Panasonic GH5S 4K / 240fps low light monster

[Vesku]

I wonder how difficult it is to test GH5S DR compared to GH5? Just shoot some high contrast scene and watch? If there is no visible difference the DR is the same.

I can see a difference between my GH4 and GH5 even when Panasonic says they have about the same DR.

[jonpais]

Did you even watch the video I just posted?

[jonpais]

Some swear by charts. Others by counting nits. Still others by shooting in a dimly lit room with a bright window and pulling up shadows or pulling down highlights. But I can just about guarantee that whichever method you use, the difference will be negligible when compared to viewing footage from your present camera on a decent display. In the YouTube HDR thread, someone mentioned that colorists over at liftgammagain scoffed at the idea of using an Inferno as a reference monitor, but I think they would be just as horrified to see what many  are using to grade their SDR content.

[IronFilm]

5 hours ago, Zak Forsman said:

dammit. had my hopes up briefly about the dynamic range. but this is the only side by side test of dynamic range i've seen and despite matt frazer's assertion that the GH5s is 14 stops at native ISOs, I don't see much (if any) of a difference here from the GH5. highlight roll-off is certainly better, but I still think there has been a miscommunication inside Panasonic that conflated "two more usable stops" due to better noise performance, with, "two additional stops" that would raise it to 14 stops total.

not a huge loss in the grand scheme of things, i guess.

Even having the same DR at 2500 as the other low base ISO is still a big win!!

3 hours ago, Mattias Burling said:

He knows way more than people give him credit for. But I don't care.

From years of watching him, that might maybe be true in the stills realm, but not in the video world.

[Vesku]

59 minutes ago, jonpais said:

Did you even watch the video I just posted?

I have seen it days before and this video shows identical DR for GH5S/GH5. 

Counting monitor nits is not relevant for captured DR. A SDR monitor can show whatever captured DR when shades are compressed/graded visible. An HDR photo can have 20 stops of DR and we can see details in highlights and shadows in rec709 display. 

[Mmmbeats]

I don't know if it's proper etiquette to post stuff from other forums here, so please forgive / delete as appropriate if it's not the done thing.

I thought this was a really good comment by a user on DVXuser (Thomas Smet) :

Quote

Yes it works when shooting V-log. Thats one of the main advantages of the GH5s over the GH5. Not so much just for shooting in the dark but you can shoot V-log at ISO 2500 with the same roughly clean look as V-log at ISO 400 which is just under a 3 stop improvement. This opens up a couple of possibilities:

1. You can shoot indoors with a f5.6 zoom for the same roughly noise quality and dynamic range for a prime lens shooting at f2.2. This opens up a whole new world of possibilities using zoom lenses in tough situations where with the Gh5 you almost always had to use primes or extremely expensive zoom lenses with a very limited zoom range. The f2.8 zooms can be as clean as a theoretical f1.1 zoom lens.

2. You can shoot 3x slow motion with roughly the same noise quality as regular motion on the GH5. Slow motion in poor indoor lighting is now somewhat possible.

3. You can use less output LED lights 3 stops lower on an indoor set which means cheaper, smaller and lighter light kits.

4. You can shoot a production outdoors at ISO 400 and indoors at ISO 2500 with the same lenses and aperture for a consistent look throughout the production. Every lens looks different at different aperture amounts. F1.4 will never look the same as f4.0 so just adjusting aperture will never deliver consistent results. One solution is to use a boatload of ND filtering outside so you can keep the lens as the same aperture as you would use indoors. That only works if you use the lenses wide open or if you use a ton of light indoors. A wide open lens is never as good as when stopped down a bit and every lens has a sweet spot. FF lenses tend to have a sweet spot around f4-f5.6. Dual native ISO means it is now much easier to target the same sweet spot aperture on a production without a massive light kit.

5. You can use a lot more diffusion on lights to help soften them. Smaller LED lights create more of a harsh light. The more the lights are diffused the weaker the output. ISO 2500 means you can diffuse the daylights out of those lights to get very soft lighting without an impact on the quality.


Dual native ISO is not just a means to suck in more light to shoot in the dark. It is a highly advanced tool that can and should be leveraged to rethink how one shoots in different environments. It really is a game changer. ISO 2500 doesn't mean no longer having to use lights for production value but it means smaller more agile lights can be used that take minutes to setup vs hours. There are a slew of affordable LED panel lights out there now with CRI 97+ accuracy, dimmable and color temp adjustable for under $100 or $200. They don't have the massive output of traditional lights or even a Dracast LED but with 2500 ISO these lights become a more realistic option.

  I'm particularly excited about points 1 and 5.  I've mentioned lenses before, but I'm constantly wishing I could add more diffusion to LED lights, but until now there's been too much of a loss of output.

[DBounce]

3 hours ago, jonpais said:

Did you even watch the video I just posted?

Here is the big difference between the sensors... if the GH5S is using the Starvis IMX294, then this sensor can capture high and low ISOs simultaneously. This is how it was designed by Sony. The way Panasonic has implemented it splits this function into two discrete events. Personally I would love to have a real HDR mode that does what the sensor was originally designed to do. And I believe that would not be hard to do. Even if it sucks more power, they could make it only function when connected to the battery grip or AC power. The option would be welcomed.

[buggz]

This is an option in ML.

When I first heard of dual ISO in the GH5s, this is what I thought of, the ML implementation of using BOTH at same time, alternating scans, actually.

Was kind of disappointed in the Panasonic method, though, it does have a working purpose...

 

25 minutes ago, DBounce said:

Here is the big difference between the sensors... if the GH5S is using the Starvis IMX294, then this sensor can capture high and low ISOs simultaneously. This is how it was designed by Sony. The way Panasonic has implemented it splits this function into two discrete events. Personally I would love to have a real HDR mode that does what the sensor was originally designed to do. And I believe that would not be hard to do. Even if it sucks more power, they could make it only function when connected to the battery grip or AC power. The option would be welcomed.

 

 

[Mmmbeats]

Is there any reliable info on whether the GH5s uses dual circuitry like the Varicam range, or if it's dual ISO is a different effect produced by the new sensor tech?

[jonpais]

1 hour ago, Mmmbeats said:

I don't know if it's proper etiquette to post stuff from other forums here, so please forgive / delete as appropriate if it's not the done thing.

I thought this was a really good comment by a user on DVXuser (Thomas Smet) :

  I'm particularly excited about points 1 and 5.  I've mentioned lenses before, but I'm constantly wishing I could add more diffusion to LED lights, but until now there's been too much of a loss of output.

As much as I respect Thomas Smet and appreciate the thought that went into his overly optimistic analysis...  the working methods he is suggesting run contrary to my own as well as those of the filmmakers I admire most. There’s no right or wrong here, it’s personal preference for sure. But before getting into that, I’m still highly skeptical that ISO 2500 is going to be as clean or have as much color depth as ISO 400. That out of the way, I much prefer the quality of fast primes to slow zoom lenses, at least in the micro four thirds system. And shooting between f/4-5.6 is like shooting at f/11-16 in full frame terms - not very attractive for what I shoot, but maybe great if you do a lot of real estate jobs. Not sure what he means by closing down the iris so much in order to reach the sweet spot of the lens, unless he’s talking about inexpensive kit zooms, because good lenses should be sharp enough wide open. Your equipment shouldn’t be dictating your choices for you, so choose gear that is right for the job. Anyhow, I wouldn’t slavishly allow test charts to determine which aperture I shoot at. I just finished watching an episode from Chef’s Table on Netflix, and I’m fairly certain the entire episode was shot at f/2.8 or wider, much of it probably at f/1.4, with what appears to be super 35 - and it’s a gorgeous look. I’ve never needed a basketful of ND filters outdoors to shoot at wide apertures like he’s talking about either, but then again, I never shoot out in the open at high noon. As far as lighting goes, I have to give him credit just for mentioning it. But using smaller, cheaper LEDs is the very last thing you want to do, even if your camera is clean as a whistle at ISO 50,000. Because very small light sources produce unpleasant light. And good LEDs are dirt cheap nowadays. For shooting interviews or beauty shots anyway. Not saying small lights don’t have their uses. I don’t know, much of his reasoning seems to be based on saving money, but anyone investing in a $2,500 camera should be prepared to purchase the best lenses available as well as good lights, microphones, etc. Why buy the most expensive m4/3 camera to date and skimp out on the essentials?

Edit: I just finished watching slashcam’s low light test, and ISO 400 on the GH5 is better than ISO 2500 on the GH5s. No surprise there.

[Trek of Joy]

On 1/16/2018 at 5:30 AM, Robin Billingham said:

The regular gh5 is a 2x crop of full frame. If both formats achieve 16×9 by cropping the top and bottom on the sensor(making the sensor smaller) then a 2x crop ratio would be maintained. However on an oversized sensor the 16×9 mode is achieved by using extra pixels available within the imaga circle, no cropping, so the ratio between the formats is changed to a less than 2x crop (in the case of the gh5s). If the full frame sensor was also an oversized multi aspect one then the 2x crop would remain between the 2. In reality this makes the gh5s crop around the 1:86-1:89 mark. Small changes for 16×9 or 17:9 etc 

Or 2x crop for stills in its native format 

I must be missing something because even 17:9 is still within the m43 image circle. Its a wider FOV compared to the GH5 which essentially crops the top/bottom off the m43 image for 16:9 video - hence the 2.18-ish crop compared to FF. But in DCI its still in the m43 image circle as seen below - which has a diameter of 21.63mm, or half FF which is 43.27. All aspect ratios are a 2x crop on the GH5s. So how is it 1.89 or whatever?

[jonpais]

Dynamic range, low light sensitivity and other cool stuff. Based on what I’ve seen so far, I wouldn’t be purchasing the GH5s for extra dynamic range. Sorry if this has already been posted. 

 

[AaronChicago]

@jonpais Looking at his DR test, the GH5s shows an extra stop in the highlights, and one less in the shadows. I think I'd prefer that.

[jonpais]

@AaronChicago He does say the GH5s is cleaner at the extremes of exposure than the GH5. It’s still so subjective. For one thing, trying to judge clips with YouTube compression. And I’m allergic to noise. ? Maybe when it’s available here, I can do a quick side-by-side test in the shop.

[sanveer]

After reading the many reviews online, these few questions popped in my head:

1. Could Panasonic have changed the dual iSO settings from ISO400 and ISO800 to ISO800 and ISO5000 instead? What about an even higher ISO setting like ISO12800?

2. I think the whole 14 stops of dynamic range is most likely bad communication of the information regarding "2 Stops more Dynamic Range" (from usable ISO3200 to ISO12800) and interpreting it to mean from 12 stops to 14 stops.

But the curious question, therefore, is, how is Sony able to extract 1-2 stops more dynamic range in their 8-bit video with much lower bitrate. Also, does the much slower read speed (indicated by the enormous rolling shutter) help with creating better dynamic range? If it is, could Panasonic use the same technology, thereby losing some rolling shutter advantage for more dynamic range.

 

[jonpais]

There’s also the issue of charts - what may appear to be a significant advantage on a test chart does not always translate into dramatically better performance in the real world. In other videos, I had a difficult time telling the two cameras apart at ISO 400. Of course, the advantage is clear at higher ISOs. 

[ade towell]

And that is a big deal for many - that the GH5s retains the 12 stops of DR at iso 2500 whereas by then the GH5 has dropped to about 9.5 stops

[Don Kotlos]

1 hour ago, sanveer said:

1. Could Panasonic have changed the dual iSO settings from ISO400 and ISO800 to ISO800 and ISO5000 instead? What about an even higher ISO setting like ISO12800?

1

The dual gain depends on the capabilities of the sensor, so my guess is that by design that sensor is fixed to 400/2500. Changing that would require changing the analog circuits on the sensor. 

1 hour ago, sanveer said:

2. I think the whole 14 stops of dynamic range is most likely bad communication of the information regarding "2 Stops more Dynamic Range" (from usable ISO3200 to ISO12800) and interpreting it to mean from 12 stops to 14 stops.

 

I agree. And that is the main advantage of this sensor, having a very good dynamic range in dim light. 

1 hour ago, sanveer said:

how is Sony able to extract 1-2 stops more dynamic range in their 8-bit video with much lower bitrate.

 

With a linear scale such as RAW the bit depth directly translates to the stops in dynamic range. To get around that, non-linear functions change the input-output relation (gamma) and companies use different names for them (S-log, V-log, C-log..). Depending on the capabilities of the sensor and the bit depth of the file, the log can be more aggressive, by compressing more the highlight information. The bitrate does not affect dynamic range. So Sony could do it by a combination of sensor performance/more aggressive log gamma. 

1 hour ago, sanveer said:

Also, does the much slower read speed (indicated by the enormous rolling shutter) help with creating better dynamic range? If it is, could Panasonic use the same technology, thereby losing some rolling shutter advantage for more dynamic range.

3

No. That is only the case when you move to global shutter. 

 

[webrunner5]

9 hours ago, Mattias Burling said:

He knows way more than people give him credit for. But I don't care.

Over two stops of DR sounds game changing since it would put it infront of the BMCC (13), Canon C100 (12) and A7s (11). Would love to see samples of that.

Doeas anybody know any good video showing it, dont feel like watching all of them?

Well just because it is say 2 stops better does not equate to 2 stops better DR if that makes sense? It means it is as good at say 6400 than the GH5 is at 1600 ISO I guess. I am surprised no one has done a serious DR test on it yet??? Should not be that hard to do sort of quickly?

15 hours ago, jonpais said:

frustrating that not a single person seemed at all interested in Mystery Box’s extremely detailed blog post about grading HDR, just question after question about dynamic range. You don’t necessarily have to be ready to grade and deliver in HDR to take an interest in the workflow. I take an interest in just about anything having to do with filmmaking.  And while I’m discovering that HDR is turning out to be more expensive than I thought at first, I find it ironic that everyone is only concerned about capturing 14 stops of dynamic range but satisfied with delivering six or whatever. To me, it’s like having spent my entire life listening to harpsichord music and being introduced to the piano. just my two cents

Hell I don't even have a TV or Monitor that has HDR or a camera that shoots it! And I doubt  not alone on here, and is probably the reason of not much interest.

[Mmmbeats]

4 hours ago, jonpais said:

As much as I respect Thomas Smet and appreciate the thought that went into his overly optimistic analysis...  the working methods he is suggesting run contrary to my own as well as those of the filmmakers I admire most...

Well, as much as I admire your photography, I'm going to have to respectfully disagree with a lot of what you've said there.

Quote

I much prefer the quality of fast primes to slow zoom lenses, at least in the micro four thirds system.

Everybody prefers the quality of primes.  I shoot almost exclusively on them.  But that's not the only consideration.  When you're filming actuality that's changing fast in front of you, you just can't afford to faff around changing lenses.  That's where a good quality (and yes, that does mean pricey) zoom wins out.  Every time.

Quote

 And shooting between f/4-5.6 is like shooting at f/11-16 in full frame terms - not very attractive for what I shoot, but maybe great if you do a lot of real estate jobs.

This comes down to the shot.  DoF is a storytelling tool like all others.  It works best when it is used with an awareness of contrast.  Shallow depth of field, after shallow depth of field, just starts to lose it's impact and meaning.  Mix it up a bit in keeping with the flow of the storytelling, is my preference as both a viewer and a shooter.

One thing I feel strongly is that super shallow DoF is destroying the relationship between the subject and their surroundings in a lot of short films I see online.

The main thing is that you should be able to make the choice based on your own creative interpretation, not just because you've run out of exposure options.

Quote

Not sure what he means by closing down the iris so much in order to reach the sweet spot of the lens, unless he’s talking about inexpensive kit zooms, because good lenses should be sharp enough wide open.

It's pretty easily demonstrable that most lenses are indeed sharper slightly stopped down (regardless of cost).  Having said that, I'm generally happy to shoot wide open on most lenses despite the slight difference.

Quote

Your equipment shouldn’t be dictating your choices for you, so choose gear that is right for the job. 

Erm... that's exactly the thinking that's driving me towards the GH5s - it opens up more creative choices.

Quote

As far as lighting goes, I have to give him credit just for mentioning it. But using smaller, cheaper LEDs is the very last thing you want to do, even if your camera is clean as a whistle at ISO 50,000. Because very small light sources produce unpleasant light. And good LEDs are dirt cheap nowadays. For shooting interviews or beauty shots anyway.

His comments apply to just about any LED panel or hard light out there (apart from the massive new 4 x 4 flexlights, or the really high output lights like the 300D or whatever).  Getting them to look nice and soft *and* retain a decent level of brightness is often a struggle.  I actually think this will be one of the biggest benefits of the system.