The way I see it is that the overwhelming majority of music out there is PCM (or analog) and there’s no way a rational person is going to ditch all that just to have a handful of DSD recordings. And yes, the recording is everything, perhaps even more so than the format itself. I’d rather listen to a well-recorded track on 16/44.1 (or even 320 kbps MP3) than a poorly recorded track on DSD.
Listening to music is a subjective experience and the competence of the engineering is critical. That said, the idea that red book is "good enough" assumes that DSD is a marketing conspiracy (mqa for example). I was shocked at the detail in the mofi dsd64 transferred vinyl of Bill Evans & Jim Hall "Undercurrent" when I upgraded my turntable with a Grado Platinum Tibre. I record my own music at 192Khz 32bit and listen to vinyl rips all the time. CDs sound amazing on high end gear but HR sounds good on just about anything worth listening too. I believe Paul and Kevin Grey when they say DSD256 is better than CD quality.
So basically all of Octave Records material has been through DXD (24/352.8kHz) which is hi-res PCM. That means at least 2 conversions from DSD capture -> DXD mixing/editing -> DSD final purchase. Why not just DSD capture (since you believe this is better) -> DXD and then sell the final album as DXD PCM without getting it back to DSD and save yourself another lossy conversion step??? This also implies that the best quality recordings from Octave Records should be the 24/352.8kHz download option, not the DSD ones!
Absolutely amazing how all these "experts" in the comment section constantly try to correct each other. I know Paul encourages people to comment on the topics and have a pleasant exchange of ideas, but it gets down right comical some times.
Excellent discussion, there is the 8bit DSD in the DSD official specification designed for editing, both 64 and 128FS , 8bits per sample is enough to do all the editing maths. It is a matter to implement it natively on the chip and it was also done a few times, Sony e-chip for instance. DSD by nature is very ambient and warm, however, it tends to saturate at high volumes just like analogue audio when PCM would exhibit digital harshness. High volume is the ultimate test to appreciate the difference and essentially hear it. 7:02
What you described as the last point is exactly what has been done as "high-end" in the analog era: "Direct-to-disc". It's basically the same principle as the early recordings - where there was no mixing at all and the needle just scratched the original recording into wax or whatever material.
@1:44 "If I said..." Paul, you never said that. @4:31 "If you start with DSD..." That is what you have said in the past. @5:40 "If you were to record..." That is also what you have said in the past.
As always I enjoyed your offering Paul. And I certainly enjoy al the comments you trigger. At the same time I wonder when some of the people reacting here will start with listening to their music and stop chasing the ultimate sound. It's almost pathetic in some cases. No matter how much money and time you spend finding the ultimate system, it's just not there. Often good is good enough.
Hey Paul... I don't know if you ever knew Brad Miller who at one time way back when ran Mobile Fidelity Sound Labs. Back in the late 80s or early 90s he came to the studio I was working at in Portland, Oregon with his Colossus Digital Recorder... a 4 channel unit with single ended (unbalanced), discrete electronics that sounded incredible. At the time we had a Mitsubishi X-86 two track that I had replaced the filters with Apogees. I have to say that Brad's recorder was markedly better, but I still believe that it was a PCM unit. Unfortunately Brad passed a few years later and I believe that nothing ever happened with the Colossus. I also remember the Soundsteam digital recorder... wasn't that a DSD unit? Anyway... great video, Paul! Cheers!
Thanks, Jim. I never knew him but knew of him. And I assume it was PCM. The first commercial DAW was the Sonoma System Sony designed, but the first standalone recorders were from Sony, Tascam, and Korg (best I know)
Its close but DSD has a more analog, real sound. When someone is singing its like they're in the room. I get the same sound I get from analog recordings from the 50's, early 60's
Thanks Paul, I had no idea that DXD was actually a PCM format. I have always mistakenly thought that DXD was just a more professional (higher resolution) of a DSD format. Thanks for teaching something.
I don't think people will ever understand the difference between PCM and DSD without explaining to people that there is a digital domain and an analog domain. In the digital domain, PCM vs DSD almost doesn't matter. It's just a band limited signal, and with enough processing power you can recover the EXACT analog waveform with either method. The trick with DSD is that you don't actually need a "DAC" per se. The digital signal can be directly used as an analog signal. Generally speaking you would want to use a lowpass filter at the end of the digital stream, but you don't have to. This is the whole magic of the direct stream digital. Spend all your time and energy perfecting the digital domain where it counts, and keep the analog stage simple and clean.
"The digital signal can be directly used as an analog signal" is the bit I find confusing, or obfuscated is that what is happening in the DAC? and i mean the appliance DAC rather than strictly the converter circuit process, that is the amplifier down stream scaling the filtered digital output to desired on off voltages for listening? and the reactance/capacitance of the analog circuits and drivers doing the smoothing of the pulse width into a waveform? Similar to PWM motor control?
Hush! Don't spill the beans! There is a whole industry and associated marketing industry built on general misconceptions around PCM vs DSD. In the digital domain both are virtually interchangeable as long as the information density of both encodings are comparable. The sensitivity to bit corruption and associated ability for error correction does differ between the two encodings.
Seems like DSD vs hi-res PCM comes down to a matter of extended frequency response (DSD) vs lower noise floor (PCM). Which is more important to you? Regarding sample vs bit rates, in general, higher sample rate yields more audible benefit (as long as you are at or above 16 bit or so and don't plan on processing or boosting the signal in any way subsequently).😮
I’ve done some extensive testing of DSD 5.1 vs DVD Audio vs Tidal ATMOS vs Apple ATMOS vs ATMOS disc on recordings I have on multiple platforms. Dire Straits Brothers in Arms, Fleetwood Mac Rumours, Van Morrison Moondance, Eagles Hotel California. Pink Floyd Dark Side of the Moon, and a few others. Play Oppo, Marantz and NAD systems on 7.1.4 Polk Legend and NAD M28 and Rotel Class D. I found that the Marantz / Oppo sounded better via HDMI DSD where the NAD is all PCM. I can’t put it any better then this, the sound quality of clarity and detail of say the song Brothers in Arms, Money for Nothing or Hotel California just outshine the streaming Atmos versions, and it’s very noticeable. However the NAD with its PCM processing vs DSD is a letdown and there is some improvement but not much. After all the DSD is lost. I have a 6 ATMOS Blu-ray Discs using Dolby True HD and these outperform the streamed versions. One such example is Air’s ‘10,000 Hz’ and Everythjng But The Girl’s ‘Fuse’ recorded at Abbey Road for ATMOS production. This is just my opinion.
when you decide your not really happy i was pretty dissapointed try the mola mola tambaqui it eats it for dinner without breaking sweat but they are unicorns to find secondhand x
The audio format considered closest to analog is PCM (Pulse-Code Modulation), as it is the standard method for converting analog sound waves into digital data, capturing a very close representation of the original analog signal without compression; you'll typically find PCM encoded within WAV or AIFF files. Fact
Mr Paul, If you were drunk now, you could have used this shortened answer; Hans of the Netherlands, Well yeah, of course I hear a difference between DSD and PCM... -they are COMPLETELY different letters. Thank you for your question Hans that'll be all for today's audio educational workshop.. Hiccup. 🎉🎉🎉
Dear Paul sir, it's my humble request to listen old 90's or 20's bollywood music and all songs i e dsd and dxd or doP will fail in battle between pcm flac. Please try 🙏
In an effort to maintain revenue Sony developed dSD many years ago when their licensing stream for the CD was expiring. PCM has improved greatly over recent years and the PCM problems that DSD was supposed to address no longer exist. DSD has serious technical problems. Such as very high noise levels just above audible range requiring sophisticated filters and the need to convert to analog or PCM to edit and mix. When a PCM file is played on a native DSD single-bit converter, the single-bit DAC chip has to convert the PCM to DSD in real-time. This is one of the major reasons people claim DSD sounds better than PCM, when in fact, it is just that the chip in most modern single-bit DACs do a poor job of decoding PCM. A good DSD and a good PCM recording can be audibly indistinguishable. Unlike PCM, DSD can produce ahigh noise floor, bad distortion and stair steps in the output analog waveform. A very high sample rate is required to make these problems inaudible. This results in a huge file compared to PCM with no improvement in audio compared to a good PCM recording.
This is quite simply incorrect. Though, a good try. All modern DACs use a form of PDM to work with PCM or DSD. It's a Sigma Delta Modulator at the heart of modern DACs (essentially multibit DSD). If you run single rate DSD then yes, the noise shifting requires a fairly steep filter to lower it. However, running 2X to 4X pushes the noise way out into easy territory. And in a direct A/B DSD sounds remarkably better than PCM.
@@Paulmcgowanpsaudio High-resolution PCM and DSD formats of comparable resolution are statistically indistinguishable from one another in blind listening tests. Loudspeakers and room acoustics are sar, sar more important.
@@geoff37s57 Well we can certainly agree on the importance of rooms and loudspeakers. Be careful swallowing the blind listening test bit. You (likely) don't know the experience level of the listeners nor the resolving power of the system, nor the environment and circumstances of the test. That all matters.
@@pascalne2414you hit the nail there have a Sony cdp-x7esd with burr brown Comparing it with the latest dacs or players with delta sigma it simply can not be beat details are missing with delta sigma
Why not record on tape, then mix analog then convert to DSD for the sacd, cd and downloads and release a limited run of pure analog signal path vinyl copies, then later release a run of dsd converted vinyl also? Could even sell needle drop downloads for those who want the vinyl character without a turntable. Surely converting analog tape to dsd is going to have an inherently more analog sound than a purely digital process that mimics analog. It is like adding digital grain to a digitally shot film. It has a somewhat analog look, but cannot beat filming on grainy film stock and creating a digital intermediate. Also by recording this way, the sacd and dsd downloads will sound very close to the analog recording and mix having only had one transfer to digital. A new and improved AAD the most pleasing spars code.
I’m wondering if there is a way to have 2 bit or more DSD? It may be technically impossible now, but is there anyone working on it especially with quantum computing and superpositioning? Just a thought.
Hey Paul... I think the letter writer may have misunderstood your talk about DSD & DoP... i forgot the video title but I remember you talking about it... so to be clear you never said what the letter writer mentioned. :)
Hey what about the DSD on the fly conversion from PCM on my Sony WM1ZM2 DAP? Is that worth anything or is it a waste of time? I can't really tell the difference if I'm being honest, but maybe I don't have the source to make it shine.
I wonder if any audio engineers mixed with tube amps. I would be interested in hearing music mixed with tube amps due to tube amps already being analog. Basically mixed digital music with tube amps for having both digital and analog. Maybe even have music mixed with tube amps instead of just relying on just using digital audio work stations or computers like the modern era while fixing the dynamic range so the audio does not end up being too fatiguing or too warm but have well balanced mastered and mixed tracks but incorporate tube amps in the mixing and mastering process. For example, use real instruments and than mix and master the instruments with tube amps.
Most of them use analog compressors and pre-amps in recording, tube or solid state. That being said, in mixing, there has been a lot of progress in digital plugins that sound like analog outboard gear, and some of the most famous mixers use mixing in the box these days. Pure analog on tape is almost gone in recording, even with the biggest bands. Did you know that The River album by Springsteen was recorded on 16bit digital equipment.
If DSD is superior, and I'm not doubting Paul's assertion, then why did Tascam and Sony withdraw their consumer recording products from the North American market? I was p!ssed when Tascam withdrew their DA 3000 and called them, and no one there knew why.
Because using DSD as a consumer is a big hustle. As simple as that. (+ in order to edit DSD, as just a random consumer, you have to convert it to PCM anyway)
@@djlolerkoster What editing? I want to archive my albums. Previously, employed a Teac 2300SD, then B 77, both biased for Grandmaster 456 tape, which I bought from Fred Locke Studios in Berlin, CT back in the mid '70's. That turned out to be a disaster for the long term.
One of the reasons is the high amount of resources required for it to become a mainstream standard. We've gotten closer to it becoming a reality, but apparently we're not there yet.
It's just a fact that you can't mix and process DSD in a digital mastering process and thus will end up with those DSD tracks becoming transcoded to PCM tracks and processed into a PCM master. And that's alright when running PCM in higher sample rates. Then when the output format of a digital mastering process is PCM, why distribute it as a lossy DSD copy? I prefer a lossless copy of the master file itself even if it's 352.8kHz PCM (DXD) or whatever. Reality is that both DSD and PCM can run with high enough sample rates and resolution to capture finer details than our hearing can detect. I believe Octave Records make amazing tracks not because of DSD but because of an audiophile passion and a goal of making the best possible sound when using an audiophile system. Most studios nowadays tend to compromise against what you get out of a cheap Bluetooth speaker.
there's nothing really about dsd that you couldn't do operations to it as dsd like mixing and cutting, keeping track of that current values supposed to be. as for why distribute in dsd.. well you could mix it in pcm at 4x(or whatever) quality/rates and then retain some of that in the dsd. not that it matters. it certainly doesn't matter if your dac takes 16/24/32 bits at a time in or such. you know what the dsd myths resemble? early mp3 myths, such that an mp3 would be impossible fundamentally to seek etc.
@@lasskinn474 I think of MQA when DSD gets hyped too much. Underlying commercial interests are skewing the message. For calculations to happen on a DSD track, you need to convert the track to PCM.
If the DAW supports rendering in PCM and DSD, the DSD version will be closer to the lossless format, and it'll sound more like what you hear when playing back the multitrack. Even if you're using plugins which require PCM conversion, there's another step of PCM conversion on the output rendering. On a DSD output rendering, you're skipping that PCM conversion and simply writing the analog output waveform onto a digital medium. At least that's my experience with trying it both ways.
@@arthurkillen396 No, it doesn’t work like that at all. You can’t mix tracks of DSD without first running a transcoding into PCM streams. So first your DSD data will have to be converted to PCM, then you can process it, and afterwards you have a PCM master. If your DAW outputs DSD, it still means it ended up with a lossy transcoding from an internal PCM stream to an external DSD stream. It might make you feel it produced DSD, but really it had to make a PCM version first and the output will only be less than that actual master PCM version. If you think there is some gear that can output DSD as the genuine master in a digital music production, name me the gear or software exactly. Theoretically, you could run a DSD-256 into PCM at 256x44.1kHz=11.2896MHz PCM 32 bits and do the math with such insane PCM frequency and convert it to DSD-256 again and claim you maintained the original time resolution. But I don’t believe anyone does that???
A decent dongle DAC and IEMS will reveal the difference. Headphones like the Sennheiser HD560s, which is usually under $200, will reveal the difference. You can even upsample PCM and hear the difference with something like Audirvana, HQPlayer, or Neutron player, though it won't be as great as music which originated as DSD.
A true DSD sounds way different then any PCM but there are only a handfull of real ones most are converted PCM crap with cracks, pops, or any other noise. And a true DSD does not have that and you only hear music in a much better way then PCM. Vocals are more direct and just better and everything is in balance not too much bass or that it sounds from within a box it is just perfect.
Paul day drinking vids? Guess I missed those? A different channel? You have been very specific about recording in DSD and then converting and editing in DXD/PCM. That some people can not detect or even might even prefer the lower quality of PCM, especially Red Book, says more about a system (or ears) that can not resolve. I still argue that DSD should be editable. Each bit is basically "On, Yes or NO". All that has to be calculated is whether with all existing data at that point in time, On YES or NO? In PCM it is adding numbers and truncating back down to the existing bit depth. When I was in Video and image manipulation, calculations often went into the time domain as well as instantaneous voltage level. Comparing one data point location (bit/ pixel) across multiple frames over time. So it would seem some form of time domain rather than voltage domain processing could be used. e.g. needing to calculate at 8 times the DSD rate for 8 inputs.
@glenncurry3041: Regarding "DSD should be editable": The problem is that the DSD signal's value at any point in time (i.e., any sample) does NOT represent the analog signal's actual (absolute) value at ANY point in time (as IS the case for PCM). It only encodes the DIFFERENCE of the current sample from the previous sample. For PCM, each sample answers the question "What is the analog signal's voltage now?", over and over and over again, 44,100 times per second (or whatever the sampling rate is), with 16 bits of resolution (65,536 different levels) or whatever the bit depth (resolution) is. You can "cut" the signal stream at any sample point and do whatever you want with it - make it louder, softer, add (mix) it with another stream, etc., all in that same digital signal processing mathematics domain. With DSD, each sample only answers the question "Is this sample louder or softer than the previous one?", over and over and over, 1.5 million times per second (or whatever the sampling rate is), with only 1 bit per sample. It's all only relative - no absolute reference level - except for the very beginning of the stream. You have to have the entire stream to reconstruct the original analog signal. You can't just "cut" (edit) the digital stream at any random point (sample) and edit it in any way (louder, softer, mix with another signal, etc.). I hope that helps.
@@boblehman1726 I am well aware of the technology behind DSD and PCM. And: "With DSD, each sample only answers the question "Is this sample louder or softer than the previous one?", is wrong. DSD is ONE bit. There is no "softer" only "louder". Each bit is determined by a comparator to determine if the voltage is higher than the existing level, YES/NO. So no you confuse rather than help. In DSD playback bitstream each bit has one task. Turn on Yes/no. If 8 streams are being mixed the output is either turn on that bit Yes/No. Not turn on 8 bits higher or 6 bits higher. Just ONE bit YES/NO. By time domain multiplexing as is done in video to remove zizzies, edge crawl, movement artifacts and such, calculations can be done outside of instantaneous voltage values. One of the early devices to do so and Ampex ADO (which I sold) had dedicated CPU power equivalent to a Super Computer of the time to do that processing. But in the end each bit of the new DSD data stream is either 1 or 0. Either that data sample is saying turn on or not.
@@glenncurry3041 Explaining how DSD works immediately gets very complex, too much so for this forum and most of its audience to be helpful for conceptual understanding (without getting bogged down into circuit modules, algorithms, noise/distortion generation, noise shaping, etc.). I only wanted to focus on the concept for why typical simple editing functions (so easily done in analog or PCM) are so difficult with DSD. Here are some simple excerpts from web sources such as Wikipedia, WhatHiFi, Motorola, Audiophile Inventory, etc. that should be helpful to interested readers: "Direct Stream Digital (DSD) is a high-resolution audio format that uses a single bit to sample an analog waveform and determine if it's higher or lower than the previous sample. ... DSD is a 1-bit Sigma Delta Modulated (SDM) audio format developed by Sony and Philips. ... DSD uses a single bit of information, and all [that] this information tells us is whether the current sample of the analogue waveform is higher or lower than in the previous one. Compared with the over 65,000 different values 16-bit PCM has, the two values (0 if the new sample [o]f the signal is lower or 1 if it’s higher) of DSD appear mighty limiting. That resolution shortfall is made up by the very high sampling rate of over 2.8 million times a second - that’s 64 times the speed of CD. ... If you could look at a DSD digital stream, it’s possible to draw the analogue waveform simply by looking at the density of 0s and 1s. The more 0s there are, the lower the waveform goes, and it’s the opposite for ones. Where there is a balance of the two values, we’re at, or close to, the zero signal point. ... That all looks good, but there are also issues with DSD. It’s not very practical to manipulate a DSD recording, for starters. All the things that are required post-recording such as equalisation, editing, dynamic range control and adding reverb usually involve the DSD stream being converted to PCM to do the processing and then switched back. That’s hardly a pure way of doing things, right? Just about every studio recording made with DSD goes through this process. It’s down to a lack of suitable equipment and processing software. You may come across the term DXD on a recording. This is where the original DSD signal has been converted to 24-bit/352kHz PCM and processed in that form throughout. While the name (intentionally or not) sounds like DSD, it is simply very high-resolution PCM. ... DSD uses delta-sigma modulation, a form of pulse-density modulation encoding, a technique to represent audio signals in digital format, a sequence of single-bit values at a sampling rate of 2.8224 MHz. This is 64 times the CD audio sampling rate of 44.1 kHz, but with 1-bit samples instead of 16-bit samples. ... DSD is simply a format for storing a delta-sigma signal without applying a ... process that converts the signal to a PCM signal. ... DSD differs from the PCM format used by compact disc or typical computer audio systems: while PCM uses a multi-bit value (representing a large range of amplitudes) at a low sample rate, DSD instead uses a single-bit value (representing an increase or decrease in amplitude) at a sample rate much higher than the signal's bandwidth. The process of creating a DSD signal is the same as ... a 1-bit delta-sigma analog-to-digital converter (ADC) ... . The short-term average of the 1-bit DSD bitstream signal is proportional to the original signal amplitude. DSD music mixing and mastering for SACD or Internet download presents challenges due to the difficulty of performing digital signal processing (DSP) operations (such as equalization, balance, panning) in a one-bit environment. Older analog recordings were processed using analog equipment and then digitized to DSD. It is also possible to avoid processing by using only the available adjustments in the studio equipment while recording to DSD. One DSP technique available is to convert the DSD to PCM and use standard PCM equipment such as Pro Tools, useful for rock and contemporary music which rely on multitrack techniques, then digitally convert back to DSD format. Some DSD proponents dislike this technique claiming that the PCM conversion to a lower sample rate reduces the sound quality of DSD. A format and set of tools for PCM processing of DSD has been developed under the name Digital eXtreme Definition (DXD). This is a PCM format with 24-bit resolution sampled at 352.8 kHz. ... DXD was initially developed for the Merging Technologies Pyramix workstation ... in 2004. This combination meant that it was possible to record and edit directly in DXD, and that the sample only converts to DSD once before publishing. This offers an advantage to the user as the noise created by converting DSD rises dramatically above 20 kHz, and more noise is added each time a signal is converted back to DSD during editing. The Pyramix Virtual Studio Digital Audio Workstation allows for recording, editing and mastering all DSD formats up to DSD256. ... The coarsely-quantized output of a delta-sigma ADC is occasionally used directly in signal processing or as a representation for signal storage (e.g., Super Audio CD stores the raw output of a 1-bit delta-sigma modulator). ... Delta modulation is an earlier related low-bit oversampling method that also uses negative feedback, but only encodes the derivative of the signal (its delta) rather than its amplitude. The result is a stream of marks and spaces representing up or down of the signal's movement, which must be integrated to reconstruct the signal's amplitude. ... 1-bit delta-sigma modulation is pulse-density modulation: In the specific case of a single-bit synchronous ΔΣ ADC, an analog voltage signal is effectively converted into a pulse frequency, or pulse density, which can be understood as pulse-density modulation (PDM). A sequence of positive and negative pulses, representing bits at a known fixed rate, is very easy to generate, transmit, and accurately regenerate at the receiver, given only that the timing and sign of the pulses can be recovered. Given such a sequence of pulses from a delta-sigma modulator, the original waveform can be reconstructed with adequate precision. ... [S]igma-delta modulation was developed as an extension to the well established delta modulation. ... Delta modulation is based on quantizing the change in the signal from sample to sample rather than the absolute value of the signal at each sample. ... In each clock cycle, the value of the output of the modulator is either plus or minus full scale, according to the results of the 1-bit A/D conversion. When the sinusoidal input ... is close to a plus full scale, the output is positive during most clock cycles. A similar statement holds for the case when the sinusoid is close to minus full scale. In both cases, the local average of the modulator output tracks the analog input. When the input is near zero, the value of the modulator output varies rapidly between a plus and a minus full scale with approximately zero mean. ... Pulse-density modulation, or PDM, is a form of modulation used to represent an analog signal with a binary signal. In a PDM signal, specific amplitude values are not encoded into codewords of pulses of different weight as they would be in pulse-code modulation (PCM); rather, the relative density of the pulses corresponds to the analog signal's amplitude. The output of a 1-bit DAC is the same as the PDM encoding of the signal. In a pulse-density modulation bitstream, a 1 corresponds to a pulse of positive polarity ... and a 0 corresponds to a pulse of negative polarity ... . A run consisting of all 1s would correspond to the maximum (positive) amplitude value, all 0s would correspond to the minimum (negative) amplitude value, and alternating 1s and 0s would correspond to a zero amplitude value. The continuous amplitude waveform is recovered by low-pass filtering the bipolar PDM bitstream. ... In pulse-density modulation, a high density of 1s occurs at the peaks of the sine wave, while a low density of 1s occurs at the troughs of the sine wave. ... Sound/air oscillation each moment has a certain position (instant pressure level). The pressure level after the microphone is converted to an electrical oscillation. The electrical oscillation at each moment has a certain position that refers to the air pressure level. Pulse Code Modulation consists of samples. The sample is a measurement (number value) of an instant position of the electrical oscillation (voltage) at the moment of the measurement. In sigma-delta modulation (a.k.a. DSD) each 1-bit sample shows a positive [1] or negative [0] change of the pressure level relatively to its previous value. And, as rule, the sigma-delta modulation is explained as a saw (level grows or falls) between samples. ... Is native DSD editing possible? Audio editing is cutting, merging, gain altering, equalizing, etc. of file(s). "Native editing" (without conversion to multibit and DSD re-modulation) is possible for cutting and merging onebit records in bit-perfect mode only. All other kinds of editing demand conversion to multi-bit values and re-modulation. In this case, losses of the editing are comparable with multibit resampling. ... Thus, audio tracks in 1-bit format should be converted to PCM audio (filtered or DXD). After it the[y] are mixed and mastered. Final mix is modulated to DSD back.
@@boblehman1726 DSD is as simple as it gets. Compare current signal level to the next sample. Is it higher? Yes/No? But to you this is "very complex, too much so for this forum and most of its audience to be helpful for conceptual understanding"? So you obfuscate as a red herring? SAD! But the length of the reply seems to make you feel better? Good for you!
Gosh dang it... I should have never clicked on THIS video, DxD???? Whaddya sayin? -you mean there's a WHOLE 'nother format, THAT YOU USE (so there must be some secret underlying reason) That you have divulged because of your totally normal and acceptable as far as I'm concerned habit of filming these segments during happy hour. I mean, we're lucky you have any time at all, who are we to judge, right? Therefore; I didn't know this video was going to spawn a litter of questions that now I have to Google the answers to... Which annoyingly enough usually just sense me back to your videos... This means, a six pack of Modelo, more codecs, probably new mastering software to switch from DSD to DxD for My: -experimental, yet still safe for human consumption ==>> "un-mix tapes" which are basically: Backups of my purchased SACD collection for playback on my 3-way tri-amplified pair of Sony loudspeakers. Each driver has its own amplifier channel left and right, six channels. *I use a software DSP to Pre-Process the DSD audio files into what I call a: ∆∆ Stereo to 'spectrum split' 6 track demux-tape, burned onto a multichannel DVD-Audio disc or Blu-ray. Basically, we have the Sony speakers that have a really bad crossover design, but really good drivers and enclosure. I don't want to mention their name because they need to stay cheap. Ahem. We have the crossover points that are optimal and suggested by many people on the oracles of audio websites. Remember, a hundred people's opinions can't be wrong. We also take the test measurement microphone and kind of make sure the sweeps are reflecting the driver curve which were also wonderfully provided by the many though various levels of quality, aforementioned oracles. I have a chrontab (a script) running on one of my NAS servers that automatically received RIPs from a SACD/DVD/BLURAY disc that is inserted into the 'Sony Blu-ray player of specialness' (complete with the extra-special sceeen saver update thumb drive), in the handy dandy DSD format. This back up folder is polled regularly by the handy script, and when an audio file pops in, software DSP will process The stereo files into the six split tracks automatically, while performing and applying the following operations/filters: *Split stereo channels at preselected crossover points/slopes with the magic of 'fancy pants math' applied and plenty of 'manilla envelooe filtration' - rhe 6 discreet channels are meant for reproduction on multichannel DVD audio or BOO-RAY players. In my experimental demo system, unlicensed and unwarranted I tell you... The 6 ch out from my Sony Blu-ray, feeds amplifiers, that each feed the drivers directly, with their own high quality low oxygen wires, so nobody has to fight about anything. It's like all the kids got their own toy to play with in the back of the car. One might process the digital stream out from the Sony with an "outboard DAC of quality and heft" as it were, -- if one were to be so rich and famous as to afford DACxation better than my Sony Blu-ray internals, that is. ™™Hey, Cheap audio man said it was okay to do it, you can't pick on me/™™ Software DSP applies correction curve for: "driver anomaly compensation" This smoothes out the Ring rings and the wawas that happen when the drivers get agitated by The amplifiers picking on how big their inductance and outductance and circle duck dialedness... These are all super tech terms that you will learn about when you're older Paul.. Next is the optional "EQ wizard acquired room correction filter." Which we usually use unless we're going to take this road on the show. No, wait strike that - reverse it. .. We are installing also a three-way system in the automobile and a portable Bluetooth streamer thingy that has a 5 channel DMX for a kicking 2.1 amazingness. No EQ wizard curve on those files for example, just the ones that are going to live in the living room. Redundantly. Since we have this kind of granular control in the pre-production mix down, We are noodling around with attempting phase / timing correction. We have a carver Sonic holography preamp/spent uranium frame over here. We were thinking of measuring the signature for creating a filter to use the DSP modelling and the data from the signal analyzer comparator thingama-bobber 5000, 😊 and you get a lawsuit and a ticked off ghost of Mr Carver, most likely. Okay we'll skip that one but you get the point. So it's your fault. All your fault. Every bit of it. Your input is appreciated. Cuz you started this. Fine sir. Cheers to you for spreading your knowledge and The bug of audioaphilia. Okay that's weird on two levels I don't like talking about bugs near my ears or something that sounds like necrophilia. We'll name it later. I have research to do and money to spend un-wisely.... sure beats Netflix though !! 😅 Muchas gracias Myke
I've found that PCM sounds better than DSD on Chord devices. I think it has to do with how their FPGAs upsample all audio. Seems to be a better synergy for PCM than DSD.
How nice it must be for you, Paul, that you get to live in physically beautiful Boulder, Colorado AND this metaphysical La, La, Land . . . at the same time.
I am not selling any DSD products, rather listening to them, and I hear immediate and huge differences between DSD and classic PCM standards (CD, FLAC, nevermind MP3). A lot of PCM stuff is very nice and enjoyable, but pretending it is perfect and unimprovable is dishonest.
And 2 this guy dont know that much about audio. More info. generally speaking, PCM is considered to be closer to analog sound than DSD, as PCM is the standard digital audio format that directly samples and quantizes an analog signal, while DSD uses a different encoding method that can be seen as a more "indirect" representation of an analog waveform.      Key points to remember: PCM (Pulse Code Modulation): This is the most common digital audio format, where an analog signal is directly sampled at regular intervals and converted into digital values, making it a relatively straightforward representation of the original analog signal.    DSD (Direct Stream Digital): This format uses a single-bit data stream at a much higher sampling rate than typical PCM, which results in a different way of representing the analog signal, often described as being closer to a "pulse density modulation" approach.
“I can hear a slight difference.” No, you can’t. Paul, you don’t even properly understand digital audio given the incorrect things you’ve said many times regarding PCM. You think you do, but you don’t. If you did, you wouldn’t be repeating some of the same nonsense that way too many others repeat. You’ve got a lot of experience and knowledge on many things to do with audio, but you do NOT understand how digital audio works. There’s only so many times a guy can tell you to go watch Digital Show & Tell until you do understand things properly, but it is clear you’d rather continue to think you already know what you think you know rather than actually learn the correct information. It’s sad, concerning you. And aggravating, concerning everyone else you tell incorrect information to. Ignorance is not a virtue.
So then why don’t you make a video on your channel about PCM, DSD etc? and spread your absolute vast knowledge of the subject so others can know the truth rather than waste it on bashing paul. I’m sure many could benefit since you have all the correct answers.
@@ClaytonMacleod well i already know it’s out there and i know the answer as well , but what’s the point of pointed out he’s wrong? Makes no sense to me unless it’s self gratifying to you? Either way I find it strange when people find it necessary to point out a mistake or wrong answer a professional has made as if it discredits their whole life’s work due to one mistake.
@@Audiofreak71 Dude, he's telling others, and others believe him even if he's wrong, because they don't know any better. So if Paul instead had the correct information and shared that instead that means not only does he now possess the correct answer, everyone he tells that believes it is the correct answer will also have the correct answer. Spreading correct information rather than bogus information is a Good Thing. TM It is better for everyone, including him, if he both has and shares correct information. Has absolutely nothing to do with me or how I feel about myself. Smarten up.
If the original recording is good there is nothing wrong with 16bit 44.1 (Redbook).I find it rich,warm and detailed and very pleasing to the ears.
Agreed. It all depends on what the recording technician did regardless of what was used to do it.
The way I see it is that the overwhelming majority of music out there is PCM (or analog) and there’s no way a rational person is going to ditch all that just to have a handful of DSD recordings. And yes, the recording is everything, perhaps even more so than the format itself. I’d rather listen to a well-recorded track on 16/44.1 (or even 320 kbps MP3) than a poorly recorded track on DSD.
Listening to music is a subjective experience and the competence of the engineering is critical. That said, the idea that red book is "good enough" assumes that DSD is a marketing conspiracy (mqa for example). I was shocked at the detail in the mofi dsd64 transferred vinyl of Bill Evans & Jim Hall "Undercurrent" when I upgraded my turntable with a Grado Platinum Tibre. I record my own music at 192Khz 32bit and listen to vinyl rips all the time. CDs sound amazing on high end gear but HR sounds good on just about anything worth listening too. I believe Paul and Kevin Grey when they say DSD256 is better than CD quality.
@@housepianist That’s an excellent point.
@@housepianist My thoughts exactly.
So basically all of Octave Records material has been through DXD (24/352.8kHz) which is hi-res PCM. That means at least 2 conversions from DSD capture -> DXD mixing/editing -> DSD final purchase. Why not just DSD capture (since you believe this is better) -> DXD and then sell the final album as DXD PCM without getting it back to DSD and save yourself another lossy conversion step???
This also implies that the best quality recordings from Octave Records should be the 24/352.8kHz download option, not the DSD ones!
Agree 100% Paul- I have quite the collection of SACD's and everyone I own from Octave has outstanding balance and resolution.
Absolutely amazing how all these "experts" in the comment section constantly try to correct each other. I know Paul encourages people to comment on the topics and have a pleasant exchange of ideas, but it gets down right comical some times.
Excellent discussion, there is the 8bit DSD in the DSD official specification designed for editing, both 64 and 128FS , 8bits per sample is enough to do all the editing maths. It is a matter to implement it natively on the chip and it was also done a few times, Sony e-chip for instance. DSD by nature is very ambient and warm, however, it tends to saturate at high volumes just like analogue audio when PCM would exhibit digital harshness. High volume is the ultimate test to appreciate the difference and essentially hear it. 7:02
What you described as the last point is exactly what has been done as "high-end" in the analog era: "Direct-to-disc". It's basically the same principle as the early recordings - where there was no mixing at all and the needle just scratched the original recording into wax or whatever material.
@1:44 "If I said..."
Paul, you never said that.
@4:31 "If you start with DSD..."
That is what you have said in the past.
@5:40 "If you were to record..."
That is also what you have said in the past.
Why don't you back up your statements with facts and give us links to prove your points? I know, it's just easier your way...
As always I enjoyed your offering Paul. And I certainly enjoy al the comments you trigger. At the same time I wonder when some of the people reacting here will start with listening to their music and stop chasing the ultimate sound. It's almost pathetic in some cases. No matter how much money and time you spend finding the ultimate system, it's just not there. Often good is good enough.
Hey Paul... I don't know if you ever knew Brad Miller who at one time way back when ran Mobile Fidelity Sound Labs. Back in the late 80s or early 90s he came to the studio I was working at in Portland, Oregon with his Colossus Digital Recorder... a 4 channel unit with single ended (unbalanced), discrete electronics that sounded incredible. At the time we had a Mitsubishi X-86 two track that I had replaced the filters with Apogees. I have to say that Brad's recorder was markedly better, but I still believe that it was a PCM unit. Unfortunately Brad passed a few years later and I believe that nothing ever happened with the Colossus. I also remember the Soundsteam digital recorder... wasn't that a DSD unit? Anyway... great video, Paul! Cheers!
Thanks, Jim. I never knew him but knew of him. And I assume it was PCM. The first commercial DAW was the Sonoma System Sony designed, but the first standalone recorders were from Sony, Tascam, and Korg (best I know)
Analog--> DSD--> high rate PCM to mix--> to (DSD or lower rate PCM or DAC pressed to vinyl) seems like the mixing stage is the critical part
Its close but DSD has a more analog, real sound. When someone is singing its like they're in the room. I get the same sound I get from analog recordings from the 50's, early 60's
So the best would be to record completly analog on twenty four tracks and them record the mix down in DSD
I would love to hear a “direct-to-DSD”, unmixed, live recording. I have a couple of analog direct-to-disc records and they are superb.
DSD sounding better than PCM is likely too small to hear.
Blind listening comparisons would be the way to test this.
its totally different on a even half decent system
You have a world of differences to discover then. :) Does not mean you cannot enjoy both systems though.
Thanks Paul, I had no idea that DXD was actually a PCM format. I have always mistakenly thought that DXD was just a more professional (higher resolution) of a DSD format. Thanks for teaching something.
That’s because fremer confuses DSD and DXD
I'll drink to that. 🍸
Here here I third the notion
I don't think people will ever understand the difference between PCM and DSD without explaining to people that there is a digital domain and an analog domain. In the digital domain, PCM vs DSD almost doesn't matter. It's just a band limited signal, and with enough processing power you can recover the EXACT analog waveform with either method. The trick with DSD is that you don't actually need a "DAC" per se. The digital signal can be directly used as an analog signal. Generally speaking you would want to use a lowpass filter at the end of the digital stream, but you don't have to. This is the whole magic of the direct stream digital. Spend all your time and energy perfecting the digital domain where it counts, and keep the analog stage simple and clean.
"The digital signal can be directly used as an analog signal" is the bit I find confusing, or obfuscated is that what is happening in the DAC? and i mean the appliance DAC rather than strictly the converter circuit process, that is the amplifier down stream scaling the filtered digital output to desired on off voltages for listening? and the reactance/capacitance of the analog circuits and drivers doing the smoothing of the pulse width into a waveform? Similar to PWM motor control?
Hush! Don't spill the beans! There is a whole industry and associated marketing industry built on general misconceptions around PCM vs DSD.
In the digital domain both are virtually interchangeable as long as the information density of both encodings are comparable. The sensitivity to bit corruption and associated ability for error correction does differ between the two encodings.
Seems like DSD vs hi-res PCM comes down to a matter of extended frequency response (DSD) vs lower noise floor (PCM). Which is more important to you? Regarding sample vs bit rates, in general, higher sample rate yields more audible benefit (as long as you are at or above 16 bit or so and don't plan on processing or boosting the signal in any way subsequently).😮
I’ve done some extensive testing of DSD 5.1 vs DVD Audio vs Tidal ATMOS vs Apple ATMOS vs ATMOS disc on recordings I have on multiple platforms.
Dire Straits Brothers in Arms, Fleetwood Mac Rumours, Van Morrison Moondance, Eagles Hotel California. Pink Floyd Dark Side of the Moon, and a few others.
Play Oppo, Marantz and NAD systems on 7.1.4 Polk Legend and NAD M28 and Rotel Class D.
I found that the Marantz / Oppo sounded better via HDMI DSD where the NAD is all PCM.
I can’t put it any better then this, the sound quality of clarity and detail of say the song Brothers in Arms, Money for Nothing or Hotel California just outshine the streaming Atmos versions, and it’s very noticeable.
However the NAD with its PCM processing vs DSD is a letdown and there is some improvement but not much. After all the DSD is lost.
I have a 6 ATMOS Blu-ray Discs using Dolby True HD and these outperform the streamed versions. One such example is Air’s ‘10,000 Hz’ and Everythjng But The Girl’s ‘Fuse’ recorded at Abbey Road for ATMOS production.
This is just my opinion.
Why not just stay analog? Seems that we have no R&D into the ADC’s
My first PS audio dac is on the way (direct stream)..stoked! I had to knwo if there really is a difference. (Bought used)
when you decide your not really happy i was pretty dissapointed try the mola mola tambaqui it eats it for dinner without breaking sweat but they are unicorns to find secondhand x
The audio format considered closest to analog is PCM (Pulse-Code Modulation), as it is the standard method for converting analog sound waves into digital data, capturing a very close representation of the original analog signal without compression; you'll typically find PCM encoded within WAV or AIFF files. Fact
Mr Paul,
If you were drunk now, you could have used this shortened answer;
Hans of the Netherlands,
Well yeah, of course I hear a difference between DSD and PCM...
-they are COMPLETELY different letters.
Thank you for your question Hans that'll be all for today's audio educational workshop..
Hiccup.
🎉🎉🎉
Dear Paul sir, it's my humble request to listen old 90's or 20's bollywood music and all songs i e dsd and dxd or doP will fail in battle between pcm flac. Please try 🙏
Interesting to find out what changes were made to have the PCM sound better and then what changes were made to have DSD sound better
Resolution increase.
In an effort to maintain revenue Sony developed dSD many years ago when their licensing stream for the CD was expiring. PCM has improved greatly over recent years and the PCM problems that DSD was supposed to address no longer exist. DSD has serious technical problems. Such as very high noise levels just above audible range requiring sophisticated filters and the need to convert to analog or PCM to edit and mix. When a PCM file is played on a native DSD single-bit converter, the single-bit DAC chip has to convert the PCM to DSD in real-time. This is one of the major reasons people claim DSD sounds better than PCM, when in fact, it is just that the chip in most modern single-bit DACs do a poor job of decoding PCM.
A good DSD and a good PCM recording can be audibly indistinguishable.
Unlike PCM, DSD can produce ahigh noise floor, bad distortion and stair steps in the output analog waveform. A very high sample rate is required to make these problems inaudible. This results in a huge file compared to PCM with no improvement in audio compared to a good PCM recording.
This is quite simply incorrect. Though, a good try. All modern DACs use a form of PDM to work with PCM or DSD. It's a Sigma Delta Modulator at the heart of modern DACs (essentially multibit DSD). If you run single rate DSD then yes, the noise shifting requires a fairly steep filter to lower it. However, running 2X to 4X pushes the noise way out into easy territory. And in a direct A/B DSD sounds remarkably better than PCM.
Not all Dacs use sigma delta conveniently forgetting R2r dacs ?
@@Paulmcgowanpsaudio
High-resolution PCM and DSD formats of comparable resolution are statistically indistinguishable from one another in blind listening tests. Loudspeakers and room acoustics are sar, sar more important.
@@geoff37s57 Well we can certainly agree on the importance of rooms and loudspeakers. Be careful swallowing the blind listening test bit. You (likely) don't know the experience level of the listeners nor the resolving power of the system, nor the environment and circumstances of the test. That all matters.
@@pascalne2414you hit the nail there have a Sony cdp-x7esd with burr brown
Comparing it with the latest dacs or players with delta sigma it simply can not be beat
details are missing with delta sigma
Why not record on tape, then mix analog then convert to DSD for the sacd, cd and downloads and release a limited run of pure analog signal path vinyl copies, then later release a run of dsd converted vinyl also? Could even sell needle drop downloads for those who want the vinyl character without a turntable. Surely converting analog tape to dsd is going to have an inherently more analog sound than a purely digital process that mimics analog.
It is like adding digital grain to a digitally shot film. It has a somewhat analog look, but cannot beat filming on grainy film stock and creating a digital intermediate.
Also by recording this way, the sacd and dsd downloads will sound very close to the analog recording and mix having only had one transfer to digital. A new and improved AAD the most pleasing spars code.
I’m wondering if there is a way to have 2 bit or more DSD? It may be technically impossible now, but is there anyone working on it especially with quantum computing and superpositioning? Just a thought.
Hey Paul... I think the letter writer may have misunderstood your talk about DSD & DoP... i forgot the video title but I remember you talking about it... so to be clear you never said what the letter writer mentioned. :)
Hey what about the DSD on the fly conversion from PCM on my Sony WM1ZM2 DAP? Is that worth anything or is it a waste of time? I can't really tell the difference if I'm being honest, but maybe I don't have the source to make it shine.
I wonder if any audio engineers mixed with tube amps. I would be interested in hearing music mixed with tube amps due to tube amps already being analog. Basically mixed digital music with tube amps for having both digital and analog. Maybe even have music mixed with tube amps instead of just relying on just using digital audio work stations or computers like the modern era while fixing the dynamic range so the audio does not end up being too fatiguing or too warm but have well balanced mastered and mixed tracks but incorporate tube amps in the mixing and mastering process. For example, use real instruments and than mix and master the instruments with tube amps.
Most of them use analog compressors and pre-amps in recording, tube or solid state. That being said, in mixing, there has been a lot of progress in digital plugins that sound like analog outboard gear, and some of the most famous mixers use mixing in the box these days. Pure analog on tape is almost gone in recording, even with the biggest bands. Did you know that The River album by Springsteen was recorded on 16bit digital equipment.
If DSD is superior, and I'm not doubting Paul's assertion, then why did Tascam and Sony withdraw their consumer recording products from the North American market? I was p!ssed when Tascam withdrew their DA 3000 and called them, and no one there knew why.
Because using DSD as a consumer is a big hustle. As simple as that. (+ in order to edit DSD, as just a random consumer, you have to convert it to PCM anyway)
I'm guessing for the same reason SACD isn't mainstream
@@djlolerkoster What editing? I want to archive my albums. Previously, employed a Teac 2300SD, then B 77, both biased for Grandmaster 456 tape, which I bought from Fred Locke Studios in Berlin, CT back in the mid '70's. That turned out to be a disaster for the long term.
@@xaviermontalban717My guess is that SACD was way overpriced. Had they used a modest up charge the market response would have been far different.
One of the reasons is the high amount of resources required for it to become a mainstream standard. We've gotten closer to it becoming a reality, but apparently we're not there yet.
It's just a fact that you can't mix and process DSD in a digital mastering process and thus will end up with those DSD tracks becoming transcoded to PCM tracks and processed into a PCM master. And that's alright when running PCM in higher sample rates. Then when the output format of a digital mastering process is PCM, why distribute it as a lossy DSD copy? I prefer a lossless copy of the master file itself even if it's 352.8kHz PCM (DXD) or whatever. Reality is that both DSD and PCM can run with high enough sample rates and resolution to capture finer details than our hearing can detect. I believe Octave Records make amazing tracks not because of DSD but because of an audiophile passion and a goal of making the best possible sound when using an audiophile system. Most studios nowadays tend to compromise against what you get out of a cheap Bluetooth speaker.
there's nothing really about dsd that you couldn't do operations to it as dsd like mixing and cutting, keeping track of that current values supposed to be.
as for why distribute in dsd.. well you could mix it in pcm at 4x(or whatever) quality/rates and then retain some of that in the dsd. not that it matters. it certainly doesn't matter if your dac takes 16/24/32 bits at a time in or such.
you know what the dsd myths resemble? early mp3 myths, such that an mp3 would be impossible fundamentally to seek etc.
@@lasskinn474 I think of MQA when DSD gets hyped too much. Underlying commercial interests are skewing the message. For calculations to happen on a DSD track, you need to convert the track to PCM.
If the DAW supports rendering in PCM and DSD, the DSD version will be closer to the lossless format, and it'll sound more like what you hear when playing back the multitrack. Even if you're using plugins which require PCM conversion, there's another step of PCM conversion on the output rendering. On a DSD output rendering, you're skipping that PCM conversion and simply writing the analog output waveform onto a digital medium. At least that's my experience with trying it both ways.
@@arthurkillen396 No, it doesn’t work like that at all. You can’t mix tracks of DSD without first running a transcoding into PCM streams. So first your DSD data will have to be converted to PCM, then you can process it, and afterwards you have a PCM master. If your DAW outputs DSD, it still means it ended up with a lossy transcoding from an internal PCM stream to an external DSD stream. It might make you feel it produced DSD, but really it had to make a PCM version first and the output will only be less than that actual master PCM version. If you think there is some gear that can output DSD as the genuine master in a digital music production, name me the gear or software exactly. Theoretically, you could run a DSD-256 into PCM at 256x44.1kHz=11.2896MHz PCM 32 bits and do the math with such insane PCM frequency and convert it to DSD-256 again and claim you maintained the original time resolution. But I don’t believe anyone does that???
Nice discussion, most of us do not have a revealing enough system to tell the difference…. Negative comments presented aggressively are not welcome..
I only have a Denon 1520ae, a Fiio K7 dac and Dali Zensor 7 and i can hear the difference clearly on this cheap system.
Even the most basic entry hi-fi system can clearly show the differences, so don't get bogged down in hardware tiers.
A decent dongle DAC and IEMS will reveal the difference. Headphones like the Sennheiser HD560s, which is usually under $200, will reveal the difference. You can even upsample PCM and hear the difference with something like Audirvana, HQPlayer, or Neutron player, though it won't be as great as music which originated as DSD.
A true DSD sounds way different then any PCM but there are only a handfull of real ones most are converted PCM crap with cracks, pops, or any other noise. And a true DSD does not have that and you only hear music in a much better way then PCM. Vocals are more direct and just better and everything is in balance not too much bass or that it sounds from within a box it is just perfect.
If you can mix with analog recordings, and DSD is very much like analog, why can't you mix with it directly?
Paul day drinking vids? Guess I missed those? A different channel? You have been very specific about recording in DSD and then converting and editing in DXD/PCM. That some people can not detect or even might even prefer the lower quality of PCM, especially Red Book, says more about a system (or ears) that can not resolve.
I still argue that DSD should be editable. Each bit is basically "On, Yes or NO". All that has to be calculated is whether with all existing data at that point in time, On YES or NO? In PCM it is adding numbers and truncating back down to the existing bit depth. When I was in Video and image manipulation, calculations often went into the time domain as well as instantaneous voltage level. Comparing one data point location (bit/ pixel) across multiple frames over time. So it would seem some form of time domain rather than voltage domain processing could be used. e.g. needing to calculate at 8 times the DSD rate for 8 inputs.
@glenncurry3041: Regarding "DSD should be editable": The problem is that the DSD signal's value at any point in time (i.e., any sample) does NOT represent the analog signal's actual (absolute) value at ANY point in time (as IS the case for PCM). It only encodes the DIFFERENCE of the current sample from the previous sample. For PCM, each sample answers the question "What is the analog signal's voltage now?", over and over and over again, 44,100 times per second (or whatever the sampling rate is), with 16 bits of resolution (65,536 different levels) or whatever the bit depth (resolution) is. You can "cut" the signal stream at any sample point and do whatever you want with it - make it louder, softer, add (mix) it with another stream, etc., all in that same digital signal processing mathematics domain. With DSD, each sample only answers the question "Is this sample louder or softer than the previous one?", over and over and over, 1.5 million times per second (or whatever the sampling rate is), with only 1 bit per sample. It's all only relative - no absolute reference level - except for the very beginning of the stream. You have to have the entire stream to reconstruct the original analog signal. You can't just "cut" (edit) the digital stream at any random point (sample) and edit it in any way (louder, softer, mix with another signal, etc.). I hope that helps.
@@boblehman1726 I am well aware of the technology behind DSD and PCM. And: "With DSD, each sample only answers the question "Is this sample louder or softer than the previous one?", is wrong. DSD is ONE bit. There is no "softer" only "louder". Each bit is determined by a comparator to determine if the voltage is higher than the existing level, YES/NO. So no you confuse rather than help.
In DSD playback bitstream each bit has one task. Turn on Yes/no. If 8 streams are being mixed the output is either turn on that bit Yes/No. Not turn on 8 bits higher or 6 bits higher. Just ONE bit YES/NO. By time domain multiplexing as is done in video to remove zizzies, edge crawl, movement artifacts and such, calculations can be done outside of instantaneous voltage values. One of the early devices to do so and Ampex ADO (which I sold) had dedicated CPU power equivalent to a Super Computer of the time to do that processing. But in the end each bit of the new DSD data stream is either 1 or 0. Either that data sample is saying turn on or not.
@@glenncurry3041 Explaining how DSD works immediately gets very complex, too much so for this forum and most of its audience to be helpful for conceptual understanding (without getting bogged down into circuit modules, algorithms, noise/distortion generation, noise shaping, etc.). I only wanted to focus on the concept for why typical simple editing functions (so easily done in analog or PCM) are so difficult with DSD. Here are some simple excerpts from web sources such as Wikipedia, WhatHiFi, Motorola, Audiophile Inventory, etc. that should be helpful to interested readers:
"Direct Stream Digital (DSD) is a high-resolution audio format that uses a single bit to sample an analog waveform and determine if it's higher or lower than the previous sample. ... DSD is a 1-bit Sigma Delta Modulated (SDM) audio format developed by Sony and Philips. ... DSD uses a single bit of information, and all [that] this information tells us is whether the current sample of the analogue waveform is higher or lower than in the previous one. Compared with the over 65,000 different values 16-bit PCM has, the two values (0 if the new sample [o]f the signal is lower or 1 if it’s higher) of DSD appear mighty limiting.
That resolution shortfall is made up by the very high sampling rate of over 2.8 million times a second - that’s 64 times the speed of CD. ...
If you could look at a DSD digital stream, it’s possible to draw the analogue waveform simply by looking at the density of 0s and 1s. The more 0s there are, the lower the waveform goes, and it’s the opposite for ones. Where there is a balance of the two values, we’re at, or close to, the zero signal point. ...
That all looks good, but there are also issues with DSD. It’s not very practical to manipulate a DSD recording, for starters.
All the things that are required post-recording such as equalisation, editing, dynamic range control and adding reverb usually involve the DSD stream being converted to PCM to do the processing and then switched back.
That’s hardly a pure way of doing things, right? Just about every studio recording made with DSD goes through this process. It’s down to a lack of suitable equipment and processing software.
You may come across the term DXD on a recording. This is where the original DSD signal has been converted to 24-bit/352kHz PCM and processed in that form throughout. While the name (intentionally or not) sounds like DSD, it is simply very high-resolution PCM. ...
DSD uses delta-sigma modulation, a form of pulse-density modulation encoding, a technique to represent audio signals in digital format, a sequence of single-bit values at a sampling rate of 2.8224 MHz. This is 64 times the CD audio sampling rate of 44.1 kHz, but with 1-bit samples instead of 16-bit samples. ...
DSD is simply a format for storing a delta-sigma signal without applying a ... process that converts the signal to a PCM signal. ...
DSD differs from the PCM format used by compact disc or typical computer audio systems: while PCM uses a multi-bit value (representing a large range of amplitudes) at a low sample rate, DSD instead uses a single-bit value (representing an increase or decrease in amplitude) at a sample rate much higher than the signal's bandwidth.
The process of creating a DSD signal is the same as ... a 1-bit delta-sigma analog-to-digital converter (ADC) ... . The short-term average of the 1-bit DSD bitstream signal is proportional to the original signal amplitude.
DSD music mixing and mastering for SACD or Internet download presents challenges due to the difficulty of performing digital signal processing (DSP) operations (such as equalization, balance, panning) in a one-bit environment.
Older analog recordings were processed using analog equipment and then digitized to DSD. It is also possible to avoid processing by using only the available adjustments in the studio equipment while recording to DSD.
One DSP technique available is to convert the DSD to PCM and use standard PCM equipment such as Pro Tools, useful for rock and contemporary music which rely on multitrack techniques, then digitally convert back to DSD format. Some DSD proponents dislike this technique claiming that the PCM conversion to a lower sample rate reduces the sound quality of DSD.
A format and set of tools for PCM processing of DSD has been developed under the name Digital eXtreme Definition (DXD). This is a PCM format with 24-bit resolution sampled at 352.8 kHz. ...
DXD was initially developed for the Merging Technologies Pyramix workstation ... in 2004. This combination meant that it was possible to record and edit directly in DXD, and that the sample only converts to DSD once before publishing. This offers an advantage to the user as the noise created by converting DSD rises dramatically above 20 kHz, and more noise is added each time a signal is converted back to DSD during editing.
The Pyramix Virtual Studio Digital Audio Workstation allows for recording, editing and mastering all DSD formats up to DSD256. ...
The coarsely-quantized output of a delta-sigma ADC is occasionally used directly in signal processing or as a representation for signal storage (e.g., Super Audio CD stores the raw output of a 1-bit delta-sigma modulator). ...
Delta modulation is an earlier related low-bit oversampling method that also uses negative feedback, but only encodes the derivative of the signal (its delta) rather than its amplitude. The result is a stream of marks and spaces representing up or down of the signal's movement, which must be integrated to reconstruct the signal's amplitude. ...
1-bit delta-sigma modulation is pulse-density modulation: In the specific case of a single-bit synchronous ΔΣ ADC, an analog voltage signal is effectively converted into a pulse frequency, or pulse density, which can be understood as pulse-density modulation (PDM). A sequence of positive and negative pulses, representing bits at a known fixed rate, is very easy to generate, transmit, and accurately regenerate at the receiver, given only that the timing and sign of the pulses can be recovered. Given such a sequence of pulses from a delta-sigma modulator, the original waveform can be reconstructed with adequate precision. ...
[S]igma-delta modulation was developed as an extension to the well established delta modulation. ... Delta modulation is based on quantizing the change in the signal from sample to sample rather than the absolute value of the signal at each sample. ...
In each clock cycle, the value of the output of the modulator is either plus or minus full scale, according to the results of the 1-bit A/D conversion. When the sinusoidal input ... is close to a plus full scale, the output is positive during most clock cycles. A similar statement holds for the case when the sinusoid is close to minus full scale. In both cases, the local average of the modulator output tracks the analog input. When the input is near zero, the value of the modulator output varies rapidly between a plus and a minus full scale with approximately zero mean. ...
Pulse-density modulation, or PDM, is a form of modulation used to represent an analog signal with a binary signal. In a PDM signal, specific amplitude values are not encoded into codewords of pulses of different weight as they would be in pulse-code modulation (PCM); rather, the relative density of the pulses corresponds to the analog signal's amplitude. The output of a 1-bit DAC is the same as the PDM encoding of the signal.
In a pulse-density modulation bitstream, a 1 corresponds to a pulse of positive polarity ... and a 0 corresponds to a pulse of negative polarity ... .
A run consisting of all 1s would correspond to the maximum (positive) amplitude value, all 0s would correspond to the minimum (negative) amplitude value, and alternating 1s and 0s would correspond to a zero amplitude value. The continuous amplitude waveform is recovered by low-pass filtering the bipolar PDM bitstream. ...
In pulse-density modulation, a high density of 1s occurs at the peaks of the sine wave, while a low density of 1s occurs at the troughs of the sine wave. ...
Sound/air oscillation each moment has a certain position (instant pressure level). The pressure level after the microphone is converted to an electrical oscillation.
The electrical oscillation at each moment has a certain position that refers to the air pressure level.
Pulse Code Modulation consists of samples. The sample is a measurement (number value) of an instant position of the electrical oscillation (voltage) at the moment of the measurement.
In sigma-delta modulation (a.k.a. DSD) each 1-bit sample shows a positive [1] or negative [0] change of the pressure level relatively to its previous value.
And, as rule, the sigma-delta modulation is explained as a saw (level grows or falls) between samples. ...
Is native DSD editing possible?
Audio editing is cutting, merging, gain altering, equalizing, etc. of file(s).
"Native editing" (without conversion to multibit and DSD re-modulation) is possible for cutting and merging onebit records in bit-perfect mode only.
All other kinds of editing demand conversion to multi-bit values and re-modulation. In this case, losses of the editing are comparable with multibit resampling. ...
Thus, audio tracks in 1-bit format should be converted to PCM audio (filtered or DXD). After it the[y] are mixed and mastered. Final mix is modulated to DSD back.
@@boblehman1726 DSD is as simple as it gets. Compare current signal level to the next sample. Is it higher? Yes/No? But to you this is "very complex, too much so for this forum and most of its audience to be helpful for conceptual understanding"? So you obfuscate as a red herring? SAD! But the length of the reply seems to make you feel better? Good for you!
Gosh dang it...
I should have never clicked on THIS video,
DxD???? Whaddya sayin?
-you mean there's a WHOLE 'nother format, THAT YOU USE (so there must be some secret underlying reason) That you have divulged because of your totally normal and acceptable as far as I'm concerned habit of filming these segments during happy hour. I mean, we're lucky you have any time at all, who are we to judge, right?
Therefore;
I didn't know this video was going to spawn a litter of questions that now I have to Google the answers to... Which annoyingly enough usually just sense me back to your videos...
This means, a six pack of Modelo, more codecs, probably new mastering software to switch from DSD to DxD for My:
-experimental, yet still safe for human consumption ==>> "un-mix tapes" which are basically:
Backups of my purchased SACD collection for playback on my 3-way tri-amplified pair of Sony loudspeakers.
Each driver has its own amplifier channel left and right, six channels.
*I use a software DSP to Pre-Process the DSD audio files into what I call a:
∆∆ Stereo to 'spectrum split' 6 track demux-tape, burned onto a multichannel DVD-Audio disc or Blu-ray.
Basically, we have the Sony speakers that have a really bad crossover design, but really good drivers and enclosure. I don't want to mention their name because they need to stay cheap. Ahem.
We have the crossover points that are optimal and suggested by many people on the oracles of audio websites. Remember, a hundred people's opinions can't be wrong.
We also take the test measurement microphone and kind of make sure the sweeps are reflecting the driver curve which were also wonderfully provided by the many though various levels of quality, aforementioned oracles.
I have a chrontab (a script) running on one of my NAS servers that automatically received RIPs from a SACD/DVD/BLURAY disc that is inserted into the 'Sony Blu-ray player of specialness' (complete with the extra-special sceeen saver update thumb drive), in the handy dandy DSD format.
This back up folder is polled regularly by the handy script, and when an audio file pops in, software DSP will process The stereo files into the six split tracks automatically, while performing and applying the following operations/filters:
*Split stereo channels at preselected crossover points/slopes with the magic of 'fancy pants math' applied and plenty of 'manilla envelooe filtration' - rhe 6 discreet channels are meant for reproduction on multichannel DVD audio or BOO-RAY players.
In my experimental demo system, unlicensed and unwarranted I tell you...
The 6 ch out from my Sony Blu-ray, feeds amplifiers, that each feed the drivers directly, with their own high quality low oxygen wires, so nobody has to fight about anything. It's like all the kids got their own toy to play with in the back of the car.
One might process the digital stream out from the Sony with an "outboard DAC of quality and heft" as it were, -- if one were to be so rich and famous as to afford DACxation better than my Sony Blu-ray internals, that is.
™™Hey, Cheap audio man said it was okay to do it, you can't pick on me/™™
Software DSP applies correction curve for:
"driver anomaly compensation"
This smoothes out the Ring rings and the wawas that happen when the drivers get agitated by The amplifiers picking on how big their inductance and outductance and circle duck dialedness...
These are all super tech terms that you will learn about when you're older Paul..
Next is the optional "EQ wizard acquired room correction filter."
Which we usually use unless we're going to take this road on the show. No, wait strike that - reverse it. ..
We are installing also a three-way system in the automobile and a portable Bluetooth streamer thingy that has a 5 channel DMX for a kicking 2.1 amazingness. No EQ wizard curve on those files for example, just the ones that are going to live in the living room. Redundantly.
Since we have this kind of granular control in the pre-production mix down, We are noodling around with attempting phase / timing correction.
We have a carver Sonic holography preamp/spent uranium frame over here. We were thinking of measuring the signature for creating a filter to use the DSP modelling and the data from the signal analyzer comparator thingama-bobber 5000, 😊 and you get a lawsuit and a ticked off ghost of Mr Carver, most likely. Okay we'll skip that one but you get the point.
So it's your fault. All your fault. Every bit of it.
Your input is appreciated. Cuz you started this. Fine sir. Cheers to you for spreading your knowledge and The bug of audioaphilia.
Okay that's weird on two levels I don't like talking about bugs near my ears or something that sounds like necrophilia.
We'll name it later.
I have research to do and money to spend un-wisely....
sure beats Netflix though !! 😅
Muchas gracias
Myke
Bernhard Roads
I think Rob watts of Chord Electronics prefers PCM to DSD.
I've found that PCM sounds better than DSD on Chord devices. I think it has to do with how their FPGAs upsample all audio. Seems to be a better synergy for PCM than DSD.
Carroll Passage
How nice it must be for you, Paul, that you get to live in physically beautiful Boulder, Colorado AND this metaphysical La, La, Land . . . at the same time.
Looks like there is 2 from bubu land here to, why write something like that?
More La, La Land, Paul! Rocky Mountain High dreams of capturing AND releasing great sound is a noble pursuit!
Sabina Creek
Jaron Loop
And 3. If you want real anolog audio get vinyl or download from a website or torrent sound literally beats dsd by far
It's sad that only people who are selling DSD products hear any difference between DSD and PCM. 😢
I am not selling any DSD products, rather listening to them, and I hear immediate and huge differences between DSD and classic PCM standards (CD, FLAC, nevermind MP3).
A lot of PCM stuff is very nice and enjoyable, but pretending it is perfect and unimprovable is dishonest.
This is not true
@edfort5704 Have you done these comparisons on the same music fone for both formats?
@@craigellsworth3952 If you mean the same songs or the same music types, yes, both.
It must suck to be an audiophile looking for new things to worry about to spend even more money or make bankers very happy.
Schroeder Corner
It's not your dac. It's your brain.
And ears, and chest, and the fun 'gland'. :)
And 2 this guy dont know that much about audio. More info. generally speaking, PCM is considered to be closer to analog sound than DSD, as PCM is the standard digital audio format that directly samples and quantizes an analog signal, while DSD uses a different encoding method that can be seen as a more "indirect" representation of an analog waveform.





Key points to remember:
PCM (Pulse Code Modulation):
This is the most common digital audio format, where an analog signal is directly sampled at regular intervals and converted into digital values, making it a relatively straightforward representation of the original analog signal.



DSD (Direct Stream Digital):
This format uses a single-bit data stream at a much higher sampling rate than typical PCM, which results in a different way of representing the analog signal, often described as being closer to a "pulse density modulation" approach.
This is the dumbest thing I have listened to in a while
Retire
“I can hear a slight difference.” No, you can’t. Paul, you don’t even properly understand digital audio given the incorrect things you’ve said many times regarding PCM. You think you do, but you don’t. If you did, you wouldn’t be repeating some of the same nonsense that way too many others repeat. You’ve got a lot of experience and knowledge on many things to do with audio, but you do NOT understand how digital audio works. There’s only so many times a guy can tell you to go watch Digital Show & Tell until you do understand things properly, but it is clear you’d rather continue to think you already know what you think you know rather than actually learn the correct information. It’s sad, concerning you. And aggravating, concerning everyone else you tell incorrect information to. Ignorance is not a virtue.
So then why don’t you make a video on your channel about PCM, DSD etc? and spread your absolute vast knowledge of the subject so others can know the truth rather than waste it on bashing paul. I’m sure many could benefit since you have all the correct answers.
@@Audiofreak71 Because the Digital Show & Tell video already exists. Pointing out that Paul is wrong is not bashing him.
@@ClaytonMacleod well i already know it’s out there and i know the answer as well , but what’s the point of pointed out he’s wrong? Makes no sense to me unless it’s self gratifying to you? Either way I find it strange when people find it necessary to point out a mistake or wrong answer a professional has made as if it discredits their whole life’s work due to one mistake.
@@Audiofreak71 Dude, he's telling others, and others believe him even if he's wrong, because they don't know any better. So if Paul instead had the correct information and shared that instead that means not only does he now possess the correct answer, everyone he tells that believes it is the correct answer will also have the correct answer. Spreading correct information rather than bogus information is a Good Thing. TM It is better for everyone, including him, if he both has and shares correct information. Has absolutely nothing to do with me or how I feel about myself. Smarten up.
@@funny0000000 Good job calling his viewers feeble minded, and encouraging harassment. Very adult of you.