neutral ground and not to be flamed at again, we can call them XC45 and CR98 if we want or whatever. Still I would say (after endless comparisons and listening) that CR98 takes something away and XC45keeps the things sounding close to their natural original sound and space ... And I personally prefer that ...

I too think the dynamic differences between the mentioned two monochord samples are truly negligible comparing to the difference in the sound character.

I have another preamp here now: Thermionic Culture Earlybird 2.2 For me, it is at the same quality level as XC45 But it DOES add something, although in very very decent way ... It makes the sound a bit more "liquid" , focused and honeyish in a way (maybe slightly bigger as well). But still very clean and detailed. But a touch of sweetness is there. I like it very much ...It is almost the same clean as Forssell, but it makes the things to sound slightly fuller and sweeter (than they really sound).

For example Lachapell adds too much to the sound for me ... or DW Fearn ... but Forssell (sorry XC45) simply sounds like a big open window to me ... as natural as possible

BTW - as for the slightly "coloured" preamp, this Earlybird has elevated me the most out of all the tube preamps I have tried so far (Pendulum, DW Fearn, Lachapell, Fetcode, Millennia M2B etc.). Pendulum was nice but a bit too "sharp" in a way to me .. DW Fearn was too coloured. Lachapell a bit too coloured and "soft" for me ... Fetcode a bit too plain ... Earlybird seems to sound fabulous so far and is in pair with Forssell (perfect companion and complementary sound). I will have very tough choice to decide which one to use for recording a choir in two weeks time ...

If I remember well, Mr. Plush acquired Earlybird too some time ago and had similar impressions (using it now as a main orchestra preamp, replacing E.A.R.)

[Marlan]

Just for the record, I actually appreciate what Ivo is trying to do and my comments were specifically directed to people on this forum making big definitive claims on recording gear
they have not used and/or tested properly themselves.

Best,

Marlan

[andy_simpson]

I think Ivo's approach is perfectly valid but only with the conscious acknowledgement that the full logic of the process is that the result is directly subject to the microphones & speakers used.

In other words, if we change the speakers or microphones, the results may change.

If we change the perspective of the question so that whatever pre-amp/converter is a constant and the speakers or microphones are variable, we might choose different speakers/microphones to suit the pre-amp/converter in hand for 'best sound'.

Andy

[~ufo~]

I for one, and I suppose I'm far from alone, would like to listen to samples of a pre amp comparison done right, scientifically. outputs matched w/ test tones etc.

does anyone know of such material existing ?

[ISedlacek]

Originally Posted by ~ufo~

I for one, and I suppose I'm far from alone, would like to listen to samples of a pre amp comparison done right, scientifically. outputs matched w/ test tones etc.

does anyone know of such material existing ?

I don't know - maybe you should make it ? But do you normally record test tones or music ?

[nas]

Originally Posted by ~ufo~

I for one, and I suppose I'm far from alone, would like to listen to samples of a pre amp comparison done right, scientifically. outputs matched w/ test tones etc.

does anyone know of such material existing ?

Perhaps not as scientific or controlled.. but this may help:

The Listening Sessions - The Sessions

[Audiop]

Test tones may actually make sense if the right one are used. Experience has shown that when it's hard to hear if a DUT affects the signal audibly with this or that piece of music there are special test tones that are very hard for the gear to handle. My point is if the DUT can handle these test tones without audible coloration one can be pretty sure it will handle any music signal you throw on it.

I intend to do some bypass testing with Line Audio 8MP, Earthworks ZDT1024 and The top of the line Forsell preamp next month.

A signal will be attenuated, fed to the preamp, and recorded. This way we can compare the original (that goes into the DUT) to the signal that comes out of the DUT.

The P48V will be taken out of the equation but the rest of the circuitry is tested.

The above mentioned testmethod is not only good when you want to investigate if the DUT is transparent or not, but you can also learn about the absolute sound of the DUT on a specific signal.

When you A/B test you never really know what a specific piece of gear does to the signal as you compare it to another unknown DUT in an unknown (strictly speaking) chain of other components. To many variables to get truly meaningful results.


Level matching was mentioned earlier and it's important of course.

Ivo mentioned 0.001dB as a kind of joke if I got it right.. but that is not very far from what's needed to exclude that variable. it's assumed that matching levels within 0.1dB is good enough but there are people that have been able to pick out two identical signals being 0.1dB apart with statistical significance. Therefore one could argue that for scientificall testing 0.05dB is necessary since AFAIK nobody has been able to detect such a difference.



/Peter

[~ufo~]

Originally Posted by ISedlacek

I don't know - maybe you should make it ? But do you normally record test tones or music ?

only test tones..... I've gone off the idea of musicality, it's overrated!

seriously though, I mean calibrating w/ testtones and then recording music.

once I shot out two pres, calibrated it by eye on the DAW meter and then normalised the tracks and posted them. someone said I should really calibrate both pre's gain w/ test tones and noise. My position was that if one pre has a faster spikier transient response and there for has relatively higher peaks, it will overload your ADC quicker, therefore the gain will be dropped accordingly in everyday use and normalising suit in my opinion is kind of fair, because the quicker pre will have to be used at lower levels to avoid overs.

so I think a scientific way of testing them is highly desirable, but I also keep a real world practice point of view. However, I'm not the one to do the test methinks, my studio and methods are hardly high end or scientific.

[ISedlacek]

Originally Posted by ~ufo~

only test tones..... I've gone off the idea of musicality, it's overrated!

seriously though, I mean calibrating w/ testtones and then recording music.

once I shot out two pres, calibrated it by eye on the DAW meter and then normalised the tracks and posted them. someone said I should really calibrate both pre's gain w/ test tones and noise. My position was that if one pre has a faster spikier transient response and there for has relatively higher peaks, it will overload your ADC quicker, therefore the gain will be dropped accordingly in everyday use and normalising suit in my opinion is kind of fair, because the quicker pre will have to be used at lower levels to avoid overs.

so I think a scientific way of testing them is highly desirable, but I also keep a real world practice point of view. However, I'm not the one to do the test methinks, my studio and methods are hardly high end or scientific.

And what practical output do you expect from that ? For me - if you cannot clearly hear a sound difference between two devices under normal, non-scientific conditions, the difference is meaningless for a practical use. Because there are plenty of much more important factors that form the final sound ... Using this or that preamp when you cannot hear any significant difference between the two, does not matter to me ...

[~ufo~]

I hear what you're saying and I sympathise.

However, the practical output, I guess, would be this:

you will ensure the the pre amps' gain are as perfectly matched as possible, in a scientific way (i.e. calibrated via testones/noise), so that any audible difference between the sound recorded must therefore be the difference in characters of the pre amps. It will hopefully eliminate any difference in the chain other than the device you are testing. like different output levels to the ADC etc.

To me, a scientific way of setting up a test setup is ultimately the best way to decide.
Especially when you get into a range where, say differences in cables or differences between matched mics are potentially comparable to the differences between pre amps.

granted, this begs the question: "then why bother?"

this exactly what I seek to answer, as I'm relatively new to the whole pre amp business.

[Audiop]

Originally Posted by ISedlacek

And what practical output do you expect from that ? For me - if you cannot clearly hear a sound difference between two devices under normal, non-scientific conditions, the difference is meaningless for a practical use. Because there are plenty of much more important factors that form the final sound ... Using this or that preamp when you cannot hear any significant difference between the two, does not matter to me ...

Ivo,

the problem (the way I see it) is the day you prefer one piece of gear to another because it sounds better to you. Let's say the price is the double for the "better" sounding unit and you shell out for it not realizing that the two pieces og gear tested actually has identical performance and you prefered the one that had 0.5dB stronger output.

This is a known fact among experienced testers and something that should not be overlooked. Sure there are times when the audible performance is so big even a level difference wouldn't stop you from telling them apart but how do we know for sure where the limit is?


/Peter

[Audiop]

Originally Posted by ~ufo~

once I shot out two pres, calibrated it by eye on the DAW meter and then normalised the tracks and posted them. someone said I should really calibrate both pre's gain w/ test tones and noise. My position was that if one pre has a faster spikier transient response and there for has relatively higher peaks, it will overload your ADC quicker, therefore the gain will be dropped accordingly in everyday use and normalising suit in my opinion is kind of fair, because the quicker pre will have to be used at lower levels to avoid overs.

Yes, recording a 1k sine at the start of every test makes sense but it's not easy when using mic's. When testing pure electronic gear it's easy and reliable though.

As for peaky transient response. Do you mean that you have measured pre's that actually has a measurable difference on the level of musical transients?

It sounds like a non issue with any well designed amp.


/Peter

[~ufo~]

Originally Posted by Audiop

As for peaky transient response. Do you mean that you have measured pre's that actually has a measurable difference on the level of musical transients?

It sounds like a non issue with any well designed amp.

no that was purely speculation on my part, but some pres are definitely quicker sounding than others, so I figured that would make a difference in the recorded attack levels.

[MrArtrock]

Hi everyone, regardless of the (usual) scientific gibbering provided above and wich I won´t comment any further except my deep respect for IVO´s attempts to show inherent sonic differences between preamps in this case.

I have, based on his and some other guys sound & profound opinions invested in my first Forssell SMP-2 a couple of weeks ago, went to Prague just before Christmas and met up with Ivo for picking up 2 such units intended for use here in Sweden, possibly the first ever!

All I can say without repeating the correct & justified superlatives already surrounding these fine preamps is, after extensive testing and running through the entire microphonepark at Camelot studios since the unit arrived here, I´ve reached upon a number of conclusions wich I never would have without these or possibly Earthworks ZDT or even DACS Clarity...

Beeing a happy Trident S80 owner things have worked just fine and for pop & rock productions this is just the best "flavoured" desk to have I believe.
Comparing the preamps in semiscientific tests last weeks between these 2 "units" in my case was just for showing these differences to myself and perhaps sharing them with the world one day.

What fascinated me most, besides the natural and clean response, was that I actually discovered how UNMATCHED my pair of AKG C 414´s were.
(sold and delivered to me as matched from AKG)
Allright, they are from 1992 ca and maybe there´s a need for recapping them etc but still, this I would never have believed!
With this I´m not saying that my highly-regarded Trident S80 is veiled or unclear in any way, quite the opposite; the reason for its reputaion as the worlds best sounding console w. such musical pre´s & eq´s etc lies of course in the character of a lifted, more shiny yet deep and rich sound.

I also tested SMP-2 against Aphex 207D wich is considered a clean preamp w. AES/EBU direct outs, again, in terms of clarity & evenness throughout the spectrum the Forssell wins of course and while the 207D is not bad either it lacks something here and adds something there thus making it harsh, even if it´s probably something "common" people would not notice...

Anyway, thanks Ivo for your warm and nice manners and for the beautiful music you make, sounds really great on CD´s I can assure the rest of you!

Will have more of these pre´s in the future for sure*