Had a chat with a shop selling the GH5 and I'm now hearing that to meet a proper broadcast/film deliverable, your minimum sensor size is Super35. If you submit anything for primary content channels at Discovery or the BBC, anything below Super35 will be rejected.

The GH5 is a tough pill to swallow. Cause the EVA1 has a Super35 sensor, EF mount and etc with a sane recording backend that's similar to the GH5, but those planning to shoot nature stuff for Discovery or BBC will find out that their MFT sensor is too small to deliver as a proper deliverable if they shot it in ALL-I mode or with a Ninja Inferno on a GH5 to shave some dollars off.

The shop even said that the GH5 was a "YouTube camera" that shouldn't be taken seriously for real production.

What do you all think? Are the standards too strict? Is the shop BSing? I might just look for a FS7 used and pair it with a Fujinon MK lens. I'd lose realtime HDMI out in 10bit 4:2:2 for the framerates I shoot at, but internally, it would suit my needs.

With exception to big-budget jobs or when a client asks for a specific camera (which these days I consider those requests to be totally arrogant, ignorant or downright idiotic considering you *cannot* tell the difference on-screen anymore with LOG profiles and grading homogenizing everything), I've been using the GH5 for nearly all our productions with fabulous results. Even national campaign broadcast commercials. But to me this is no big deal: We produced similar projects on the HVX200 P2 camera back in the day...

The GH5 is in point of fact the only DSLR on the market that shoots the same codec, color space, data rate and formats as digital cinema cameras. Period. No other DSLR even comes close. And with the advent of firmware 2.2 that gap between the GH5 and it's photography-based brethren grew even wider.

The GH5 being a "YouTube" camera is total BS; that's what a salesperson says when they're keen on selling you something else - more expensive. Or they're ignorant of the full capabilities of the GH5. But as I mentioned above, the little Panny simply stands-out head-and-shoulders above any other DSLR.

The BBC's nose-to-the-clouds standards have never meant anything to me, but if you intend to submit to them or their affiliates then you just don't have any choice, you'll need to step-up. If you want to play in their sandbox, then it's by their rules.

However with a major project coming up I'm seeing the limitations of the MFT format myself, and mostly from a DOF and angle-of-view perspective. Something that S35 addresses. And, I'll be moving up to an S35 platform in the very near future and leaving behind the GH5/MFT format altogether.

To that end if you're seriously considering stepping up I can tell you with utmost certainty that the most logical step-up from the GH5/s isn't the EVA-1. Yikes, I can't believe I said that being a dedicated Panny shooter ever since P2 days, but I just got hands-on experiences with the Ursa Mini Pro, C200, EVA-1 and FS7 MkII. The clear winner was the Ursa Mini Pro - for several reasons which I won't bore you with here.

The FS7 MkII outpaces the Ursa Mini just by a slight margin - in certain areas, but I can't justify the price-point for it's output and feature set.

The simple fact is that even our venerable GH5 is limited to a 10-bit output, whether it's ALL-i or not. And when you see files from a 12-bit, 14-15 stops of dynamic range and compare it to the GH5... it's just no contest. There's simply more color information - a LOT more and it shows in color-finishing/grading.

Don't get me wrong, the EVA-1 is a great camera and because it's a Panny chipset the colors between the GH5 and it merge nicely in post. BUT, footage from the Ursa Mini Pro is just... that much more stunning. It's palpable.

Technology being what it is currently, I don't ever see a MFT camera coming close to having a 12 or even 14-bit output. It would require a much bigger heatsink on the chip than a GH5-sized body would allow. Not to mention the 50% percent crop from "full-frame".

So if you're like me and in a place that's forcing you to consider stepping up to a true digital cinema body, then take a good look at the Ursa Mini Pro. The guy "Wolfcrow" on YouTube did a fabulous job of laying out a comparison between all the sub-$10,000 s35 cameras, check it out.

Is it your intention to deliver to the BBC or discovery channel?

The BBC accepts DSLR as long as they "have EBU R118 test report results that meet the UHD Tier 2 or HD Tier 2L requirement".

UHD1 Tier 2 allows 10 bit sensors as small as 1/2 inch and even smaller if tested positively.

What kills most cameras from the EBU and BBC's point of view is PSNR more than pixels and sensor size. Noise levels in images aimed at broadcast need to be tightly controlled due to the signal path losses in amplifiers, attenuators, signal splitters and combiners, lossy cables etc in the transmission path.

Tier 2 UHD requires a signal to noise ratio of 44dB at 0dB gain. The best test results I have seen from a GH5 are around 41dB which is if you understand the dB ratio curve tells us that this 3dB difference in SNR means a doubling of noise. For broadcast this is a major influence in what cameras are acceptable.

https://www.dxomark.com/panasonic-lumix-dc-gh5-sensor-review-best-performer-in-the-lineup/

EBU Tiers here:

https://tech.ebu.ch/publications/r118

Most decent 3 chip 2/3" cameras such as a Sony PXW-400 can deliver a PSNR of up to 62 dB in the Y channel with noise suppression on. In reality as an RGB output signal this would be closer to 55-57dB. On that basis all the DSLR/mirrorless cameras are way outside the higher end tiers for broadcast acceptability. Even the venerable old Sony EX3 1/2" chip cameras delivered 54 dB in the Y channel. Canon's new C200 delivers 53dB (Typical) in 50Hz mode. Sony’s FS7-F5-F55 range deliver 57dB.

Basically if whoever you are shooting for is going to use the EBU Tier system to determine what camera level they want you to shoot with and want you to meet Tier 1 in either HD or UHD the camera needs to deliver better than 48dB. Most S35 single chip cameras need to be delivering figures well in excess of 48dB for a good SNR in a 10-bit 4:2:2 recording.

Why is this? As we can see from the table below to record a true full 10-bit image we require a dynamic range of 60dB. To deliver a full true 8-bit signal we require a dynamic range of 48dB. Cameras that deliver 10-bit 4:2:2 recordings really require a SNR of 60dB to ensure that most of the information is signal rather than noise in that 10-bit container. In practice and reality this is not the case but one cannot get away from the fact that the better the SNR is of a camera the lower will be the noise component of that signal. Quite well put here:

https://www.provideocoalition.com/s_n_ratios_demystified/

SNR is expressed in units of power or decibels (dB).

SNR (dB) = 20log (Signal e- / Noise e-)

8-bit 256:1 48dB Dynamic Range
10-bit 1024:1 ( 4x Increase) 60dB Dynamic Range
12-bit 4096:1 (16x Increase) 72dB Dynamic Range
14-bit 16384:1 (64x Increase) 84dB Dynamic Range
16-bit 65536:1 (256x Increase) 96dB Dynamic Range

Chris Young

Interesting there is a little footnote on negative gain settings. That has been a debated topic in the past as to whether negative gain actually gives you more usable SNR or if it really just brings the same range, shifted down by the 3dB for example.

True Kevin you are correct technically but in practice I've found most cameras that are capable of anything delivering around ten stops or better with a dynamic range in excess of 50dB suffer very little if anything visually when going out to a six stop Rec 709 range. I've worked on plenty of live OB's and visually seen nothing when we have tried negative gain. Sure from a post point of view the quieter the image the nicer the look generally.

Chris Young

Has anyone looked at the GH5s SNR numbers from both RAW bayer data and the video modes via HDMI capture?

The BBC have specs for different purposes.

You are simplifying this far too much. It is not just sensor size, but the whole style and system. BBC's Marcella, for example was shot on Panasonic Varicam 35 - and there is no way in the world a production like this would have considered a DSLR as serious competition to it. There are minimum standards for technical specs for these kinds of programmes, but far more to it than just basic techniques specs. The glass on the front of the varicam, for example.

Exactly Paul.

It was always the case with the Beeb when I trained there and basically it is still the case years later. There are certain standards they and most broadcasters have to adhere to for a number of reasons. One is that a broadcasters technical delivery requirements are based around the fact that its whole chain of delivery needs to dovetail into the interoperability requirements for the UK DTT platform. All these parameters are laid down by Ofcom in the UK.

https://www.ofcom.org.uk/consultations-and-statements/category-2/broadcast-tv-technical-codes

With regards to DSLR, and this applies to Mirrorless as well, the BBC are quite adamant about the use of these cameras. This information can be found on page 8 of the

TECHNICAL SPECIFICATION FOR THE DELIVERY OF TELEVISION PROGRAMMES AS AS - 11 FILES

This also applies to the following channels in the UK.

BBC, BT Sport, Channel 4, Channel 5, ITV, Sky, STV and TG4

As for the 60Hz world most of that is governed by SMPTE spec so that may be totally different. That said I've seen all manner of "non-broadcast" spec cameras used in all manner of broadcast productions. I do have to say though that the Beeb were always one of the toughest to get program tech clearance through. A few years back we produced a 13 ep series for Granada in the UK and whilst I always aimed for the maximum delivery quality I have rarely ever had a series accepted without some kind of a tech query but in that case all went through without a question.

My motto is if you can get a type of camera and kit or anything else accepted go for it but do your homework first. If necessary discuss with the broadcaster and submit any supporting tech material for the kit you want to use along with some footage samples If they sign off on that all well and good. If you can deliver quality like this off a GH5S I would be asking for sure:

Ettore's Stargate (The Journey of Nau-chan) on Vimeo

http://dpp-assets.s3.amazonaws.com/wp-content/uploads/specs/bbc/TechnicalDeliveryStandardsBBCFile.pdf

Chris Young

If I was shooting for broadcast, I'd probably only have the GH5 as a B camera for shots where a smaller camera was needed. However there is plenty of video needed that is not going to end being broadcast. It's not the be all and end all of video production.
Still this story emphasises my gear buying policy of doing my own research and then buying online. Rarely do I go into a store and if I do, I don't ask the store to tell me what I should already know myself.

Brick and Mortar is still good for bartering. I still tend to prefer brick and mortar for demos.

As long as my SNR on the GH5s is equal or better than my EX1R, I'm satisfied.

Brick and Mortar is still good for bartering. I still tend to prefer brick and mortar for demos.

At least you're in a place where there ARE retail stores that carry these products to check out in person!

The BBC are getting into hot water by equipping their journalists with phones for video shooting. Camera people are incensed that the Beeb's own standards are perfectly bendable when immediacy triumphs over quality - even worse when many of these phones only shoot in 30 fps! On one hand they have serious and uncompromising standards for some things, yet accept mediocrity when it suits. Journalists with hand held stabilisers for iPhones!

I agree with Paul about this duality of standards in the BBC.

Was filming something in Norwich recently & I noted the BBC journalist covering the story near me had “the iPhone kit” - and he was literally surrounded by people like me shooting on far better cameras. I watched the news broadcast that night and sure enough it looked really mediocre - and jerky (no doubt because of 30fps broadcast at 50i here in PAL land).

I accept that fast breaking news will use on-the-scene viewer/public supplied phone clips - but reporters shooting pre-planned news items on iPhones is just cutting standards too far in my book/something I find disappointing. Ho hum...

I think I'm pretty much down to a GH5s with a Speedbooster XL, or the A7sIII if it has 10bit HDMI. If the A7sIII still lacks 10bit HDMI, and still uses Micro HDMI, It would be the Speedbooster XL, a Tamron 24-35 for dramatic productions, and a stock 12-60 2.8-4 for Vlogging.

for Vlogging.

So the guy in the shop was right after all? ;)

OK,...this is completely unethical and I am NOT saying people should do this. In fact, I am saying "do not"........

However,...the image quality form the GH5 at 1600 ISO or below at 400Mbps will match the quality of any super35 sensor camera out there today. I mean it this way....There is no chance in Hell that if you shot wolves eating a dead bison at Yellow Stone and you shot it correctly with a GH5....that ANYBODY is going to say; "I dont know,...this does not look like a Super35 image sensor to me". This isn't going to happen! Not even the BEST trained eyes can tell a super35 image from an MFT image. (if shot correctly at 1600 ISO or lower.) This GH5 image at lower ISO's is spectacular and has been compared to RED and ARRI cameras in video all over the place. There are videos that compare it to the EVA-1 and some show the GH5 can be just as clean or even cleaner.

I can absolutely tell you as a 100% complete, absolute FACT that Discovery DOES take 8bit 4:2:0 footage and airs it. We also know the BBC has used the A7S-II 8bit bit for nature shots in low light and so does CNN for it's Wonder List series.

Look at sports. They often use tiny 2/3 sensor ENG cameras!

Ironically, all broadcasters are only capable of airing 8bit 4:2:0. It's funny how deliverables are so much higher than anything they can actually deliver themselves. Over the air, satellite and cable or only 8bit 4:2:0 in MPEG-II or H264 at 20Mbp/s or as low at 9Mbp/s in some cases by the time it hits your TV.

It is wrong to do this but if you shot something good on a GH5 and "said" you used an EVA-1...nobody will ever challenge it.

Again,...do NOT do this. I'm not saying anyone should.


FYI....George Lucas shot huge budget Star Wars movies (prequels) on Sony 1440x1080 cameras with sensors much SMALLER than Micro 4/3 on cameras that were WAY,...and I mean WAAAAAY less sophisticated than the GH5. The GH5's processing power is much more powerful than George's Lucas' Sony F900's...by orders of magnitude!

Wasn't the highest grossing movie in Hollywood history, "Avatar" shot on tiny 2/3 inch sensors? The BBC can't broadcast "Avatar" because it wasn't shot on cameras that meet their standards?

CT

This is what I meant in another one of my replies (about the 2/3" 4K cameras) about how companies fret too much about WHAT equipment you're using rather than what you can DO with the equipment. However I just flipped through a couple published technical specifications from some big names and they spell out pretty clearly (https://www.corusent.com/technical-specifications-content-delivery-broadcast/) that if you expect to get paid for the work or otherwise receive financial backing for the project - you meet their approved equipment specs. The only exception is if you can explain that lives will be endangered while using the prescribed equipment in a given scenario, then they might allow non-approved devices (such as action cameras). The consequence is that it affects what equipment models people will purchase, and therefore what manufacturers will focus on developing.

On the other hand in business: there's specification, and then there's what actually gets implemented, and sometimes they're not always the same. i.e. This is reserving their rights to pull the 'technical specification' card at any time to reject something, even if it may actually be for other reasons.

No doubt, all media companies have their standards for deliverables. Discovery has it's Gold, Silver and Bronz standards. I can only tell you that they ABSOLUTELY will allow you to go outside of it. You have to know who you are dealing with at the company too. If you captured something SPECTACULAR like a shark jumping or a lion fighting a crocodile or Northern Lights or anything that is really compelling or causes people to say "WOW!" and you did it in 8bit 4:2:0?....trust me, you are FINE!

In these cases, the content is WAY more important than the spec.

They just dont want entire production are not done that way.

Do you think Discovery would have an issue with the "toy effect" or "fantasy" filter setting on my gh5? I kinda like those presets, I could always say it was shot on a Alexa just to clear some doubt and I have some killer footage from my cat eating a dead mouse.

If your cat was eating that dead mouse and it suddenly became alive again and attacked your cat, killed it and swallowed your entire cat?....if you got that with your old flip phone camera....Discovery would take it and air it! The more amazing it is, the higher the chance that it will get aired no matter how bad the quality its. Discovery's own standards do allow for certain percentages to be non-standard and still make it to air. Like I said, the BBC used the A7S and so did CNN.

Did you know the original episodes of "Deadliest Catch" were shot on Sony 8bit, 1440x1080 HDV camcorders? The boats would leave the docks with 6-8 camcorders on board and only 3 or 4 would make it undamaged. They don't meet the broadcast standards but they were considered "disposable" cameras and that is why they used them.

Anyway,...yes, the standard is what they demand. but, if you have something incredible you shot on a GH5, they absolutely WILL take it if it's good.

That's all I'm trying to say. And, I do NOT advocate lying about it because absolutely nobody will be able to tell the difference! No way.

Cliff I think you will find they were EX1s which were a full 1920 x 1080. The "Alaska the Last Frontier" series is being shot on Canon XF 300 series in 1920 x 1080i MPEG-2 50-mbit 60i.

As you say the story is where it is at. If you had the leader of your country stretched out in the gutter, out like a light with a bottle of booze in his hand and you had shot it on a piece of string you would be $$$s in the bank.

Did you see Black Swan by Darren Aronofsky. It was shot on ARRI Super 16 using Fuji film... plus some Canon DSLRs. A beautiful production.

BLACK SWAN | Official Trailer | FOX Searchlight - YouTube

Technically not good enough for Netflix or Discovery delivery. Give me break many would say!

I understand standards, BBC's specifically as that's where I trained. Believe you me if it's good it will get used regardless of what the tech standards are. In most cases common sense will rule when a judgement call has to be made. If it is a commissioned program then that is a different situation. You are are granted the commission with all the tech specs ironed out beforehand. If you accepted the commission on the basis of S35 or 2/3" acquisition then that is what you are expected to shoot on. Today you would not get anything commissioned unless it was S35 or three chip 2/3" with a minimum of 50-mbit in interlace or 100-mbit in progressive. Exceptions can and are made but the basis of most commissioned work is based on the two criteria outlined above. In the bulk of cases you also have to supply the original camera media and or files.

Chris Young

It is wrong to do this but if you shot something good on a GH5 and "said" you used an EVA-1...nobody will ever challenge it.

Again,...do NOT do this. I'm not saying anyone should.



In the bulk of cases you also have to supply the original camera media and or files.


Oops, there goes your chance in lying about what camera you used :) It's good that some "facts" are set straight again now and then.

Cliff I think you will find they were EX1s which were a full 1920 x 1080. The "Alaska the Last Frontier" series is being shot on Canon XF 300 series in 1920 x 1080i MPEG-2 50-mbit 60i.

As you say the story is where it is at. If you had the leader of your country stretched out in the gutter, out like a light with a bottle of booze in his hand and you had shot it on a piece of string you would be $$$s in the bank.

Did you see Black Swan by Darren Aronofsky. It was shot on ARRI Super 16 using Fuji film... plus some Canon DSLRs. A beautiful production.

BLACK SWAN | Official Trailer | FOX Searchlight - YouTube (https://www.youtube.com/watch?v=5jaI1XOB-bs)

Technically not good enough for Netflix or Discovery delivery. Give me break many would say!

I understand standards, BBC's specifically as that's where I trained. Believe you me if it's good it will get used regardless of what the tech standards are. In most cases common sense will rule when a judgement call has to be made. If it is a commissioned program then that is a different situation. You are are granted the commission with all the tech specs ironed out beforehand. If you accepted the commission on the basis of S35 or 2/3" acquisition then that is what you are expected to shoot on. Today you would not get anything commissioned unless it was S35 or three chip 2/3" with a minimum of 50-mbit in interlace or 100-mbit in progressive. Exceptions can and are made but the basis of most commissioned work is based on the two criteria outlined above. In the bulk of cases you also have to supply the original camera media and or files.

Chris Young

Oh, they were HDV alright. How is this for "deliverables"? We, in fact had crews onboard the boats "delivering" their dailies by sealing HDV tapes into floating plastic throw-overs being tossed straight into the ocean! We then had crews waiting in small boats immediately to snatch them up and race the tapes back to shore!

Yes, other cameras were used in different seasons too but lemme tell you, that show was a camera grave yard in the first seasons!

On "actual" media straight from camera the being delivered? No, we have TONS of old content down the hall from vendors that sent deliverables on Digi SR tapes and even USB hard drives. Discovery also buys fully edited shows from vendors too and of course, we reversion them also. Not everything is 100% done in-house. So it's a very wide range of stuff.

OMG...I haven't thought about those days in a loooong time!

And yes, you guys are right, "content" always wins over "specifications" every single time. They will never say "no" to any material that is really good....no matter what. The standards are guidelines or best practices and exactly what the company is asking from everybody. I'm only saying that those rules can be and are often broken, you just need to have a good reason to deliberately break them.

CT

Okay, sounds like I will start with a GH5s then, if Sony doesn't answer back with 10bit in the A7SIII

if Sony doesn't answer back with 10bit in the A7SIII

They won't.

They won't.
They might add 10 bit, actually I think it is likely they will.

We will find out in September.

My guess;

In order for the A7S-III to have 10bit internal recording, Sony will need move the A7S-III out of XAVC-S and give it full XAVC 10bit. (this uses an .mxf wrapper today)

If Alpha gives the A7S-III full XDCAM 10bit CODEC, this will CERTAINLY piss off the XDCAM Pro division! They probably will protest the Hell out of that in Sony management meetings..lol . Seeing an Alpha camera with an incredible image sensor getting 10bit sampling would drive them NUTS!

I think in order to keep the Sony eco-system "balance" and keep everybody happy, Sony will keep the A7S-III with the usual XAVC-S 8bit 4:2:0 CODEC. I think they WILL lift the 29 minute recording limit and WILL allow 10bit HDMI output for external recording only. This will probably keep all the Sony marketing experts happy and they will not lose too much sleep if it is done this way.

I'm expecting the A7S-III to be a phenomenal performer in a big rig situation with a Shogun Inferno.

I think Sony will still clock the sensor readout pretty show though for these reasons:

1.) Clocking it slow will help control heat as we know that faster scan speeds make hotter and noisier sensors.
2.) Keeping a high degree of rolling shutter will keep the XDACM managers very happy.

I think that people don't realize just how much of a roll that Sony internal "politics" plays in the features that each camera is allowed to have.

We might find out this Sunday at Sony's NAB press meeting???

CT

I'm expecting the A7S-III to be a phenomenal performer in a big rig situation with a Shogun Inferno.

Not if it is 8 bit.

They should read the memo: the 10 bit HDR recording era has started.

Not if it is 8 bit.

They should read the memo: the 10 bit HDR recording era has started.

Here's a hint of what's to come:

the new Sony RX0, which cannot even shoot 4K internally, puts out a clean 10bit 422 4K signal via HDMI. It is Sony's first consumer cam to do that. The spec is confirmed by Atomos and by an independent test based on a Shogun recording from the camera, which can detect how many colors are actually recorded not just count bits.

Not if it is 8 bit. They should read the memo: the 10 bit HDR recording era has started.

So you're saying all the current 8-bit Shogun recorded A7S and A7R II videos aren't very good? Do you have specific examples in mind?

So you're saying all the current 8-bit Shogun recorded A7S and A7R II videos aren't very good?
Not for Log and HDR.

Well? High signal to noise ratio is still high signal to noise ratio, no matter if its 8bit or 10bit.

Now, I love 10bit just as much as anybody here. I love 12bit raw sensor data too. But I will never say that the A7S-II or A7S-III looks terrible because its only 8 bit. Its a stunning low light performer, even in 8bit. Nobody questions that. Not Doscovery, not CNN and not the BBC. They all know you can get night shots with that camera that you cant do with any other camera. (Before the GH5-S)

We'll see if Alpha can fight Sony politics and put out their first consumer camera with 10bit HDMI output.

Not for Log and HDR.

So you can point out the A7S/A7R Slog2 footage out of this trailer? Can you point to specific segments where the 8-bit shows where it's lacking?

A STREET CAT NAMED BOB Official Trailer (2016) - YouTube

That trailer is already botched, every fifth frame is duplicated.

What does frame cadence have to do with color bit depth?

99.999% of everything we watch today is literally 8bit color depth. Even for us guys that love and shoot 10bit, everything is delivered and displayed in 8bit anyway.

Have you ever watched a Blu-ray that you thought looked good?

CT

What does frame cadence have to do with color bit depth?

99.999% of everything we watch today is literally 8bit color depth. Even for us guys that love and shoot 10bit, everything is delivered and displayed in 8bit anyway.

Have you ever watched a Blu-ray that you thought looked good?

CT
You made your point, I made mine.

I think you should not use 8 bit devices for Log or HDR.

I seems to me we have to agree to disagree! :)

You made your point, I made mine. I think you should not use 8 bit devices for Log or HDR.

What's the last project you shot in Log and delivered, either in HDR or plan 'ole SDR?

What's the last project you shot in Log and delivered, either in HDR or plan 'ole SDR?
When I make a recording I record it with an eye to the future.

But since you ask I shoot everything in log and most of the delivery I shoot is in HDR first and SDR second.

When I make a recording I record it with an eye to the future. But since you ask I shoot everything in log and most of the delivery I shoot is in HDR first and SDR second.

And everything you have ever shot in Log is in 10-bit?

And everything you have ever shot in Log is in 10-bit?
Yes, except for testing in 8 bit.

In my opinion 8 bit log is muddy.

Yes, except for testing in 8 bit. In my opinion 8 bit log is muddy.

What do you mean by "muddy"? "Muddy" usually means that fine detail is lacking, both from macroblocking within low-bitrate 8-bit files (like AVCHD), or an aggressive in-camera or post noise reduction pass. Neither of those two suggest anything to do with Log.

Also, now that you've brought up that 8-bit log is muddy, you should be able to point this out in the A7S/A7R Slog2 shots in the "Street Cat Named Bob" trailer, because the muddiness should be exacerbated by YouTube, not alleviated (duplicated frames should not have anything to do with this, so that complaint is moot).

Folks feel free to use 8 bit for log and HDR, I certainly won't and won't recommend it either!

I am not on trial here for my opinion and I don't want to waste my time on this any further, if you don't like it so be it!

Folks feel free to use 8 bit for log and HDR, I certainly won't and won't recommend it either! I am not on trial here for my opinion, if you don't like it so be it!

I don't care one way or the other. What I don't like are people who give dogmatic recommendations when they don't have any reason why, and you clearly don't know, other than that 10 is a larger number than 8, which is probably the fundamental part of the opinion.

You made your point, I made mine.

I think you should not use 8 bit devices for Log or HDR.

I seems to me we have to agree to disagree! :)

Dont get me wrong,..I think we all know that 10bit is the optimal color depth sampling we's like to use with any log gamma. Yup,..if given the chance, I'll take 40 times more color space in a log recording ANY day! AGREED!

However, this does not mean that you "can't" get very good results with 8bit. You still can. And, if you use recorded 8bit color output to ProRes HQ, your results improve even more. May people blame banding on 8bit stair steps when the REAL problem is h.264 Log GOP compression instead.

When H.264 Long GOP is stressed, it clusters pixel values into single value "blocks". It can represent,...let's say 8 pixels and assign them all ONE single color value. This "looks" like 8bit banding but it's really not! H.264 will also sacrifice gradients and shadows first. Yes,..shadows WILL look "muddy" at times during highly complex scenes that stress the bandwidth cap.

Again, this is not 8bit's fault. 8bit's in ProRes HQ is actually pretty darn durable. No, it's not 10bit and yes, it's a "thinner" space,...but you can EASILY still get great really good results with logarithmic gamma color grading.

CT

May people blame banding on 8bit stair steps when the REAL problem is h.264 Log GOP compression instead.

Except it's not Long GOP compression, it's bitrate. AVCHD gets stressed easier than XAVC S. Both are H.264 LongGOP, but one tops out at 24Mbps while the other tops out at 50Mbps. At some point, you can throw enough bitrate at LongGOP, where it won't get stressed.

When H.264 Long GOP is stressed, it clusters pixel values into single value "blocks". It can represent,...let's say 8 pixels and assign them all ONE single color value. This "looks" like 8bit banding but it's really not! H.264 will also sacrifice gradients and shadows first.

This true of H.264 Intra as well, if you don't have enough bitrate for it.

May people blame banding on 8bit stair steps when the REAL problem is h.264 Log GOP compression instead.

When H.264 Long GOP is stressed, it clusters pixel values into single value "blocks".
All DCT based compression uses blocks whether it is H.264 or ProRes.

H.264 can compress in addition to intra-frame compression the motion compensated difference between frames.

Given the same bitrate inter-frame and intra-frame compression combined will provide a superior image compared to only intra-frame compression.

The only disadvantage to files that have inter-frame compression is that it is hard to edit because the decompression depends on multiple frames.

H.264 is neither better nor worse than ProRes, it all depends on the bitrate and you can encode H.264 all-intra and also use lossless compression.

So you can point out the A7S/A7R Slog2 footage out of this trailer? Can you point to specific segments where the 8-bit shows where it's lacking?

A STREET CAT NAMED BOB Official Trailer (2016) - YouTube (https://www.youtube.com/watch?v=JNJY0vXnEjQ)

Gary, with respect, it would take a damn good pixel peeper to point out 8 bit slog footage from a Trailer such as this. Fast clips that barely register on screen hardly allow time for the human brain to process what they are seeing without judging the quality of the footage. By all means, show a long clip of the movie and ask the same question, but this Trailer..... I'm not a robot you know. :)

Personally I prefer using 10 bit for Log and HDR; I just find it grades easier. That said, I've used Log for 8Bit and it ain't bad.

That said, with A Street Cat Named Bob, they were using an Atmos Shogun, shooting at 4:2:2 colour, at a high data rate. So perhaps the argument is, that shooting Log on 10 bit isn't the issue, but shooting with 4:2:2 colour is important. With my GH5, that means shooting 10 bit.

All DCT based compression uses blocks whether it is H.264 or ProRes.

H.264 can compress in addition to intra-frame compression the motion compensated difference between frames.

Given the same bitrate inter-frame and intra-frame compression combined will provide a superior image compared to only intra-frame compression.

The only disadvantage to files that have inter-frame compression is that it is hard to edit because the decompression depends on multiple frames.

H.264 is neither better nor worse than ProRes, it all depends on the bitrate and you can encode H.264 all-intra and also use lossless compression.

Understood. I understand DCT compression and cluster averaging. My only point on that os that ProRes HQ is nowhere near as agressive in this process relative to H.264 Long GOP at a low bit rate. The lower the bit rate your cap is and the more complex your scene is the more and more macroblocking you get. And vice versa as you move back "up" the bitrate ladder.

Yes, given enough bit rate, Long GOP can be tough as nails too. No doubt about that.

My point was that this process "can" look like banding and is easy to blame 8bit sampling for. I have done test recordings with 8bit XAVC-S and 8bit ProRes HQ and the stretched the Hell out of both and seen the h.264 "band" way before the same shot did in 8bit ProResHQ. This happened because what I saw was not "true" 8bit banding but compression "banding" (macroblocking because of a sressed CODEC)

Thats my only point.

And yes....we ALL love 10bit here. Im sure nobody is arguing that 8bit is just as good. I think we are mostly just trying to say that 8bit CAN still give very good results too.