Buying guides 2015

News Reviews Articles Buying Guide Sample Images Videos Cameras Lenses Phones Printers Forums Galleries Challenges

Showing dynamism: EOS 80D breaks new ground for Canon low ISO DR

Published Mar 21, 2016 | Richard Butler, Rishi Sanyal

1 Raw DR: Exposure Latitude

Our testing shows that Canon's EOS 80D features a sensor with greatly improved Raw dynamic range. The results appear to confirm the company has moved to a new sensor design with lower read noise.

Previous Canon sensors have conducted the analogue to digital conversion step away from the sensor, an approach that contributed noise that limited dynamic range at low ISO settings (while still allowing the excellent high ISO performance that Canon's CMOS chips built a reputation with). The sensors in the EOS 80D and EOS-1D X Mark II appear to have an on-chip ADC design that conducts the conversion within the sensor, shortening the electronic path and preventing this noise building up. Let's take a look at the 80D's Raw DR performance in a couple of our studio tests.

Exposure Latitude

In this test we look to see how tolerant of pushing exposure the 80D's Raw files are. We've done this by exposing our scene with increasingly lower exposures, then pushed them back to the correct brightness using Adobe Camera Raw. Examining what happens in the shadows allows you to assess the exposure latitude (essentially the dynamic range) of the Raw files.

Because the changes in this test noise are primarily caused by shot noise and this is mainly determined by the amount of light the camera has had access to, the results are only directly comparable between cameras of the same sensor size. However, this will also be the case in real-world shooting if you're limited by what shutter speed you can keep steady, so this test gives you an idea of the amount of processing latitude different formats give.

As you can see, the 80D is contributing less noise to its images than the 70D did, and this difference will be evident when you try to pull shadows up. It isn't quite a match for the Nikon D7200, but it's enough that in real-world use, the files should have similar - if not slightly more - processing flexibility than the Canon EOS 5DS, despite the latter's larger sensor!

See our ISO Invariance test results

Tags: cameralens-features, canon, technology

TABLE OF CONTENTS

Comments

Total comments: 843

good as it can be from Canon and still years behind Sony
kudos to Canon who has waked up from the deep sensor sleep and ugly read out noise

1 upvote

excelente camara!!

0 upvotes

The photographs are even more interesting than I thought. As long as you are pushing shadows three stops or less, the 7D II actually looks cleaner; compare the ISO 400+3 stops for example. I do think this confirms that while the 80D may be more ISO invariant and have more dynamic range, you'll probably see better results with the 7DII so long as you are not heavily DR limited.

0 upvotes

I can't select 7D II on this page. Where are you comparing the two?

0 upvotes

Page 2, ISO invariance.

0 upvotes

Hopefully this is not the only set of images for the Studio tests.
I downloaded the JPGs both from the 80D and the 7DII and noticed
that the 7DII test images are better lit than the 80D, so the
"results" from the 80D test images are not an accurate
account of the camera's ability to produce comparative
images.

Download them and take a look. If you look at the bottle
Just to the right of the 20 Schilling note you'll see that on the
80D image the bottle top and neck are lit to the right.
On the 7DII image in the same location the bottle is equally
lit on the right and left of the bottle top and neck.

You'll also see that the Test image for the 80D is darker
on the left side of the entire test image.

I can only hope that DPR will reshoot this test for the actual
review, if not then there is some marketing play going on and
the 80D may be steeling too much of the 7DII's thunder.

Hopefully not. Just wanted other members to be aware of this
and to download images for themselves.

0 upvotes

Interestingly, the 80D's sensor itself appears to have slightly higher read noise than the 7D Mark II. The ISO 3200 images on the 7D II look cleaner. Perhaps due to the addition of the ADC circuitry on sensor?

0 upvotes

@PetaPi
"Look at what Magic Lantern does to a Canon. All that could be in the camera tomorrow if only some committee at Canon would have some guts and trust in the future.
There's nothing they can say that will convince me it's not possible to put more features in the cameras.
There's going to be some company out there that's not going to live in fear that if they put feature X in the camera today there's nothing for them to put in 3 years later. For now, Canon manages them and give incremental meagre features.
Feel free everyone to flag my comment."

Fair is fair, you have good and valid points here.
I happen to be using the Magic Lantern software on my Canon.

Edited 2 times; latest 4 minutes since posting

2 upvotes

So 90% of the DSLR are using Sony sensors, but Canon is Canon. So at least one company is doing it right for how many years.

1 upvote

800+ comments and coming. Very impressive. I would expect less comments if Canon would announce a 1GPx FF sensor :) So here is the plan now:
1. Buy an 80D
2. Start taking underexposed images and lift the exposure by 5 stops in PP
3. ?????
4. Profit!

Edited 5 minutes after posting

3 upvotes

You and the underpants gnomes think alike.

0 upvotes

it is to show that there is more details in shadow areas. If you can't understand that. You have to read your own comment one more time. And think: Am I proud about my comment?

0 upvotes

You can't understand jokes, can you? You seem to be very proud of your comment. Achievement unlocked, congrats!

1 upvote

Hi.... Interesting electronics etc 80D vs the 70D which I own, Also interesting how the image quality can be increased, lets hope the future chips get even better and better.

However:- the burning question is this.....

Does this on the 80D really make the images so massively better then the 70D ?

Personally I feel the 80D is an incremental improvement not a proper upgrade, Which would make EVERY 70D owner want to upgrade asap....

What do you think.... :-)

0 upvotes

I would have been jumping for joys. Unfortunately it is a little too late after a damage done by an A7R2 and its stellar lenses.

1 upvote

Yep. The stellar 200 f/2L IS, stellar 300 f/2.8L IS II, 600 f/4L IS II etc... All Sony native E-mount. Oh, wait, that's not Sony lenses? The Sony 24-70 f/4 aka "Zeiss", that one is stellar for sure ;)

0 upvotes

@AngularJS While I agree the lens selection is a serious issue for E-mount, the examples you gave just aren't. Sony has a 135mm f/1.8, 300mm f/2.8, 500mm f/4, all working with the A7RII, stabilized. Those all go pretty well head-to-head with the Canon. There's no exact match for the 200mm, but the other two are almost identical. Once you're going large aperture telephoto, the A-mount/E-mount adapter has basically no downsides -- it adds little size or cost relative to a $6000 5.6 pound lens. Indeed, Canon has to make some compromises in optical design for long/wide-aperture IS, and for such lenses, IBIS is a clear win.

If you had mentioned Canon's perspective control lenses -- especially their completely unique wide angle perspective control lens -- that'd be one thing. But the ones you mention have reasonable equivalents for the A7RII.

Edited 27 seconds after posting

0 upvotes

Theoretically you can adapt pretty much everything on the A7r2, but it won't ever work as a native lens (tried already :). Yes, Sony offers similar "big whites" for the A-mount, but look at the 500mm f/4 prices: Canon - $9k, Sony - $13k. $4k diff means another high-end body. I shoot both Sony and Canon, so far kept just the 55 f/1.8 and 35 f/2.8. The rest of the lenses is "less stellar". The new GM lenses look good on paper, looking forward to see some actual tests though.

0 upvotes

What do you mean "it won't ever work as a native lens?" My understanding was that most of the A-mount lenses, specifically, adapted through the Sony adapters, worked as well as native. The same is supposed to be true about a small number of Sigma lenses adapted through the Sigma adapter. The autofocus system, in reviews I read, was different, but neither better nor worse (and I expect in future bodies, will be the same).

1 upvote

I'm sorry - it was really unclear, I agree. I meant focusing speed, especially in dim light. The SLT-based EA4 adapter works reasonably fast (sacrificing some light), but the EA3 is a disaster. The AF is very unreliable if my focus point is not in the center, and the AF-C mode with any adapted lens is unusable (to my standard). No Eye-AF feature via an adapter neither. With all that said I don't think the A-mount lenses can be considered as a full equivalent of whatever is missing in the E-mount.

0 upvotes

And I am still waiting for dual pixel RAW mode before buying a Canon....

1 upvote

I am now very happy with my decision not to get a 5DsR last year. Give me a 5D4 with this sensor tech, say 28 mp and that is it. It would give very, very high quality 50x75 cm prints from a single frame. Stitch to get gigantic enlargements and panoramas.

I sold my 1DsMkIII in anticipation 6 months ago and have been shooting with a pair of 6D bodies since.

5 upvotes

And hopefully Canon will continue to improve its sensors.

Maybe they will do BSI for APS-C & 35mm sensors, eventually doing stacking tech for them, but it will maybe take a couple of gens to get there - but their lens are arguably equal to the very best and definitely are cheaper for great quality compared to Zony lens for example !!

And touch screens really do have a nice advantage to non touch, Sony does not use these much yet.

0 upvotes

Yes, I think that would be the sweet spot for me too. Particularly if it had 4K.

1 upvote

Cannot see myself that 80D is better @100 iso than the 5DS/R from ev+1 to ev+2 nor that it is better than the 5DS/R @ev+5 at same file size?

0 upvotes

We are talking about a (moderate pixel count) Canon full frame with this sensor tech.

0 upvotes

The 5dsr is a great high resolution camera if you 1) have high resolution lenses 2) if you like to see fine details. I'm super happy with the 5dsr and if one day Canon decides to update it with a newer sensor tech - I'll be first in the pre-order line. Not having a 5dsr now while waiting for a "better camera" means lost possibilities for me.

0 upvotes

For me, I have the A7RII and A7SII, as well as my Canon gear, but one Sony top of the line camera has too many pixels for my everyday photography (I sometimes welcome it but not always) and the other has too few. although it shoots great video.

To me 20-25 mp is the sweet spot

Edited 24 seconds after posting

1 upvote

So do we know if this is a FSI or possibly a BSI sensor ?

Superficial research shows that FSI with reflective lens (to bounce the photons down to well and prevent reflections from metal layers) can be nearly as efficient as a BSI, especially w/o stacking (the 'nearly' would explain the 'not quite as good as Sony/Samsung performance)..

Stacking does have an advantage beyond the obvious, in that the size of the analog layers topology can be larger than is needed for the digital layers, allowing for optimal silicon characteristics (dark noise, etc), as well as mixed substrate material.

I would hope we are all happy that Canon is working to increase the low ISO DR!

PS: To me it seems the 80D is ~2ev worse (than Nikon 7200) but is a large improvement over the 70D!!

Edited 5 minutes after posting

0 upvotes

Conventional wisdom with Canon sensors has been to over-expose a bit because you can pull details out of slightly blown highlights where it's traditionally been difficult or impossible to pull details out under-exposed shadows.

Does this approach need to be revisited with this new sensor?

0 upvotes

No. Most digital sensors behave linearly at the upper end (at least, the portion of sensor response that's used in the analog-to-digital mapping), so there won't be huge differences in response at the upper end from camera to camera. That's not a Canon thing. The ability to pull detail out of blown highlights depends on (1) how many channels are blown and (2) how good your Raw converter is at doing this.

2 upvotes

But... But.. But... All I hear from the Canon apologists on here is that Canon's products are fine with the dynamic range they have. No improvement is needed!

I guess we can now safely say that even Canon disagrees with ALL of you.

9 upvotes

Then I would suggest you have reading comprehension issues. "All I hear from the Canon apologists..." Your use of "all" means everybody, but at that point I chalked it up to a manner of speaking. But when you capitalized it and restated your position at the end of your post, I realized you truly meant everyone and that you have trouble reading/comprehending.
Allow me to explain it a little better...
The overwhelming vast majority of Canon users acknowledge that Canon has been behind in sensor technology when it comes to noise/DR. It's not even up for debate. There existed a large discrepancy.
The vast majority would welcome less noise and more DR. Some don't care because their style of shooting doesn't require that level of technology. And then some (a minority) have expressed that GREAT results are possible with even lesser tech than Canon has historically offered. This too, is not up for debate - as all famous photographers made their name prior to modern sensor tech.

14 upvotes

Absolutely. But it is too early to shout "victory" for Canon in the sensor battle.

Edited 14 minutes after posting

1 upvote

I use Canon, Sony A7RII, Sony A7SII, Olympus and Panasonic, for just a few.

What you ACTUALLY hear from Canon users if you don't have comprehension issues is that the dynamic range issues affect less than one in a hundred situations where we might photograph, that other cameras' dynamic range "advantage" occurs only at base ISO which we rarely use, and that other cameras also have disadvantages that make them, in many situations, less reliable or give less good IS, than our Canons.

Sony's "colour" affects EVERY shot I take with it. Yes, I CAN make it better, and do, but why should I need the extra effort on EVERY shot? Its focus speed is NOT up to DSLR levels whatever the true believers spout. And maintenance and servicing is behind Canon, as is its reliability.

DR is just the stick some brand bashing buffoons use to beat Canon. The main advantage of this upgrade is that now they will have to find another one. They seem pretty frustrated right now that this one is disappearing.

17 upvotes

Incidentally, I would add that in many ways the DR obsession isn't really much more advantageous than the previous megapixel race was, and I am often amused and horrified in equal part to see the "Look ma, I got HDR" images provided by many Sony users who want to prove it by pushing up shadow areas and dragging down highlights until every image has milky browny blacks, no "snap", and looks like it was shot by fogged film sometime in the 1950s.

I am constantly double checking myself to make sure I have not been suckered into that particular DR "vision".

And let's not even get into what my Sony sensors do to detail in blue artificial or LED light unless I push the colour temperature way, way, up and then compensate for it in post.

To which true believers will say: "None of this matters: dynamic range, dynamic range, dynamic range, dynamic range ......"

Yes, I probably will be selling one of my Sonys when the 5D4 comes out. The death of DSLR has been much exaggerated.

Edited 3 minutes after posting

17 upvotes

DR matters, but it is only part of the equation. Resoliution matters too, but also here it's only part of the equation. I'm still saying that my D20's 8 Mpix are enough for getting correct and awsome pictures. But I still porefer my current 5DMIII for its fast AF, high ISO, higher resolution (yes, even if 8 is OK, 20 is better) and higher DR.

It's not a religion, it's simply technological progress.

5 upvotes

Balanced and sensible post.

0 upvotes

It looks like this sensor is up to the level with Sony sensors of 2010, and that is pretty good. Practically a ISO-less camera.

1 upvote

To my eye it's better than the Sony A6000 and that is a 2 year old camera.
I would say that's pretty damn good!

9 upvotes

Doesn't seem any better than the 6000 to me, especially blacks and deep shadows where the reverse seems to be true? Maybe I'm looking at it wrong.

3 upvotes

The comments section could be greatly simplified.
All we need are two posts

1. Its a Canon. Its rubbish
2. Its a Canon. Its fantastic

Then people simply like the one they prefer.

16 upvotes

Or: It's a whatever. They ALL have issues and strengths.

1 upvote

How about "It's a fantastic rubbish" ? :)

0 upvotes

I think Canon paid a lot of money to get a positive article about the 80D.
Or did some favours to Amazon.
Grasping at straws.
Canon now has better blacks ? Is that it ?

5 upvotes

Ridiculous, and slanderous. Obvious that taking ADC on sensor would have a significant impact. In terms of the blacks, actually compared to the other APS-C cameras, certainly. D7200 excels at noise reduction but has a magenta hue. As for grasping at straws, only you are, in your effort to malign an obvious improvement in Canon sensor tech. The actual images speak differently. Yes, images do count for more than hyperbole. Looking forward to seeing much more images from Canon users, should be very interesting indeed, looking very positive. http://www.dpreview.com/forums/post/57489564

Edited 3 times; latest 6 minutes since posting

13 upvotes

@PetaPi - how many are you paid to publish negatives on Canon? Don't be a fool and make a fool of yourself.

When I bought my first Canon, it was the best of his class. Then, later, Nikon and Sony (Minolta) took over the lead. I had invested in lenses, so I stayed with Canon. It's not a religious mantra for me to stay with Canon, but only to protect my invest in lenses. If Canon can now take over again, it's a standard situation in a race, where nobody will ever see the finish line.

11 upvotes

I think Sony/Nikon paid a lot of money to PetaPi to troll the comments section of this article. Or their userbase did PetaPi some "favors".

3 upvotes

Paid a lot of money for dpr to say "much better, but still behind the best"?

That's just silly.

4 upvotes

@PetaPi,
Essentially, you are accusing DPR staff of having been bribed by Canon to publish this positive review.
You don't have any data, information, or a proof to show the world that the DPR staff is essentially, according to your statement, a bunch on unethical folks.
You have the right to dislike Canon, but your are making false statement, and the facts (presented by the article) are far louder than your shoot-from-the-hip false statements

3 upvotes

@PetaPi: understanding bias is not easy. You should upgrade your knowledge. Try some other articles, eventually oriented to another camera brand. Then, make a comparison. And, if you don't care about bias, fine, let's try marketing - and marketshare. So, Canon's (visual) marketshare on this site is on the higher or on the lower side? What do you think?

0 upvotes

Ok, maybe I went to far. Don't have any proof.
New camera generations used to be ground breaking. Now, DP makes a full fledged feature out of something that Canon itself forgot to mention in the press release.
If I was the CEO of Nikon and said that DP is biased, maybe I could be called slanderous, but I am a mere mortal with an opinion.
DP, Amazon, Canon don't make money of average reviews or cameras. Less so from products negative reviews. Amazon could have bought ads in DP but they choose to buy the cow.
I think most of reviews are very good on DP, but DP has no teeth, no bite. I can't remember how it was before. I don't like it when everything is pink.
I never touched a Sony. I am not that popular to be paid for trolling.
This is would be poor quality trolling. No better than Russian propaganda.
There is improvement in Canon, but this is the niche in a niche.
I wish there was a way do delete the post. Don't really enjoy this much controversy.

0 upvotes

Petapi, the groundbreaking days of digital cameras are gone for the foreseeable future, for the next several generations of cameras it will be about refining what already exists.

0 upvotes

That's like saying there's no music after The Rolling Stones or The Beatles, no better engine than the gas engine, and only lame drama after Shakespeare or maybe all the good movies have been made already.
Look at what Magic Lantern does to a Canon. All that could be in the camera tomorrow if only some committee at Canon would have some guts and trust in the future.
There's nothing they can say that will convince me it's not possible to put more features in the cameras.
There's going to be some company out there that's not going to live in fear that if they put feature X in the camera today there's nothing for them to put in 3 years later. For now, Canon manages them and give incremental meagre features.
Feel free everyone to flag my comment.

2 upvotes

Fair is fair, you have good and valid points here.
I happen to be using the Magic Lantern software on my Canon.

1 upvote

Well done Canon, about time.

My first DSLR was a 20D and it was a very fine camera indeed, 'well sorted' as DPR used to say.

I don't shoot with Canon any more, but kept some of my L lenses in case I returned one day, and I could be very tempted by this 80D.

But there are so many great choices these days, it's so hard just to pick one or two.

The Pentax's with the great pixel shift feature, K1 and K3ii, both on the short list.

The new 24MP version of X-Trans, finally delivering on the X-Trans promise, when that comes to a DSLR style body, it will be on the shortlist.

The excellent 1" Sensor in the Nikon J5, if we get a V4 body with that, for a very compact option, on the shortlist.

An updated Olympus E-M1 with pixel shift and the new 20MP sensor, that would make the shortlist too.

Canon, Nikon, Pentax, Fujifilm and Olympus, it's suddenly becoming a very long shortlist!

3 upvotes

Does this mean Canon is using a Sony sensor? ;-)

2 upvotes

No, b/c Nikon 7200's SONY sensor is much cleaner than this Canon for example.

I'm very glad to see Canon uses 1" SONY sensor for its P&S, otherwise it will add POOR IQ to already:

- poor AF system
- poor battery life
- Poor processing power

Canon should use SONY sensor before knowing how to make a decent sensor.

Edited 27 seconds after posting

12 upvotes

Instead of just maligning with irrelevant hyperbole, it would be better to understand the changes that make the new generation of Canon sensors reflect significant improvement. I am no scientist, but understanding the changes is very easy, once a little commonsense is used.

3 upvotes

To ImageAmateur,
"Instead of just maligning with irrelevant hyperbole"

I did write it with a wink. :-)
Canon is copping the flack because they've been riding on their 'big' name for some time & as nice as it is with the improvements they've made it's also fun to give them a bit of stick instead of certain other smaller companies.

0 upvotes

Seems you do not know much about how a Canon works. High ISO was not an issue to start with (if you expose correct and utilize the analog amplifiers of the Canon design) and since nowadays some potential buyers think that a camera system is crap because it cannot push low iso shadows by at least 5 stops now Canon implemented that too. That is just a win win for anybody really interested in Canon. If you push your shadows regularly for 6 stops well go buy a D810 or learn a bit more about how to get the shot right to begin with. I would think the latter will have the biggest impact.

Edited 52 seconds after posting

4 upvotes

Urghh the same pathetic counter arguments again and again and again and again.

Please explain how "getting the shot right to begin with" can magically compensate for the camera having insufficient dynamic range to capture the scene? I've seen about 20 people give this argument yet none appear to be able to answer this simple question.

1 upvote

Exactly. The issue isn't so much with the pattern noise rearing it's head when lifting shadows, it's that it's there to begin with IN the shadows. If you shoot something low key, you have to spend time in post ensuring black is black with Canon where with the Nikons it isn't there at all. The dislike of Canons pattern noise has jack all to do with lifting an exposure 5 stops, if a camera cant produce CLEAN SHADOWS then it's no good for someone that needs clean darks. Period.

2 upvotes

Yup and I run into in to this problem all the time doing landscapes, if you have portions of an image that aren't exposed enough to overpower the noise you can see it at ISO 100 quite clearly with ZERO processing. Then you do something mild like adjusting levels and curves and because the software doesn't discriminate between actual image information and noise grain you end up making the noise more prominent too.

I have shots with the setting sun in frame, glaringly bright and then a few inches away an area of shadow with visible noise, it's not because the shadow areas are inappropriately exposed, it's because the shadows always have that noise present.

The higher the noise floor, the brighther the tones will be and still exhibit visible noise.

Edited 1 minute after posting

2 upvotes

The counter-argument has always been that Canon is the best and use proper exposures, and there are no need for better DR. I'm a little curious why Canon now suddenly invest money in a sensor with lower readout noise?? Better DR
and my sarcasm is based on a large number of argument from Canon fan boys during the years who have declared my arguments for a greater DR, lower banding etc etc as idiotic

Edited 8 minutes after posting

7 upvotes

I doubt anyone said "NO" reason. It's not the only reason to buy one camera versus another one. And how good is the dual pixel on your Sony/Nikon? How about lenses?

3 upvotes

Mikael, you belong in a mental asylum. Now that you've posted the same comment at least three different times, it looks like YOU are the one who is the fan boy (of anything that's not Canon). Anyone who argued that they didn't need anything more from Canon were probably people who actually use their camera to take photographs, instead of forum trolls who circle jerk about Sony with the rest of the DPReview staff.

22 upvotes

Good lord, is this guy still banging on?

I swear I go away for months, if not years, and I come back and he is still bashing a camera he does not use (or like).

I don't use Nikon. (One of the only brands I don't) Time spent going out of my way to bash it = ZERO.

I avoid people muttering to themselves in the street ...........

Edited 4 minutes after posting

17 upvotes

Easy. No Canon users can buy because they have already said they dont need it.

4 upvotes

Yes, Canon!
AE-1 PROGRAM

1 upvote

@GaryJP.... I hear ya!

2 upvotes

So now you can use a Canon either with proper exposure or go wild with post process pushing. What is not to like? Yesterday claims have already been eaten by today's development. Sometimes the bear ...

Edited 2 times; latest 2 minutes since posting

3 upvotes

How does "proper exposure" compress the dynamic range of the world so that it fits the dynamic range of the camera?

People keep saying "proper exposure" makes cameras with superior dynamic range unecessary, someone tell me how to properly expose those sunset seascapes with my Canon so that both highlight and shadow detail is maintained please? If I learn how to properly expose then I can stay with Canon.

Edited 2 times; latest 2 minutes since posting

2 upvotes

Fatdeeman, it doesn't. It is simply the mantra used by a lot of Canon users who don't understand metering, and use the term "shadow lifting" when it has nothing to do with shadows.

2 upvotes

@fatdeeman
Some posters here just do not understand DR and metering.

1 upvote

Any particular reason for all the sarcasm? I never find the need to go to camera/manufacturers reviews and bash it. Seems like passive aggressive online chest pounding. Oh well.

8 upvotes

Plus Nikon doesn't even make their sensors, so Sony should get the credit.

4 upvotes

well David, I have been written about the poorer DR in Canon since 2008 and I wonder, WHY are Canon now changing their sensor read out and to less noise ?During all the years I have received the answer- Canon's DR is good enough, expose properly from the fan boys, and to me it make no sense to invest money in something Canon fan boys already said is good enough ( or is it as long Canon has no answer to Sony)

Edited 6 times; latest 12 minutes since posting

1 upvote

Being "good enough" doesn't mean an improvement won't help their market share and profits. My Ferrari is "fast enough", but I wouldn't mind if it was faster (like my neighbor's Lamborghini). Can you understand that simple logic or is it too complex for you?

7 upvotes

"I've been complaining about it for years and now they do it and I'm asking why".

What a god awfully foolish question that was already answered by the person who asked it.

Really, who actually complains about sensors getting better? Canon muat be in your nightmares of iso noise :)

I read and hear about m-43 all the time and while I would never ever buy a camera with a 2X crop that's not a powershot comma I still feel no reason for me to bounce around for rooms taking potshots at. It just seems totally pointless to me especially when the answer is most obvious. Well that's my 2c, see good day.

3 upvotes

That's a lot of worrying! Poor boy.

0 upvotes

Well, the Canon sensor was good, now it's better! No need for a polemic around this. If you look at the tests, you allways got a mixed image. If you look at the comments, there is a lot of superflous, meaningless oppinion, including yous and thios one.

1 upvote

It is not shortening the electronic path or even the integration of the converter that achieves the improvement. It is the fact that they have increased the effective number of bits (ENOB) of the camera by going to entirely different converter approach. The ramp converter approach is very accurate, but very slow so in order to use it they need to use a lot of them and the easiest and least expensive way to achieve this is to put them on the same die with the sensor elements.

Shortening the path does bring benefits but this was never the root cause of the issue people have been whining about. The root cause always was that they put an ADC with an ENOB of 12.5 bits in what they claimed was a 14 bit camera. The reason that they had to do that was that converters that were fast enough to digitize just a few analog lanes from the sensor design they had, were pipeline designs that sacrifice accuracy to achieve speed.

Edited 2 minutes after posting

4 upvotes

@David Hull
Do you think Chipworks will examine the 80D sensor and associated electronics for a definitive analysis? Canon is being extremely quiet about the sensor improvements aside from DPAF.

1 upvote

@bernardly - DxO already did.

0 upvotes

Is this speculation or do you have inside knowledge to the ADC's used by Canon ?

Also, there is no reason to have more bit accuracy than the signal you have if using a standard DR step is there - IOWs, if you have only '12.5' steps of signal using a 14 or 16 bit ADC will make NO diff.

But if you decrease either/both the analog noise via NR techniques OR/& reduce noise amp noise before the ADC by getting ADC closer to last analog stage, then you will have better conversion capability, yes?

As to the quality of the ADCs, I saw pics of a 35mm Canon cam (5DIII or 1Dx) showing Burr-Brown ADCs which in the hifi audio world are considered (or at least were) quite good/excellent (and analog signals are analog signals whether music or static analog sensor pixel values).

So unless you can fully state the model of ADCs that are/where used by Canon that are 'effectively' 12.5 bits vs full 14 bits then you are just speculating.

So please, give a url to the external ADCs past used by Canon, thx

1 upvote

I checked DxoMark and they do not have any measurement results for the 80D yet, though no doubt, they will be forthcoming given the importance of this new Canon camera. But keep in mind DxoMark just analyses data from RAW files.

ChipWorks examines hardware after device disassembly and uses electron microscopy to analyse very fine electronic structures of unknown components. Given that ChipWorks has analysed Canon sensor technology in the past, it would be expected for them to apply their reverse engineering expertise to new technology from Canon.
http://www.chipworks.com/about-chipworks/overview/blog/full-frame-dslr-cameras-part-ii-canon-stays-course

Edited 2 times; latest 5 minutes since posting

3 upvotes

Coyote_Cody --

Rather than "speculate" you can simply "Calculate". Let's look at the 5Ds. This camera has a firing rate of 5 frames per second at 50x10^6 pixels per frame. That means that you are going to have to make 250 million conversions per second therefore needing a digitizer that can support that conversion rate. You decide on a 14 bit converter and you decide to use four quad ADC packages for a total of 16 converters (we know this from the canon white paper). So you need the converters to be capable of 250/16 = 16 MSPS. So you pick something like the Burr Brown (Now TI) ADC3441. This is a 25 MSPS Quad 14 bit ADC that can do the job.

If you pull up that data sheet on the TI site you will see that it has an ENOB = 12 bits (despite being a 14 bit design). If you calculate ENOB = SNR(dB)/6 you get 74/6 = 12.33 ENOB.

So call it "Educated Speculation".

TI DS: http://www.ti.com/lit/ds/sbas671c/sbas671c.pdf

Edited 1 minute after posting

1 upvote

But once again speculating on which ADC model they use and using the current worst/hardest example - a 50mp sensor & not the 80D sensor mp.

TI also makes 16bit ADC of sufficient speed, so depending on the chosen ADC, Canon could be getting real 14bits-correct? Also 14bit ADCs with full 14bit resolution, still implies speculation w/o actual model used!

And if fast enuf, ADCs can perform multiple conversions on a single sample as well & correlate the results.

Quite a range of models by TI, model-slas760, is a 16bit with a 14bit low power mode w/ 85-dBFS SNR at f in = 3 MHz, 100 MSPS.

So unless the model is know then - still speculating !!

I have read that Sony was using a 8bit ADC then a 12bit ADC on some recent sensors, yet no one is speculating that they do not provide true 14bit raw.

Of course, if the noise level swamps the lsb's, then sure ENOB is less.

Still seems speculation!

0 upvotes

Cody --

I just picked the most recent camera as an example. We know that they are using QUAD ADC packages (based on one of the tear-downs and their white papers) so it is a matter of taking a guess as to which one it might be.

Without a schematic and BOM, everything is speculation.

I am very confident in the system analysis that I sent you in the PM, take a look. I think it tells the story pretty well.

One problem with going to a 16 bit converter, would be the noise sensitivity at it's input. This would be a situation where the signal lengths would probably be a problem as Rishi alluded in the header.

Probably the best part of the new architecture is that they don't have to worry about shielding and routing as much.

As you point out, there are a number of plausible approaches, the trick is to find one that meets the cost and power budget and still does the job.

Edited 3 minutes after posting

0 upvotes

expose properly, what is that? something Canon has invented?
does not answer my question, why a new sensor in Canon, there are no need for better DR due the Canon boys

3 upvotes

Can we not just accept that there are people who do not want or need to recover shadows by that many stops? Just as we can accept that a lot of People will take advantage of the increase base ISO DR.

Edited 8 minutes after posting

4 upvotes

If you expose properly, the dynamic range of any given scene magically compresses to match that of the camera being used. You can also reduce the dynamic range approx 2 stops if you "learn to process properly"

If you spent more time taking photos instead of engaging in a circle jerk with your fellow Sony and Toshiba fanbois pushing photos 13 stops you might know that!

5 upvotes

Hell, at that point, why ILCs at all? All of our smartphones apparently have enough DR for shooting a black cat backlit by a supernova with full fur detail.

0 upvotes

Too bad they suck at posting messages an appropriate number of times. :p

Edited 2 minutes after posting

0 upvotes

Again: because they can do it now!

0 upvotes

I have a novel idea. Expose your images properly and not push shadows 5 stops in some ridiculous test to compare noise. Call me old school.

8 upvotes

Ever shoot a sunrise/sunset with sun and foreground detail in a single image? Hell, even Nikon is confident enough in their sensors' DR these days to include a highlight priority metering setting, that nearly assures you'll have to do at least a little, if not a lot, of shadow pushing in post. Not like there wasn't a bunch of darkroom manipulation going on to help extend film's dynamic range back in the day, either.

5 upvotes

Ever hear of ND filters I would say pros never push 5 stops, except maybe as an absolute last resort They were doing this since film. ND's give better results.

3 upvotes

the point is that if 5 stops looks horrible, a more normal 2 stop push will look mediocre. They push 5 steps to make it easier to see. The example above (lake and guy with soccer ball) is pretty noisy for example if you click on it to view full size.

0 upvotes

No. I don't have any need to shoot that way. If I was going to do that I'd probably bracket and/or HDR. I am a fan of more dynamic range. But IMO it's just not as big of a deal as some make a fuss over. I have done quite well shooting with much less DR than in pretty much all recent bodies. For me, measurbating is fun...but splitting hairs and bashing one over another is something for people that don't actually use their equipment to do.

5 upvotes

I do get that some situations like the example above would benefit from more DR. It's just not something that is keeping me down at the moment.

3 upvotes

"Proper exposure" (whatever the hell that is anyway) doesn't prevent a particular scene having too much contrast for the cameras dynamic range.

Why is it that all these counter arguments are so weak? What's to stop a Nikon user "exposing properly" too? All the arguments in the defence of Canon can be applied to any system out there but the strengths of other systems cannot be applied to Canon.

Edited 12 seconds after posting

2 upvotes

ND filters have never been an ideal solution because they are totally inflexible when it comes to the shape of the horizon. Anything that isn't perfectly flat and protrudes into the transition is going to get darkened too.

2 upvotes

why put a new sensor inside a Canon ? there are no need to have a larger dynamic range according many comments during the years here at Dpreview and at other Canon sites. why not settle for the higher noise and banding as usual

Edited 1 minute after posting

9 upvotes

What happened to cigarette lady?

Please bring back cigarette lady.

2 upvotes

like my Riesen better

Edited 19 seconds after posting

2 upvotes

Ah that's a dog without looking close its shape looks like a camera. A fuzzy camera at that lol.

Cheers

0 upvotes

a Riesen camera, and no banding

2 upvotes

A bit snarky...aren't we.

1 upvote

Pffffff.... If you have nothing intelligent to contribute, then stay quiet!

1 upvote

Why are these raws so small. Imaging Resource posted 80D raws a few days ago, and they're on the order of 32MB to 46MB per file?

0 upvotes

Canon Raws usel lossless compression (like ZIP files).

In general files with more real information compress less so are bigger. You can use the size as a rough guide to information (Image) content.

These were under-exposed ISO test files and have lower captured information. You can see the size increasing from 22MB to 30MB as each file is less underexposed and therefore contains more image detail. A correctly exposed image will be larger still.

Also the captured scene will play a role. An image with a large chunk of featureless blue sky will compress more than an image full of intricate detail.

1 upvote

Yes, but I've never seen variation of 100%. And the biggest files from IR are twice this size--still boring studio shots too.

0 upvotes

Are you sure you are looking at the right files?

This one is fully exposed and is 32MB:

http://www.imaging-resource.com/PRODS/canon-80d/80DhSLI00100NR0.CR2.HTM

On the other hand, if you are reading the size inside PS or LR you will be seeing the uncompressed size.

0 upvotes

The IR files vary in size up to 46MB.

If you read my comment, I very clearly said 32MB to 46MB.

The DPR files are on the order of 23MB.

Reading the files size in Windows.

Ya know, I'm not new to looking at raw files.

Edited 25 seconds after posting

0 upvotes

Cool, let us know when you figure it out.

0 upvotes

Naj:

When I handle a D80, I'll look for compression settings.

But I'm not likely to bother to tell you.

0 upvotes

Lossless compression is right. The phenomena of smaller files happends with less noise...

There are no compression options for raw files...

Edited 57 seconds after posting

0 upvotes

Finally! I'm glad not to have to change brands after 17 years of Canon. It's time to start saving to replace my old 7D by 7D3 and 6D by 5D IV.

3 upvotes

Finally Canon appears to get that its historic sensors have lagged current technology. Now just bring it to the professional APS line, the 7DII successor, to compete with the Nikon D500.

3 upvotes

I would agree that Canon will likely put this sensor into the new 7D3. Possibly a few other surprises as well to compete against D500. Put a rotating screen on 7D3 and a lot more video DSLR users would buy a 7D3, as well as high speed photo shooters, and keep both CF and SD like current 7D2. Will Canon put 4K in the next 7D3, who knows? Maybe, if price matches D500. They do not want to loose sales to 5D3 or higher end models. Maybe 5D4 will have 4K first?

Edited 48 seconds after posting

0 upvotes

Still about 1 stop behind the 2 years old Samsung NX1 :P

10 upvotes

Right, NX1 a camera so successful, it's already out of stock.

3 upvotes

NSA must have deleted MG's post, which made me write my post. May as well delete my post as well, since it doesn't make any sense anymore.

0 upvotes

I suggest you display four settings in the image comparison tool, ISO 100, 1/160 s, +5 EV

upper left: Canon EOS 80D, upper right: Nikon D7200
lower left: Canon EOS 70D, lower right: Nikon D7100

The images speak for themselves.

Edited 4 minutes after posting

8 upvotes

Absolutely. And now I'm even more satisfied with my D7200. It does make a difference when trying to recover detail out of the darker areas of a bird's feathers, especially when the environment is not a studio and light is often not perfect.

6 upvotes

For me the 80D looks better than D7100 and it's about 1 stop noisier than D7200 which is not too bad and probably hard to notice outside the lab.

Definitely something to be happy about.

The 80D has other advantages also, like the dual pixel, F8 AF points, great live view and the most important, the lens system.
Where else you can get a wide angle as good and as affordable as the 10-18?

14 upvotes

At four stops the D7100 clearly beats the 80D. At 5 stops they both have different issues.

1 upvote

At 4 stops you can already see some banding on the D7100 image and i prefer the little bit more noise of the 80D but no banding.

But really, none of this is really important for photography, we are talking about lab results here. There are 1000 other more important aspects which make a camera better.

I actually used to own the D7100 with 17-55 and i loved it, a great camera. But i prefer the Canon lens system.

13 upvotes

blackcoffee. I beg to differ on the importance of low noise floor in photography. And people keep comparing the 80D to the 7100...I had a mullet in the 90's, then I got a haircut...you know what I mean?
The 7100 is an old sensor, that's why they have the 7200 now, but that's fine keep comparing canon 2016 with nikon 2012. That's about where canon's at with sensor tech.

4 upvotes

photomedium: I said it's not the most important thing. We are talking about 1 stop difference when you push the shadows at low ISO and that's compared with the best APS-C sensor.

Someone else before me compared to D7100, i just pointed out that in my eyes that camera is not better(DR).

Why people like you are so butthurt about everything Canon releases like your life depends on it?
For you it's important the best low ISO DR then buy the best camera for that and for me it's more important the high ISO and the amazing tele lenses Canon has, i will buy that.

Everyone is happy.

4 upvotes

Id be VERY happy with no banding, and one stop under the 'best' apsc-c sensor performance.

Ive been doing great with canon's crap sensors for years...amazing, eh?

Long live the 80D (or whatever comes after it :))

What I really wish was a FF verison of this same camera. After my 5d3, im not sure how much i can commit to even a really good apsc. They are not that much smaller than an FF, which is the greatest factor.

Ah well, ill figure it out.

1 upvote

The new 1DX also has this kind of improvement, but of course that is top of the range.
The new 5D will probably also have the new sensor tech.

0 upvotes

Set all 4 cameras to push +3 and the difference is much less, however 70D noticeable worse. At 100 iso and +3 push, small difference between 80D, D7200 and D7100.
How many of us are pushing files past +3 regularly. Worth mentioning, as how important is pushing it further except the odd file that was exposed too far off.
The difference is indeed noticeable at +4 and +5 exposure push for noise in 80D, 70D or lines in D7100 sensor show.
Unless going to 100 iso and pushing +5 is used to expand Dynamic range and hold bright highlights as dpreview suggests in a bright contrasty scene for example on the Canon 80D, now possible compared to 70D which was far worse.

Edited 3 minutes after posting

0 upvotes

@Blackcoffee / Adrian Van,

Agree. +3EV, 80D has some noise, but black is black, noise is uniform and no color blotch. D7100 already has hints of banding, D7200 has smallest noise pattern but color is going magenta, A6000 noise is stronger and color blotchy. This 80D sensor is better than both D7100 and A6000, both at low ISO and high ISO. Different than the D7200 low ISO, not sure if worse per se, due to the D7200 color shift, D7200 better noise control at High ISO, but 80D holds its own in color retention. All those who remark that it is still behind a few years are way off the mark. It looks like a good sensor.

Edited 1 minute after posting

0 upvotes

Yup. There are horrible stripes in the Nikon 7100 image.

0 upvotes

Pathetic! Not the camera, but the way us men (I'll give the ladies credit for having more sense), can get so worked up over a product release to the point we trade insults with each other.

Ease up folks and enjoy what you have. Others would.

24 upvotes

Won yer a go eye cudnent evn spel cannone an now all eye dew ish camplain abut it.

2 upvotes

Have you ever seen them fight over a dress or a pair of shoes?

8 upvotes

Wow a lot of bitternes here folks. As we said before we can not please everybody. IMHO, DPR should give as sample DR using the latest DPP version. Coz I knew that everything will be a different story. This Camera is awesome for amateurs, and it will make them excel from their field. Canon in every way always deliver a greats tools for us. If you think this is not for you them simply look for another brand and enjoy clicking and pushing shadow to suit your needs. Have a nice day everyone.

17 upvotes

Well you know there are certain topics you never discuss in polite company: religion, politics and camera brands.

3 upvotes

Awesome for way more than amatuers.

Amateurs are NOT worried about DR...they are worried about taking a half decent picture.

4 upvotes

Richard Butler: You fundamentally misunderstand many things. The true reason for variable ISO is to accommodate a range of light intensity. The ISO setting establishes the mathematical mapping from analog voltage reading at a sensor element to quantized binary numeric value. But you write this, which is all nonsense except for the first phrase of the first sentence: "A camera with a very low noise floor is able to capture a large amount of dynamic range, since it adds very little noise to the detail captured in the shadow regions of the image. This has an interesting implication: it minimizes the need to amplify the sensor's signal in order to keep it above that noise floor (which is what ISO amplification conventionally does). This provides an alternate way of working in situations that would traditionally demand higher ISO settings." You are confused at a very fundamental level, not just here, but most of your explanations are jabberwocky.

3 upvotes

Wow, unbelievable. You should work as a writer.

0 upvotes

Obviously you don't know what an ISO-invariant sensor is. It's precisely that, the ability to do away with analog amplification and do only numerical amplification of the signal. It's the future of sensors, and it's already present in some cameras in more or less advanced forms. Canon lags behind because their noise floor precents numerical amplification and requires analog amplification.

So, to paraphrase your own words: you fundamentally misunderstand many things.

Edited 3 times; latest 3 minutes since posting

16 upvotes

To Kaiser, I suggest you read more and write less.

5 upvotes

keiser snooze,
you maybe a bit confused yourself.

2 upvotes

Kaizer, unfortunately for you, the misunderstanding is on your side. Please Google "ISO invariant sensor".

1 upvote

JABBERWOCKY - LOL

luv that

0 upvotes

Watching "The Usual Suspects" taught us to be suspicious of anything Kaiser Soze says...

0 upvotes

Its not Mr Butler who invented the term 'ISO invariance''. Looks like Its the new dude Rishi wrote most of it. He kinda write all sort of jargons no one can understand... but will look cool high tech stuff...😉..never mind..lol..a.hardly anyone can make out anything from those sentences.... now that is very common ..I remember good old days from 10 years ,this site used be a photographers site..aesthetics color. Now this has become a testing lab for sensors.

Edited 2 times; latest 12 minutes since posting

2 upvotes

Yeah, I mean it's not like sensors are used to make images or anything.

0 upvotes

I guess this puts to rest concerns from the minority of Canon users who plan to shoot and process with a +5EV exposure push in mind.

For everyone else, if you often find yourself making a -5EV metering error, keep in mind that the Auto mode should prevent it from happening again.

26 upvotes

Not if your scene exceeds 10 or 11 stops or so, which in certain environments isn't uncommon.

15 upvotes

Good point! The next time I lock my ISO at 100 and underexpose just for fun, I can feel safe in knowing I can finally recover the picture I completely screwed up.

I also love that I have countless Sony/Nikon/Fuji users to have online chats with while I process the aforementioned screwed up photos for hours on end. This is awesome!

7 upvotes

I wonder why those silly film manufacturing suckers ever bothered to make the 15-stops Vision-3 color negative film, it seems like a foolish stubbornness.

And you know, that silly Emmanuel Lubezki who won 3 cinematography oscars in a row, hellbent in shooting with that 18 stops of DR camera. I wish he knew how to meter a scene.

17 upvotes

@Don Sata.. comparing motion capture techniques and products to stills photography... okay...

Please tell me what Lubezki would have done if the scene had more DR than the camera could handle... the same thing someone with a lower DR camera would do right?

2 upvotes

@Don Sata
Good point, Don. People want to make excuses for DR not being important. It is important and the more the better. With digital the extra DR is in the shadows so underexposure is used. With film it is in the highlights so overexposure is used. Some criticise the colour shifting with digital underexposure and subsequent shadow pushing. Colours also shifted with overexposed film but that never stopped photogs from using the technique to capture more DR when needed. It is just another option,

With digital the response of the pixel to exposure is linear. What sensors need now to increase the highlight dynamic range over metered middle gray is a nonlinear pixel response whose sensitivity to photoelectric conversion decreases as the saturation point is approached. This would emulate the shoulder highlights that film had. Another solution would be to have a clock at each pixel that would record the pixel exposure time close to saturation.

Edited 2 times; latest 9 minutes since posting

7 upvotes

"It is important and the more the better." is not entirely true: the best sensor is the one that has the same DR as the scene you want to capture. If your sensor has more DR, it is not used to its full potential.

1 upvote

!?! StefanD, did you read what you wrote? :))

5 upvotes

@bernardly: Doesn't the nonlinear tone curve applied during RAW processing already emulate film as far as possible based on the sensor DR?

BTW, the MIT researches a "modulo camera" which will provide an "Unbounded High Dynamic Range":
http://web.media.mit.edu/~hangzhao/papers/moduloUHDR.pdf
I am very curious whether (and when) this will be implemented by a camera manufacturer.

2 upvotes

This comment only highlights photographic ignorance, nothing more.

5 upvotes

@bildsee
Thanks for the fascinating MIT link which I will read when I have some time.

Yes, the nonlinear tone curve applied to the linear data from the sensor does emulate the tonal response of film. But in terms of highlight exposure from the sensor it is limited. That's why I think it would be fantastic if sensor design could be revolutionized. I think it is only a matter of time before such new technologies become widely available. Probably in 10 years our current state of the art sensors, stacked and BSI designs, will seem very crude.

There's also the nanotechnology product by InVisage called QuantumFilm that holds a lot of promise.

Interestingly, smartphones with their huge demand for sensors which are also comparatively small are proving to be an incubator for new technology.

2 upvotes

Photokhan, lack of understanding tech aspects of sensor technology or whatever is not "photographic ignorance".

I am an engineer myself and photo hobbyist, and i also like the tech side, however i have a few artist friends, with some amazing photographic and editing skills, who would make some jaw dropping images without caring a bit of what the signal-to-noise ratio ever means, or how much DR at base ISO your camera has... those are photographers.

3 upvotes

There's always one smart-aleck who says "stop complaining about dynamic range, learn how to meter" but I always wonder if that person ever actually takes photos in high-DR conditions or ever feels like trying to achieve natural tonality and roll-off, or whether they just blast the contrast in post and pass off the limitations of their gear and technique as "their aesthetic choice..."

1) There is no disadvantage to having more DR.
2) Many scenes require heavy PP to avoid exceeding capture DR.

This example is about at the limitations of my M4/3 camera, but I underexposed as much as my sensor could tolerate pushing the shadows. While LR sliders don't directly correspond to stops, this is about a +2 EV push in the shadows, and a -1.5 EV pull on the highlights. So the metering is "wrong" and I needed to "rescue" the image, but it was done entirely intentionally.

http://i.imgur.com/nCKr8JN.jpg
http://i.imgur.com/YXr7TSd.png

HDR bracketing and Grad-NDs both have severe limitations.

Edited 2 times; latest 20 seconds since posting

4 upvotes

@Androole
Good examples!

1 upvote

@bildsee: Actually, no, the nonlinear curve applied during Raw conversion to JPEG does the opposite - it increases contrast, thereby decreasing the effective captured dynamic range, to basically deal with the limitations of our low dynamic range, dim monitors. It chooses 'pop' over global dynamic range.

The most successful photographers are able to do this conversion better themselves: keeping the global dynamic range but increasing contrast only in midtones, to introduce the 'pop' factor while not sacrificing global dynamic range. It'll be easier once we have brighter / higher DR monitors, where you won't even have to do this 'tonemapping' (in this case, mostly shadow pushing, & then reintroduction of blacks to maintain the perception of contrast).

In fact, on those monitors (some already here, but more coming in the future), you won't even have to do a +5EV shift to see the noise in noisy shadows of limited DR cameras exposed for the highlights.

2 upvotes

Rishi, good point with the DR of monitors, but still most of the photo exhibitions are still on printed paper, and the DR of paper is not going to change very soon.
But maybe we'll give up completely on printing at some point when we'll have integrated high-res displays inside our walls...

0 upvotes

Clearly you never shot an image with both sky and shadows in it at the same time.

0 upvotes

While the dynamic range of output media (be it low-DR monitors, HDR monitors, gelatin silver prints, or inkjet prints) is an interesting discussion, I feel that it is slightly different from the input / capture dynamic range at hand.

I've often heard people say "oh, there's no point in having more than 7 stops of DR because you'll never be able to print or display that much." But the purpose of having high capture dynamic range is so that you can gracefully compress all of those tones into the required output range. If the original scene has 18 stops of dynamic range, I can do a 5-shot bracketed capture, merge to a DNG in Lightroom, and then compress all of those tones into the visible 7-stop range so that it looks natural without immediately clipping to black or white. It obviously won't retain all the tones of the original because that's limited by the output, but it will look infinitely more natural than something that couldn't capture that DR in the first place and simply clipped.

Edited 31 seconds after posting

3 upvotes

57even, was your "smart" answer directed at my remark? If so, you maybe should read it again...

0 upvotes

badi, pretty sure that 57even's response was addressed at fmian who posted the initial comment.

0 upvotes

@Androole
Very good explanation on why high input DR is still very important for low output DR displays and prints. A+ :)

2 upvotes

@Badi

No, it was addressed to the OP. But the DR of paper is irrelevant - you can still map 14 EV input into 7 EV output if you control the mapping selectively. Many movies have scenes with more than that, usually shot on film because digital was quite limited until recently.

Just shooting a backlit subject could require 14EV to leave sufficient tonal range in the subject's features after adjustment.

0 upvotes

Exactly what Androole and 57even said. Effective artists 'tonemap' all that extra dynamic range into a lower DR output. The best ones are able to do it while still maintaining the impression of contrast (i.e. avoid a 'flat' look).

In the future, though, I doubt much of this tonemapping will even be necessary due to the advent and popularization of high brightness/contrast display technologies.

1 upvote

Is it coincidence that Canon has suddenly upped its DR game, about the same time Sony began manufacturing dual pixel af sensor for Samsung? Patent swap?

5 upvotes

You are not alone in that guess. I thought precisely the same thing, and since have found a couple mentions of the possibility. It would seem to make sense.

0 upvotes

Cool.

0 upvotes

Still so far behind Nikon that Canon engineers should be embarrassed to let people see the comparisons.

8 upvotes

Except Nikon uses Sony sensors, so this comparison is not apt - Canon is behind Sony

0 upvotes

For anyone who can't be bothered to read all the comments it goes like this:

A lot of people said something along the lines of "it's an improvement but still not on a par with Sony sensors which is disappointing"

And then the some other people said stuff like this:

My grandpa took photos with a potato and he managed to sell photos to magazines so what's the point in Sony sensors?

Sony sensors are only better if you push shadows 17EV so what's the point in having anything better than a Canon sensor?

I sell my photography for a living and I shoot Canon so the superiority of Sony's sensors is irrelevant!

Dynamic range is a FAD!

If you knew how to process photos properly the higher noise floor and smaller usable dynamic range of Canon sensors would magically vanish!

If you knew how to expose properly it would magically cancel out technical shortcomings in Canon sensors!

If you want the best sensor you're not an artist!

Whatever, Canon sell the most so they're still the best!

Edited 32 seconds after posting

65 upvotes

Wow @fatdeeman, you are really buttthurt aren't you? What happened, did some guy with a personality and a Canon steal your girl away so now you hate everything Canon?

Seriously dude, your posterior expression goes beyond just being a Sony fanboi so what's up?

14 upvotes

Actuallly @fatdeeman - I just love ironic humour. I just noticed your challenge entry. Your Sony image lost to a 7D image. Yes, your camera with so much better DR and pushable shadows lost to a camera with a sensor that was unanimously bashed for being noisy even at base ISO and don't even think about pushing shadows! So much for technicality eh?

Kinda makes your rant above seem like an 8 year old in a grocery store where the parents refuse to buy him some sugary cereal and he has a hissy fit.

27 upvotes

@Interestingness: No... Flippin... Way!! LMFAO! That's awesome!

2 upvotes

Congrats on your 8 Likes.
Let us know if you can barter
them for a cup of coffee.

0 upvotes

Nailed it. People are so predictible. Good job

2 upvotes

As the responses to fatdeman's amusing piece demonstrate, it's not just the latest technology that many Canon users are missing out on, it's a sense of humour.

18 upvotes

Love it. Thanks for the laugh :)

1 upvote

With that long list, fatdeeman still missed the point.

Its not that Canon sells the most so they're the best (although this does sound like a typical childish DPR comment.)

It's that Canon sells the most, despite not being the best.

And...with the arrival of the internet and super useful sites like DxO (joking; smiley face) chances are, most people know Canon is not the best.

2 upvotes

Talk about a sense of humour fail.

Sony fanboi? After all that stalking you failed to even check my gear list? I have WAY more money invested in Canon than I do Sony, that's the truly funny thing and I think that's what bothers some people because I have absolutely zero incentive to unjustly criticise Canon.

If I was going to have a hidden agenda surely it would be in Canon's favour?

And as for losing the challenge, what is your point exactly? I guess I should have used my Canon instead and then I would have won, right? Or maybe it's just a bit of fun and completely irrelevant to making a judgement on the technical qualities of an imaging sensor. Is that really the best you could come up with?

Edited 5 minutes after posting

7 upvotes

@fatdeeman... You have loads of gear invested in two or more brands but you seem to spend more time writing comments here than taking photos, enjoying your gear. Are you camera tester?

Edited 26 seconds after posting

3 upvotes

Zzzzzz oh look another 90 degree subject change.

3 upvotes

@AbrasiveReducer
'It's that Canon sells the most, despite not being the best.'

The best at what?

'And...with the arrival of the internet and super useful sites like DxO (joking; smiley face) chances are, most people know Canon is not the best.'

From what I can gather that people 'know' it's that DXO's measurement of one tiny aspect on a camera in an extremely limited situation shows that it's behind the pack.

It would probably be just as (if not more) relevant for me to say Canon is the 'best' because it has an 8-15mm circular fish-eye lens. Because no one else makes one like it.

2 upvotes

Good improvement Canon, although still lag behind for low ISO. This is the best all round camera for video, still and action/wildlife.
So far non of the camera is perfect.

0 upvotes

Canon rings in the New Year...well, er, the year 2011, at least.

10 upvotes

The Nikon D7100, to which the sensor mostly compares (actually the 80D sensor is a bit better), was issued in 2013.

;-)

1 upvote

I do not think/see this Canon sensor can beat 2011's SONY sensor which is NEX7.

Poor DR in low ISO is Canon's signature feature for years, so does this one.

1 upvote

The D7000 (2010) is a much better comparison, being made by Sony, and ISO invariant. The D7100's sensor was a failed attempt at a successor by Toshiba, though they completely passed with full extra credit on the D7200's sensor. :)

The NEX-7's sensor was kind of meh, getting beaten by the NEX-6's, which is probably why Nikon went over to Toshiba instead of sticking with Sony. Sony did make up for it with the D5300 and D5500 sensors, though they were still well behind Toshiba's second attempt.

0 upvotes

Compared to any Canon, a comparable Nikon is such a boring choice.

2 upvotes

Well, the D7000 is 6 years old...

0 upvotes

I'm all for boring choices. They tend to do what you expect of them, don't get in the way and don't require lengthy excuses.

0 upvotes

For me - lets not forget that due to DualPixel design we actualy compare 48 Mp 80D sensor with very tiny photodiodes with competitors, thus gap in performance looks even smaller.
Still, what about ability for exposure correction in M autoISO mode... Feature, which absence is ridiculous in my opinion.

0 upvotes

You can use your AEB tool.

0 upvotes

No, I cant, for example, in video. And again, AEB have another purpose, I just want add constant bias. This feature is common on many cheaper cameras, and is purely software. So it can be implemented even through firmware update. And many others things too...

1 upvote

You can apply exposure compensation in shutter + aperture priority mode (M mode with auto ISO). P. 197 in the manual.
Too bad, it doesn't have the 7DII's lever though and you'll have to customise the SET button for that.

2 upvotes

Good point, thanks. I've missed the fact, that manual exists already. It'll be interesting to read. Lets hope that above is true for video also. Now waiting for full review.

0 upvotes

this is the ultimate proof that Canon users were wrong all along

8 upvotes

This is the ultimate proof that a very small minority of Canon users were wrong all along.

*there, fixed that for ya

3 upvotes

I thought it was about taking photos, and not you being right while everybody else was wrong.

7 upvotes

No, it is the ultimate proof that it was worth to stay with Canon. I tried Nikon and Sony (had Nikon, borrowed Sony). Personal taste... but for me Canon is a tool that does not stand in my way. The need to use proper ISO was and is still better for me than missing or screwing up the shot.
Sensor is just part of the package, and most people tend to focus on it too much, just because it is measurable/comparable numerically.

7 upvotes

in the beginning most Canon users didn't want to believe it and blame dxomark of favoritism. But the last few years, I think most Canon users admitted to it (the eyes don't lie), and asked canon to improve their dr. So both statement above are correct, just when it happened.

5 upvotes

@Spectro - I'm an even different Canon user. I acknowledged that other than Canon's current sensors (6D, 1DX, 20-24 MP APSC), their sensors of the past also exhibited banding to greater or lesser degrees. Usually greater. This banding made the usability in demanding situations even less than DXO would indicate with their scores, because banding wasn't taken into account.

6 upvotes

Sure, keep being a fanboy trolling for attention by stoking brand war fires. Real photographers were itching for better DR but know that good skill gives good results 98% of the time. Not worth switching systems for that extra 2%. Anyway, you'd better get back to shooting your crappy "painting-like" HDR crap.

2 upvotes

@Jonathan, banding do happen if you push pull and do heavy editing, but isn't as bad on the newer sensor. I remember somebody uploading a d700 file with heavy banding of the sky, we told him to get a new camera.

0 upvotes

The D7200 is even more impressive. At this rate of improvement, the D500 should be a DR and high ISO beast next to Canon and the rest of the APS-C competition. Fuji being a close 2nd place. No doubt in the mirrorless arena Fuji is top dog for APS-C!

8 upvotes

Agreed, when you do a comparison with the D7200, +1EV on the 80D is approx. the same as +3EV on the D7200. Nikon still smokes it BIG TIME!
Matter of fact, even the 6D is not as clean as the D7200.

Edited 2 minutes after posting

6 upvotes

The dual pixel liveview full time AF smokes the contrast detect liveview of D7200 if smooth video focusing is your thing. If only photos, not video, are your thing the D7200 rocks. For being quite good in both photo and video, 80D presents a compelling argument to excel at both in single camera body. In most types of general uses and with fast lenses 80D will be very good.
Canon's new 18-135mm lens is far better than previous lens with better glass design optics inside and likely outperforms Nikons 18-140mm lens for sharpness also useful for video or stills in moderate walkaround zoom.

Edited 2 times; latest 4 minutes since posting

4 upvotes

Slow down there cowboy, we don't even know what the video quality is like yet, if it's anything like previous models the resolution could be disappointing, despite canon bodies being capable of stunning RAW video through magic lantern they tend to prevent their DSLRs from threatening cinema EOS sales by restricting the quality.

Some people don't use autofocus for video at all but it will be interesting to see how dual pixel is coming along with the 80D. The Sony A6300 has shown astonishing promise in terms of video AF, fast and accurate but most importantly SMOOTH and with the mirrorless knack of tracking exactly what you ask it to. If the Canon can do the same then we may see autofocus becoming more widespread even for more serious use.

0 upvotes

The issue up until now has predominantly been the fact that camera doesn't know the operators intent which can lead to focus not being where it's needed which of course looks unattractive and unprofessional. Cameras are now getting to the stage where they are able to stick to specific patterns and shapes with greater tenacity and could suddenly improve upon manual focusing in some situations.

0 upvotes

Who ever made sensor for D7200 is really good. I know it's not Sony because it smokes all the Sony crop sensors...

0 upvotes

For run and gun video like fast moving pace wedding and event video having phase detect on sensor is far better than contrast detect alone. Manual focusing has its pros and cons, however both the 70D and 80D and 7D2 are built for video in motion where people are moving quickly such as dancing at the reception. Sony is also, with A77ii and also FF A7 series mirrorless. D7200 while it has decent HD video, probably is not the first choice of a serious part time videographer using DSLR on a budget. 80D and 70D would be in apsc and for FF, lots of pros like Canon 5D3 or Sony A7 mirrorless. Nikon cameras chosen for serious pro video are D810 or D750 and D500 because of video features and video quality. D7200 can do video, great for hobbyists and advanced consumers, but not usually first choice among serious users, as other cameras offer a lot more video features in camera, like the ones I mention. D7200 excels in apsc image quality and DR for photos, although manual focus video is possible.

1 upvote

@ LO

You MUST be kidding.

7200 cleaner than a 6d...ahahha...that's hilarious.

0 upvotes

@EDWARD 6D is very unclean very noisy in one to one comparison with D7200 . 6D is noisier than 80D too. While 80D is superb improvement, it looses against D7200 by some margin.

1 upvote

The A6300 will be very close to the D7200 (Toshiba) in mid/high ISO, though the D7200 beats it in low ISO DR due to the nature of the sensor.

The D500's characteristics are very much up in the air. If it follows the D5's lead, it will be quite flat in DR at low ISO, in order to exchange for extended DR and lower noise at high ISO. The A6300 reacts in a similar fashion (via Aptina's dual gain conversion), but nowhere near as extreme since the architecture is different.

0 upvotes

@ EDWARD ARTISTE Please explain how it's hilarious, all you have to do is select D7200 and 6D, at 0 through +2EV, they are about comparable, with the D7200 just slightly cleaner. At +3EV and above the D7200 is much cleaner than the 80D and the 6D. I think you are mistaking this for a HIGH ISO test, when this is a DR test. Get your info straight before posting.

2 upvotes

Here's how I have my info straight:

Using the 6D in multiple countries. Actually owning one.

Never worrying about iso - the performance in that regard is spectacular.

Thinking that the 7200's image quality will beat the 6D- LOL. Even with these lab exsposures, do you REALLY think the 7200's IQ will be better?

Are you even taking into account the MP differences, which will directly affect the grain?

Man, you guys are really stretching this iso argument.

How about you guys show it in your own pics....yeah, I thought so.

0 upvotes

@EDWARD,
What ever you wrote is NA for the those who you think as recipients. Your writings are applicable to you. You have to lol at yourself and mock yourself.

0 upvotes

@jay

With all respect due, hey, if I think some piece of information is off base, it should be noted. With that said, I actually own the camera. You guys are looking at lab exposures on the web. The data is hugely useful for sure, but does not tell the whole story.

Im not a huge fan of the camera itself, with its pathetic AF system - however the iso and IQ (when it focuses properly) is spectacular.

The iso performance is so good in fact, its the only reason i still own the camera. For landscapes, its invaluable. Its the first camera body Ive used that I never worried about iso, because it always looked good.

And really, if a m43 sensor gets better iso than a 6D, what does it matter? You STILL wont get better overall IQ.

If you guys want to go around saying you'll get better pictures from the 7200 than a 6D, by all means, knock yourself out. But sometimes its actually good to get info from people who have actually owned and used said hardware, for whatever that's worth :)

1 upvote

Much better than before... thank God. Too bad I already gladly dumped my canon gear for superior Nikon/Sony sensors. I do shoot a lot at low ISO from a tripod so the swap was incredible for me.

If I were still using Canon I would be pleased (but extremely annoyed it took this long and who knows how long they'll be stuck here). Plus... you now have to buy another camera just to be only slightly behind the competition (on this one metric alone)

8 upvotes

You shouldn't change your gear according to what gearhead fanboys says. They have nothing to do with photography but sit in front of a screen gazing at pixels nothing else. If you constantly need to to push shadows +5-6 stops fine. Do you? What about lenses, usability, ergonomics, view finders, colours the list goes on and on.

4 upvotes

Maybe he just likes clean low ISO files.

5 upvotes

Then any 1'' compact will do fatdeeman

3 upvotes

I would also think that if youre pushing iso 3+ stops regularly, your probably doing it wrong.

2 upvotes

Or maybe just doing something they CAN.

4 upvotes

Why push 3 stops when you can have white skies!

Said nobody ever. -____-

1 upvote

@tags how old are you? Hope you asked your dad first to dump his gear ;)

0 upvotes

Interesting results when you select the Olympus M5ii to compare.

0 upvotes

That is because you see +0EV comparison in olympus, while canon is set to +5EV. Set both to same.

Edited 25 seconds after posting

1 upvote

I was debating on coming back to a large non-mirrorless camera like a k-3 or 80D,,, why would I ever now my em5 mkII destroys it in low iso, it's not even close. Can't wait to see what the Omd-Em1 mk2 brings.

0 upvotes

Both set to +6EV. Results can't be right! If they are correct the Olympus is impressive.

0 upvotes

They might've improved the low ISO performance slightly but at 3200 it's actually worse than Canon 760D and Nikon D5500.

http://www.dpreview.com/reviews/image-comparison/fullscreen?attr134_0=canon_eos80d&attr134_1=canon_eos760d&attr134_2=nikon_d7200&attr134_3=nikon_d7100&attr134_4=canon_eos7dii&attr134_5=nikon_d5500&attr136_0=18&attr136_1=18&attr136_2=18&attr136_3=18&attr136_4=18&attr136_5=18&normalization=full&widget=328&x=0.150140509&y=0.5334473

0 upvotes

Strange, it clearly outclasses the D7100 at ISO6400. http://www.dpreview.com/forums/post/57483256

1 upvote

When you compare it to a Nikon 7200 well guess what it still is not as good. Nice try Canon. Better than before, but still behind.

I have been a Canon user for years and years, but 3 years ago I just gave up on their new stuff. They are living in the past. And still are.

They make darn good lenses, better than Nikon, and have more of what you need but...

Edited 1 minute after posting

11 upvotes

.. but the small difference in DR performance still renders their entire camera system utterly useless despite having amazing lenses and more of what you need.

20 upvotes

Agreed, Canon is just not on par with Nikon in terms of DR. 0 to +3EV, even the Nikon D7100 is cleaner than the 80D.

4 upvotes

''even the Nikon D7100 is cleaner than the 80D.'' Where did you get that, not true at all.

http://www.dpreview.com/forums/post/57483256

0 upvotes

@ImageAmateur ummm...the DR results that this post presents!!! This is a DR conversation. IT IS NOT a high ISO convo! Do your homework before posting! Better yet educate yourself as to what the post is about.

Edited 7 minutes after posting

1 upvote

that really isnt true. both systems have their strengths and weakness. while canon has better lenses on some focal lengths, nikon has on other... and even then, the difference is really really small. you can clearly see that here:
http://www.dxomark.com/lenses/brand-canon-nikon/launched-between-1987-and-2016/mount_type-Canon_EF-Nikon_F_FX#hideAdvancedOptions=false&viewMode=list&yDataType=global

0 upvotes

Anything DXO Mark is immediately dismissed!

0 upvotes

yeah yeah.. i know, conspiracy theories on the internet and such.. but the truth is, dxomark is a really good comparison tool...

1 upvote

Scientifically, I completely agree, but everything is ultimately judged visually, not by graphs. So, whatever looks sharper is better not what ever graphs better. I watched so many YouTube videos trying to convince everyone that DXO Mark is the end all be all, but I don't make purchases by what a graph says. I make them from the results I get from the equipment, as I assume you do too. That's all I'm saying.

0 upvotes

well my real world findings are more or less comparable to dxomark. so it is a good start to go there for a fast comparison. that said - i do have a niche in photography that dxomark doesnt cover and to that i do refer to other websites. but for most users - dxomark is good enough

1 upvote

The ITCH buddy, it's called the itch..

0 upvotes

To me great DR is one of the last things missing on Digital cameras yet.

We have pretty good high ISO on most. If you can shoot at ISO 6400 you got most stuff covered. Banding seems to be a thing of the past. Just about every company now makes a camera that is darn good at picture making.

But Only Nikon seems to have really figured out how to do High DR in their cameras well. And that is the difference between getting the shot you Have to have to put food on the table.

And I don't care what anyone says Canon has still not figured it out, and will be 2 to 3 generations behind Nikon.

I want a camera that takes what I see in the viewfinder, not 2/3 of it. Nikon and Sony are top dog in it and they are getting my money, not Canon. I am sure if you could take all your Canon lenses and put them on a Nikon body with no penalty, Canon would would sell one heck of a lot less cameras..

0 upvotes

As a rat shooter I can't wait to try this in my cave !

2 upvotes

Suggestion for DPR Challenge: Low Light Rat Capture
Shoot in very dark environment, room, or cave. Use low ISO and underexpose. The rear LCD monitor should be a dark frame. Then brighten the shadows in post with a 6 EV push. If a rat is revealed submit the shot to the challenge.

2 upvotes

Total comments: 843

www.dpreview.com

Editorial content

Product database

Community

About us Work for us Advertise with us FAQ Feedback Interest-Based Ads Privacy Legal

All content, design, and layout are Copyright © 1998 - 2016 Digital Photography Review All Rights Reserved. Reproduction in whole or part in any form or medium without specific written permission is prohibited.

Mobile site


The Canon that can: Canon EOS 80D Review Apr 27, 2016	Upwardly mobile: Sony a6300 Review Apr 6, 2016	Retro through-and-through: Fujifilm X-Pro2 Review Mar 14, 2016	Taking it easy: Canon EOS 80D shooting experience Mar 24, 2016

Body Only, Base In stock Usually ships in 24 hours	From $1,199.00 »
18-135mm, Base In stock Usually ships in 24 hours	From $1,599.00 »
18-55mm, Base In stock Usually ships in 24 hours	From $1,349.00 »


Medina of Meknes by cand1d from Streets #1 : Streets in colour	Cain square by jonkoi from Dog challenge - Crop photo to a SQUARE	Cohen by richiebee from Light It Up!	White Muscovy Duck by RWWatkins from Abstract Animal Photo

62	Still not enough lenses for FF Sony?
60	Post Images of May 9, 2016 Mercury Transit Here
57	PART 2 This Week Through your m4/3 2016.05.07
53	Your Very First Digital Camera
49	Weekly X Series Photo Thread 8 May - 14 May
47	DPR has a unpleasantly high ratio of quotes to contents.
45	*Weekly Wildlife Thread: May 8th - 14th*
43	One thing I find interesting about the MFT format.
42	D500 Airshow! Sharpness? Pft.. (some other concerns though...)
38	I need wide angle 4/3 lens advice for Panasonic and Olympus lenses on Olympus M5 Mark II camera
35	X-Trans color gamut
34	Best Buy has D500 bodies in stock.
30	my portraits are too clinical and sharp
30	Mothers Day - Indoors Living Room Shots - Auto Setting a Flop?
30	DX f1.8 vs FX f2.8
29	I'm new to 4/3 cameras. I find Olympus lens names confusing.
27	Comparison Tests of Leica 100-400, Fuji 100-400, Olympus 300 & Nikon 80-400 VRll
27	Windows 10 controversy
26	NX-KS Mod pack - installation
26	X70 - stumbling behavior