John Curl's Blowtorch preamplifier part IV

Status
Not open for further replies.
You create an impasse here, the industry at large or "we" have to spend 100's of man hours and 1000's of dollars to refute casual uncontrolled listening impressions of a few people. How about you finance a program to support your claims.

Already been mentioned numerous times, roughly 1/3 of audiophiles are highly neurotic.

Don't know anything about the tests you refer you, what the research question to study was, who you tested, or what equipment and methodology, etc., was used.

Research on humans is complicated and expensive, no question.

Early Geddes came up with the 5% number, probably given the usual 95% number once thought sufficient for human studies.

I don't know what the right number is. I do know that there is some portion of the population who have never been the focus of study. Not enough of them to matter in high volume markets, I would imagine.

However, for long term listening a larger number people probably are affected by some aspects of sound quality. It doesn't have to be at a conscious level to influence listening habits. People are influenced in many ways without conscious awareness having a clue that influence and behavior change occurred. It is of course complicated to study like many other aspects of humans.

In more direct response to your the issue you raise, I have already stated I am not interested in spending the rest of my life involved in studying human perception to settle an argument with a few people. Its not what I want to spend time and limited resources on. I have other priorities.
 
Last edited:
PET is not super accurate either when the scan goes longer.

These are both metabolic indicators anyhow, which requires a priori knowledge of the neural pathways around that area to have a remote chance of telling us much. Injury and decay and gross physiological function? Absolutely useful.

I would just love some sobriety being brought forward. Just freaking enjoy the music, folks.
 
Lots of hate on op-amps from the discrete crowd, no matter how it performs.
There are historical reasons for this. As the first solid state amplifiers on the market were sounding bad, hence the preference of some audiophiles for tubes, the first 741 and TL07x were not fantastic at all.
Too, the word "perform" question me. Isn't the quest for amazing performances on paper another form of audiophile snobbism?

On this point, I see no difference in behavior between the extremist objectivists and the subjectivists of the extreme opposite.
 
Last edited:
Administrator
Joined 2004
Paid Member
At the time of their introduction, the TL07x series sounded amazing in a great way. Much lower hiss levels and distortion. It was a miracle part. I used many back then.

The 741 didn't sound much different from early solid state. It made some new circuit possible. It really did spur designers onto a better part. I remember using 301 and 709 op amps for audio instead of the 741 in my designs that weren't discrete.

-Chris
 
At the time of their introduction, the TL07x series sounded amazing in a great way. Much lower hiss levels and distortion. It was a miracle part. I used many back then.

The 741 didn't sound much different from early solid state
I used a lot (thousands) too, but, Lord, how to compare any of the numerous mixing desks using them with, by example, any NEVE, that are, yet today, considered as one of the best sounding mixing desk ever. The Rupert Neve desks were discrete solid state. The problem of TL07x was noise and lack of current.
741 was a real mess: not enough slew rate, noisy, lot of IM.
That you can not notice as obvious, using one or two in a preamp at home with low gain, can be catastrophic when you have hundreds of them playing with real direct sounds of instruments.
 
Last edited:
Yep, because the op amp is a black box with little to no room for a story to be created. In high end audio, performance doesn’t sell, stories do.

To reiterate: the vast majority of recordings people are using to evaluate their hi-end playback systems have been passed through many op amps before being delivered in a consumer playback format. This does not prove that all op amps sound great all the time, and there is little doubt that highly tweaked discrete circuitry can perform as good or better than some op amp circuits. It does however mean that whatever the "special" quality that people are able to extract with their discrete playback circuitry has not been eliminated by properly engineered studio equipment employing op amps.

Howie
 
Member
Joined 2002
Paid Member
The very best sound I have heard has been in various recording studios, I don't think people appreciate how much limiting and compression go into a final mix in almost all recordings...
Thank you Howie. It seems people who form the music do understand that the majority of the end users do have restrains of highly dynamic content playback and they act accordingly.
As for the target range, a dated paper but I guess for the most part still valid
https://pdfs.semanticscholar.org/dc75/f7f82a580fe00cdb701ed6d30efe01ae6cd5.pdf


Thank you Allen!

George
 
www.hifisonix.com
Joined 2003
Paid Member
@billshurve “Thank you for reminding us of this Howie. Easy to forget that even recordings that say 'no compression' have probably had some gain riding. I suspect/fear that quite few recordings that people report to sound 'dynamic' are anything but in reality but have the right secret sauce applied in the studio.”

My Claude Adabbo 1976 LSO recording of Stravinsky’s Firebird Suite has one of the largest dynamic ranges of any record I have. I often wonder if they also did some gain riding - hard to tell if they did because if it was it’s been done very well.
 
I sometimes read strange comments on limiters / compressors used in studios. Their use on an instrument or a voice is not to reduce the overall instant dynamic, they can even increase it if the attack time is long. It is more to increase the presence and the intelligibility of the voices, for example, which allows them to be moved back into the mixes, leaving more room for the other instruments. Or to "sculpt" the sound of electric guitars, for example, like guitars bump (cocotes)* or the typical sound of Mark Knopfler.
Basically, the idea is not to reduce the peaks, but to raise a little the low levels without touching the instantaneous dynamics too much. And it is rare that they compress the overall dynamic of the mix.
Most sound engineers I know love the max instant dynamic they can have. Don't put the blame on them.

It is at the time of the mastering that the massacre often takes place with fast limiters which notch out the instantaneous pics, and compressors on the global mix which reduce the dynamics so that the "product" goes louder on the radio or in noisy environments at low listening levels.

Listen to this tune, setting the voice and cymbals at normal level: The brass section will kill you like in real life if your system is loud enough to afford them (Scott ;-):

YouTube.

* YouTube
 
Last edited:
In more direct response to your the issue you raise, I have already stated I am not interested in spending the rest of my life involved in studying human perception to settle an argument with a few people. Its not what I want to spend time and limited resources on. I have other priorities.
But you seem happy to spend your time repeatedly saying you perceive something as better whilst apparently understanding that it is meaningless to anyone but yourself
 
Thank you Howie. It seems people who form the music do understand that the majority of the end users do have restrains of highly dynamic content playback and they act accordingly.
As for the target range, a dated paper but I guess for the most part still valid
https://pdfs.semanticscholar.org/dc75/f7f82a580fe00cdb701ed6d30efe01ae6cd5.pdf
George
George,
At that time, almost 30 years ago, they were struggling with 20 bit A/D’s that probably even had no real 20 bits DR.
And when mixing their N digital channels, noise increased with 10logN accordingly.
I’m almost sure that with the recording equipment used today and refined noise shaping algorithms, we have enough DR to deal with.
But isn’t the tendency nowadays to compress (non classical)music and make it as loud as possible instead of using the whole DR ?

Hans
 
.
And when mixing their N digital channels, noise increased with 10logN accordingly.
Not really, Hans. For several reasons. If, on the multi track, instruments are all recorded at max level on the tape, they are played in the mix at various ones. When you mix two of them at equal volume, the overall level increase of 3dB, and the noise equally: the signal noise ratio remain unchanged.
If you add the fact that the tracks are muted between the moments the instruments plays, the final mix has a better signal noise ratio than the worse original track. Considering the mask effect, it explains why you do not notice the amp noises of the electric guitars amplifiers, by example, that are obvious when played alone.
Digital noise is no more a concern, and Mister Dolby can take his retirement.
If most sound engineers ask for as much bits as possible, it is not because a noise issue, it is because they worry about the definition of instruments played at low levels.
 
Not really, Hans. For several reasons. If, on the multi track, instruments are all recorded at max level on the tape, they are played in the mix at various ones. When you mix two of them at equal volume, the overall level increase of 3dB, and the noise equally: the signal noise ratio remain unchanged.
If you add the fact that the tracks are muted between the moments the instruments plays, the final mix has a better signal noise ratio than the worse original track. Considering the mask effect, it explains why you do not notice the amp noises of the electric guitars amplifiers, by example, that are obvious when played alone.
Digital noise is no more a concern, and Mister Dolby can take his retirement.
If most sound engineers ask for as much bits as possible, it is not because a noise issue, it is because they worry about the definition of instruments played at low levels.
Sorry,
You’re the expert in this field, but I have problem understanding your “improving S/N” explanation
When using two microphones, why should the S/N level increase in case only one instrument close to the mic is producing sound, even when muting the other channel? Nobody worries about S/N in loud passages, exactly the quiet ones are important
And worries about Definition instead of Noise, what does that mean ??

Hans
 
Last edited:
Member
Joined 2002
Paid Member
George,
At that time, almost 30 years ago, they were struggling with 20 bit A/D’s that probably even had no real 20 bits DR.
True, Hans. I tend to forget how far back in time 1990 was :D (*)
Although in technical terms studio situation is very advanced compared to 1990, the methodology hasn’t change (record/mix at as high bit depth as available, then master and deliver at 16bit for the final DR target of around 100dB with the help of dither)

But isn’t the tendency nowadays to compress (non classical) music and make it as loud as possible instead of using the whole DR ?
Yes and this can work as a reminder that we should not confuse high bit depth (and the high dynamic range that it has the potential to provide) with the high SPL level.
The benefits of high bit depth were/are evident at the low level information. At recording/mixing, this is advantageous. At playback, high bit depth can be a blessing or a curse.


(*) The technology of that era has made what is available in CDs of all the classical/opera/jazz great performances of 1940 t0 1990. We have more than enough broad base now to conclude that bad CD transfers were due to bad practices/faulty decisions and not due to inadequate technology (**).
(**) Sony and lately Warner have made some excellent work on remastering renown performances of great artistical value. So, some dubious remastering past work can go where it belongs, to the trash bin (call me ###Classics)

George
 
Member
Joined 2014
Paid Member
As for the target range, a dated paper but I guess for the most part still valid
https://pdfs.semanticscholar.org/dc75/f7f82a580fe00cdb701ed6d30efe01ae6cd5.pdf
George


Is that the same Mr Dunn of the jitter test waveforms?


My Claude Adabbo 1976 LSO recording of Stravinsky’s Firebird Suite has one of the largest dynamic ranges of any record I have. I often wonder if they also did some gain riding - hard to tell if they did because if it was it’s been done very well.
I'll assume autocorrect is playing up today and you meant Abaddo? It's an interesting question as some on the internet claim the late 70s was a low point for DG as they were using huge mixing desks, spot mikes everywhere and even polishing the laquers to reduce HF peaks. The end being what the conductor wanted the soundscape to be not what was there (at least with HvK). Not Decca phase4 but possibly not realistic either.



Interesting one to persue. That recording hasn't been uploaded to the dynamic range database, but the most dynamic firebird they have is RR-70, which did win a grammy in 1977. annoyingly its HDCD so you'd have to rip and then decode to get the full effect.



-------------------------------------------------------------------------------- Analyzed: Oue (Eiji), Minnesota Orchestra / Stravinsky: Firebird, Song of the Nightingale, Rite of Spring -------------------------------------------------------------------------------- DR Peak RMS Duration Track -------------------------------------------------------------------------------- DR18 -0.19 dB -25.73 dB 21:15 01-The Firebird Suite (1919 revision) DR21 -0.86 dB -30.80 dB 20:13 02-The Song of the Nightingale DR16 -0.01 dB -22.82 dB 33:30 03-The Rite of Spring --------------------------------------------------------------------------------
 
www.hifisonix.com
Joined 2003
Paid Member
No, not auto correct - just my spelling. Yes it’s Abaddo

I’ll have to try the Minnesota one.

Agree some of the DG recordings can be flat. I’ve got the Berlin SO doing Mozart’s 40 and 41 conducted by von Karajan and I do find it very flat. But the Abaddo recording i mentioned earlier is a different league. Another good one is Tchaikovsky’s Swan Lake/Sleeping Beauty- also Berlin SO.
 
Status
Not open for further replies.