Paul explains digital audio with a new term

[@rayl6599]

I feel the analogy is not so great on the reproduction end bec video actually shows the frames in quick succession whereas audio uses noise shapers to low pass filter the output. So if you were to re-capture a single point in time, for audio, you might capture a sample between two origianl PCM samples, yet this is not possible for video. i.e. video playback relies on your eyes/brain to do the "low pass filtering" equivalent, but audio playback uses hardware to do it instead of using your ears/brain.

[@boblehman1726]

Excellent point! Paul's answer is pretty good, but this important point needs to be included. Because of video & film's dependence on the eye/brain illusion, some "under-sampling" errors are very common (e.g., stationary or jittery helicopter blades, backwards-rolling wagon wheels), and we've all learned to live with such errors/artifacts. Similar errors/artifacts COULD happen with digital audio, but they're rather effectively eliminated (below the threshold of hearing) by the anti-aliasing recording and reconstruction playback filters.

[@PetraKann]

Interesting points that you raise

[@spentron1]

Yes. And even a tiny moment of audio is many waves, and the sampling has to keep up with the waves much faster than moments. When I finally understood that the A->D sampler can't be allowed to see frequencies AT ALL above half its sampling frequency, then I understood how it can reproduce that exactly.

[@TheDanEdwards]

Reasonable take. Yet also we can think of the brain (and associated sensors) as a low-pass filter. We have a limit on the speed at which we can take in information. So that is why we think video is motion instead of a set of still motion.

[@stefanblutke8053]

I‘m gonna use the film reel as new analogy picture for my students at school now… 😊

[@MD-md4th]

This isn’t quite right. We do not here a bunch of individual snapshots of voltage. Yes, that is how the recording is made, but the DAC uses interpolation to create a continuous waveform of varying voltage out of those snapshots of voltage, which is then sent to the amplifier by the output stage.

[@kaichiehchang2289]

thank you….

[@jimw5165]

@@kaichiehchang2289the filter does the smoothing

[@jkwhtsll]

Paul, let me try this one more time. Your train analogy is not bad. The stream of data are points taken at intervals separated in time by the sampling rate frequency. Assuming the signal is constantly changing, the then decoding the data stream back to voltage variations MUST be stairstepped as there is no information in the stream as to what the signal voltage was between the sample points. However, the resulting stair step represents a signal variation at the frequency of the sample rate. The low pass filter removes that frequency variation. All of the demonstrations that there is no stair steps show a pretty sine wave. But what if the the input was not a sine wave? Surely your engineers can show us the signal coming from a MK2 with the filter turned off in the software.

[@Oystein87]

Video is usually 24, 30 or 60fps. Normaly around 30fps as you record it in from your iPhone. Movies are mostly 24fps.

[@unity1015]

Filters and mechanical properties of the speaker smooth things out.

[@jkwhtsll]

Video rate for professional movies is 24 fps. Paul, that we do not perceive the stair steps does NOT mean they are not there. To repeat myself, digitize a 6k sign wave at 24k starting at 0. You get 4 train cars: 0, max +, 0, max - do the same for a square wave with the same maximum amplitude. The data will be identical even though the analog sources are distinctly different.

[@Fastvoice]

There is no "video rate for professional movies". It's called "frame rate". Videos have different frame rates than movies. You can watch some videos here on YT with 60 fps.

[@paulchristman5875]

There are videos on KZbin showing a bird flapping its wings. The speed of the flapping wings and the video frame rate are exactly the same speed, making the bird look like it's floating there magically.

[@googoo-gjoob]

thank you, young man

[@puglife6291]

Is interesting how interlaced video tape has a smooth continuous look to it as more change is happening more often even if the change is only is less. Also how a plasma TV display has a pleasant smooth yet detailed image quality because the pixels are slower at transitioning than an lcd. Is this possibly similar to a tube amp? The tubes blend the incoming signal slightly due to being less efficient and apart from pleasant distortion could this blending help create that warm syrupy sound we like. Perhaps we need to stop worrying about digital VS Analog and look at ways to convert digital signals in a most pleasing way. It will never be vinyl as the mechanical nature is entirely different. But we can make it just as pleasing sounding just different.

[@Ed.T-p1b]

Just wonder how fast the coil in the woofers and tweeters have to response, and how many mixed of frequencies they have to handle effectively at any instant.

[@razisn]

The maximum speed of oscillation of a driver is exactly the same as the highest frequency it is designed to reproduce.

[@D1N02]

Video used to be 50 or 60 times a second depending on the TV standard in your country. PAL or NTSC. NTSC was faster but less resolving, than PAL. It was linked to the AC power frequency.

[@Fastvoice]

AFAIK it was only half of a picture every 1/50 or 1/60 second (interlaced) - so in fact it had 25 or 30 frames per second.

[@Oystein87]

No, 24, 25 or 30fps. You are thinking of the Hz

[@NoEgg4u]

@3:45 "...play 'em real quick..." Not mentioned is that the timing of each snapshot needs to be extremely accurate. Otherwise, the playback will be jerky or unnatural. With audio, we would hear jitter. In fact, with audio, nearly every stereo has noticeable jitter, and yet it goes unnoticed -- until you virtually eliminate the jitter. An A/B comparison allows you to hear the differences. It is sort of like growing up watching TVs with tubes, and most people thought that they looked fine. But compare that to a high definition TV, playing a 1080p movie, and you can immediately see the difference. Then bump that up to a 4k video on a 4k TV, and you can, again, see the improvement. But without ever seeing the improvement, each generation seemed fine. The same goes for nearly all of our stereos, when it comes to jitter. We play music in our homes and in our cars and with ear-buds in the street, etc. It sounds great. Well, when you then minimize the jitter that you never realized you had, you will hear a nice improvement. Minimizing jitter is an engineering accomplishment that requires a quality transport. You are not going to find one of such quality in cars or portable stereos. For the home, there are many to choose from, and the best ones are expensive. Price does not always dictate quality. But the ones that really do the best job will cost quite a bit. And so it goes with our host's point @3:45. Quick is important. But the timing, the unison presentation of each frame, plays an important (perhaps critical) role in the quality of digital reproduction. The precise timing from frame-to-frame (from one digital sample to the next digital sample), when done right, with a quality transport, will bring your digital listening experience to a new level of sound quality. All digital music players (like CD players and streamers) have built-in transports. If you have one of the few high-end CD players, it will have a great transport. I have never listened to our host's CD players. He has extolled the virtues of his products. Perhaps his CD players and his streamers have the jitter problem well in hand. One day, I would love to hear one. In my local high-end store, I have heard an Aesthetix Romulus Signature DAC that includes a Teac transport, and it sounded glorious. I had music files on a laptop, and fed the music to that Romulus box, directly from the laptop's USB port, via an Audioquest Diamond USB cable, using JRiver's Media Center. It sounded very good. Then I played the exact same music files, after they were burned from that laptop to a CD. The CD was placed into the Romulus' CD tray (the Teac transport). Well, that CD which had the exact same music files were clearly better sounding when played from the CD. Ergo, that Teac transport was doing a remarkable job, minimizing jitter. And if you only heard the music directly from the laptop, you would have thought that it sounded great -- and it did, until the jitter was minimized with the Teac CD tray transport, and you could hear the improvement. We all have jitter, and I am not sure if affordable, consumer level audio gear will ever address the jitter, because nearly no one knows that they have jitter, and nearly no one cares. The A/B comparison that I heard was the first time I got my brain to detect jitter. Now I know what to listen for, so I am cursed. ;-)

[@Spractral]

Now how does analogize it when it's a concert film? Things to think while I start watching Brahms 2nd at 445am

[@kimbaleon27]

Now that its 8:34am... In terms of digital media, when you're watching that concert, you're watching one of those flip-books (or film reels), while you're listening to another _separate_ flip-book (or film reel). I'm sure the processing the equipment does is a _bit_ 😉 of a different story.

[@davidstevens7809]

Do i get an interview yet paul

[@Raymondey]

Video rate is 30, 50 and 60 fps, also 120, 240 etc. fps if you want to make slow motion video afterwards

[@karthikeyan-lv5on]

Should we buy DAC, Pre amplifier, streamer, Equalizer, DSP, dividing networks, power amplifier, processor, etc., separately in the name of building separates system!? If it is about lady and separates system, I'll agree! But not ready to buy every capacitor and IC, etc., in the name of separates system. This is atrocious!?

[@stevenholquin2127]

Side Note: If You Notice Paul and Everything Else in This Video Episodic Is In Focus The Mid Ground The Back Ground and The Fore Ground Back When The Camera 🎥 Lenses Worked Like a Human Eye 👁️ Now The Camera 🎥 Lenses Are So Sophisticated Everting is In Focus Some IPhone 📱 Programs You Can Adjust The Camera 📷 Lens’s Speed You Can Slow it Down or Speed It Up Yet BlueTooth Can Travel at 2.1 Mbps I Mean This is Moving Information at Millions of Frames Per Second….This is Why If You Think About It Bluetooth Speakers Are Great Because They Are Referring To Frequency Hopping at Millions Of Bits Per Second Faster Than Any Conventional Speaker 🔈 Wire Sorry Folks I Got Carried Away Could You Imagine Paul on a Box Of Nodoze and Coffee ☕️ I Can Hear The Comments Now I Better Bounce Before I Become Persona Non Grata

[@Bassotronics]

Wtf kinda post is this? What weird grammatical output.

[@stimpy1226]

These cars are called data packets

[@johannesnel8060]

Most videos run at 25 frames per second. But higher quality videos can run at 50 or 60 frames per second.

[@MarcoRistuccia]

In such case our eye-brain system represents a biological low-pass filter.

[@user-od9iz9cv1w]

I was just thinking the same thing. The brain assumes these stairstep sounds or fast frame pictures are analogue and it works to smooth out the sound. Problem is, the harder it has to work the more taxing it is. Bad digital is irritating and exhausting. Much less accurate vinyl or tape sounds soothing. Great digital can be even similar. And a real instrument is always best.

[@boblehman1726]

@user-od9iz9cv1w The reconstructed sound with audio is not "stairstep".

[@user-od9iz9cv1w]

@@boblehman1726 It is a sequence of discrete samples, right? To some extent some DAC's employ filters and I imagine even an R2R DAC like the TDA1541a naturally smooths the signal to some degree. But it is by definition an inaccurate facsimile of the original signal that the brain works to interpret. If it was perfect there would be no difference between 44.1 or 176.4 PCM or DSD512 sample rates.

[@boblehman1726]

I didn't say it was perfect. Nothing in sound capture/recording/reproduction is perfect, analog or digital - far from it. Digital signal processing is a complex multi-step process. ONE SINGLE important aspect that Paul and others of us have been trying to explain in relatively simple terms, in which various significant details are often lost, especially when analogies are used, is that there is NO "stairstep" effect whatsoever in the reconstructed analog output from a DAC. NONE. But that seems to be the hardest misunderstanding to eliminate. Many folks seem to get stuck, forever, on some form of "stairstep" concept that they've seen in the analog sampling or quantization stages of the A-to-D process (before digital encoding even takes place). For absolutely all PRACTICAL purposes, the storage and later retrieval of the digital coding is perfect. In a DAC, the perfectly retrieved stream of digital codes, at their sampling-rate time intervals and bit-number quantization amplitude levels, are once again in a stair-step representation. But unlike film video, of flip-card image "samples", that stair-step digital audio signal MUST be run through a bandwidth-limited low-pass filter to form an analog signal. In Red Book PCM (16-bit coding, 44.1 kHz sampling), any imperfect artifacts of the processing would sound like hiss at upwards of 80 to 100 dB below the intended signal level (depending on the headroom allowed, noise-shaping used, etc.). Unlike with the analogy with discrete film/video/flip-card images, the human ear/brain has no need to "smooth out" any audio signal stair-steps - that is done by the DAC's low-pass filter, up to about 22.05 kHz, down to about -90 dB, well beyond what we can readily perceive. I hope that helps. If not, seek out some excellent KZbin videos that graphically explain it extremely well.

[@MarcoRistuccia]

Inside DACs there is a physical low pass filter which transforms an impulse series (not a stairstep) signal back into a smooth one. That's the reason why we don't see separated impulses when we measure the output signal of a DAC with an oscilloscope. I was saying that for cinema on film there is no corresponding circuitry. Signal is effectively output at 24 fps, it is an impulse series. Our eye-brain system turns the series into a stairstep due to a biological behaviour of our retina which is able to hold an image for a small amount of time. This avoids to some extent that we see abrupt changes when moving from one frame and the next one. However, if we pay attention, we still see the steps, especially when we see out of the corner of our eye. So, to be more precise, our eye-brain system is not exactly a low-pass filter, it is more of a latch/buffer which turns impulses into a stairstep. It lacks the "smoothing" part. So the signal is something in-between an impulse series and a smoothed signal: it is indeed a stairstep. On the contrary, our ears do not even feature such latch/buffer capability. So we need for sure a physical low-pass filter.

[@stevenholquin2127]

Back Before Digital Cards Where Used Camera 🎥 Frames Where 20 to 24 Frames Per/Second So Let’s Jump To The Present Cameras 🎥 Use a Digital Card and The New Digital Cameras 🎥 They Shoot at 120 Frames Per Second So What Can Possibly Go Wrong….Everything! You See at 60Hz or 60 Cycles a Motion Picture Studio Lights are Plugged Into a Supply or Line Power The Lights Are The Load AC Voltage So The Lights Are Plugged Into AC Voltage at 60 Hz or 60 Cycles Yet Today’s Cameras Shoot at Twice That Speed of 60 Cycles or 60 Hz at 120 Frames Per Second So What Can Possibly Go Wrong Well The Camera 🎥 Is Shooting at Twice The Speed As The Lights at 60 Hz So You See The Lights Flashing Because The Camera Is Shooting at Twice The Speed of The Lights….I Won’t Use a Train Analogy I’ll Make it Even More Simplistic If You Watch Any Home Made KZbin Videos That Where Recorded on a IPhone 📱 And Lights Are On In The Background That’s Another Subject You Will See The Lights Flashing Because Even You’re IPhone Can Shoot at 120 Frames Per Second and This Causes The Lights To Look 👀 Like They Are Flashing But They Are Not Now The Camera Lenses Now Everything is in Focus The Mid Ground The Back Ground and The Four Ground They Used To Work Like The Human Eye 👁️ So Ladies and Gentleman Check Out Some KZbin Videos and You Will See The Lights Flashing Because The IPhone Is Actually Shooting Your Video Faster Than The Power That’s Running The Lights at 60 Cycles I Don’t Like Thinking 🤔 This Early in The Morning I Put Paul On So I Can Fall Asleep 💤 Not Wake Up 😊 So Good Night Miss. Calabash…..

[@jimmygigga6819]

blah