kye

Tomato - tomato!  🤣
 
The fact you say "the noise that comes from adding gain in post" makes me suspect it was more the issues with the equipment you used (low end consumer goods with noisy preamps) rather than the file format you used. 
The fact that everything with 32bit (MixPre series / Zoom F6 / etc) today has rather good preamps could lead you into thinking it was the file format which made the biggest difference rather than simply the general upgrade to newer modern semi-pro equipment. 
 
I disagree. Just because you're on a scripted drama series doesn't mean you necessarily get proper rehearsals, and neither are you able to go up to the 1st AD and ask "hey can we redo that because I blew up the mix?" (well, you could, but too many of those and you'll be out of a job!)
Plus I work on plenty of documentary or reality tv shoots were there are no retakes and everything is totally uncontrolled surprises at times. 
I do try to keep that in mind, but I do work on a much more diverse range of productions than you think, not just larger budget productions but also a lot of "zero budget" shoots too. 
24bits of audio has 144dB of dynamic range. When did  you last record anything louder than 144dB? (120dB is already going past the limits of human hearing, at this point it can be painful and doing damage to you. The sound has been weaponised against you!)
Thus my point before, it it like people having their Sony PMW-F55 with 16bit of raw and complaining it can't do 20bit raw instead? Errrr... isn't it already enough? 
I think a lot of desire for 32bit is really a desire for better equipment. Upgrading from an H4n to a Zoom F6 then you get the majority of the benefits from the quieter preamps (and TC/metadata/etc). 
If the original H4n had preamps as quiet as a F6, then people would likely peak their audio a lot less often than they did back then. 
Should a Sony PMW-F55 owner shoot their interviews in 16bit raw "just in case they get it wrong" because they're "also juggling audio and lighting themselves as well" so as to "give themselves a fighting chance"? 
It just seems to me like serious serious overkill, that's "fixing" problem that really shouldn't exist. If you're paying so little attention to your audio that you've got no idea at where even your very rough general levels are at, I just feel that there could be other serious issues happening as well! 
Just like if a cameraman can't even roughly get their exposure right (thus "needs" 16bit raw), you'd worry also about if they are framing in the subject at all, have they got focus at all, are things in the frame that shouldn't be? etc etc etc 
I'll happily record whatever they tell me to record in! And are paying for. 
But just like very very very few Sony F55 owners are ever being told to record in 16bit raw, but instead their bread and butter day to day shoots would likely be in 10bit, so the same is true for me. Will be many more years until the expectation changes from 24bit to 32bit audio files for me to record in. (am even a bit doubtful if this will ever happen in my life time. But "never say never", who knows, maybe in 2055 we'll all be filming even basic interviews in 32K of video resolution)

I'd take recording onto a FX30/FX3 directly over using a Nagra D!
Four channels of 24bit audio with the FX30:
https://www.bhphotovideo.com/c/product/1729318-REG/sony_fx30_digital_cinema_camera.html 
 
  MixPre6 & Wisycom MCR54 could fit under your jacket without anybody noticing, and there is four channels of lavs being recorded with ease. 
 
I'd imagine the Marantz PMD661 rarely gets used now you've got the Zoom F4. 
Am not sure if I'd even bother with a MixPre6 if your Zoom F4 is working fine for you. I'd rather aim a step higher and get a Zoom F8n, or a MixPre10. (or if doing this professionally... an 833 / Nova2 / Cantar Mini )

Couple of points to keep in mind: 
1) the professional market moves at a slower cycle than the consumer orientated market, look at how a new "Sony a7 something" comes out every time we blink, while the Aaton Cantar X3 is 8 years old (well, depending on how you count it... is over 8 years since it got shown off in early 2014, but orders didn't ship until later on in 2014 I believe) and yet the X3 remains today the very best you can get! A bit like ARRI who saw success so long with their original ALEXA sensor, only just recently bringing out the new ARRI 35 sensor. 
2) the Sound Dept moves on even a bit slower cycles than the Camera Dept does 
So yes, that's why even ancient recorders like the Sound Devices 788T can still be relevant today (especially if you're not working on the higher end productions. In which case, it is high time you upgrade to a Cantar! Or at least a Scorpio or 888). 
 

Remember kids...  when venturing into new camera territory always wear your protective equipment - you never know what dangers lurk in the dark...

Has anyone seen a glimpse of the dreaded hammer?

Not an audio expert by any stretch but it does feel like there are so many really great but also really affordable options out there. I still use a Zoom F4 and an older Marantz PMD661, but aside from keeping my eyes out for a great price on a used Mixpre-6 II - I think I'm good with what I have until one or all of them break. So @IronFilm I do wonder what that means for the audio recorder market - bcuz it does feel like it might've hit a brick wall?

Impressive Strides sounds like the name of a shop in 1960s Carnaby Street where Austin Powers would go to buy his best kecks.

The original claim by @The Dancing Babamef was....
My reply was simply that YouTube is good enough to show the difference between:
Resolution X sensor -> Resolution X timeline -> Resolution X YT upload  Resolution >X sensor -> Resolution X timeline -> Resolution X YT upload  It doesn't matter if X = 4K or 1080p, YT quality on both of them is enough to see differences.
I think people form the opinion that YT is completely useless and don't think it can do anything because of the way it handles grain (i.e. a bloodbath) and the high quantity of very low quality uploads where the footage has been crunched way before it made it to YT.
I'm yet to understand what makes a YT upload look worse or better, but there is definitely a lot of variation across different channels / videos.
In terms of a 4K sensor -> 1080p timeline -> 4K export-> 4K upload, that can benefit hugely from a little bit of sharpening.  Most of the people that think that 1080p is completely inferior to 4K are seeing the extra YT bitrate from the 4K YT stream, or are simply looking at sharpness, which is easily compensated for in grading.  

The original claim by @The Dancing Babamef was....
My reply was simply that YouTube is good enough to show the difference between:
Resolution X sensor -> Resolution X timeline -> Resolution X YT upload  Resolution >X sensor -> Resolution X timeline -> Resolution X YT upload  It doesn't matter if X = 4K or 1080p, YT quality on both of them is enough to see differences.
I think people form the opinion that YT is completely useless and don't think it can do anything because of the way it handles grain (i.e. a bloodbath) and the high quantity of very low quality uploads where the footage has been crunched way before it made it to YT.
I'm yet to understand what makes a YT upload look worse or better, but there is definitely a lot of variation across different channels / videos.
In terms of a 4K sensor -> 1080p timeline -> 4K export-> 4K upload, that can benefit hugely from a little bit of sharpening.  Most of the people that think that 1080p is completely inferior to 4K are seeing the extra YT bitrate from the 4K YT stream, or are simply looking at sharpness, which is easily compensated for in grading.  

Your perspective comes from a much more controlled environment than what the majority of people out there are working on. That's not to say your perspective isn't welcomed, because it is an interesting one even if most of us on this forum will never work at that level, but sometimes I think you forget that.
@kye's question was whether 32 bit float gives you headroom in the event you need it, not whether 24 bit is sufficient when you're on a set with a dedicated sound guy. (The answer is yes, btw. You're only limited by your microphone, as there isn't anything on earth that is loud enough to distort 32 bit audio. As long as the sound isn't too loud for your microphone you're good to go.)
The Tascam X8 (or other low end 32 bit float recorders) won't give someone audio that's better than a professional sound guy with professional gear can achieve. But for low budget projects or hobbyists, where it's not incompetence but the inability to juggle everything at once that necessitates recording on 32 bit, it's a HUGE deal and gives them a fighting chance. And eventually even sound professionals will embrace it once they let go of their pride. 😉
Me? I'm just happy that I don't have to worry about the 300 pound pro-wrestlers clipping the audio when they're yelling and hollering because I'm nowhere near the recorder and even if I was it'd be a nightmare to adjust the sound while also filming something so chaotic and unpredictable. Now I can get the shot AND get good, clean audio. 😊

Thankfully I'm not in the film industry but in the video production industry. 😉
32 bit float has been liberating as a one (or two) person crew. Setting my levels once (I still choose to), and not having to worry if someone (or something) gets too loud is beautiful because I have the range to still pull it down if they clip. It's not perfect, sounds can still be too loud for the mic itself obviously, but I've yet to experience that. I've just experienced lovely audio so far. 
I've never had enough hands to properly mix or adjust audio live the way I wish I could. Before I'd have to set my levels lower to be safe because I wasn't in a position to adjust them on the fly. Now I'm able to record two wireless handheld mics, the soundboard, and a crowd mic all at once with no clipping and without the noise that comes with adding gain in post. 
Everything gets written off as marketing or a gimmick at first. And for a lot of people it is completely unnecessary. But it's a game changer for a lot of people. 

Let's give a camera analogy to help you understand:
Let's pretend you have a camera that can record 16bit raw video files. (such as the Sony PMW-F55) Would you need 20bit raw video files?? 
No, of course not. (heck, even 16bit is overkill for most people! With 10bit being probably the most common bit depth) If someone is regularly messing up their exposure so badly with 16bit files that "it isn't enough", then I think they've got much bigger issues than what bit depth they're recording in!
Likewise, the amount of dynamic range in 24bit audio files is huuuuuuuge, you've got to be quite incompetent to not find that to be enough. (it isn't like back in the bad days of 16bit, when you had to be very careful to have your levels well above the noise floor while still being low enough not to peak, which meant for a fairly narrow window you had to aim for)
Let's have another camera analogy: if you're recording with a camera that only ever outputs 8bit 420, then why on earth would I want to record externally with a 12bit 444 recorder??? 
That's the case with audio, when 100% of the dialogue is being captured wirelessly. 
Thanks 🙂 Maybe in January I'll make a 2024 prediction! haha

Here is a short demonstration of the Cr channels: https://drive.google.com/file/d/1veiqGodXKxerdNF_ksqLlAnFWbGQUtEe/view?usp=sharing
8bit is rubbish, never ever use it. 10bit is somewhat better than on the X-T3/4, but the chroma smoothing is still there, ProRes doesn't seem to help either. Interestingly the X-H2s seems to have more smoothing than the X-H2.
As much as I love Fuji for stills, I think I'll stick to other brands for video as I'm very sensitive for this kind of stuff, but for 99.9% of Fuji video shooters this is still a non-issue I guess.

..yeah I'm an 80s kid!

Aftern hearing about this company in another topic after a few years, they've released some software to help people shooting cDNG. I've not yet tested it, but it looks pretty good. Essentially it's just playback and light raw controls for now, but their roadmap suggests they're working on exporting to ProRes, which will be great for cameras like the Sigma FP, that only record uncompressed.
https://www.octopuscinema.com/wiki/index.php?title=OCTOPUS_RAW_Player#Installer

This is an interesting camera that is now on Kickstarter.
Takes Canon EF lenses and shoots Fuji Instax Mini film.
Not sure about naming it after Prince Andrew though..
 
\\
 

Nobody is allowed an opinion if they haven't first developed, designed and marketed a cinema camera.
I guess we should all just shut up then
Apart from tupp?

Well I got this right. The Nova2 didn't come out in 2021! But it did in 2022... which was a surprise to me. 
https://www.jollypostie.com/store/gotham-sound/meet-the-zaxcom-nova-2-and-new-zmt4s-62f2ea42cfd0e
However the Nova2 was a relatively "small" update. (but very handy for those who it does scratch that itch the original Nova wasn't managing to reach)
Well, I guess I kinda got this right? Sort of. 
The Zoom F8n Pro was formally hinted at in 2021, and officially announced early in 2022. 
But the F8n Pro was also a very small update, not the bigger "F10" update I was hoping for. 
Ah well, but in general my predictions ("not much will happen") for 2021 I got right! 
 

No more of him on this forum please @kye

I was hyperbolizing it a bit but what I said is true to an extent. 4K to 1080p is what you do when you are playing the file locally but uploading to Youtube it's better to leave it as 4K and add a slight NR to it. 
 

Plus you get a lot of popular 2022 convenience features at a very low price point. You really can’t get full frame 10 bit PDAF and IBIS for this cheap otherwise

Yeah Kye I never got that from your posts.  My whole point was that downsampled 1080p looks very similar to 4k even when viewed on a 4k screen, and because of that fact, the Atomos Ninja Star makes the Nikon Z6 a pretty great camera in 2022, as you get a great image but much less inconvenience compared to most external recorder/monitors.

My vote is for Cinemartin 🤪

 

1. The pixel-level read node design is similar to IMX610, only the implementation is different and dictates that it has fast conventional read speeds.
2. The physical pixels are 49 megapixels and the entire sensor is natively designed to be 49 megapixels, reusing the 4.2um BSI pixel design and using both analogue binning and digital binning together to achieve a 12 megapixel sensor.
3. IMX510 disables the all-pixel readout mode, so there is no possibility of an all-pixel readout, and naturally it cannot achieve 2x2 OCL AF, which is the biggest difference between it and IMX472.
4. Dividing IMX510's logical pixel into four physical pixels. The readout is achieved using 1:2 analogue binning before PGA and digital binning after ADC. The specifics are:
 A. 48 physical megapixels, divided into upper left, upper right, lower left and lower right in-group pixels.
 B. 24 mega pixels are read, binned in the form of upper left + lower right, upper right + lower left, dual stream 14bit readout.
 C. 2:1 pixel binning in the digital domain to generate 12 megapixels at 15bit, discard 1bit to 14bit output.
 D. The readout speed is around 21ms, approximating 48fps; when the precision is reduced to 12bit, 96fps can be achieved.
 E. The ADC does not have an 11bit mode, so it cannot achieve the faster 24M 11bit -> 12M 12bit.
 F. The digital binning discards 1bit of precision regardless of the mode of output, a waste of performance deliberately designed into IMX510.
 G. In one video mode, the internal readout of two 3.84K/128fps 12Bit ADC streams are digitally binned, but the resolution is not twice that of 3.84K. The precision and resolution are wasted.
5. Due to the uniqueness of the readout mode, IMX510 cannot achieve any 2x1 OCL in-group AF, and the orthogonal readout pixel groups cannot be used for phase detection. Therefore the only phase focusing design for the IMX510 is masked PDAF. A focusing method using 2x2 OCL AF will only be available when the all-pixel readout mode is unlocked.
6. The hardware performance of the sensor goes well beyond the limitations of the "IMX510" name.
7. If this sensor were to be a normal Bayer sensor, the readout speed would depend only on the total number of analogue pixels before the ADC, due to the pixel readout design. Thus 48 megapixels at 24fps 14bit. For this a Modified Bayer CFA can be used, which is suitable for pixel designs with 2:1 analogue signal binning - maximising its performance and enabling dual mode switching between high resolution and oversampled high speed shooting:
 Crop to 16:9 to achieve 8.5K/57.7fps 12bit, 2x4.35K/115.5fps 12bit respectively; crop slightly to 7.68K/64fps 12bit, 2x3.84K/128fps 12bit.
 Notably, its 2x3.84K/128fps 12Bit readout truly has double the resolution and achieves IMX301-like oversampling performance (Sony F65RS).
8. The readout speed of any column-parallel ADC design of an image sensor must be scaled by the line readout speed, by the total number of pixels multiplied by the number of columns, and at the same level of precision comparing:
 A 48 megapixel 14bit 24fps sensor reads at a larger scale than a 12 megapixel 14bit 48fps sensor, but we cannot call it a greater total number of pixels read out, but rather a faster readout, measured by miliseconds.
 
 At a given precision, readout time (the time taken to read a frame), readout scale (how many pixels are read in a second), and readout speed (how many rows of pixels can be read in a second by a column of ADCs), are three dintinctly different and important metrics.

Here's a handy little tip for, I guess, any V-Log shooters. If you use the default V709 monitoring lut using the V-Log View Assist function then you still get a relatively washed out image. Not only is the contrast level quite mild, but the black never becomes true black on the LCD or viewfinder.
It makes it a bit harder to eyeball exposure as two stops under or two stops over still looks decent on the screen.
I made a VLT file (link below) with much stronger contrast - around 7.5 stops, which is the average scene brightness range. If you stick this on your card and load it up in the menu and use it you will get true blacks.

But it also makes it much easier to eyeball if you're slightly under or over as the contrast level is quite higher than the Panny default.

https://drive.google.com/drive/folders/1qYrz5lq4CUgWkKgudGI24f2VWp3CdkCu?usp=share_link
Note: it does make banding more visible as it's a low resolution LUT. That banding is "there" in the default View Assist LUT as well, it's just not as noticeable. But obviously it has no effect on the footage recorded.
 
 

OK, I just popped outside the back door and pointed the camera at a street light at night at base ISO 250 and DR Boost 2000 and there was no streaking visible in the viewfinder.
Honestly, it's such a non-issue. Don't let it put you off. Like I said, I had totally forgotten about streaking until now, because I simply haven't seen it since the time I succesfully attempted to reproduce it.

There's a lot of opinions about how good YT is or isn't, but I'm not so sure.
Here's a video from ARRI that is "only" uploaded in 1080p, but just looks fantastic...
I own both the OG BMPCC and BMMCC cameras which are native 1080p sensors, and even shooting in RAW or Prores HQ and processing them in post, I still seriously struggle to get an image as detailed as the above, even though the above has been seriously compressed by YT.  The BM 1080p cameras have a slightly softer pixel-to-pixel transition, simply because they're not 1080p 4:4:4, whereas downsampling cameras are all going to be 1080p 4:4:4, and YT has enough quality at 1080p to show these differences.
With my GH5, the difference in resolution between the 4K mode that's downsampled from the 5K sensor and the 4K 1:1 mode is definitely noticeable, even though the 4K 1:1 is a very small crop and even if you adjust ISO, SS and aperture to create the cleanest and sharpest images possible.