DXOMark rates GH4 at 13 stops dynamic range - 1 stop better than 5D Mark III

[Emanuel]

Sony's A6000 is so much better than GH4 as a stills camera.

I'm pretty sure that it is very close in video as well and all that for less than half the price!

Sony is gonna say good bye and get past everybody else pretty soon...

 

oh God, a camera is NOT only sensor or test rates!

 

At times, I wonder if this geek idea of a capture device has anything to be with real photography... (E :-)

[Emanuel]

PS: I can give an example. My favorite camera is GH series, but before I was using my 5D2, out of order later on when I replaced it for a GH1. Before I had read 4/3" format was not the best for still photography...  because of this misconception exactly about what photography is. No more no less in any way other as the exact acquisition device from several variables which conduct to a possible outcome.

 

Try to go with a RED ONE to the middle of the crowd and you'll see the street photography you'll get ;) Who doesn't understand this, it can only be a good theoric sit behind the desk :)

[Don Whitmore]

oh God, a camera is NOT only sensor or test rates!

 

At times, I wonder if this geek idea of a capture device has anything to be with real photography... (E :-)

 

You just came to a forum posting about technical specs and complained about people caring about technical specs. We don't need another obligatory "it's not the tool that matters, it's the artist" post.

[demetris]

oh God, a camera is NOT only sensor or test rates!

 

At times, I wonder if this geek idea of a capture device has anything to be with real photography... (E :-)

 

Did you read the article in the first place? 

Hello, it is about sensor ratings...

[demetris]

well i think its absolutely great what sony comes up with on paper. no other company brings out cameras with such amazing specs in the smallest possible bodies. i often like to compare Sony with RED. not that their cameras are comparable in any way, but both companies seem to have the exact same goal of pushing the limits of whats possible with todays cameras and always being the first of introducing new standards and features. 

 

on the other hand theres Panasonic, who would never come out with a camera that isnt rock solid in its performance. FF 4K camera that overheats after 30min of recording? it would never see the light of day if Panasonic built it.

 

its really up to you what you choose, but id always go with a rock solid camera. just like i would always choose an Alexa over an Epic.

 

Talking about rock solid performance please do not forget that GH2 had a dreadful battery life and GH3's viewfinder and screen was practically unusable for many.

GH4 has come a long way and it really looks like the first DSLR(mirrorless) truly made for video.

However, there are many reports about an audio bug regarding GH4...

Overheating has been a big problem on certain Sony cameras so far, let's see about A7s, but I wouldn't exactly call Panasonic's GH line performance rock solid.

You are right about choosing a rock solid camera and that's why companies still offer professional video cameras and not only cheaper hybrid alternatives.

[hmcindie]

Sony NEX7 scores the same amount of points on DXOMark as the 5dmarkIII. But I can say (I have both here in my hands) that the images from the NEX7 are considerably worse.

 

How do those measurements work??

 

And why DXOMark seem to go out of their way to avoid measuring for example Fuji sensors?

 

I will bet my ass that the Sony A7s will be the best sensor they've ever tested just because they don't really test anything else except dynamic range.

[marienho]

What I found interesting to see is that they measured the greatest dynamic range and best Signal to Noise at ISO 100 (measured at ISO 88), you have only access to ISO 100 when you turn on the Extended ISO range, something I never did before now :-)

 

I have just published a blogpost comparing the usable dynamic range of the BMPCC, Nikon D800 and GH4. I also show how I conducted these tests. In my opinion the usable dynamic range is around 7 stops. If I did something wrong in these tests I would really like to know how I could improve it.

[Andrew Reid]

What I found interesting to see is that they measured the greatest dynamic range and best Signal to Noise at ISO 100 (measured at ISO 88), you have only access to ISO 100 when you turn on the Extended ISO range, something I never did before now :-)

 

I have just published a blogpost comparing the usable dynamic range of the BMPCC, Nikon D800 and GH4. I also show how I conducted these tests. In my opinion the usable dynamic range is around 7 stops. If I did something wrong in these tests I would really like to know how I could improve it.

 

Nice test.

 

Pretty much backs up my own less scientific 'by-eye' comparison between my BMPCC and GH4. Usable dynamic range is almost identical.

[Emanuel]

You just came to a forum posting about technical specs and complained about people caring about technical specs. We don't need another obligatory "it's not the tool that matters, it's the artist" post.

 

 

Did you read the article in the first place? 

Hello, it is about sensor ratings...

 

 

I surely read it. Reason why someone needs to introduce some common sense and real scope over such geek questions. If you are in these forums for a decade, you should have crossed yourself with my posts for sure. And also prevailing the technical aspects of business when needed, aside others (other aspects count: they must be seen altogether; that's an usual mistake). Take a look in one of my links at my signature or both.

 

A few of you were probably in the high school, I was feeding myself from high end technology inside this industry. Trying to convince people to shoot 1080p 3:1:1 rather than 35mm film.

 

This doesn't mean we have to lose the sense of proportion of technical aspects. That's the whole point. When we lose it, we lose any basis for our wondering. A discussion about cameras serves if balanced. If not, if we distort the blanket only for one side of the bed, we will lose the whole purpose for.

 

E :-)

[dahlfors]

 

I have just published a blogpost comparing the usable dynamic range of the BMPCC, Nikon D800 and GH4. I also show how I conducted these tests. In my opinion the usable dynamic range is around 7 stops. If I did something wrong in these tests I would really like to know how I could improve it.

 

Just a note: D800 sensor has higher dynamic range at ISO100 compared to ISO200 (around half a stop difference).

 

However, I highly doubt you'd see some really noticeable difference in your test at ISO100 vs ISO200, since the line skipping and how the video processing functions is what limits the DR in video output on D800, not the sensor itself.

 

Also, seems like the GH4 really performs well!

[Emanuel]

Here's a pretty fair concept: usable... ;-)

 

 

Usable dynamic range is almost identical.

 

 

[Michael1]

What I found interesting to see is that they measured the greatest dynamic range and best Signal to Noise at ISO 100 (measured at ISO 88), you have only access to ISO 100 when you turn on the Extended ISO range, something I never did before now :-)

 

I have just published a blogpost comparing the usable dynamic range of the BMPCC, Nikon D800 and GH4. I also show how I conducted these tests. In my opinion the usable dynamic range is around 7 stops. If I did something wrong in these tests I would really like to know how I could improve it.

I really like your camera test.  Great info there.  I, too, have been wondering about the dynamic range of the cameras tested.  I hope you test some more cameras, even though I know you thought it was boring. :)

 

This got me thinking.  If you are using an 8-bit Codec, wouldn't the maximum number of stops of dynamic range be 8, perhaps 9 if you throw in gamma?

 

Michael

[leeys]

How accurate are these tests though really? When I tested the a6000 (13.1 evs) and gx7 (12.2 evs), I took some raw photos and used Lightroom to push the shadows and pull the highlights on both at different ISOs. The  latitude on both ended up being the same (same highlight recovery, same amount of noise in the shadows). In fact slightly worse on the a6000 because at high ISOs the shadows had a purple cast.

 

And if you use the dpreview camera comparison tool, you can see that the Canon 6d clearly has less noise than the Nikon d800, yet it's rated lower on DXOMark.

 

These tests are pretty good, but you must understand them, and never ever look at the summary alone. For one thing DxO judges based on 8MP images, which when downsizing might affect the final image quality.

 

I've Shot with 6d and own d800.  Chalk and cheese.  I found the 6d noisy and a bit gnarly for a 20mp camera.  THe d800 shadow noise cleans up very well. 6d still had that typical low frequency fuzzy purple and magenta chroma noise that is hard to clean up.  The d800 continues to blow me away with how much range is usable.  I wish as a company they had just the teensiest bit of vision.

 

I do like what the D800 sensor can do. It's an amazing sensor. Actually, even the 24MP APS-C sensors are quite amazing too. It's really Canon that has dropped the ball here.

 

Sony NEX7 scores the same amount of points on DXOMark as the 5dmarkIII. But I can say (I have both here in my hands) that the images from the NEX7 are considerably worse.

 

How do those measurements work??

 

And why DXOMark seem to go out of their way to avoid measuring for example Fuji sensors?

 

I will bet my ass that the Sony A7s will be the best sensor they've ever tested just because they don't really test anything else except dynamic range.

 

As I was saying - look at the entire suite of measurements across all ISOs. It's actually not that straightforward some times. eg. a D4 has worse dynamic range than a D800 at first glance, but at high ISOs it easily pulls away, making it such a great high ISO camera.

 

I really like your camera test.  Great info there.  I, too, have been wondering about the dynamic range of the cameras tested.  I hope you test some more cameras, even though I know you thought it was boring. :)

 

This got me thinking.  If you are using an 8-bit Codec, wouldn't the maximum number of stops of dynamic range be 8, perhaps 9 if you throw in gamma?

 

Michael

 

The tone curve plays a part too, so some settings let you eke out a bit more. The main thing here is that with raw you get a lot of data that you can post process with. Not so with baked images/video.

[leeys]

I'll have to have a look at the Nikon 14mm F2.8 lens.  I don't see it on their website, though.

 

Michael

 

Oh, and that 14mm has been discontinued, with the 14-24 taking its place. I never shot with it, but it's not a very highly regarded lens.

[Don Whitmore]

I surely read it. Reason why someone needs to introduce some common sense and real scope over such geek questions. If you are in these forums for a decade, you should have crossed yourself with my posts for sure. And also prevailing the technical aspects of business when needed, aside others (other aspects count: they must be seen altogether; that's an usual mistake). Take a look in one of my links at my signature or both.

 

A few of you were probably in the high school, I was feeding myself from high end technology inside this industry. Trying to convince people to shoot 1080p 3:1:1 rather than 35mm film.

 

This doesn't mean we have to lose the sense of proportion of technical aspects. That's the whole point. When we lose it, we lose any basis for our wondering. A discussion about cameras serves if balanced. If not, if we distort the blanket only for one side of the bed, we will lose the whole purpose for.

 

E :-)

This is the real problem in this industry, in any artistic industry tbh. Injecting your ideals of how anyone should go about learning their craft, with such brazen arrogance and reverence for your experience, aren't the kind of conversations we should be having. 

Again, if you want to talk craft, go to a forum post about craft. If you want to provide meaningful conversation, to the point at hand, then chime on in. I for one appreciate these kind of tests. The bulk of my knowledge is in computing. As I transition to film, namely digital film, I've found dynamic range information incredibly useful.

[Simon Bailey]

GH4 does not have 13 stops of DR for video. 

 

DXOMark's dynamic range measurement is only for stills. If their test was for video too then if you look at their Nikon D800 tests that would mean the D800 had a DR of 14.4 and its only 12 for video.

[Andrew Reid]

DXOMark's dynamic range measurement is only for stills.

 

No need to state the obvious!

[hmcindie]

 

DXOMark's dynamic range measurement is only for stills.

 

They did test the Red Dragon

 

http://www.dxomark.com/Reviews/RED-Epic-Dragon-review-First-camera-to-break-the-100-point-DxOMark-sensor-score-barrier/Dynamic-range-and-color-depth-First-place-in-both-categories-14.8Ev-26.5-Bits

 

It's funny how it's dynamic range is "only" 0.4 better than the D800 considering how much Red was touting a huge dynamic range before release (something like 16-17 stops). Though DXOMark can't get the actual sensor data out of the Dragon so I don't know how comparable those numbers are.

 

The A7s will sour above the Dragon in dynamic range and also create lots of articles for people to click in the near future.

[ND64]

you confused Ev with Stop. 14.4 Evs on D800 is equal to 13.5 stops. so 12.8 on GH4 is much lower. however, DxO measures the DR down to 1:1 signal to noise situation, that is not practical (an image which has same amount of noise as it has signal, is useless for you, no matter how much DR is there). I usually ignore 3 stops of the score, so 10.5 stops for D800 (still) image could be a safe bet.  

[jgharding]

They did test the Red Dragon

 

http://www.dxomark.com/Reviews/RED-Epic-Dragon-review-First-camera-to-break-the-100-point-DxOMark-sensor-score-barrier/Dynamic-range-and-color-depth-First-place-in-both-categories-14.8Ev-26.5-Bits

 

It's funny how it's dynamic range is "only" 0.4 better than the D800 considering how much Red was touting a huge dynamic range before release (something like 16-17 stops). Though DXOMark can't get the actual sensor data out of the Dragon so I don't know how comparable those numbers are.

 

The A7s will sour above the Dragon in dynamic range and also create lots of articles for people to click in the near future.

 

Yes the Red does compressed raw video so its the same as stills. Imagine if Nikon did that...

 

bear in mind pushing Red to its limits in post doesn't look anywhere near as nice as pushing say, the Alexa in post. Here's a great test of usable overexposure for various cine cams, where Alexa takes it again: http://www.cinematography.net/edited-pages/NZCS-Contrast.html

 

That's what 14 nice stops allows you to do in video. Red may be catching up but even the Dragon will probably look far less nice than Amira, is my guess.

 

Sony have a high end line to protect, so the A7s in video mode will not match cinema cameras for dynamic range, they can't really allow it to.

 

Here's how far over you can push the wonderful Alexa and still get it back:

 

http://www.cinematography.net/alexa-over/alexa-skin-over.html

 

When we have this in DSLR we're laughing. Hell, when we have it in any other camera we're laughing! I think you may be able to do similar with 5D MKiii raw but I've not tested. The Alexa's dual-gain sensor structure remains unique though AFAIK.