That was just the mainstream taste in that era and probably done on purpose. There's no intrinsic reason why a digital mixer should inadvertantly change the EQ.
Could be. However, my Uncle who has mixed on early digital boards disagrees. I don't know if this violates the member rules but my Uncle Jack is going to respond. He understands this stuff better than me. I asked him to join but he is too busy at the studio. He will not be able to respond afterwards so this is just a guest engineer talking about his experiences. My uncle dropped out after Grade 6 so his written English is....not up to snuff. He typed it, E-mailed it and then i copied it into here. "......Hello everyone. I am Jack Morris audio engineer of 53 years. I hope you are all well. Old digital boards could sound nasty if you pushed the eq to hard. a lot of engineers said the digital mixers colored the sound of the mix. I found them to be cold and without feeling. Spec wise yes very good but those very early digital boards did something to the midrange. Hard to put into words. If you want 3 db of 8 000hz eq it could sound like you used twice the eq on early digital boards. it was not something you could measure. The eq was very angry. Very agresive. They wanted them because very quiet. No hiss or noise. Full automation without VCA. The time during 80's was to mix very loud. You get tired soon. Companies put out monitors that had a easier sound at the top end and mids to ease mixer burn out. On these speakers you would mix heavier in the top end because the monitor was laid back in the 8 to 20khz area. With monitors everything is backwards. Mix on bass heavy bright speakers and the mix will sound bass shy with little brightness. So this could be reason for bright mix of album too. .." Sorry about the English. I don't know how much sense that made. All the Digital mixers I mixed on (in or out of the box) sounded great. I had no experience on first generation digital mixers. But my Uncle did.
All early Polygram CDs have silver painted edges. Probably because many early Philips CD players had the CD visible and therefor prone to outside light coming in and disturbing the readout. So yes, edge treatment can have an effect as even Philips aknowledged that by painting the edges.
Ive never understood that! Surely with digital music it's either there or it isnt. You can't enhance a Bit of music at the reading stage.
I always thought that early Cd's were generally mastered bright because record companies wanted so show off the new format and show how crisp and detailed it was compared to the old analogue formats? Not sure if the early 16 track Sony digital recorder/mixer that was used on the Dire Straits -Brothers In Arms recording sessions was first generation? I presume it was. Because that recording (especially the WG copies) sound superb and as good as anything recording or mixed on later digital equipment.
That's a fair description of the Telarc sound. Telarc recordings have macro-dynamics that will impress. But miss out on the subtle and the things that can help make a recording sound real. The macro-dynamics will impress (those bass drum hits!). But the rest of the music ends up boring and lacking.
Please explain how oversampling in the DAC solves the design problem for the real-world anti-alias filter needed in the ADC (or decimator if recording at a higher sampling rate)? You first have to create the 16/44.1 signal before you can put it on a CD. Firstly, will extra bits make any difference, since your position up to now has been 'no'? The word-length defines both the dynamic range and the resolution, which is the smallest signal change which can be coded, which is separate from the noise floor. You are convinced that the 98 dB dynamic range of CD is overkill for domestic situations, so how much dynamic range is needed? 80 dB? 70 dB? 60 dB? If we apply your logic, and assume 80 dB, we would need 13-bits, and if all the word-length is doing is defining the noise floor, we would not be able to distinguish between 13-bit and 16-bit audio at the same sampling frequency. Do you agree? Using your logic, I assume you do not believe in dither either? Which is why I specifically referred to quiet rooms and headphones, however, see the above thought exercise on reducing word-length to fit the dynamic range of the listening environment. You did not respond to my question on why you mentioned needing an anechoic chamber? No one has responded to the paper I linked to showing that listeners can distinguish hi-res from CD quality. I wonder why!
Oversampling ups the internal sampling rate by artifically computing a sequence with higher sampling rate from the original samples, simply by interpolating. The DAC then actually operates at that new higher sampling rate. This relaxes the requirements for the analogue filter stage, because the interval between the audio you want to keep and the undesirable artefacts widens considerably. In this respect it acts exactly as if the audio had been sampled at the high sampling rate originally, but without the need to lay down unnecessary data and all while remaining compatible with conventional CDs. What's your point? No it hasn't if you read carefully. I said several times that additional bits lower the intrinsic noise floor of your system. What you need to understand is: Those are two different words for the same thing. I understand how intuitively you think that the audio is defined "less precisely" with fewer bits. In a way that is true, but the "lower precision" manifests itself by adding more noise. That is all it does. Subjectively you hear the music just the same without perceived distortion, but the hiss gets louder. That's a grey area question and to an extent a matter of taste, so my answer would be as good as anyone else's. What I'm saying is that 96 dB should satisfy the most ambitious demands, it is not reasonable to demand more for home listening. Not entirely. We would hear the different noise floors under the right circumstances, for example in a very quiet room or with good headphones. How does that follow from anything I said? Dither is actually a necessity in a properly designed DAC, and I'm assuming it in place. I meant to say a room devoid of ambient noise. I think it was discussed here before. They will be able to distinguish the ever so slightly higher noise of CD, if the volume is cranked up really loud. That is not disputed. Otherwhise, doubtful. There have been many tests showing that a 16/44 ADC/DAC chain is transparent to the majority of people under normal conditions.
Out of curiosity, I exported a few songs out of Audacity as an 8-bit PCM file. The higher noise is immediately noticeable, but what's remarkable is how much it sounds like tape hiss. It's almost on par with the tape hiss present in old jazz records like Kind of Blue. If you take that 8-bit PCM file and invert it relative to the 16-bit original, what you're left with is nothing but noise. There's no extra "reverb trails," no "magic" or "warmth" left over. Just noise. The 8-bit PCM file has managed to capture every bit of the original sound, minus its own intrinsic noise. I don't know how people can say analog tape has all this magic built into it, when really the tape hiss you're hearing is the analog equivalent of quantization noise. I would go out on a limb and say all vinyl records can be transparently represented by 10 bits of resolution, and all albums recorded to tape with no NR can be represented by 12-14. The ol' Audacity inversion test reveals a lot about what you think you might be hearing. If you can't invert and isolate the "magic" you get with higher bit depths, is it really there?
Actually it was vinyl that was mastered bright, to make up for the upper midrange droop in the response of many phono cartridges. But CD players had no such deficiency, so when the same master was put on a CD, it sounded strident due to that unnecessary midrange boost.
Immediately noticeable? With noise-shaped dither, 8 bit can sound pretty good: brothers-8bit.flac Imagine what it does for 16 bit...
Throw the baby out with the bath water very often? You're sure putting a lot of faith in your test. Now care to take a SACD version of Kind Of Blue and do a listening comparison with it dumbed down to oh, maybe redbook? See if you can hear a difference when its the same material, a higher rez version and the same thing pulled down to redbook. You might be surprised. CJ
There's a constant high-pitched tone, so I don't know if you did your dither process properly. If I really was throwing the baby out with the bathwater, I would have converted all of my music library to 12-bit FLACs. As far as listening is concerned, I don't have a SACD player, but I do have a DSD-capable DAC. I used to stream the ripped DSD files to my DAC through my PC. These days I just put the discs in my CD player, because I really couldn't tell a difference, and I like the tactile process of spinning CDs better... Not really "dumbed down" at all; at least, not dumb enough for human ears to tell. Hundreds of comprehensive scientific listening tests done by others seem to support this idea too.
There's not a constant tone, but there is noise expected in an 8 bit file, just shifted up 14k-22kHz (the noise-shaped part) where the threshold of hearing makes it less objectionable. Good ears on this one, but also showing that our ears have little ability to discern the character of high frequencies, as they are all sine waves when there's no harmonic that can be heard.
All you have to do now is find a paper that shows there are a significant number of such listeners, and another paper to show that those listeners actually prefer the sound of hi res to CD, and you’ve won!
At the beginning CD players were aimed at audiophiles too and I don't know in the US but in Spain and Europe in general they cost a leg and an arm back in the 80's so not everybody could afford them. I got my first CD player back in 1988 when I was almost 15, I spent two years saving money from a weekend job I got and even with my two years of savings my father paid half of this CD player, which was a Top loader Toshiba I fell in love with at the moment (I actually didn't fall in love with it, I just wanted to se the CD spinning at a high speed). If memory servesme well Rush' Presto was a digital recording from the late 980's so no Hi Res here. The differences you say you're hearing on Mr. Hoffman's hybrid SACD of it are surely different equalization moves for each version. Mr. Hoffman doesn't do a DSD mastering for SACD and this DSD mastering is donconverted to 44.1/16 for the red book layer, he masters each resolution/layer separately and most likely differently.
I suspect you might be wrong there, as IIRC, the signal was split (after SH had finished mastering it) into the two formats - same mastering, different formats. Please, somebody, feel free to tell us if my memory is wrong here.
Why on Earth would be do that? What possible reason would there be to have the two layers sound noticeably different? Unless he just like twiddling knobs, of course.
Guys, really with this stuff? I master something once. As the analog song is being mastered, a split feed takes the signal to the DSD and the PCM units at the exact same time. Nothing else is done. No different mastering for one or the other. Always the same, one is just higher def than the other. The way I've always done dual layer SACDs. Is we clear??
Haven't you gone on record saying you have to use special tricks to keep the CD layer from losing reverb trails, or something to that effect? I know you've used the word "tricks" before when referring to CD mastering. Surely you don't also apply the same tricks to a DSD transfer? I mean no disrespect.
Steve posted this almost 11 years ago as a way to hear the difference between redbook and DSD hi-res of a single mastering (link here): This would seem to me his assertion that there is more in the signal that hi-res can capture than 16/44.1 redbook can.
