Speculate: How much noise improvement from 5D3 to 5D4?

[VelocideX]

I realise that the 5D4 hasn't been announced yet, but I'm curious on views about how much improvement in noise we might expect going from the 5D3 to the 5D4 (assuming it comes out in the first half of this year).

I note that the 7D2 has about 1/3 stop noise improvement (via increased sensitivity) from the 7D, which is disappointingly low.

I love low-light photography, and am really hoping for 1/2 to 1 stop improvement (although I appreciate 1 stop improvement would be a huge stretch). Getting ISO 25,600 to the point where it is seriously usable (like ISO 12,800) would be fantastic.

Many thanks!

 

[Don Haines]

I realise that the 5D4 hasn't been announced yet, but I'm curious on views about how much improvement in noise we might expect going from the 5D3 to the 5D4 (assuming it comes out in the first half of this year).

I note that the 7D2 has about 1/3 stop noise improvement (via increased sensitivity) from the 7D, which is disappointingly low.

I love low-light photography, and am really hoping for 1/2 to 1 stop improvement (although I appreciate 1 stop improvement would be a huge stretch). Getting ISO 25,600 to the point where it is seriously usable (like ISO 12,800) would be fantastic.

Many thanks!

There is signal, and there is noise.

With the improvements in quantum efficiency lately, I would expect to see about a 1/5 stop improvement in signal strength, but as to noise, it's very hard to say... perhaps another 1/5 stop if the existing A/D system is tweaked from the 5D3, but the intro of a new technology could drastically change things....

That's a fancy way of saying probably 1/3 stop, but who knows....

 

[BozillaNZ]

High ISO noise or low ISO noise?

High ISO noise maybe 1/2 stop better

Low ISO noise has a lot of room for improvement. If we are luckly (not sure atm), it could be 3-4 stops better :-X :-X :-X If not, then it's 0 stop, which means no improvement :'(

Or if Canon can employ the on sensor, dual gain ADC, then it will be 5 stops better. One could dream... ???

However I've already given up on chasing the shadow pushing mania. Even if it's clean when pushed 5 stops, the tonal range is severely limited. Better learn how to properly bracket and merge. Unless someone uses 16-bit ADC (or more) and gamma mapped raw encoding, bracket is a better option.

 

[RGF]

I realise that the 5D4 hasn't been announced yet, but I'm curious on views about how much improvement in noise we might expect going from the 5D3 to the 5D4 (assuming it comes out in the first half of this year).

I note that the 7D2 has about 1/3 stop noise improvement (via increased sensitivity) from the 7D, which is disappointingly low.

I love low-light photography, and am really hoping for 1/2 to 1 stop improvement (although I appreciate 1 stop improvement would be a huge stretch). Getting ISO 25,600 to the point where it is seriously usable (like ISO 12,800) would be fantastic.

Many thanks!

To be precise, based upon all available information to date, 2.34 averaged over all ISO and sensor signals from 70 to 95% of maximal count.

 

[Marsu42]

I note that the 7D2 has about 1/3 stop noise improvement (via increased sensitivity) from the 7D, which is disappointingly low.

What do you expect from a crop sensor? Even Sonikon doesn't do better. Given the development with this sensor size, most people (including me) were pleasantly surprised there was progress at all.

Getting ISO 25,600 to the point where it is seriously usable (like ISO 12,800) would be fantastic.

Btw: You do know that iso >6400 on 5d3 is simply digitally amplified, and you only loose dynamic range when shooting raw and going higher? The high iso values are for jpeg, video, marketing - but not for hq stills shooting.

I love low-light photography, and am really hoping for 1/2 to 1 stop improvement (although I appreciate 1 stop improvement would be a huge stretch).

On *high* iso? Imho: forget it. One stop would mean 200% the efficiency of the current design. What you can expect is less banding in the shadows like on 1dx and 6d, i.e. you can really make max use of the dynamic range and software denoising w/o running into artifacts.

Personally, I expect this the only gain on the 5d4 - they'll probably raise the resolution a bit and stay about the same on snr at 100% crop. Most folks will be fine with this, there are plenty of other recent innovations to make the 5d4 sell.

 

[LetTheRightLensIn]

Hopefully 3-4 stops DR at ISO100. Likely 1/3 stop better DR at ISO100 : :-\ :'( ???. (Or if following 5D2 to 5D3 pattern, likely -1/4, minus yes, stops DR at ISO100 ??? ??? : : :-\ :-\ :'( :'().

Fantasy 8+ stops better SNR ;D. More realistically 1/3 to (pushing it) 1/2 stop better SNR.

 

[tcmatthews]

To early to tell ask again in a year.

 

[Mt Spokane Photography]

Don't expect high ISO noise improvement, Canon is going for low ISO improvements, better color, things like that. Any High ISO noise improvement will be slight. Of course, the new Digic 6 processor will produce jpegs at higher ISO settings and less detail, but the actual raw sensor output will be a incremental improvement.

The Dual Pixel sensor does have promise if Canon can do more things with it than they have shown on the crop cameras. I suspect that there are some more tricks they can use and are reserving them for the more expensive models. That is the standard sales tactic, whether its a iphone, a new car, a TV set, or a Canon DSLR.

 

[BozillaNZ]

Don't expect high ISO noise improvement, Canon is going for low ISO improvements, better color, things like that. Any High ISO noise improvement will be slight. Of course, the new Digic 6 processor will produce jpegs at higher ISO settings and less detail, but the actual raw sensor output will be a incremental improvement.

The Dual Pixel sensor does have promise if Canon can do more things with it than they have shown on the crop cameras. I suspect that there are some more tricks they can use and are reserving them for the more expensive models. That is the standard sales tactic, whether its a iphone, a new car, a TV set, or a Canon DSLR.

There is no any indication that Canon will be working on improving low ISO DR. They said they are working on a high resolution sensor (no DR mentioned), they also stated that their sensor is the best (sigh...). All those EXMOR wannabe rumors are more like a wishlist than actual information, that's why it seems very depressing in Canon land.

I think from current reliable source, it's best to hope for the next FF sensor to be a FF 7D2 sensor: 52MP dual pixel, no banding, same 11.5 stops low ISO DR. Then you can only get positive surprise if it turns out to be more than that.

I'm currently trying hard to master the way of shooting sky separately (2 shots bracket, -2 to -3 EV) and combine in post. If done right it looks like you have all the DR you really need. The only situation I find that needs more DR is when bright sky is present.

 

[unfocused]

Okay. First of all. Sensor technology has advanced to the point where future improvements will be incremental at best. And that is across all brands, despite what the Sony fan-boys would have you believe.

That said, the 7DII sensor is, by all accounts, quite a significant improvement over the 7D sensor and even a slight improvement over the 70D.

If you think the improvements of the 7DII were insignificant, then you will likely view the improvements of the 5DIV in the same light. If you appreciate that Canon did improve upon the sensor for the 7DII, then you'll likely appreciate any improvements that materialize with the 5DIV.

The same people who whine on this forum about Canon sensor tech will continue to whine, regardless of what improvements may occur.

Now, as someone who uses the 5DIII in frequent low-light situations, I am completely blown away by the capabilities of the Canon sensor. Results matter, and the results I get leave me very impressed and pleased with the performance of the 5DIII. We are playing in the margins and it is nearly impossible to see significant low-light improvements. For my purposes, the Canon 5DIII is the best available camera for all-around use, reaching a reasonable compromise between resolution, dynamic range and ISO sensitivity.

I am one of those who has no desire for high megapixels or extreme dynamic range and I believe Canon will continue to focus the 5D series on this all-around capability, making marginal improvements across the board.

Finally, it would be a real surprise to see a 5DIV in the first half of 2015. As Maeda Masaya alluded to in his interview, product development cycles are getting longer as the technology matures. It would not be at all surprising to see a 5DIV unveiled at the 2016 Photokina.

 

[sanj]

No speculation here. Around 1/2 stop.

 

[Keith_Reeder]

I note that the 7D2 has about 1/3 stop noise improvement (via increased sensitivity) from the 7D, which is disappointingly low.

And that's disappointingly inaccurate.

I own both cameras and - depending on the ISO - I see the very best part of a stop in noise improvement.

Seriously - treat anything you read from/about DxO as a work of fiction compared to Real World experience...

 

[Keith_Reeder]

That said, the 7DII sensor is, by all accounts, quite a significant improvement over the 7D sensor and even a slight improvement over the 70D.

Exactly that - I see the exact same thing (I've got a 70D too).

The same people who whine on this forum about Canon sensor tech will continue to whine, regardless of what improvements may occur.

And especially that.

 

[GMCPhotographics]

High ISO noise or low ISO noise?

High ISO noise maybe 1/2 stop better

Low ISO noise has a lot of room for improvement. If we are luckly (not sure atm), it could be 3-4 stops better :-X :-X :-X If not, then it's 0 stop, which means no improvement :'(

Or if Canon can employ the on sensor, dual gain ADC, then it will be 5 stops better. One could dream... ???

However I've already given up on chasing the shadow pushing mania. Even if it's clean when pushed 5 stops, the tonal range is severely limited. Better learn how to properly bracket and merge. Unless someone uses 16-bit ADC (or more) and gamma mapped raw encoding, bracket is a better option.

What he said....a lot of truth here.

 

[danski0224]

I realise that the 5D4 hasn't been announced yet, but I'm curious on views about how much improvement in noise we might expect going from the 5D3 to the 5D4 (assuming it comes out in the first half of this year).

The difference between the 5DIII to 5DIV isn't as great as the difference between the 5DIV to the 5DV. I'd wait.

The direct print button finally works too.

 

[Sabaki]

I note that the 7D2 has about 1/3 stop noise improvement (via increased sensitivity) from the 7D, which is disappointingly low.

And that's disappointingly inaccurate.

I own both cameras and - depending on the ISO - I see the very best part of a stop in noise improvement.

Seriously - treat anything you read from/about DxO as a work of fiction compared to Real World experience...

I concur

I've learnt big time with my 7Dii that it's not just noise but the type of noise which is different. So much easier top clean up which results in a cleaner, sharper image after post.

In respect of the 5D4, I really, really would like to see less noise at base ISO, with the same manageable type of noise that the 7Dii renders.

 

[whothafunk]

I note that the 7D2 has about 1/3 stop noise improvement (via increased sensitivity) from the 7D

this statement is complete and utter bullcrap.

 

[Don Haines]

I note that the 7D2 has about 1/3 stop noise improvement (via increased sensitivity) from the 7D, which is disappointingly low.

And that's disappointingly inaccurate.

I own both cameras and - depending on the ISO - I see the very best part of a stop in noise improvement.

Seriously - treat anything you read from/about DxO as a work of fiction compared to Real World experience...

I concur

I've learnt big time with my 7Dii that it's not just noise but the type of noise which is different. So much easier top clean up which results in a cleaner, sharper image after post.

In respect of the 5D4, I really, really would like to see less noise at base ISO, with the same manageable type of noise that the 7Dii renders.

it is right and it is wrong...

The 7D2 is about a third of a stop more sensitive than the 7D and that gives you more light.... that part of the statement is right, but the electronics of the A/D system also seems to be better, and that gives less noise. We have more signal and less noise... according to many reviewers it is roughly a stop better. My personal experience would be to say 1 1/3 stops better, but that's anecdotal, not controlled testing..... but the DXO claim of 1/3 stops is rubbish. What else would you expect from a company that ranks the 50F1.8 as a superior lens to the 600F4II???

 

[Marsu42]

The 7D2 is about a third of a stop more sensitive than the 7D and that gives you more light

Hmmmm, I don't know if this statement is precise, but I know what you're trying to say :->

In fact, what I am still missing and would like to have as a feature on the 6d is a second dial which controls the ambient light - dial left = sun goes up, dial right = sun goes down

 

[sdsr]

One aspect of this that may matter is the resolution of the 5DIV's sensor and how you view your images. If you look at your photos at, say, 100%, higher resolution images are more magnified than lower resolution images - and of course that includes any noise. I mention this because I own both a Sony a7r and the much-touted high ISO king, the Sony a7s. If you view at 100%, the a7r's images look much noisier than the a7s's, but that's largely because of the difference in magnification between 36MP @ 100% and 12 MP @ 100%. View them at the same scale and the differences diminish drastically - or so it seems to me (I'm referring to comparisons of RAW files here). In terms of noise, the differences among those two and the 6D (which I still own) and 5DIII (which I recently sold) strike me as pretty trivial in terms of noise. You can try this for yourself with the DPreview comparison tools. (There may be more significant differences among these cameras in terms of color and detail as the ISO gets higher, but those differences seem minor to me too.)

Of course, I haven't a clue when a 5DIV will appear, or how its sensor will perform in practice. But I wouldn't be surprised if, assuming it has much higher MP count than the 5DIII, that some/many will buy one, take high ISO images, view them at 100%, and complain about the noise....