I am new to recording but have some thoughts afte watching recent mofi debacle, on their recent original master record claim,
t is know that back in the reel to reel tape Era everything is recorded analog is all the neessary harmonic is retained, when we are going to digital era due to sampling their information in the recording is missing.
now I was wondering in todays digital age is one microphone recording all the freqency specturm is not the corect way to record, we should use around 4 microphone ,1 is the cordination recording all frequency, and 3 other miciphone recorde at the freqency specturm they are prefect at. After that the recorded cue will be mixed toether . Or is the r2r and delta sigma ADC method is incorrect
This is a similar idea of sigma camera optical sensor (Foveon), which it has 3 /4 layer which readers a specific light wave in a window then combined togthers (analogue recording), while traditional bayer sensor use interpolation method (digitaltized recording)., however HDR recording is easier at bayer sensor , so differnt optical exposure is used when taking picutre and combine to former an image which rivial Foveon sensor)
t is know that back in the reel to reel tape Era everything is recorded analog is all the neessary harmonic is retained, when we are going to digital era due to sampling their information in the recording is missing.
now I was wondering in todays digital age is one microphone recording all the freqency specturm is not the corect way to record, we should use around 4 microphone ,1 is the cordination recording all frequency, and 3 other miciphone recorde at the freqency specturm they are prefect at. After that the recorded cue will be mixed toether . Or is the r2r and delta sigma ADC method is incorrect
This is a similar idea of sigma camera optical sensor (Foveon), which it has 3 /4 layer which readers a specific light wave in a window then combined togthers (analogue recording), while traditional bayer sensor use interpolation method (digitaltized recording)., however HDR recording is easier at bayer sensor , so differnt optical exposure is used when taking picutre and combine to former an image which rivial Foveon sensor)
Nothing is missing on a digital recording - quite the contrary - digital recording is superior in every sense to the analog.
Microphone technique and recording/storage mechanism has no relation. Your camera sensor analogy is not correct as I see it - both are as digital as can be.
Can you please point to an r2r ADC product?
//
Microphone technique and recording/storage mechanism has no relation. Your camera sensor analogy is not correct as I see it - both are as digital as can be.
Can you please point to an r2r ADC product?
//
thx for the reply, I am not proficient on the subject of ADC products so I can't name one.
To answer /clarify the camera analogy a bit more yes both senosr technique are digitization but the implemtnation result is different. However the overall image , depth 3D, colour accuracy of Sigma sensor (Foveon sensor) is far superior than the the bayer sensor and closer to tradiation phtography but Sigma sensor has lesser resolution. So I am thinking we have been implementing the wrong way in the process of recording and digitalitizng sound wave else we shuoldnt been taking 30 years to achieve the near quality of reel to reel.
So the camera analogy should be
Flim base photography = reel to reel /vinyl
digital based technology = 16bit PCM /dsd
I am not very good at ADC hopefully what I am thinking and have research os making sense.
To answer /clarify the camera analogy a bit more yes both senosr technique are digitization but the implemtnation result is different. However the overall image , depth 3D, colour accuracy of Sigma sensor (Foveon sensor) is far superior than the the bayer sensor and closer to tradiation phtography but Sigma sensor has lesser resolution. So I am thinking we have been implementing the wrong way in the process of recording and digitalitizng sound wave else we shuoldnt been taking 30 years to achieve the near quality of reel to reel.
So the camera analogy should be
Flim base photography = reel to reel /vinyl
digital based technology = 16bit PCM /dsd
I am not very good at ADC hopefully what I am thinking and have research os making sense.
I agree with TNT. Current digital recording is far mor accurate than analog.
You may have a personal preference for one or the other, and they can certainly sound different, but the numbers are what they are.
Reel to reel can sound very nice but not due to accuracy, it sounds nice for several reasons, one is the gradual overload characteristics.
Overload characteristics of digital recording tends to be harsh, but a good recording egineer knows how to avoid it.
Jan
You may have a personal preference for one or the other, and they can certainly sound different, but the numbers are what they are.
Reel to reel can sound very nice but not due to accuracy, it sounds nice for several reasons, one is the gradual overload characteristics.
Overload characteristics of digital recording tends to be harsh, but a good recording egineer knows how to avoid it.
Jan
In photography, it's like in audio - once you have run it through Photoshop, there is no difference and you can create any look (inc. Foveon) as long as the source material is as truthful to the original as possible. Photoshop = EQ ;-)
The Foveon has some (good?) distorsion, thats why we like it - this is also a common aspect with sound - the more 2nd (good?) and 3rd and some added bass , the merrier 🙂
//
The Foveon has some (good?) distorsion, thats why we like it - this is also a common aspect with sound - the more 2nd (good?) and 3rd and some added bass , the merrier 🙂
//
Well recorded CDs, ones made with some of the best recording equipment made today, can sound quite good. In other words, the technology exists to encode digital data with very high quality. Cost may be a factor in doing it well, however.
The other part of digital reproduction, dacs, seems to be where more technical problems remain. Cost is also a factor if subjective sound quality is more important than time-averaged steady-state FFT measurement numbers (because we know such measurements are not fully predictive of SQ). Some remaining problems include resolution limits of R2R dacs, and audio-signal-correlated modulator RF noise present in the analog outputs of sigma-delta dacs (because output stage opamps vary in tolerance for RF noise).
The other part of digital reproduction, dacs, seems to be where more technical problems remain. Cost is also a factor if subjective sound quality is more important than time-averaged steady-state FFT measurement numbers (because we know such measurements are not fully predictive of SQ). Some remaining problems include resolution limits of R2R dacs, and audio-signal-correlated modulator RF noise present in the analog outputs of sigma-delta dacs (because output stage opamps vary in tolerance for RF noise).
Last edited:
The never-ending argument over 1's and 0's as opposed to analog.......and if it's better.....
Sure, there are "apparent" differences.
But we hear, speak, and are born with the ability of analog sound, naturally.
You choose your own poison.
Sure, there are "apparent" differences.
But we hear, speak, and are born with the ability of analog sound, naturally.
You choose your own poison.
There's two things wrong here. Firstly, a digitized signal doesn't look like that, it's just a drawing convention. A common fallacy.
Secondly, hearing is intrinsically a digital process. Haircell neurocells firing pulses that vary with intensity and frequency, a mix between PPM and PAM. You can't get much more digital than that.
If you start with wrong understanding then you get to a wrong conclusion.
Jan
Last edited:
A lot of people may indeed prefer the sound of original recordings done on analog gear and multi-track tape. The recording engineers knew how to get the most out of the equipment available at the time. A lot of the "sound" was analog processing, mic input preamps, compressors, EQ and effects. Even today a lot of engineers will record digitally but prefer to use some key outboard analog gear (or digital VST plug-ins that emulate them) to "warm up" the recording. At 24 bit 96 kHz sampling I cannot tell the difference between a live feed of a mic'ed guitar and the recorded playback. It sounds very present, accurate and live.. which is okay technically but not necessarily what you would want to hear on every instrument all the time throughout an album. At 140 dB of dynamic range that level of detail may be too tiresome to listen to..but it is brilliant in capturing the recordings almost perfectly. It's how you treat them in the mastering stage and what the ultimate purpose of the recording is. Producing a blues-rock album, recording a live symphony orchestra performance, or archiving the sound of a stradivarius violin for analysis, these all have different end goals. but most recording engineers would use digital recording over analog tape these days.
No razorblades needed for editing, singer doesn't have to sing in tune, drummer doesn't have to play in time...
Would you count an FM radio signal passed through a limiter (in the RF sense, a hard clipping amplifier) as digital? I wouldn't, because it is only (more or less) discrete in momentary value, but continuous in time.
I don't agree. I find digital recordings mimic what I hear in a concert hall to a greater degree than analogue ones.
//
You appear to be mixing concerns over recording (signal capture technology), with recording (and reproduction) technique. As others have correctly pointed out, digital is far better performing at recording/signal capture, than is analog. Where I suspect there may be opportunity for significant improvement is in the area of recording (and reproduction) technique. Such as, perhaps, Ambisonics, and the like. Which are not simple surround sound. While I'm not knowledgeable enough to offer any intelligent suggestions for advancement in this area, it seems to me that we have, essentially, reached the limits of reproduction verisimilitude enable by a simple two-channel final mix, and playback.
IME it's very easy to make a superb sounding digital recording. However it seems that many people put a huge effort into making them sound bad. 😛
Digitally, you can't record anything above half the sample rate, because when you try, it will alias to some frequency below half the sample rate.
There is a long debate going on about whether what happens above 20 kHz is at all relevant for human listeners and if so, what is to be preferred: as extended a response as possible, smooth roll-off with a short impulse response or minimal ultrasonic preringing?
With 192 kHz sample rate, you could theoretically record up to 96 kHz, but 88 kHz is a more practical figure because of filter transition bands.
I don't see what the advantage of your four microphone technique would be. You do need four full-range microphones to characterize the sound at one point in space including all directional information (scalar sound pressure and three-dimensional volume velocity), see Ambisonics.