Based on sonics... which do you prefer ?

Based on sonics which do you prefer.

  • Ruby

    Votes: 14 42.4%
  • Opal

    Votes: 19 57.6%

  • Total voters
    33
  • Poll closed .
Status
Not open for further replies.
This may indicate why, in another threads, master tapes are regarded superior or vinyl is regarded superior above even hires digital. IMO, it is the smaller information content what is in fact preferred. Hiss (dither) or 1/f groove noise mask the real signal details, + added distortion in both tape and vinyl adds originally non-existent information. I may be wrong, but it is my opinion.

Technically your argument is more logical. But from what I heard, listening perception only, Ruby has much more "information". Is the "noise" the information? Doesn't have to be. If the "noise" is 50% music + 50% junk then removing the "noise" is removing the music.

I know the perceived sound of noise in amplifiers. It is different from the "information" I can hear in Ruby. Too many actually. The emotional content and the nuance of the vocal makes the singer in Ruby more like human than "human in a box/can" the singer sounds in Opal (which could possibly be just the sound of one bad capacitor)
 
For that sample I'd want it to go through the tape deck a few more times 🙂 There are some bursts of sound that still have a very hard edge and are not pleasurable at all.

That was a creative decision, they are supposed to be hard edged. 🙂 The cassette may even have made it subjectively harder by distorting on peaks: IIRC I drove it right into the red for more compression.

Also IIRC, I used the same BASF Chrome tapes available from LIDL as shown earlier in the thread, and the deck was one of these: Uher CR-240 Stereo Single Cassette Deck Buy Now
 
Last, the perceptual impression depends on audio chain one uses, including sound card.

Yes, most Fostex speakers used by tube amp users are even more disturbing to my ears. But from my experience, my perceptual impression is not much affected by source or even amplifier but speaker.

Well, certain amps, like some CFA, in simulation perform equally well with different load variations, and in real life they work very well even tho the speaker is cheap.
 
There are are at least 3 different 'quality levels' I can listen comfortably to: first is everyday sound - car radios, small TVs, record store speakers, old cassette recorders; next up is decent hifi, as in products from mainstream Japanese manufacturers which are not pushed to higher volumes; and finally is highly optimised, tweaked audio which achieves convincing realism. Each can be enjoyed for what it is, one doesn't have expectations beyond what past experiences have told one is possible in each area - and so one is quite happy with each performing within those limitations.

The worst offenders, in my experience, are the more expensive, pretentious, ambitious setups which scream at you to pay attention, they needle every nerve ending, and are extremely uncomfortable and unpleasant to have around ... they make listening to music a chore ...
 
What's the point of that burst at 50s? Were you involved in creating the original or did you just have to try to turn it into something marketable?

I created the whole thing for fun as an amateur electronic musician. I wasn't really planning on marketing it.

Getting involved in music production is an eye-opener. I think every hi-fi enthusiast should see where the mystery meat in the sausages comes from. 🙂
 
........So far I have not taken part in these tests. Two reason for this:
1. I have no convenient way of playing downloaded files except through my laptop itself and I certainly would not expect to be able to distinguish anything but the worst problems with that..........

You could burn them to CDRW to make an audio CD using media player which in fact is what I do for some of the tests.

Not necessarily too late, IMO.
I have posted 2 files recently, record from a cassette tape deck and CD data from the same live concert.
Cassette file has audible hiss, it has measurable roll-off above 10kHz and it has smaller dynamic range. Despite of that, there is a preference (heb1001) for the cassette file. This may indicate why, in another threads, master tapes are regarded superior or vinyl is regarded superior above even hires digital. IMO, it is the smaller information content what is in fact preferred. Hiss (dither) or 1/f groove noise mask the real signal details, + added distortion in both tape and vinyl adds originally non-existent information. I may be wrong, but it is my opinion.

I have heard of this too, that the ear can seemingly winkle more information from the noise. The pianist Alfred Brendel remarked that he preferred his earlier analogue Beethoven cycle to the later all digital edition and that the analogue master and slight hiss added a "patina" to the playback experience that he found preferable.

Technically your argument is more logical. But from what I heard, listening perception only, Ruby has much more "information". Is the "noise" the information? Doesn't have to be. If the "noise" is 50% music + 50% junk then removing the "noise" is removing the music.

I know the perceived sound of noise in amplifiers. It is different from the "information" I can hear in Ruby. Too many actually. The emotional content and the nuance of the vocal makes the singer in Ruby more like human than "human in a box/can" the singer sounds in Opal (which could possibly be just the sound of one bad capacitor)

Remember that the overall response of tape has quite wide amplitude vs frequency characteristics. Even on an optimally set up machine such as this the response will vary considerably as equalisation and bias anomalies vs tape formulation come into play. A characteristic of the ferrite heads on a deck such as this are that they can saturate relatively early. One reason for me not recording at too high a level.
 
heb1001 said:
If you are forced to make a choice based on an emotional response but can't differentiate on an emotional response then you are forced to differentiate some other way which can give spurious results.
You were not forced to make a choice. You were asked to make a choice, based on sound alone. If you could not make a choice on sound alone then either you say nothing or you say so, at an appropriate point in the proceedings which will not disturb things for others. Analysing the files and desperately trying to make a choice on other grounds is not what was asked for.
 
Getting involved in music production is an eye-opener. I think every hi-fi enthusiast should see where the mystery meat in the sausages comes from. 🙂

Do you think this is true for classical music as well?

There's maybe a limit to how much effort you should go to to reproduce the sound of something that has been looped through an old tape recorder but I've noticed that since I built my amp and bought my first set of good speakers I've started to see the point of classical music and actually bought some.
 
You were not forced to make a choice. You were asked to make a choice, based on sound alone. If you could not make a choice on sound alone then either you say nothing or you say so, at an appropriate point in the proceedings which will not disturb things for others. Analysing the files and desperately trying to make a choice on other grounds is not what was asked for.

You should read the thread again. Mooly explicitly said that those people who wanted to look at the waveform should go right ahead.

And I did think I had already said I had no preference and felt I was being pressured to make a choice.

We're all interested in different things. I'm interested in learning how to relate the subjective impression of the sound with the analysis of the files because it's going to help me build better amplifiers. If you are not interested, there's no need for you to reply to my posts.
 
Cassette vs. CD

This analysis might be interesting as a CD x cassette difference. Red line i cassette, blue line is CD. It is that Clapton file that I posted as cassette and CD recordings. FFT over complete song, with peak hold. See how central frequencies are quite equal in spectrum, but both high frequency and low frequency roll-off for the cassette file.
 

Attachments

  • overall_fft_clapton.PNG
    overall_fft_clapton.PNG
    90.7 KB · Views: 96
Do you think this is true for classical music as well?

I only have experience with rock, pop and electronic music, but I would say no. In modern music the recording studio is practically an extension of the instruments. In classical music there is a very clear divide between the playing of instruments, and the capturing of the sound that comes out of them. I would say you don't "produce" classical music, you "record" it.

There is still a lot of creative freedom in terms of mic choice and placement, but classical music listeners probably wouldn't accept anything that sounded too different to the live sound you hear in a concert hall. The original hi-fi enthusiasts were arguably trying to recreate the concert hall experience in their living rooms.
 
Isn't this the whole point behind the Shannon-Nyquist theorem? You send a series of samples, then reconstruct them using the equation sin(x)/x. The reconstruction gives you back the original analog signal, or as much of it as will fit inside the Nyquist bandwidth. The analog reconstruction filter is what performs the conversion between discrete and continuous time.

You have the same "Nyquist bandwidth" problem with representing the music as a gigantic equation. The greater the information content of the music, the more terms your equation needs. (What is the equation for 3 seconds of white noise? What would it look like for different noise bandwidths?)


I find the recent listening tests interesting. The impression I get is that nobody has a clue what is going on until the results are revealed, at which point it degenerates into a frenzy of ***-covering. This is exactly what the "rationalist" theory of audio predicts. 😛

In my previous life as an electronic musician, I have "mastered" tracks to cassette to dirty them up. 🙄

Darn! You're quite right. I'd still have to pick an upper frequency limit for the equations, so it gets right back to Nyquist. Damn you Nyquist!

I still think it would be nice to have a standard direct recorded test piece for these tests. The same original would be used for all tests by anybody who wanted to contribute. The originals here have distracting flaws that everybody is focusing on, whereas the interesting stuff, at least to me, isn't being talked about. Something with acoustic instruments, plus a level tone for matching and a pink noise segment at the end for SA. Something recorded with a large diaphragm condenser mic, because they can have a crazy low noise floor. Alas, I play no instruments, only records, so I can't supply this.

Way back in college, and after, we recorded a lot of LPs to cassette. Not super expensive cassettes, TDK SA on a Harmon Kardon CD91 or similar. With everything set up well, those (Dolby) recordings were nearly hiss free and matched the LP play almost perfectly, and my hearing was much better back then. I think one has to be careful with vintage equipment because it may no longer perform as new. I know I can't even get belts for my HK that are quite the same as the originals, and they don't perform as well.
 
Last edited:
Hi been following this also interesting thread, think both the results and all nit picking is interesting. Think the nit picking is inevitable when done with members that have knowlegde in both circuit creation and studio production. But as said by others it would make best and real progress to poll if only listening/judge/vote is done honestly at first, then after vote had been giving, one can use analyzing tools. Looks like in threads that analysing tools have been used before members is voting, feel strange if one can't listen and vote without using visuel tools.
Thanks Mooly and PMA for assess to good old tapesound. Does any share access to PMA's files, don't know if it's IE11 strangeness but i cant download from "Filefactory".
 
Last edited:
Conrad - yes, they matched LP's 😀

In 1977, as a university student, I worked about 2 months in Supraphon recording studios. Once we made a recording in a church. The live music was recorded to Studer A80 tape recorder and then, a copy was made to one of the best SONY cassette recorders available at the time. You would not believe how horrible was the sound from SONY in a direct comparison with Studer A80.
 
Way back in college, and after, we recorded a lot of LPs to cassette. Not super expensive cassettes, TDK SA on a Harmon Kardon CD91 or similar. With everything set up well, those (Dolby) recordings were nearly hiss free and matched the LP play almost perfectly, and my hearing was much better back then. I think one has to be careful with vintage equipment because it may no longer perform as new. I know I can't even get belts for my HK that are quite the same as the originals, and they don't perform as well.

Yes, exactly. The ruby and opal samples don't sound like tape as I remember it.
 
Status
Not open for further replies.