Jump to content

Fuji GFX 100 loses to Huawei P40 Pro Plus smartphone for dynamic range


Andrew Reid
 Share

Recommended Posts

  • Administrators

DSC00002-scaled.jpg

Have smartphones overtaken the best medium format cameras for dynamic range?

This is a comparison that should worry everybody in the camera industry.

Full blog post:

https://www.eoshd.com/news/fuji-gfx-100-loses-to-huawei-p40-pro-plus-smartphone-for-dynamic-range/

Link to comment
Share on other sites

Isn’t the smartphone just doing auto-magic photo bracketing? Which could be done manually with the GFX 100, albeit with a lot more work. Seems kinda unfair to the GFX 100 merely on a technical standpoint (not on a user experience standpoint). Which means the smartphone took more than one photo, perhaps up to 5, compared to the GFX 100’s 1 RAW.

It’s surprising to me camera companies have not embraced this “in camera” processing more, even for special “auto” modes where the user just wants the camera to do everything.

Regardless, this is a good example of why camera sales have plummeted in the consumer market. It all comes down to the user experience and the smartphone, in this case, clearly provided the better user experience. If you were to repeat the test and bracket 3 RAW photos from the GFX 100 the results would be superior.

Link to comment
Share on other sites

  • Administrators

Yes of course it's multiple frames at different exposes fused together. Tap the shutter button and it's done for you, by a neural network no-less, with AI. And certainly more than just 3 frames fused together, due to the speed of the sensor.

The GFX 100 burst isn't fast enough for this to work.

It does show that small, fast smartphone sensors can overcome shortcomings like dynamic range, with computational photography.

They have never been able to beat a medium format camera, massive 44x33mm sensor with 100 megapixel before.

The Huawei is the first to do this.

Link to comment
Share on other sites

  • Administrators

The medium format level dynamic range seems to extend into portrait / aperture mode as well on the P40 Pro Plus... which is very unusual for a smartphone.

The depth map computations AND the multiple frame merges all done at once in a fraction of a second.

Here's the result...

IMG_20201106_010909.jpg

Link to comment
Share on other sites

Since this is definitely the result of the smartphone combining multiple images, competing with this in stills is a matter of bracketing.

For video, I imagine there may be various schemes. Picture a big ugly Franken-Camera with a main sensor and a smaller secondary sensor to grab the highlights. Something like an A7s iii with a NEX riding in a ‘sidecar’. Unwieldy, complicated, and totally uncompromising; exactly the king of thing that is at home on a film set.

I am also reminded of Fuji’s old multi-size pixel sensors in the Nikon-bodied Finepix line. They were stills only, but they were claimed to be magic sauce by many wedding photogs in the mid 2000s.

Link to comment
Share on other sites

Ah. Maybe if a more mainstream manufacturer tried the outcome might be different.

I have always quite fancied a high end phone with lens attachment and grip etc, but then the cost wasn't that different from am equally capable compact which in itself, is more convenient.

But it's an inevitable fact that tech in phones is closing the gap on 'large glass' camera/lenses so the 'answer' for the industry is on the skills side, - being able to offer something that tech cannot compensate for.

Link to comment
Share on other sites

Impressive. 

And how is it for video? 

Probably cannot do all these tricks on the fly in video mode yet and still look great. But... Give it a couple of years and it will be able to. 

Computational cameras are the future. And everyone will have an alexa micro in their hands in a couple of years. 

With probably a lot of tools that aid regular folks to make better shots. 

Link to comment
Share on other sites

1 hour ago, Nerv said:

Again, the lens is the most important part of an image

Respectfully, it’s not.

The lens is just one part of the equation.

You can put great glass on a shit camera and it’s still a shit camera just as a shit lens on a great camera is still a shit lens.

A great photographer with an average camera will always out-shoot a complete muppet with uber-expensive kit.

I will put skills before kit every time.

But phones, some of them are looking a bit good and I’m one of those people that went to just using my phone over a camera...but gone back to camera despite the ‘inconvenience’.

Can’t deny the capability of some phones these days in the right hands however. There’s some great stuff in another thread on this forum.

Link to comment
Share on other sites

  • Administrators
4 hours ago, Stab said:

Impressive. 

And how is it for video? 

Probably cannot do all these tricks on the fly in video mode yet and still look great. But... Give it a couple of years and it will be able to. 

Computational cameras are the future. And everyone will have an alexa micro in their hands in a couple of years. 

With probably a lot of tools that aid regular folks to make better shots. 

Yes it does all the tricks on the fly. Video dynamic range is very similar to the stills mode.

Doesn't shoot hybrid LOG gamma like the Xperia 1 II though.

Link to comment
Share on other sites

I think modern phone sensors actually use single shoot, but with varying exposure time, eg quarter pixels exposed for +1ev, another quarter pixels exposed for -1ev etc. Therefire much less ghosting compare to traditional bracketing. Then HDR-processed by smartphone's powerful ISP. If the scene is bright enough eg daylight, bracketing still used in tandem to further increase DR.

Link to comment
Share on other sites

  • Administrators
31 minutes ago, Wan Taquddin said:

I think modern phone sensors actually use single shoot, but with varying exposure time, eg quarter pixels exposed for +1ev, another quarter pixels exposed for -1ev etc. Therefire much less ghosting compare to traditional bracketing. Then HDR-processed by smartphone's powerful ISP. If the scene is bright enough eg daylight, bracketing still used in tandem to further increase DR.

Thanks for the explanation.

It's a 50MP Quad Bayer sensor in the phone, almost 1" size.

Quad Bayer can use Dual ISO in one exposure, the pixels at different exposures are merged in the 12MP shot.

Also it is RYYB. The yellow diodes let in more light than traditional Green ones with bayer sensors.

Then there is neural network and AI processing side.

And probably a very high speed readout in some modes with multiple frame merges.

Smartphones are leading the way for camera technology at the moment.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...