Quick review of A7S and GH4

[Guest Ebrahim Saadawi]

Philip Bloom's original 4K to download from the A7s recorded to blackmagic ultrastudio 4K!

What really astonishes me is the amount of detail (seems more than the GH4) and the dynamic range, I mean this kind of shadow and highlight information is nothing I can get even close to with the GH4

[noone]

on the plus side, I bet slo moving interviews shot in the dark of night with no lights will look fantastic.  OK I'm being sarcastic.  forgive me.

Then again, if you came upon Elvis and Amy Winehouse doing a duet in a dark dingy club, what would you want then?

[Jacek]

on the plus side, I bet slo moving interviews shot in the dark of night with no lights will look fantastic.  OK I'm being sarcastic.  forgive me.

 

Nice complement to "TheISO100SunnyDayAtNoonButBewareShadesCamera" GH4 ;) .

[hmcindie]

We have to keep in mind that if we shoot slog2, the base iso is 3200, so in bright conditions a7s loses lots of DR compared to GH4 or itself.

 

No it doesn't. It's just a measurement thing. It measures iso 3200 with slog2 but DO NOT compare it to other cameras. There is no DR loss compared to anything.

[jcs]

GH4 with Voigtlander 25mm @ F.95, ISO 1600, Natural picture profile (+ post CC, Neat Video noise reduction)

 

A7S with Metabones v4 adapter and Canon 50mm F1.4 @ F1.4, ISO 6400, picture profile 1 (+ post CC)

 

The A7S really shines in low light (kind of like having Neat Video built into the camera). The GH4 with a very fast lens isn't too shabby, and cleans up well with Neat Video. I still prefer the GH4 color science, though I am still experimenting with the A7S settings (seeing too much green/yellow in skin tones).

 

It's possible to get 20% slow motion with the GH4 4K: shoot 4K @ 30p and reinterpret as 24p in post (a subtle but useful effect).

[Amro Othman]

I think it's something Sony should really work on- the green/ yellow in skin tones. I think that's a major reason why many use Canon DSLR's for video... because they have a better colour science and makes skin tones preferable.

[maxotics]

For what it's worth, I believe the idea of "color science" differences between these cameras is over-estimated.  When cameras create JPGs, which one will usually see when software shows a thumbnail, the manufacturer can easily balance warmer ( like Canon) or cooler ( like Nikon).  But the RAW files have no real "science" behind them.  All the chips are essentially the same.  Yes, the camera makers can adjust the "raw" values before saving, but not in any way that you can't debayer them to get any color you want.  DR is slightly different, and Nikon seems to have found a way to re-adjust high and low values from RAW data to preserve more detail than Canon.  But again, in the video world, I don't see that it caries down to 8bit.

 

Video is a bit different, of course, because you're not working with RAW data (which is why I don't pay attention to any of these DxO scores which really have no bearing on compressed video IMHO).  They're only relevant to RAW images.

 

With a significantly larger sensor, the A7s will have superior light gathering power over the GH4, however it will be more prone to aliasing, which has a large color aberration component due to mosaic (bayer) color sensors.  

 

In good light, where shallow DOF is not necessary, there is no doubt in my mind that the GH4 is the better camera.  1. 4K downreses to true 1080 and 2.  Panasonic focuses on video compression to do in camera what one might do in post with RAW.  

 

In weak light, the a7s if going to have less noise and greater DOF than the GH4, even with a weak video CODEC.  

 

There's a lot of wishful thinking about cameras.  You want to make what you have do what a camera does, that you don't have.  In a sense, that's the whole point of this site--getting the most out of consumer-based digital cameras.  However, I feel we have to keep things in perspective.  If you shoot primarily in low light the a7s is going to do better for you than the GH4 with all the noise reduction in the world.  The real question, to me, is what situations do the cameras visibly diverge?

 

Comparing the GH4 to the A7s in good light is a bit pointless.  To GH4 owners it will show who the GH4 is superior in resolution, to the a7s owners they will grant resolution is better, but will think the shallower DOF worth the trade off.  To each his own.

 

How little light must there be before the GH4 grants the a7s has more light gathering power?

[Michael1]

 

With a significantly larger sensor, the A7s will have superior light gathering power over the GH4, however it will be more prone to aliasing, which has a large color aberration component due to mosaic (bayer) color sensors.

Maxonics, if both cameras are doing one to one pixel mapping at 4K, why would the a7S be more prone to aliasing?

 

By the way, everyone, Sony calls this camera the "a7S", not "A7S" or "A7s".  Don't shoot the messenger.  :P

 

Michael

[Inazuma]

What really astonishes me is the amount of detail (seems more than the GH4) and the dynamic range, I mean this kind of shadow and highlight information is nothing I can get even close to with the GH4

 
I think you're hyperboling a bit there sir.
 
This is from my GX7 with shitty Variable ND that creates a bit more noise than there should be.
 


And here it is with an Auto Light Balance effect.

 

 

if I used a better filter and graded it properly, the tones and colour would look better. If I had the option to use a better codec than the 28mbps 50fps then the result would look better again.

 

Still, the A7s no doubt pulls off very natural looking DR. But the GH4 isn't a camera that "cant even come close" to getting nice range. Unless youre telling me the GH4 has a worse sensor than the GX7..

[Guest Ebrahim Saadawi]

Very impressive for the Gx7 Inazuma, but have tried grading those 4K files off the a7S?  it's scary how good it looks for me certainly! 

[Oliver Daniel]

The dynamic range on the a7S and low light looks incredible, however, as I'm 95% handheld that rather crappy rolling shutter is going to ruin a lot of my shots. I can't keep still. I need to jump into the action!  :D 

 

... So I'll hire the a7S when its on a tripod and its REALLY REALLY dark. It's no replacement for actual good lighting. I use lights kind of very very much because, that's what makes a dynamic picture, lights right?  :P

 

Can definitely see the cameras place in the market. (cave shooting! Right I'll stop with the bad jokes. The a7S does look great. Honest!)

[Wulf]

 

A7S with Metabones v4 adapter and Canon 50mm F1.4 @ F1.4, ISO 6400, picture profile 1 (+ post CC)

 

The A7S really shines in low light (kind of like having Neat Video built into the camera). The GH4 with a very fast lens isn't too shabby, and cleans up well with Neat Video. I still prefer the GH4 color science, though I am still experimenting with the A7S settings (seeing too much green/yellow in skin tones).

 

 

everyone who's not colourblind would agree about the green and yellow in her face - but sry to ask, wft was this for a lightening situation?

your gh4 slomo looked really great, w/ doubt, but this pic just sucks ... no hard feelings please, but I really think you could do much better. :-)

[jcs]

everyone who's not colourblind would agree about the green and yellow in her face - but sry to ask, wft was this for a lightening situation?

your gh4 slomo looked really great, w/ doubt, but this pic just sucks ... no hard feelings please, but I really think you could do much better. :-)

 

That was shot in a very dark scene, just a few low-output incandescent lights above- way darker in real life. I let Photoshop try to auto-white balance (camera was set to AWB). Looking at the vectorscope in PPro, the skintones are pretty close to the skin tone line (but rotated slightly- not directly tracking the skintone line)... Something about the Sony sensor/color science is biased towards green (and when mixed with perhaps another red-bias we get yellow). I didn't spend much time trying to fix it to make a point: Sony color science is a challenge (have the same issue with the FS700). The A7S is a useful tool despite the issues- I'll spend time figuring out how to work around them (something I wouldn't need to do with Canon). Once I found that the Natural profile worked well with the GH4, and to not use the curve controls, I've gotten generally good results without much effort with the GH4. Hopefully we'll find similar settings with the A7S.

 

Here's straight from the camera with just saturation reduced slightly:

 

Always looking to learn new things- if anyone has a process to make this look better, would like to hear the steps (will definitely experiment with camera settings more- prefer in-camera accurate color vs. trying to fix in post (another reason why I prefer not to use anything like Slog unless no other choice due to conditions).

 

Note also the Philip Bloom video above: it has a greenish-yellow tint.

[jcs]

The gen 1 SpeedBooster works on the A7S! (lol). It goes into APS-C mode (when in AUTO APS-C). The quality is pretty good in crop mode- perhaps a tiny bit of aliasing (and apparent increase in sharpness from said aliasing). Tested with the 24-104 F4L at 'F2.8'- worked pretty well. After shooting everything so far with the 50mm F1.4, the IS works great handheld. With APS-C mode turned OFF, one can use the 'Crazy Wide Super Vignette' mode ;). Gotta try this with the 16-35 F2.8. With A7S+SB that extra stop of light is like double the ISO, right, so ISO 410K is now 820K? ;)

[maxotics]

Maxonics, if both cameras are doing one to one pixel mapping at 4K, why would the a7S be more prone to aliasing?

 

By the way, everyone, Sony calls this camera the "a7S", not "A7S" or "A7s".  Don't shoot the messenger.  :P

 

Michael

 

The GH4 is saving those pixels internally.  Later, you can sick a powerful PC on the task of down-resing the image to 1080 :)  The Sony is supposedly binning the pixels internally and I suspect they aren't downscaling through true binning but throwing out data, which leads to aliasing.  

[jcs]

I'm not seeing any aliasing or moire at all in 1080p FF mode (only minor aliasing in crop mode and a bit more in 60p). The internal scaling from 4K FF (24fps) to 1080p is pretty good: looks like a decent scaler (at least bilinear, perhaps better (cubic, Lanczos etc.)).

 

An interesting note regarding noise on the A7S: when looking at the display or viewfinder in low light, much noise is visible. As soon as video recording starts, the noise level drops significantly (appears to be temporal NR).

[maxotics]

Sorry for any confusion I caused, I was talking about my experience with the a7.  I'm glad to hear this is little aliasing in the a7s (makes more sense with the lower pixel count)...Ooops, I meant a7S :)

[GMaximus]

No it doesn't. It's just a measurement thing. It measures iso 3200 with slog2 but DO NOT compare it to other cameras. There is no DR loss compared to anything.

Would be great, a real-world test is the answer anyway.

[Wulf]

 I didn't spend much time trying to fix it to make a point: Sony color science is a challenge (have the same issue with the FS700). The A7S is a useful tool despite the issues- I'll spend time figuring out how to work around them (something I wouldn't need to do with Canon).

 

 

 

Weird skintones... obviously not only an issue, but a very emotional too: Nobody wants to sit hours after hours desperately trying to fix stuff like this... :-)

 

Sry again for my harsh words, JCS, but I was really shocked at first look! Thx for the warning, the picture illustrates your point perfectly (unfortunetly, one might add).

 

Perhaps I was a little biased by the a7s review of the camera store, fimed with the GH4  - here it's vice versa, the Pany looked very unnatural in comparision to the footage of the a7s. On the other hand, we are seeing even very unpleasent still pics every day out of canikons too, so as allways in life every detail needs a bit love...

[Quirky]

just a few low-output incandescent lights above

camera was set to AWB

Something about the Sony sensor/color science is biased 

 

Whilst I think that ignoring the loud and pedantic online critics is a feasible idea, can't help but wondering if you see any contradiction above? At least to an outside observer it would seem so.

 

Or maybe it's just me, but in case those incandescent lights were just ordinary household/office fluorescent lights or LED lights, using "incandescent lights," "AWB" and "biased colour science" in the same sentence wouldn't sound too logical to me. 

 

Suffice to say photos/frame grabs taken under such circumstances may look nice and could be used to highlight, compare or even judge a number of things, but surely not reliable 'colour science?' Or even accuracy of skin tones, for that matter. 

Just sayin.'

 

Once I found that the Natural profile worked well with the GH4, and to not use the curve controls, I've gotten generally good results without much effort with the GH4. Hopefully we'll find similar settings with the A7S.

 

I'm sure you will, and the results will look nice, especially in better lights. Carry on posting your findings, despite the grumpy critics.

Don't let your inner 'Canon bias' slow you down, either.  :P