GH5 GH5S versus EVA1...

I think due to the nature of the large pixels on the GH5s low light will definitely be better. There is a misconception that sensor size itself is what determines sensitivity but that is not true at all. As amazing as the EVA1 is it packs a ton of extra pixels into that s35mm sensor which means each pixel on the sensor is smaller than it could have been if it was a true 4k sensor.

The EVA1 was designed for quality and dynamic range first and low light second. Shooting in the dark is not its main purpose. The GH5s on the other hand is designed specifically for low light shooting while sacrificing some dynamic range and quality at higher ISO levels. The GH5s also has the ability to use a focal reducer like the Speedbooster which brings its FOV and DOF in line with the s35mm sensor on the EVA1. Sensitivity is increased another stop so that advantage the GH5s had is now even greater.

A camera like the GH5s also uses a much more aggressive form of noise reduction than a cinema camera would typically use. Thats why even the GH4 could beat RED in low light tests. Didn't mean the RED used a horrible sensor but that the camera was using little to no noise reduction keeping in mind when shooting raw users were likely to use noise reduction in post.

At the end of the day both cameras are exceptionally good compliments to each other. The GH5s has a bit of an edge shooting in the dark and the EVA1 has a bit of an edge when it comes to dynamic range. Beyond that it really does come down to the ergonomics and form factor of the camera and if you really need external raw. The external raw combined with an Inferno and the new ProRes raw is likely going to be the most comfortable, easy and performing raw workflow the industry has ever seen. If anybody really wants to get into raw or needs raw then the decision is easy, get a EVA1 and enjoy the most solid raw workflow the world has ever seen.
 
Cary Knoop said:
The V-Log coming out of the GH5(s) is V-Gamma/Rec.709-Gamut, the V-Log coming out of the EVA1 is V-Gamma/V-Gamut.
Click to expand...

Got it. And what are the real world manifestations of this? The way I'm reading what you wrote is that it's a consistent variability, so wouldn't any kind of necessary conversion be fairly easy to integrate into a workflow? Or do you have to go clip by clip?

TheDingo said:
In the YouTube comments the shooter lists which camera shot every scene. ( GH5s definitely looks better for the night footage )
Click to expand...

Yeah, what I meant by "had trouble" is I watched the whole thing through and made notes of which clip I thought belonged to which cam. Then I cross checked against that list and found that a few times that I thought it was the EVA it was actually the GH5s. I had previously felt I could distinguish them 100% of the time, but when tested "blind" I was around 80-90.
 
dcoughla said:
Got it. And what are the real world manifestations of this? The way I'm reading what you wrote is that it's a consistent variability, so wouldn't any kind of necessary conversion be fairly easy to integrate into a workflow? Or do you have to go clip by clip?
Click to expand...
Real world manifestations? Saturated colors.

It is consistent.

I use mostly DaVinci Resolve.

So in Resolve in color managed mode I convert the gamma to Panasonic V-Log but leave the color space Rec.709 for Gh4/5.

For ACES it is more of a pain, the right way would be to have a separate IDT but you can fix most of it by having a color space transform from AP1 to V-Gamut (and ACEScc to Gamma 2.4) and then a color space from Rec.709 to AP1 (and Gamma 2.4 to ACEScc). [Of course you might use ACEScct and use another Gamma than 2.4].
 
Last edited:
Thomas Smet said:
if that is the case then how can the same V-log lut work on both V-log and V-log L material? Wouldn't using the same lut cause some jacked up color? I thought the whole point of V-log L was that it did work perfectly in a V-log world even if that meant sacrificing some of the range in a 8bit and 10bit color space to make that perfect match happen.
Click to expand...
I agree. Of note however, is YUV header information contained in the prores files for both V-Log/V-gamut from the EVA1 and V-Log/L from the GH5s are reporting bt709 primaries and matrix coefficients. Although it would certainly support the inter-operability between V-Log and V-Log/L, it also means the transfer from camera weighted RGB matrices to prores YUV is only using a 709 subset of V-gamut; same for both EVA1 and GH5s. How can this be? Is it possible to get the full V-gamut; it has to come from raw?

For comparison, the YUV header information contained within S-Log2/S-gamut from a 10 bit XAVC-I file from the Sony PMW-F55; color primaries and matrix coefficients are *not defined*.

With the S-Log2/3 matrix coefficients and primaries undefined, the codec could allow anything; wider than 709 color gamut, whereas with V-Log/V-LogL the 709 matrix coefficients and end primaries would restrict the EVA1/GH5s prores to the boundaries of bt709 color.

I'm sorry for such a complicated discussion, but it's important for us to understand what is happening. But for sure, I agree with your point that GH5s and EVA1 both play nice with the RCM V-Log/V-gamut input transform in Resolve, interchangeably with no color shifts that jump out at me.
 
Tom Roper said:
I'm sorry for such a complicated discussion, but it's important for us to understand what is happening.
Click to expand...

Don't apologize for this! I know f*ck all about color and this chain is making me realize it's time I learn more about it. Do you (or anyone) have any handy links/books/videos to check out that would help me decipher posts like the above? As of now I'm just googling every term I don't understand but if there's a more efficient set of resources, I'd really appreciate them.
 
dcoughla said:
Don't apologize for this! I know f*ck all about color and this chain is making me realize it's time I learn more about it. Do you (or anyone) have any handy links/books/videos to check out that would help me decipher posts like the above? As of now I'm just googling every term I don't understand but if there's a more efficient set of resources, I'd really appreciate them.
Click to expand...

A good place to start is with any of Barry Green's books, simple easy to understand, readable.
 
Tom Roper said:
I agree. Of note however, is YUV header information contained in the prores files for both V-Log/V-gamut from the EVA1 and V-Log/L from the GH5s are reporting bt709 primaries and matrix coefficients. Although it would certainly support the inter-operability between V-Log and V-Log/L, it also means the transfer from camera weighted RGB matrices to prores YUV is only using a 709 subset of V-gamut; same for both EVA1 and GH5s. How can this be? Is it possible to get the full V-gamut; it has to come from raw?

For comparison, the YUV header information contained within S-Log2/S-gamut from a 10 bit XAVC-I file from the Sony PMW-F55; color primaries and matrix coefficients are *not defined*.

With the S-Log2/3 matrix coefficients and primaries undefined, the codec could allow anything; wider than 709 color gamut, whereas with V-Log/V-LogL the 709 matrix coefficients and end primaries would restrict the EVA1/GH5s prores to the boundaries of bt709 color.

I'm sorry for such a complicated discussion, but it's important for us to understand what is happening. But for sure, I agree with your point that GH5s and EVA1 both play nice with the RCM V-Log/V-gamut input transform in Resolve, interchangeably with no color shifts that jump out at me.
Click to expand...

You said ProRes files. Do you mean from an external capture?
 
With RAW, you can do 4K @ VFR on the EVA1. You can also do true VFR 10-bit HD internal on the EVA1 (although at a lower bitrate than GH5s, which is odd...). So if those VFR rates are important to you, that is nice.

ISO 2500 or lower, EVA1 has an overall nicer image to it (I wouldn't say 3x nicer, but definitely nicer). V-Gamut and Wider DR and just a cleaner overall better rendered image.

The GH5s wins in lowlight for sure (ISO 3200+).

The GH5s UI is better IMO, EVA1 is a bit clunky.

It's a tough call. Get both ;) Overall EVA1 makes me feel much more comfortable on higher end shoots than being limited to a GH5/GH5s pairing.
 
Both of the test clips I downloaded for the EVA1 I found not to be straight from the camera as hoped, therefore the metadata listing the primaries and matrix coefficients as bt709 are not confirmed. If anyone can point me to a 2-5 "second straight from the camera" V-Log/V-gamut clips, I'd appreciate it.

As an aside, the EVA1 clips that I have downloaded are simply beautiful. Compared to Arri, there is still a magenta hue to skin tones. On a scale of Arri to my F55, I'd put it about in the middle, less than F55 but more than Arri. Either one is easily correctable in post, but straight from the camera is what I'm describing here. Below are a couple of screen grabs from EVA1 clips I have downloaded, the one on top is original and the one below I adjusted the skin tone as I would for the F55.

Orig_1.1.3.jpg
Adjusted_1.1.4.jpg
 
Looking at the EVA1 myself at the moment and really need to see some side by side shots from both the GH5s and EVA1 to make an informed decision.

Ideally the way to go would be to rent and test myself, but here in Spain, if I (even) find a EVA1 to rent is hard.

I get a feeling its a close call in purely picture quality aspect (ignoring usability / features etc) if we are just dealing with online distribution.
 
Hey Rich,

differences are mainly ergonomics, ND filters, SDI and target users. Image is very good on both. BUT GH5s has 4K 60p which the EVA1 hasn't! ;)

BTW I'll have for sale my full GH5s kit with cage and MB XL 0.64x once I wrap up this project to fund post in a couple of days, if you're interested let me know, ok? I'm in EU.
 
:) I already have a GH5 and GH5s with a Ninja Inferno. But just wanting to see from a purely image point of view, how the EVA1 is any better.

Lots of online reviews of the GH5 vs EVA1 but nothing really with GH5s
 
The classic answer is "Client Perception" :D I couldn't turn up to a job with a GH5 - bigger camera/bigger budgets.
Haven't used a GH5, we have a GH4+LOG upgrade/EVA1 we shoot multi cam with and results are good.
Plus we do Motion Control Rig projects and couldn't use the GH series on those jobs.
Image from the EVA is great, manual controls (in the main) are well mapped out across the body, dynamic range and smooth roll off, strong codecs, S35 5.7K sensor, lovely colour science.
Really lightweight body gives me more options in terms of moves/handheld
only drawback is the LCD/Touchscreen function - both are rubbish!
 
One of the best things about small cameras is building them up when you need to.

I think most would agree that many times clients frown upon smaller cameras or when you're using what looks like a stills cam for video, but once you start throwing on those cages, mics, cables, external recorders (anything and everything), the "client perception" is no different.

It only matters when the client is knowledgeable (and asks for something) or when you're getting paid like $10,000+ for the day, ha.
 
put it on a gimbal with a monitor some coiled cables

QUOTE=NorBro;1986764882]One of the best things about small cameras is building them up when you need to.

I think most would agree that many times clients frown upon smaller cameras or when you're using what looks like a stills cam for video, but once you start throwing on those cages, mics, cables, external recorders (anything and everything), the "client perception" is no different.

It only matters when the client is knowledgeable (and asks for something) or when you're getting paid like $10,000+ for the day, ha.[/QUOTE]
 
dia3olik said:
Hey Rich,

differences are mainly ergonomics, ND filters, SDI and target users. Image is very good on both. BUT GH5s has 4K 60p which the EVA1 hasn't! ;).
Click to expand...
This is a profoundly inaccurate assessment of the differences between the EVA1 and the GH5S.

I could go on a big rant, but I will just say that the EVA'S sensor is twice as big, the resolution is much higher, and the sensor delivers 14 stops of dynamic range.

And the EVA does do 4k/60p, both internally and as raw output. And when in raw, it is cropped - to the exact same size and pixel count as the GH5S.

These cameras are not the same, not by a long shot. There are many good reasons why the EVA costs more than twice what the GH5S does.

The GH5S is a hell of a camera at it's price point. But it's no EVA.
 
dia3olik said:
Ha ha I meant RAW which is the main reason to upgrade...it's shitty and cropped...check it yourself ;-)
Click to expand...

??
EVA1 in RAW does 5.7K30p, 4K60p, 2K240p.
GH5S does not do RAW video.
Yes 4K RAW is cropped from 5.7K, it is pixel for pixel 4K.
4K RAW sensor area is still a bit larger than GH5S sensor.
 
You must log in or register to reply here.
Back
Top