Видео

I won't be surprised if there's a class action suit in the future. Anyone who bought that mic based on the claim that it makes the gain knob obsolete should be able to get their money back.

The weakness of consumer cameras like the Gh6 are their pre-amps. They aren't designed to amplify large diaphragm dynamic mics, for example. Would require bigger cameras and more power draw. They work best with small high sensitivity condenser or electrocet mics or with near line inputs from a wireless system of external recorder (which will have its own power supply for the pre-amps). In short, if want to pre-amp a mic that needs significant amplification you need external. If you're just using a wireless system you'll see no benefit. All ND filters inherently screw with colors, some not as much as others. I guess get better quality? Or just ignore it ;)

I've heard this before. 2 ADCs, 4 ADCs. Sound Devices has a patent on the idea but a patent doesn't mean you have a working circuit. Microphones are amplified by multiple transistors, probably around 10 or so. That's how weak their original voltage is. At a certain point the voltage is great enough to split it into two with no real lose of signal and both can be put through separate ADCs, BUT, and this is a big BUT, is that better than one ADC? I don't believe it is, for 2 reasons. First, the latest ADC chips don't claim, with data, graphs, that they do it and it has better resolution. Second, if such an approach worked why wouldn't we hear a demonstration of it? I don't discount that some day it may be possible. I just see no evidence for it and when the manufacturers, like Rode, says "the gain knob is obsolete" which is bald-faced lie my question to you is, why do you believe them?

@ they claim to use 2 ADCs. I have a mixpre 6 v2. I have tested with the same gain (high) in both 24bit and 32 bit floating. Playing high sound from speakers for both recordings, the 24 bit clipped. 32 bit float could be rescued. That said for my studio I use high end RME converters with 24bit and have no real need for 32bit in that environment as the quality of the record chain is very good with low noise anyway.

@@asmundma Do a test where you try to prove there is no difference then get back to me. As I explain in this video, it doesn't matter how many ADCs you have if the voltage out of the microphone's preamp can't surpass 24 bits even in a lab environment. Further, it's one thing to look at a meter, which is seriously dumbed down abstraction of the data (and what you hear) and think you see a difference when there is no difference to be heard. If you'd like to put some of your test files in google drive I'd love to analyze your claim! If one pays a lot of money for something one wants to believe it IS better than something cheaper. I have no beef with Sound Devices. They're rugged and look super sweet. But sorry, they have to obey the laws of physics, and limitation of material science, just like everyone else ;) All that said, Sound Devices probably has excellent pre-amps and that's worth paying for.

@@MaxoticsTV remember you put a gain in the pre-amp before it hits the Adc. Different mics have different output voltage, dynamics mic typically need higher gain to store the same “bit-value”. If you gain perfectly you may not need 32bit float to store without clipping. Some interfaces have a function to automatically lower the gain if a peek goes above e.g. -6dB. What you are probably missing is that if you gain too much you actually don’t have 144dB of dynamic range, you may start storing values very close to the upper limit. So let say your mic can deliver 0 to 1 volt. Your adc sores a 144dB value if it get 1 volt. Bit if you gain a 1 voltage signal in the preamp more it vil clip.

@@asmundma You don't need 24-bit, let alone 32-bit! At 144db you'd suffer hearing loss. Do you think you can sing that loud? Perhaps a scream could get to 120db. But if you set the gain to keep that under clipping you wouldn't hear anyone talking quietly into the mic. In my study of the data most 24bit is usually less then 16-bits effectively. Often, the computer just shifts the data over 8 bits because there isn't anything in that data. The Rode NT1 spec for example is 72db of SNL. Well under 16-bits. Their data, not mine! You're assuming that the ADC records the mic's maximum voltage at 144db. It doesn't. db is OUR scale. voltages are linear. Wherever it stores the data in the 24-bit range, there is generally only around 65,000 usable values (16-bits) that is above either nothing or noise.

I wouldn't know where to begin! Billy Hume seems to have done a fair amount of work for rappers and he seems very nice so you might ask him in a comment on his channel.

This is not software for professional video editors. As I said in the video, I'd done a lot in Premiere but when I haven't used it in a while and I want to edit together something fast then Filmora is quick and easy. I don't need to remember the shortcuts. Okay, I'm a dummy! Won't argue with you on that ;) Why can't premiere have both? What you're saying is exactly the problem. "Normal" people don't work with keyboard shortcuts. They just want to touch or click. My friend has done videos that he wouldn't have been able to complete in Premiere because Premiere would have overwhelmed him. Thanks for comment!

I agree with you! You can actually program any software to do it with a single key--in app, AutoHotKey, actionscript, etc. When I first saw the scissor icon I laughed. But like I said, when you want to do something without thinking it's surprisingly low stress and makes quick edits easy and quick! I'm a dummy, what can I say David ;)

@@MaxoticsTV yesss, programming adobe to do my bidding is my current state as a video editor. i use a python package called pymiere, but I should check those out too. Mac OS has the "automator" app as well

I never saw him comment on the forum. For all we know he doesn't go there either haha!

I get this all the time. "you don't understand the proprietary technique that makes 'x' possible.' They don't need to publish how it's done. All they need to do is provide reproducible tests and, well, hearing will be believing! I've explained in various videos why it isn't possible, though I'm sure it's not for trying. Again 1) no explanation in chips that the manufacturer would love to sell everyone and anyone 2) no empirical proof that 32-bit float recovers more microphone data than 24-bit. Thanks for comment!

​@@MaxoticsTV As I mentioned above, I think you've missed the point. The 32 bit float format has a dynamic range which exceeds 1500dB .. which is astonishing ( and unclippable! ) 24 bit PCM is limited to 144dB (max). The electronics interfaced to create audio files in the float format may well clip -- depending on how things are set up. The 32 float files themselves will exist intact, unclipped. Where they really excel is at the lower signal level settings. 24 bit fixed point recordings will start to degrade, noise wise, at the lowest levels. Not true for 32 float files. This video (from 5.38 onwards) illustrates some of the dramatic advantages of 32 bit float: ruclips.net/video/y4oNd1RgGL0/видео.html I suspect you've got your Rode set up wrongly. You say you can't see the reason for the gain control. It's just to allow sensible monitor levels . Any clipping will be a function of the Rode mic electronics, not the 32 bit float file format itself.

@@arjay1949 Everyone missing the bus here ;) I understand the range of 32-bit float. There are some issues with 32-bit float but yes, practically speaking, it is unclippable. My guess is you don't work with data. Forget dbs, they are a scale that allows us to treat the data in a way that we hear (exponentially). Curtis was fooled by the software, but I won't get into that. Suffice it to say, all ADCs output data in linear format. In 24-bit thats 000000000000000000000000 to 111111111111111111111111, or −8,388,608 to +8,388,607 (it's split in half for negative/positive air pressure voltage) You can draw your x/y axis anyway you want, scale it in dbs or whatever, but THE FACT IS AN ADC can only output BITS! Wherever you set your gain you'll even end up with 4 bits (if not 8 bits or even 10) of noise or "clipped" data. I didn't create the world, this is how we found it. This information is found in every mic specification. When you read 60db of dr think 10bits. When you read 120db, think 20bits, but seldom is such a range used for a variety of reasons. You have it right in your last sentence. Clipping is a function of what the pre-amp sends the input of the ADC. You simply CANNOT get 24-bit of usable data into the pre-amp. Your argument is is you have 32-bits (float) that it changes that. How? Whether a ruler is 12 inches or 24 inches it will never make 10 inches more than 12 inches ;) Cheers!

@@MaxoticsTV Probably best we agree to differ? The 'internet' seems very divided on the usefullness of the 32 bit float format. Some folk (like me) find it useful .... Others (like you) don't. Problem is with the internet is that it's almost impossible to know which technical comments are valid, and which are not. The only results you can really trust are the ones you deduce for yourself! ...... (I imagine you'll be selling or returning your Rode NT1 Gen5 ?)

@@arjay1949 I haven't finished my analysis yet! So who knows! I might be wrong about the usefulness of 32-bit float recording. I'm working on it some more right this second. Keep an eye on my channel if you're interesting. My favorite mic is the Electrovoice RE20, so I can't see using the NT1, but again, I try to keep an open mind!

Hi. Someone doesn't understand something here ;) Who uses 2 ADCs? Which chip? Can the extremely small voltage coming from a dynamic mic get split into lower and higher signals. Where is that done? Where is a schematic. Please don't be offended. You're taking as fact, fictions put forward by various RUclipsrs and marketing nonsense. So what am I misunderstanding. What chip from THAT, Cirrus, Analog Devices, etc., do what you say? None of the audio recording equipment companies like Sound Devices make their own chips--that I'm aware of.

I subscribed right away

@@kevinlentz7604 Thanks for subscribing to my all-over-the-place channel. It doesn't matter what equipment Hallelujah is performed on, it will always hit me. It's an interesting thing about music. You can't cheat it. Motown began in Detroit, not the home to the most advanced recording studios and musicians of NYC.

YOU might be an engineer ;) I used 1 volt because it's around the 0.7 volts I think you need at minimum to write computer memory. I didn't want to get into the weeds, just keep to the point that a microphone voltage is super, super low. And it's more than just range. This trips a lot of people up. As you point out, you're getting noise at the low end and non-linearities at the high end. I think it's a mistake to look at audio only in decibels which is really about psychoacoustic properties. The voltages are measured in bits (or transistor on/offs). They're just numbers. The wider range of numbers you try to work with the worse they are at the extremes. Is that fair to say?

@@MaxoticsTVto be fair i didn't look up the signal levels on common audio converters either. but it should be specified in the spec sheet of every intermediate to pro level audio interface. i know what you refer to, 0.7v is the point where silicon starts to conduct, which is the most important property of it being used as a semiconductor material. but that does not mean the audio signal level has to be over 0.7v to be converted. to your last point, the degree to which it works on a wide range and at the extremes is measureable and is specified on the lower end by the noise/thd level, which is commonly also in the spec sheet. the upper level limit is of course the 0dBFS max input level, after which, of course, clipping distortion occurs. but beyond noise and distortion there is no other limitation just because of wide number ranges.

@@fitzeflinger There is no signal without voltage. Most circuits work with what, around 3.3 volts. The signal must get there somehow ;)

​@ the adc has - as the name suggests - part analog and part digital circuitry. to send digital data and to write memory etc., you are right, there is a certain minimum voltage required, but that has absolutely nothing to do with the required level of the analog input signal. those are two independent circuits.

@@fitzeflinger The point I was trying to make is that the ADC needs a minimum voltage to perform quantization. They can certainly be two separate circuits with separate power supplies but I think it makes more sense to think of them as one circuit. One must bring up the mic voltage to reach the strength that the Analog part of the ADC requires. In short, the voltage output from the pre-amp must play by the ADC's rules ;) Therefore, to me, they have everything to do with each other! Hope that explains what I was trying to get at.

For a year I've ask anyone to provide a recording of the same source test sounds (controlled variable) in both 24-bit and 32-bit and to prove the 32-bit float contained data the 24-bit did not. So far, though some have sent files, none have proved it. My tests haven't proved it. I don't have a MixPre. People just SO WANT TO BELIEVE in technology. Fascinating!

I think that's part of it, but there were pianos in almost every house when I was young. Band was a big thing in schools. Talent contests, etc. So I believe the biggest spread of mediocrity is the internet in general. Easy for anyone to find an audience. You don't need to beg an A&R guy to come to your show. My epiphany came when my brother introduced me to Eric Sanko. We must have been in our early teens. Eric played "Donna Lee" on his bass. Time for me to find something else to be good at ;)

@@MaxoticsTV Yes internet ! Good point. Thanks for your answer.

Thanks! I hope to help with the theoretical stuff. The science of it. But I can barely record a child's birthday party ;)

I speculate a bit in this video. The manufacturers test to show some benefit but no one is paid to test whether the benefit is ever noticed. I believe the only reason some old Audio Interfaces begin to fail is discrete components like capacitors surrounding the chips, not the chips themselves. Again, I'm not an electronics engineer. I think it's easy to digitize any line voltage. What's difficult is interfacing with PC, Apple, Linux, Android, iPad, etc. Audio remains a second class citizen on most operating system--I mean for the audio engineer. But with the Behringer UMC1820 for $190 what more does a home studio really need in brining in 8 channels of analog from mics? Want another 8? For $150 Behringer has the ADA8200? There's so much of this equipment out there. Lost of people raving about their gear, but none saying theirs converts to digital better than the other. Again, I'm just looking at it as an outsider.

@@MaxoticsTV thanks - that's all good to know. 👌There does seem a bit of a gap in the market for budget ADCs with say 8-16 line level inputs/outputs without any preamps ? Ferrofish and Motu seem to have them at I guess mid level price.. but not really budget - Then there's the pro level converters with that weird plug interface. The Behringer stuff would do but its full of preamps, whereas I'd be happy with just mono jacks line in/out without a pad, etc. Guess they're about - I'll keep looking and hope the Gen 1 keeps going ..😬

@@pch554 Who in the home market has 8 people over playing at the same time? ;)

@ well for me, it’s convenience . Set up a couple of mics, mic an acoustic & a cabinet and stereo lines from a keyboard and some sort of audio in and bang - there’s 8. I really like to have everything ready to go before that fleeting flash of inspiration flies past. I know that’s common - sure not all projects are the same but being ready to go for most things (for me) helps a lot👀🔊

@@pch554 For me it's 8 blinky lights make me happier than 2 ;)

Genius does what it must, talent does what it can afford ;)

Good point! I certainly think I had a better listening experience than kids these days.

Hi. Thanks for comment! I've refined my thinking a bit since then. Even though 32-bit float isn't precise at the extremes, it probably wouldn't be noticed if you COULD fill those extremes with data. The bigger problem is the preamp couldn't give you accurate data at the extremes. However, I make the point about 32-bit float's inherent imprecision to sort of call out people who think they understand how it works but don't.

Hi! How can you know for sure unless you recorded simultaneously in both 24-bit and 32-bit float? This is why the scientific method is so important. We (including me!) come up with theories that something "saved" us when with proper testing we realized it didn't. Not saying you're wrong, only that it isn't proof for me. Hope you understand.

@ how can I know for sure what? That’s it’s saved me a number a times? Or that the mixpre is a good device? Or are you asking how do I know that 32-bit float offers the same fidelity as 24-bit? If that’s what you’re asking, I don’t know, but I also didn’t make any claims about that. I do know the device makes good sounding recordings with nice preamps for the size and price.

@ oh you are asking if it saved me. I know because, a handful of times when recording full concerts, I thought I set the gain correctly, but it turned out unexpected loud signals caused the channel to clip. With traditional 24 bit recording, there is no way to recover that information; the best you can do is use something like Rx de-clip. With the 32bit float sound devices, you can bring back that information that would have otherwise been lost.

@@miniatureface I was asking if your experience can be replicated (scientific method). Record yourself talking, or use test tones, whatever. Play them on your speaker and record in both 24-bit and 32-bit with the same gain settings (that should make it clip in 24-bit) then listen to the files and if you can put them in a shared folder so I can study them. Mine are here: drive.google.com/file/d/1w1CqIbgPhZq4SQfL7QTu-PxELvBPL_ol/view?usp=drive_link I show the process in the video posted before this unhinged one ;)

@ I can easily do this test but it’s also entirely unnecessary to prove my point. I’m specifically talking about audio that is clipping. You import 24 bit audio into pro tools that is clipping. You get a chopped off waveform at full scale and clipping distortion. There is no way to recover the waveform that exceeded full scale; it simply was not recorded. With 32-bit float on the other hand, you bring it into pro tools, you still find that it is clipping at full scale in just the same way. But then, you can go to audiosuite > other > gain/normalize, and you can render the audio file at a lower gain, and lo and behold, the information that exceeded full scale is actually there. This is a core difference between 24 bit and 32 bit float.

Thanks for comment! My rant is only on microphones, especially dynamics. I haven't played electric guitar in 30 years ;)

@MaxoticsTV it's a very good rant! My thought is that the concept of gain staging is the same between the two, guitar specifically is very challenging to get a di signal that has an accurate frequency response to what an amp will see. That requires both having a good pre, a to d converter, and knowing how your a to d converter responds. I only worry when my students get hung up on the gear. The cheapest is worth avoiding, but I believe good pre amps can be inexpensive. $350 for the Steinberg i have isn't bad at all. I think it's 97% of the way there with most mics, and for $6000, you get 99.9% of the way there. The difference is pretty big, but nominal in a mix, and I'd prefer a better performance with a lesser recording.

@@JewettMusic Agree! Tech, like art, as soon as you start taking it too seriously you've lost the thread.

My rant is based on the frustration of Rode, Tascam, Zoom and Panasonic saying 32-bit float "makes the gain knob obsolete" and "32-bit float eliminates clipping". I've made many videos on that. So one night I thought, I'm going to rant from the other side, the analog side. Explain why all the problems are in analog end, not digital. Digital can't fix what's missing in analog, it can only bandage it up. Other than that, I have trouble keeping my ideas in a straight line. I don't think in linear ;) Thanks for comment! Let me know if you have further thoughts in that context.

Cool. Let me know!

Once a microphone signal is pre-amped and quantized there is no "amplification", there are only numbers. -8,388,608 to +8,388,608. Nothing more, nothing less. Just one of those numbers every 48,000th of a second if your sample is 48K. One can scale those numbers to look like clipping, or re-scale them to "remove" clipping if they were improperly scaled in the first place. I look at the data, not at "charts".

@@MaxoticsTV I'm literally trying to explain you the truth about what they are talking about, the grain on truth in their marketing BS. If you TAKE that recording you made... in Audacity, highlight it... do to the effect called "amplify" and turn it down -6 db or -12db, the loudest part will not just be a flat line like it is where you see it hitting zero... there IS information there that you cannot see... THAT is what they are talking about when they claim their mic doesn't clip... it's not the mic... it's the 32 bit .wav file that has information beyond 0 DB that exists without being hard clipped and flattened. Try it and you will see for yourself that THAT is the only true part of what they are claiming, it's something anyone who produces music digitally knows.

@@RooftopKoreansMusic I did that and it is flat, clipped. Here are the files if you want to look at yourself: drive.google.com/file/d/1w1CqIbgPhZq4SQfL7QTu-PxELvBPL_ol/view?usp=drive_link

@@MaxoticsTV Something else is going on with your recordings... there must be a limiter also... I did my own tests just now... the 32 bit recording that is over modulated dose retain SOME details when reducing the gain of the recorded clip as I thought it would, although the obvious over modulation is still audible in the audio... that's something they don't address, even if the clipping doesn't occur the over-modulated sound does and is not desirable. I'm using Ableton 12 to do my tests, I'm thinking Audacity has a limiter at zero built in. Maybe the rode software doesn't limit the audio when you record so saving a 32 bit file with information above 0 is possible

@@RooftopKoreansMusic I recorded that with Reaper using Rode's ASIO driver. Again, the results are exactly what I expect. Clipping happens in the preamp long before 24-bits are reached but it does write some garbage in 32-bit float past that--I think--not sure yet. I plan to do some tests comparing both Rode's ASIO and WASAPI.

No one likes a party pooper I guess ;)

Thanks! You have no idea how lonely it is to be me talking about this stuff ;) I've given up on the video LOG and am now trying to explain why 32-bit float does not improve analog recording. It's not that I want everyone to agree with me or to shoot non-LOG. I truly believe one would be a better filmmaker if they understood how a camera works. When I was a child I thought "The Emperor's New Clothes" a true fairy tale. An exaggeration. Now I see it every day. Again, THANKS for comment and your detailed explanation of the issues involved.

I did a video a few days ago where I went over the details of the NT1 5th gen but no one watched it. So I ranted. Sadly, this is the natural state of my brain--star shaped topology, not linear. As you know, these topics ARE very complicated. One has to think through a lot of mechanisms. And then one has to fight through the vernacular.

If no one notices the bottleneck does it exist? ;) It wouldn't bother me. To me, the worst part of the mic is putting the USB port within the XLR port. The general engineer literacy of the buying public keeps sinking to new lows haha.

Some do talk about it as some objective quality. That's who I'm gently mocking. The science is pretty standard for all CMOS chips, unless I'm missing something. "Color Palettes" would have been a better term. Thanks for comment!

That's why I love him! ;) I totally relate to his attitude. Not bragging, my kids are often not amused !

HAHA. Good one. For me it is the first take. I only go down from there ;)

These kind of comments really crack me up. What about musicians who cover songs? Or do cover bands? Are they suppose to do that? Shouldn't everyone ONLY play their own music? I NEVER claimed to be a professional RUclipsr or a producer. I do this as a hobby. I'm clearly using his name to get views. I DO know something about writing/getting read/seen, etc. He CLEARLY used everyone's using Macs to do a video he knew would get views--complaining about all the cables. I didn't hold it against him! Quite the contrary, I wanted to hear another view. Same here, if you want to hear another view about Rick's stuff, I have one. If you have the emotions to leave this kind of comment you have the emotions to make RUclips videos. So create, don't hate ;)

Yes, but I do have something to say. RUclips is also a place for conversations. Rick talks about other channels and Instagram accounts too.

I was into it then. MacIntosh pre-amp, Polk speakers, etc ;) I screwed up not making that clear in this video. I was talking more about how cheap electronics changed music distribution.

And Cole Porter, Irving Berlin, Gershwin fading even further

That's the stuff that fascinates me to no end!!! Too many rules and structure leads to bad music but so does 100% freedom. Looks like we're the same generation. What pisses me the hell off is I wanted to fight the "establishment" and here I am saying we need it back ;)

@@MaxoticsTV When there is order, people seek chaos in their art; when there is chaos, people seek order in their art. People are comfortable hearing conformity in their music; artists are comfortable feeding the hungry people. The irony is that the rage against the machine is a self-loathing rage against their own power. I could be wrong. There is a lot of hate and power to go around, and I could be as blind as the next person in seeing all of it, but at least I see some of it everywhere which seems to be more than most can see.

Thanks for that story! When I took photos back then I thought long and hard, is this worth 1 of 24 or 36! Soon no one will be left who remembers those days when a producer's ass didn't clench up when someone said, "let's do another take" ;)

@@MaxoticsTV The session to record "Alfie" with Cilla Black at Abbey Road Studios had Bacharach himself conducting/playing piano (did the orchestration too). Huge full orchestra. George Martin was the producer. Bacharach made Cilla do close to 40 takes. Finally George Martin piped up to Bacharach and asked, "What exactly are you looking for?" Bacharach said, "Magic." To which Martin replied, "I think we got that on take 4."

Yes, I think this is true for all the arts. Don Quixote is still considered one of the best novels ever written. I'm a film buff. The movies of the 30s and 40s are peerless IMO.

On the aspect, I don't get it either!!!

Like in music, some mind it some don't.

Yes, I didn't make the distinction clear enough. Certainly, we had good music playing devices, but I believe it was new tech in 1980s which gave new life to those musicians in the 1960s and 1970s which made them seem better. Not that they weren't good enough for me ;)