Previous news story    Next news story

Canon EOS 5D Mark III low-light ISO series samples

By dpreview staff on Mar 2, 2012 at 14:07 GMT

We've had a chance to prepare a set of full-resolution studio-based JPEG samples from a pre-production Canon EOS 5D Mark III, at all available ISOs from 50 to 102,800. We've chosen to shoot under low level halogen lighting, which provides a stern test of the camera's sensor and noise reduction systems. Note that these samples are not intended to replace, or even be comparable to our standard studio tests, which we'll publish as usual when we get our hands on a production camera.

At a colour temperature of about 3000K and low light intensity (approx 3 EV), these images are also neutrally white-balanced. This requires extreme amplification of the blue channel (and to a lesser extent the green) relative to the red, accentuating any noise or banding that might be introduced by the camera. This should therefore be considered close to the worst-case scenario - under many situations you'll see less noise at high ISOs, for example when shooting indoors in daylight. Also remember that this is a pre-production camera, so may not be entirely representative of final image quality. 

The test scene used for these samples; focus is on the figurine lower left. The lens used was an EF 100mm f/2.8 Macro USM at F11 - due to the close distance (approx 1.5m) depth of field is limited.

Comments

Total comments: 344
123
Alejandro del Pielago
By Alejandro del Pielago (Mar 2, 2012)

Nice job, DPR !!!! No doubt everybody hopes the deep review... But meanwhile, that 3200 image could be very sexy..., let´s see the RAWs !!! Regards from Lima.

0 upvotes
marinabay
By marinabay (Mar 2, 2012)

where are the 100% crops?

0 upvotes
Dylthedog
By Dylthedog (Mar 2, 2012)

You can view full size if you scroll down

0 upvotes
focuscz
By focuscz (Mar 2, 2012)

I have found another noise test CANON EOS 5D MK III http://www.canonklub.cz/clanky/prvni-test-eos-5d-mk-iii-na-sum
The first noise test of CANON EOS 5D MK III See our fan site

See facebook fan page CanonClub, https://www.facebook.com/CanonClub.cz

0 upvotes
seta666
By seta666 (Mar 2, 2012)

better noise examples than here at Dpreview; but still not raw
http://www.canonklub.cz/sites/default/files/recenze/EOS5DMKIII/final-562x872+1132+844.jpg
I still think that ISO advantage over the 5D mkII will be bellow 1 EV

Comment edited 2 minutes after posting
2 upvotes
marike6
By marike6 (Mar 2, 2012)

I don't think to samples you linked to or the ones here on DPR are doing Canon any favors. Not the performance I expected at all.

2 upvotes
Sdaniella
By Sdaniella (Mar 2, 2012)

thx 4 url share:

ISOs up to 1600 not a problem

ISO 3200 starts to look slightly different (as expected now that we've seen a few examples)
ISO 6400 noise definitely creeping in, but can be usable

ISO 12800 noise is obvious everywhere but much better than on the PowerShot G1X (which is comparable to 60D/7D)

and from 25600-102400 noise escalates to unusable, even for me, i still don't like the 25600.

but far better than it was on the 5D2.

i used to use ISO 800 as my favorite on my 5D2 for most shooting conditions, day or night, outdoor or indoors, now, ISO 1600 looks clean enough to use all the time instead... that's a clear +1 EV difference improvement.

at the moment, for my 5D2, i still don't like to venture much beyond ISO1600 2000/2500/3200 is just not clean

i used to max out at ISO 6400 as 'tolerable' on my 5D2
but i don't think going as high as 12800 could be my new 'tolerable' max if i had the 5D3, but maybe lower, say ISO 8,000 to 10,000 (have yet to see, to be sure)

Comment edited 2 times, last edit 10 minutes after posting
1 upvote
Hannu108
By Hannu108 (Mar 2, 2012)

Thanks for posting, focuscz!

The ISO crop series confirm what wasn't so easily seen in DPReviews's test shots: up to ISO 3200 everything well. At ISO 6400 Noise increases alarmingly despite of NR!

DPReviews should have used object with finer details, i.e., hair of bristle in the test. (The stuffed rat is in out-of-focus area!) This would have revealed the noise issue. The test is more promotional rather that scientific & objective :(

This is harms Canon users, too.

0 upvotes
ageha
By ageha (Mar 2, 2012)

Looks good to me.

1 upvote
Edmond Leung
By Edmond Leung (Mar 2, 2012)

Excellent control in noise reduction at extreme high ISO.
Good Job!

1 upvote
harold1968
By harold1968 (Mar 2, 2012)

that 102400 sample is insanely good

1 upvote
cameragirl12
By cameragirl12 (Mar 2, 2012)

just curious on that last shot....the camera details say ISO 65535 but the title of the photo says 102400. Are the camera details automatically from the metadata?

0 upvotes
Andy Westlake
By Andy Westlake (Mar 2, 2012)

Yes, the gallery reads from the files metadata and that's how it's recorded in the EXIF. I'll correct it manually.

0 upvotes
cameragirl12
By cameragirl12 (Mar 3, 2012)

I'm confused...if the EXIF originally said 65535 then was that the actual ISO shot? I'm assuming that the title of the image was typed manually but the camera details were a dump from the camera. So was that shot actually 65535? Why would you change the auto data from the camera? just curious! :) the iso performance overall is pretty awesome. it would be cool to replicate it in candle light which is what a lot of low light event coverage needs.

0 upvotes
mr moonlight
By mr moonlight (Mar 2, 2012)

A bit too much NR, but still you can sort of get an idea of what the RAW files will look like without all the processing if you consider their default settings for their other cameras. It looks pretty impressive. From these pics it looks like ISO 12800 may have a slight edge over ISO 6400 on the 5DII, but we really need to see the RAW files to get a definitive answer. In the end I was happy with what I've been getting from my 5Dii so any improvement is just a bonus.

1 upvote
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 2, 2012)

First tests, not useful for focusing problems and using JPEG, but still, a good effort from DPREVIEW. That said, what we are seeing with the release of the 5D3, is that Canon is way behind in sensor technology nowadays. They couldn't release a 650D for the same problem, and we can speculate that the results we see in this test are due to processing and not the quality of the sensor.

Canon... wake up... you cant ask such prices without underlying value, and obsolete sensors is not the way to keep customers.

6 upvotes
Gothmoth
By Gothmoth (Mar 2, 2012)

well become a professionell and i may take your opinions serious. but then... no.. not when you write such nonsense.

4 upvotes
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 2, 2012)

lol right.. first, you don't know what I do or my expertise level, and second, give a counter argument instead of babbling, that way I can take you seriously

4 upvotes
Gothmoth
By Gothmoth (Mar 2, 2012)

well you are not an pro photographer that´s for sure....

three weeks ago you did not know what the auto lighting optimizer is.... now you talk about sensor technology as if you know something about it.

Comment edited 3 times, last edit 4 minutes after posting
3 upvotes
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 2, 2012)

oh, an Ad-Hominem, good, that's an old logical fallacy used when you don't have real arguments. Interesting. Yes, it was my first time dealing with that technology, and so I ask and make my own tests to see how it works, its called learning. Now, what can you tell us about sensor technology? why are you so averse to Nikon's? are you other than a Canon fanboy? please, explain.

2 upvotes
kbozen
By kbozen (Mar 2, 2012)

Bodhi Dharma Zen, since this is a new sensor, I'm very very interested how you can judge form a down-scaled (and in-camera noise-reduced)Jpeg the quality of the sensor.

0 upvotes
Xellz
By Xellz (Mar 2, 2012)

It looks more like you're Nikon fanboy. Why you think Canon is behind on sensor technology? Because it has lower res than D800? This time canon and nikon switched places. Before canon had high res body, but lower high iso perfomance, now it's opposite. D800 native iso 100-6400, extendable to 25.600 compared to 5d III ISO 100-25600 native, 50-102,800 expanded

Tell me again, why they are behind? You totally don't make sense, how you can make such remarks on sensors when camera was only released and there are no proper tests on raw, not to mention you didn't even handle this camera yourself.

0 upvotes
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 2, 2012)

there are other examples appearing on the Internet as we speak... take a look at them.. its the same technology they have been using for a while, they don't have any really new sensors to offer, currently Sony is ahead of them in terms of high ISO, Dynamic Range and others, check DxO for instance, Nikon is using Sony sensors and look how far they can go, more than 1 stop in DR and better high ISO performance...

2 upvotes
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 2, 2012)

Xellz, NOT AT ALL, I'm a user, not fanboy of ANY brand. I have been using Canon for years because, in the past, they used to have the best sensors out there, sometimes way better than any other brand. But now, they depend on their DIGIC processing to make JPEGs look nice... no, I will not change to Nikon, I'm invested with Canon as I have some lenses, so, it is just a big disappointment.

Comment edited 1 minute after posting
2 upvotes
kbozen
By kbozen (Mar 2, 2012)

you still not explained how you can deduce by jpegs from the net to the sensor architecture quality ...

0 upvotes
Alejandro del Pielago
By Alejandro del Pielago (Mar 2, 2012)

Dear Mr. Bodhi Dharma Zen:

Sure I will agree with you, but you may develope technical arguments or show evidence.

You just point the samples and give them some adjetives ( "what we are seeing with the release of the 5D3, is that Canon is way behind in sensor technology nowadays"). Really those samples are not enough like you know and DPR warns; but, then those images are not enough to say so severe critics.

Best regards.

0 upvotes
Henry M. Hertz
By Henry M. Hertz (Mar 2, 2012)

he canß´t he is a noob with a 1100D.....

0 upvotes
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 2, 2012)

Alejandro, if you agree, why would you need technical arguments? Still, the answers are there, look at DxO graphs, do a comparison regarding Dynamic Range and high ISO performance, compare Canon (specially mid and lower ranges) and you have the appropriate facts to state that current Canon sensors are mediocre.

http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Ratings/(metrics)/

And here, a concrete example (and for me a fact) about the quality of ISO 25600, you tell me about if it is usable or there is any night and day difference with actual Canon sensor's performance:

http://cpn.canon-europe.com/files/education/technical/inside_canon_eos_5d_mark_iii/11_cinc_big.jpg

Henry, learn to argue first, ad-hominem is an old logical fallacy used by people unable to state coherent and logical arguments. Read about it in Wikipedia, I'm sure you will find it interesting.

2 upvotes
Xellz
By Xellz (Mar 2, 2012)

I will agree on DxO when you actually show me, where is there 5d Mark III? Why you bring it up, when there is totally no data on this sensor performs? It's simply your opinion without any actual information and evidence. That's all.

0 upvotes
Josh152
By Josh152 (Mar 2, 2012)

Unless canon has started using Sony sensors there is no reason to think the 5D3 DxO results will suddenly beat the Sony/Nikon sensors which also have been improving.

Canon sensors have been lagging behind Sony's for years when it comes to DR and Noise. At this point Canon would have to completely change the way they design and manufacture their sensors to bring them up to Sony standards.

The only thing I have seen an any 5D3 high ISO samples is lots of NR smearing. So much that you can't even use them to accurately predict the High ISO performance of the camera. Why does Canon feel the need to be so heavy handed with the NR? I doubt it is because they have finally beaten Nikon at high ISO and DR.

2 upvotes
nathanleebush
By nathanleebush (Mar 2, 2012)

It's been pretty well documented by DxO Mark (though of course that's not the end all be all) that Sony's high end sensors outperform Canon's in most areas.

2 upvotes
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 2, 2012)

Exactly Josh152 and nathanleebush, that's exactly the problem with the new Canon... I imagine they are working like crazy in new sensors, that's probably the reason they didn't presented the successor for the T3i, they didn't have anything new to show, they couldn't stick the same sensor as the 5D3 and so they are (IMO) in trouble. Sure, the 5D3 is a great camera, yes, there are other factors to consider than the quality of sensor alone, no, I wont sell all my gear and buy Nikon or Sony cameras... but... we users should notice they reached the limit of their current technology, and instead of applaud them for asking $3,500 for an obsolete sensor, keep them focused in that customers demand no less than the best.

2 upvotes
Jexmark
By Jexmark (Mar 4, 2012)

LOL @ Bodhi. You can NOT judge a new sensor by looking at OLD sensor statistics. Until we have a bunch of RAW-files taken under different conditions the truth is you have NO CLUE what the hell you are talking about and your mindless blabbing about old sensor etc etc is just opinion and has NOTHING to do with facts.

And guess what? Most real photographers doesn't give a rats ass about DxO Marks or any other lab measuring sites for that matter. At least not anyone I know that makes a living with their camera. None.

0 upvotes
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 5, 2012)

I wonder where the ad hominem "critics" are now... so fast to attempt to castrate. First, look for the facts, analyze them, escape your bias, and then give an opinion.. but do not judge others, that's one of the MAIN problems with people in general.

Canon desperately need new sensors, I bet this is they main concern... what we public can see are the efforts of their millionaire marketing efforts, and marketeers are speciality is to sell, not to inform, they will make all sort of abracadabra's just to push something to the market, even when the heads of the company know their product is, at best, competitive, but certainly not better than the competition.

1 upvote
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 5, 2012)

Jexmark, extrapolating from known data is how one can make hypothesis with know facts (the pictures that are getting to the surface, that they didn't increment resolution, and so on). Now regarding the "professional photographers" you mention, you are right... the fact that someone earns his living from taking pictures of weddings says NOTHING about their understanding about the complexity on the field, , still, when using Canon, all of them are suffering from about 2 stop LESS Dynamic Range than other photographers using other gear...

In the end, all my ramblings are because I'm loosing those 2 stops, and I don't like having a mediocre product in my hands when I work hard to buy the best I can, I expect a company (any company) to GIVE ME ABSOLUTELY THE BEST THAT IT IS TECHNOLOGICALLY ACHIEVABLE, I demand that and no less. But big companies are concerned about costs, about making their investors happy, not about the general public...

0 upvotes
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 5, 2012)

for those who don't have a clue on what DR is and how badly it affect us (Canon user's) take a look at this:

http://testcams.com/blog/2011/05/03/nikon-dx-vs-canon-aps-c-dynamic-range/

0 upvotes
Bodhi Dharma Zen
By Bodhi Dharma Zen (Mar 5, 2012)

for those who don't have a clue on what DR is and how badly it affect us (Canon user's) take a look at this:

http://testcams.com/blog/2011/05/03/nikon-dx-vs-canon-aps-c-dynamic-range/

0 upvotes
Gothmoth
By Gothmoth (Mar 2, 2012)

for photographer (i guess we have a few here) not pixelpeeper:

http://blog.jeffascough.com/photographers/2012/03/canon-eos-5d-mark-iii-review.html

2 upvotes
sandy b
By sandy b (Mar 2, 2012)

Paid canon user

2 upvotes
Henry M. Hertz
By Henry M. Hertz (Mar 2, 2012)

unlike joe mcnally you mean? who has written about the D4...

i trust that professionells tell the truth.... unlike noobs on forums who don´t have the money to buy the gear they whining about.

Comment edited 30 seconds after posting
1 upvote
sandy b
By sandy b (Mar 2, 2012)

Really? best camera ever?? Real pro talk there. have you seen the samples he took? Best ever?? 2 stop improvement? what a laugh.

1 upvote
Henry M. Hertz
By Henry M. Hertz (Mar 2, 2012)

well i look at your images and i see only crap.... now what?

he is a explorer of light and you are a loser.. i believe him.

Comment edited 14 seconds after posting
3 upvotes
Vidar NM
By Vidar NM (Mar 2, 2012)

The comment stream. What a fabulous tool for people to interact and participate in discussions. Looking at skilled people revealing their insight and politely exchanging viewpoints is just wonderful.

6 upvotes
AbrasiveReducer
By AbrasiveReducer (Mar 2, 2012)

I'm afraid you're in the wrong place. Everyone here is a "professional" and this informs their comments.

3 upvotes
tkpenalty
By tkpenalty (Mar 2, 2012)

These JPEG samples are very clean, however for a full frame camera I'm pretty sure you're supposed to get more detail than this; theres way too much processing. However colour rendition at higher sensitivities is a very strong point; theres little colour shift, although there is a bit of dynamic range loss.

I'd wait for RAW before making any judgement.

1 upvote
Josh152
By Josh152 (Mar 2, 2012)

Considering that there is very little in focus fine detail and that there appears to be heavy NR applied to what little fine detail there is, it is impossible to judge the iso performance of the camera from these images.

Comment edited 1 minute after posting
1 upvote
Jack A. Zucker
By Jack A. Zucker (Mar 2, 2012)

those 6400 samples look horrible

2 upvotes
Sdaniella
By Sdaniella (Mar 2, 2012)

perhaps...
flip side of losing fine details on alternative models
dpr://galleries/8134818847/photos/1779165

1 upvote
Robert Hoy
By Robert Hoy (Mar 2, 2012)

I think you've been bashing this camera all morning, why spend so much time on something you don't like? That isn't normal.

3 upvotes
Gothmoth
By Gothmoth (Mar 2, 2012)

well jobless nikon fanboys.. what shall they do with their time? ;)

3 upvotes
Guy Swarbrick
By Guy Swarbrick (Mar 2, 2012)

Shoot clean high ISO images that don't have all the detail smudged out of them with NR, perhaps? Shoot 36MP images with no AA filter? Lots of things for Nikon fanboys to keep themselves busy with...

1 upvote
Jack A. Zucker
By Jack A. Zucker (Mar 2, 2012)

i haven't been bashing the camera at all. I'd love to buy one.

0 upvotes
tiklid
By tiklid (Mar 2, 2012)

ISO800: http://web.canon.jp/imaging/eosd/samples/eos5dmk3/downloads/01.jpag
ISO6400:
http://web.canon.jp/imaging/eosd/samples/eos5dmk3/downloads/02.jpg

Chroma Noise visible over the snow at ISO 800.

2 upvotes
Sdaniella
By Sdaniella (Mar 2, 2012)

ISO 800 is an 8" shot (35mm f/2.8)
ISO 6400 is a 0.6" shot (24mm f2.8)

the ISO 800 shot is quite bright, the entire landscape, both snow mountains in the background as well as the water in the foreground is reflective of a lot of the Aurora colors above it... as well as in the skies in between brightest wisps of Auroral plasma, so, how do you separate Auroral glint from the snow mid-ground from chroma noise?

i'm may not see the colors as well as you might, what colors are the chroma noise specifically apart from the obvious colors of the Aurora on varied shades of ice?

Comment edited 60 seconds after posting
2 upvotes
marike6
By marike6 (Mar 2, 2012)

I'm not sure about the chroma noise, but I downloaded the the ISO 6400 Aurora image last evening. I couldn't believe this image camera from a 5D Mk III. The lack of detail, presumably from NR, makes the ice look soft and overall the image doesn't look like it was produced by a FF camera in 2012.

Also a surprise was the ISO 100 image of the female model in the red jacket did not nearly as good as what I was getting from my 5D Mk II. In fact it doesn't look like it's from a 5D camera at all. The official D800 samples, on the other hand, were spectacular.

http://forums.dpreview.com/forums/readflat.asp?forum=1032&thread=40778609

1 upvote
Josh152
By Josh152 (Mar 2, 2012)

@mairke6

I agree, there is so much NR going on, even in the DPR samples that it is really impossible to tell the true iso performance.

If this is what canon felt the "standard" NR setting on the camera needed to be for jpegs it doesn't bode well for the actual noise preformance though.

0 upvotes
Henry M. Hertz
By Henry M. Hertz (Mar 2, 2012)

"...The official D800 samples, on the other hand, were spectacular. .."

i call that bull.... some of the D800 images are just bad.

0 upvotes
Sdaniella
By Sdaniella (Mar 2, 2012)

ISO 50-1600 looks excellent (ISO 1600 now as good as previous ISO 1600... or from what i can tell)

btw... someone cleaned the dust off of the Chinese warrior figurine, less dust on its base...

ISO 3200 noticed tiny NR going on altering something, but IQ excellent
ISO 6400 tiny NR again, IQ good

both of these are much better than before

ISO 12800 NR, IQ 'okay', noise is there but very usable, more usable than PowerShot G1X

nice to see usable ISO 12800

ISO 25600 NR can hardly help, the noise is definitely all over
ISO 51200 too noisy for my liking to ever use it
ISO 102400 so noisy, why have the setting? ... ok..
... I realize the noise will differ with different lighting, and exposures, so we'll see how it fares under other conditions.

thx for sharing... it has been a long Friday morning for you guys at DPR...
looking forward to all the other preview shots even if it is a pre-production model on hand.

Comment edited 13 minutes after posting
1 upvote
seta666
By seta666 (Mar 2, 2012)

"ISO 3200 noticed tiny NR..." Come on, NR is very heavy; most of the wood textures have dissapeared at iso 3200

1 upvote
Sdaniella
By Sdaniella (Mar 2, 2012)

keep it in context, pls:
"noticed tiny NR going on altering something"

no, the wood grain was SPECIFICALLY what i was scrutinizing at the focal point, it was at ISO 3200 that the details shifted ever so slightly, and no, the details were not massively altered by NR in an 'impressionistic-a-la-NEX-7' way... but very much intact... so, just look again, the wood grain is mostly intact, especially if you didn't have ISO 50-1600 to compare it with (from 50-1600, it looks practically identical, until ISO3200)

it is only from the transition from ISO1600 to 3200 does it look 'different' ever so slightly. not talking about areas that are known to be OOF, they would be blurry anyway, but looked at the finer details instead

we could surmise that from ISO1600 - 3200, more 'tiny' shifts of NR altering the details, but we don't have those in between ISO shots to see that.

i was merely giving my fair assessment of what was available and what i could see and deemed acceptably excellent, good, etc.

1 upvote
Alejandro del Pielago
By Alejandro del Pielago (Mar 2, 2012)

The sectors that are in focus (at least a few) show texture... And I am seeing ISO 6400... Maybe just two guys notice that....

0 upvotes
echelon2004
By echelon2004 (Mar 2, 2012)

1600 look ok, the rest is just smudge. So, not really à jpg camera?

4 upvotes
Gothmoth
By Gothmoth (Mar 2, 2012)

you are not really that clueless are you?

3 upvotes
rusticus
By rusticus (Mar 2, 2012)

what is that on the left side of the Black?

0 upvotes
Michael_13
By Michael_13 (Mar 2, 2012)

Reflections of the black background cloth.

0 upvotes
Kaj E
By Kaj E (Mar 2, 2012)

What is the blob above the owl pillow in all shots?

0 upvotes
Michael_13
By Michael_13 (Mar 2, 2012)

Reflections of the black background cloth.

0 upvotes
Kaj E
By Kaj E (Mar 9, 2013)

Looks like shadow noise blotches to me

0 upvotes
D200_4me
By D200_4me (Mar 2, 2012)

I think it's a given that these days, any high end camera Nikon and Canon produce will have outstanding high ISO performance. I'd like to see the same from Sony and the 4/3 makers, then we'll be in business! (lots of great choices for everyone)

0 upvotes
Gothmoth
By Gothmoth (Mar 2, 2012)

imo canon has by far the better lineup in 2012.

with the coming 4K video DSLR, the high megapixel 5D X(?) and the C300 that can take EF lenses the canon system is way ahead of nikon when it comes to video.

i rather buy into EF lenses i can use on a C300 in the future (when i upgrade) then a D800... even when it has uncompressed video.

the high iso example from the 5D MK3 look impressive.
6 FPS are more worth to me then 14mp more.

the build quality is impressiv too, i heard that from many sources.
much better then the already good 5D MK2.

now im only curious how good the new AF system is.
i know it will be great but how great, that is the question.

1 upvote
powertoold
By powertoold (Mar 2, 2012)

An extra 2fps is worth more than an extra 14mp? What do you shoot? lol

3 upvotes
Gothmoth
By Gothmoth (Mar 2, 2012)

well it may come as an suprise to you .... but 36 MP are pretty useless when you don´t print billboards or crop like mad.

and i do not crop images to 25% of it´s original resolution nor do i print billboards. so 22 MP are perfect for me.

but i do shoot burst a lot... and 1 image per second more can be the keeper.

2 upvotes
ZAnton
By ZAnton (Mar 2, 2012)

Wanna talk about lineup?

Nikon has 12 MP and lots of fps AND 36 MP and lower fps and both with excellent AF.

Canon has 21/22 MP, 4/6 fps the only difference - AF - crap/good and overprice for that AF - $1300.

0 upvotes
cesaregal
By cesaregal (Mar 2, 2012)

Can we have same test for D700 and 5DII?

3 upvotes
Josh152
By Josh152 (Mar 2, 2012)

and D800

3 upvotes
Ithackermike
By Ithackermike (Mar 2, 2012)

I agree.

0 upvotes
makofoto
By makofoto (Mar 2, 2012)

Since low light ISO is so good now, there should be a setting where one can set say F2.8 and 1/125th of a second, and the ISO auto follows along.

Sport shooters would also like a Focus Trap mode, where the shutter releases when best focus is achieved, ie. one should be able to pre-set focus at a certain point, and the camera will auto trip when the subject matter does through that point.

0 upvotes
Redteg94
By Redteg94 (Mar 2, 2012)

That setting does exist; its called auto ISO in manual mode.

0 upvotes
Michael_13
By Michael_13 (Mar 2, 2012)

Many cameras can do this already, even compacts.

0 upvotes
mkln
By mkln (Mar 2, 2012)

that setting exists since forever and is one of the few features that should justify an upgrade from 5d2 to 5d3 for an expense of $1500+

0 upvotes
seta666
By seta666 (Mar 2, 2012)

Wow!!! incredible strong noise reduction is what I see

Totally useless test!!!!

11 upvotes
Josh152
By Josh152 (Mar 2, 2012)

Exactly! Just like the images on the canon japan site. If the NR is set to standard in the camera for these images than Canon is being way to heavy handed with the NR.

The real test will be to see high iso images processed from RAW files by someone who knows what they are doing.

Comment edited 1 minute after posting
2 upvotes
seta666
By seta666 (Mar 2, 2012)

Yes, the Aurora 2 high iso image looks like a painting instead of like a picture; so strong noise reduction. I bet that iso improvement over the 5d mkII will be around 1/2 stop. However there is quite a lot of room for improvements in DR at low iso.
A 2 stop DR improvement is almost mandatory after APS-C sony/Nikon cameras getting 13 EV DR at iso 100

1 upvote
Josh152
By Josh152 (Mar 2, 2012)

Yeah I have a feeling all these claims of "2 stops better ISO" are based off of jpegs with very heavy noise reduction applied. My guess is the most it is actually improved when looking RAW files is one stop. Plus I seriously doubt it will have better DR or even Equal DR to the D800.

Comment edited 2 minutes after posting
1 upvote
Gothmoth
By Gothmoth (Mar 2, 2012)

im waiting that dpreview will make these test with the D800.

i have seen a few low light/high iso example wiuth the D800 and they looked worse. and the shooting circumstances where better for the D800.

0 upvotes
Robert Hoy
By Robert Hoy (Mar 2, 2012)

It's a pre-production model, shooting JPG with default noise reduction shot in VERY low light with strong blue light if you read all the info provided by Dpreview. It's an extreme test in other words. Wait until you see a production unit's studio test scene shot by Dpreview via RAW before you make your final decisions.

0 upvotes
kuklukklak
By kuklukklak (Mar 3, 2012)

Hehe, people only see what they want to see. ITs JPEGs so of course it have NR. Or may be your Nikon Jpegs have no NR at all, theyre just "naturally" noisier in Raws. :))

Comment edited 39 seconds after posting
1 upvote
Lea5
By Lea5 (Mar 2, 2012)

25.600 is still impressive. (at least for me)

0 upvotes
Sdaniella
By Sdaniella (Mar 2, 2012)

we all know this is a mini-version of a quick ISO range check...
it's appreciated nevertheless

1 upvote
Daniel from Bavaria
By Daniel from Bavaria (Mar 2, 2012)

Would say 12.800 is roughly usable, 25.600 is roghly usabel with lots of headache and the rest is marketing.

For JPEGs quite good and for sure much better than my mkII.

BG
Daniel

1 upvote
HowaboutRAW
By HowaboutRAW (Mar 2, 2012)

Did you run a 300dpi print of the 12800ASA jpeg? Even a nice letter sized crop would suffice at 300dpi. Remember to include shadow areas in the test print.

dpi=dots per inch=122 dots per cm

0 upvotes
nicolaiecostel
By nicolaiecostel (Mar 2, 2012)

The test is no good. I am an event shooter for many years and I have yet to shoot a box of crayons lit by an incandescent lamp, with the camera on a tripod. The real low light is in the evening, in a pub, or at a disco, at a pool-alley, on a walk down the Thames with the girlfriend. I shoot at wedding receptions, at meetings, in dimly lit orthodox churches, not in a studio. That is why these images are no good, a compact camera could perform flawlessly in a studio too.

I want to see images with backlighting, with spots of different colors on the subject, shot at f 2,8 handheld.

5 upvotes
Henry M. Hertz
By Henry M. Hertz (Mar 2, 2012)

well genius a test has to be reproducable.... a pub is a changing subject.

and that a P&S camers can produce "flawless" high iso images in studio shows that you have no clue what you are talking about.

just check the high iso examples from P&S cameras here on dpreview and you will notice that....

Comment edited 2 minutes after posting
14 upvotes
Alan Brown
By Alan Brown (Mar 2, 2012)

No offence but your thinking needs an overhaul:

you said 'I want to see images with backlighting, with spots of different colors on the subject, shot at f 2,8 handheld.'

I can hear the screams on the forums, now. If you did this as a test.. who could possibly say that this or that camera produced more or less soft images because you had a drink on one day and left the ale alone for the next camera test?

repeatability is the key to any testing.. otherwise it's worthless.

BTW. the crayons don't know what light is shining on them.. just the camera.

Peace :)

0 upvotes
bcalkins
By bcalkins (Mar 2, 2012)

Repeatability may be the key to any testing - but do you want a test ? Or to see what this camera does in the hands of a photographer? A test of a studio shot with strobes at base ISO is interesting to me. But useless for comparing high iso shots for many reasons:

1. If I'm using high ISO I'm not on a tripod (though I can think of people who would be).
2. I'm using a fast lens wide open, not a super sharp macro at f/11.
3. Noise reduction behaves very differently when you have a little subject or camera motion. If you take a very sharp shot with a lot of detail you can hardly see the noise reduction even at very high ISO. A very slightly blurred shot of a face in low light is a whole other story.

I agree with nicolaiecostel - this is not a great way to compare High ISO, and certainly wouldn't make me switch brands :)

0 upvotes
Gothmoth
By Gothmoth (Mar 2, 2012)

well to compare different cameras it has to be repeatable.. period.

if you want to see other images you don´t have to look at studio tests.... you can just wait for such images to show up.

0 upvotes
Alan Brown
By Alan Brown (Mar 2, 2012)

By bcalkins (26 min ago)

Repeatability may be the key to any testing - but do you want a test ? Or to see what this camera does in the hands of a photographer?

I understand your reply but the only one who could make this 'subjective' test is yourself.. it would be irrelevant to anyone else.. more so if someone other than you did it. .the variables are off the scale. I know i have good and better days handholding equipment. Why should I trust what someone says when this is subjective to the person doing the test.

The point that I take issue with is that nicolaiecostel said the 'test is no good' when in fact it's the only 'non-subjective' data that can show relative strengths or weaknesses.Everything else is opinon.
This test is not the be all and end all by any stretch of the imagination but it's the only constant we have here. We must make up our own minsd based on actual use..After that, no one elses opinon really should matter.

0 upvotes
Richard Butler
By Richard Butler (Mar 2, 2012)

These were shot under low-light and under the sorts of lighting you're likely to be shooting under - these are not colour-balanced studio lights.

0 upvotes
kevindar
By kevindar (Mar 2, 2012)

thank you for posting, and the images do look about a stop better than my 5d2. However, as a humble suggestion, could you please shoot at 35mm, or 24mm f11 next time? the most useful item in the image is the rat, which is out of focus. the dof for 100% view, where we can look at amount of detail smearing is just not there for the rat. that is my only gripe with this test.

1 upvote
mariuss
By mariuss (Mar 2, 2012)

OK ... I think with this camera is time to remember how is it to use JPEGs.
Perhaps, RAW is not "the only one choice" anymore.

0 upvotes
HowaboutRAW
By HowaboutRAW (Mar 2, 2012)

huh?

re-articulate that point, clearly.

RAW is the only answer for colour, exposure and high ISO work.

2 upvotes
visionaer
By visionaer (Mar 2, 2012)

didn't you see the heavy nr?!

1 upvote
mariuss
By mariuss (Mar 2, 2012)

More clearly, I can not. Maybe should I repeat instead?! Would this help?!
"RAW is the only answer for colour, exposure and high ISO work."
- and for low ISO and BW photos?!
... sarcasm out.

Comment edited 2 times, last edit 3 minutes after posting
0 upvotes
HowaboutRAW
By HowaboutRAW (Mar 2, 2012)

mariuss:

Try again.

"sarcasm out" is meaningless.

jpegs=a formula for blotchy BW.

0 upvotes
yuyucheu
By yuyucheu (Mar 2, 2012)

Dispointed, It's not 32MP.

1 upvote
HowaboutRAW
By HowaboutRAW (Mar 2, 2012)

Why, pray tell?

1 upvote
Anadrol
By Anadrol (Mar 2, 2012)

Buy a D800.

1 upvote
Michael_13
By Michael_13 (Mar 2, 2012)

Buy a compact. ;-)

1 upvote
HowaboutRAW
By HowaboutRAW (Mar 2, 2012)

Why on earth is anybody commenting on noise in high ISO jpegs?

Wait for the raws, and the software to extract those raws.

Hard to take anyone who makes this jpeg comment mistake as being seriously interested in digital photography. It's a big mistake, and shows profound ignorance.

(If I missed someone who commented, "well these high ASA jpegs are noisy, I'll wait for the raws and good extraction software", then my comment doesn't apply to that someone.)

2 upvotes
makofoto
By makofoto (Mar 2, 2012)

Some shooters can't take the time to process 1,000's of RAW images ... we have to shoot jpeg

0 upvotes
HowaboutRAW
By HowaboutRAW (Mar 2, 2012)

makofoto:

That's fine if you don't care about color or exposure.

But drawing noise conclusions from jpegs is preposterous.

Not saying don't shoot jpeg, saying don't draw general camera conclusions from jpegs.

1 upvote
Gothmoth
By Gothmoth (Mar 2, 2012)

well if only JPG matters for your kind of work .... why should you bother with raw.. makes no sense.

so if someone shoots only JPG and that looks great, then RAW performance is pretty theoretical for him...

Comment edited 10 seconds after posting
1 upvote
Michael Barkowski
By Michael Barkowski (Mar 2, 2012)

The scene is worst case, as the article stated, and by using OOC JPEG we have worst case processing also.

The worst case examples do reflect something about the potential. No final conclusions can be drawn, but the samples are very impressive. If this is the worst case, it's good news for Canon.

Even when JPEG-from-RAW-via-ACR examples are posted, they will not show the full potential, since as dpReview admits repeatedly, they use a pre-canned process which is not fully optimized.

Comment edited 1 minute after posting
2 upvotes
HowaboutRAW
By HowaboutRAW (Mar 2, 2012)

Gothmoth:

It's the drawing of conclusions about the camera based on jpegs that's the problem.

Michael Barkowski:

In the actual review, DPReview is likely to post raws for download, that then can be extracted in ACR. Check the reviews of approximately the last 2 years, downloading the raw files is an option. Try it with some camera that interests you. You'll need raw extraction software--ACR, Bibble, DXO.

Anyway, I'll be trying RAW to Tiff first, once someone supplies some raws and the extraction software.

0 upvotes
Gothmoth
By Gothmoth (Mar 2, 2012)

well RAW can look better or worse but it will not matter if i shoot only JPG.

if i (would) shot JPG only, i decide what i buy after the JPG quality, no matter what RAW looks like.... that´s what i mean.

maybe different for pixelpeeper but im a photographer. :)

Comment edited 1 minute after posting
1 upvote
HowaboutRAW
By HowaboutRAW (Mar 2, 2012)

Gothmoth:

If you only shoot jpeg fine, but that also means color control, noise, and exposure control don't really matter to whatever kind of photography you do.

It doesn't take peaking at a couple of hundred pixels to tell that raws extracted to tiff print better than jpegs.

Comment edited 4 minutes after posting
0 upvotes
Artistico
By Artistico (Mar 2, 2012)

processed jpegs don't say as much about noise as unprocessed raw files. Processing changes, so comparing jpegs to jpegs is pointless unless both go through the same processing. That said, I think this definitely is better than my 5D classic. I didn't want to go above ISO 1600 on my old 5D.
I've been using MFT the last few years, the GH2 being better than 5D at high ISOs, though the 5D images, even with the same lens, would look better, more natural.
With the auspicious combination of selling my flat and the 5D MkIII being released, however, perhaps I'll go back to Canon 5D when my bank account soon fills up.

2 upvotes
zsozs
By zsozs (Mar 2, 2012)

OMG !!!!! dpreview.com !! You cant take a foto ????

What are you doing ???

Shutter speed !!! Shutter speed !! Look !

No body use the Shutter speed 1/10 in real life !

The test ist not good ,if you use higher shutter speed have much NOISE in low light (1/60,1/100) and its sure iso 3200 much noise you will see !

Comment edited 25 seconds after posting
0 upvotes
chlamchowder
By chlamchowder (Mar 2, 2012)

This is a test, and for the test to be accurate, a proper exposure must be made at each sensitivity. Whether you use a shutter speed of 1/10 or 1/100 doesn't matter, as long as you let in the same amount of light (i.e., shutter speed/ISO/aperture combo matches the lighting situation).

While a longer shutter speed could increase noise, you'd probably have to make a 10 minute (or longer) exposure to see any difference.

0 upvotes
zodiacfml
By zodiacfml (Mar 2, 2012)

Are you serious????

1 upvote
tiberiousgracchus
By tiberiousgracchus (Mar 2, 2012)

Welcome to DP...troll

2 upvotes
tiberiousgracchus
By tiberiousgracchus (Mar 2, 2012)

.

Comment edited 2 times, last edit 2 minutes after posting
0 upvotes
ZAnton
By ZAnton (Mar 2, 2012)

Using low light conditions DPReview has certainly improved the ISO tests, although there are no similar shots for other cameras for comparison (except for G1X).
Second problem is, that all objects there (except the rat) are big (compared to the pixel size) and of the even color or moderately changing color.
Such conditions is excellent for the NR, but they rarely seen in the real life, (unless you are making a macro of a pencil in the dark toilet. :-))

The real troubles come on the small objects (hairs, leaves if they are far away enough) etc.
So guys, put the rat in the focus please. It looks like "rat quality" degrades very fast even here.

Comment edited 53 seconds after posting
6 upvotes
Henry M. Hertz
By Henry M. Hertz (Mar 2, 2012)

zanton the nikon fanboy.. be more concerned about the D800 iso performance. LOL

5 upvotes
ZAnton
By ZAnton (Mar 2, 2012)

@Henry M. Hertz
what?

Comment edited 33 seconds after posting
0 upvotes
Marcin 3M
By Marcin 3M (Mar 2, 2012)

ecactly the same suggestion here.

0 upvotes
canonikony
By canonikony (Mar 2, 2012)

those images are pre-production, not yet final, so read comprehensively

Comment edited 26 seconds after posting
2 upvotes
Thomas Richter
By Thomas Richter (Mar 2, 2012)

Completely agree that the setup is not good for showing ISO performance. No in focus low contrast detail, so NR can go in and smooth away all the noise and images look fine, except for appearing somewhat less 3-dimensional due to reduced color differentiation. But human eye is very forgiving w.r.t color.

4 upvotes
ZAnton
By ZAnton (Mar 2, 2012)

@canonikony
@Henry M. Hertz

my post concerns the test method itself, not the Canon's results.
If insult - is the only thing you can say - means, that YOU declare, that canons performance is not good.

0 upvotes
tiberiousgracchus
By tiberiousgracchus (Mar 2, 2012)

Wow these came out quick. Really impressed with ISO 12800. If you need i think it can be used. I dont think i will need such ISO but its definitely a step up from the 5D II. Me ? well im not pro so buying this is way out of my budget the younger brother is still my target. but will the prices settle down in a month or so..?

0 upvotes
ZAnton
By ZAnton (Mar 2, 2012)

Month? You kidding?
1 year is the time when you can start waiting the prices to come down.
2 years - is when it will really fall to a reasonable value.

3 upvotes
ZAnton
By ZAnton (Mar 2, 2012)

@Henry M. Hertz,
honey, stop the hysteria and look on my gear.

Comment edited 14 seconds after posting
1 upvote
tiberiousgracchus
By tiberiousgracchus (Mar 2, 2012)

im not talking about real reductions just slight ones that will have a knock on effect on the 5D II :P no i cant afford this one either

0 upvotes
Gothmoth
By Gothmoth (Mar 2, 2012)

well i can put a porsche carrera in my signature but that does not mean i really own it. the internet is full of lies.

sorry i don´t want to disturb your discussion guys.. just my 2 cents. ;)

Comment edited 30 seconds after posting
0 upvotes
ZAnton
By ZAnton (Mar 2, 2012)

First of all the internet is full of idiots.
One idiot claims that I don't have money even for p&s.
The other believe him and but doesn't believe me.
So what can I do? I just leave you alone.

0 upvotes
gm1974
By gm1974 (Mar 2, 2012)

Wow. Those are impressive. I'll have to wait for the price to come down and the first couple firmware tweaks (6 months or so? ;) ), but I'm definitely hoping to make the jump to Full-Frame with this one, from my 7D.

2 upvotes
Lillrobert
By Lillrobert (Mar 2, 2012)

Clearly visible noise already at ISO 6400 is not ok for a camera in that price range.

2 upvotes
photofan1986
By photofan1986 (Mar 2, 2012)

Yes, it's terrible! I've never seen so much noise in a picture!

1 upvote
D1N0
By D1N0 (Mar 2, 2012)

I think it looks impressive at 6400 I have to stop at 640. At 25600 you can see real noise. Below that it is not a real issue.

1 upvote
Henry M. Hertz
By Henry M. Hertz (Mar 2, 2012)

compared to the D800 i say it´s way ahead.

good work canon!!

3 upvotes
D1N0
By D1N0 (Mar 2, 2012)

so you have a d800 do you?

2 upvotes
Marcin 3M
By Marcin 3M (Mar 2, 2012)

A statement like from cheap ad... On available hi-iso d800 samples fine details are not as washed as on those samples from 5D3. Target doesn't allow to estimate, how NR destroys details, but some artifacts typisal for excessive NR are visible - if one has good monitor and good will to dee the true.

3 upvotes
JakeB
By JakeB (Mar 2, 2012)

Look at Henry's posts.

He's a bitter crank. Ignore him.

2 upvotes
Natan Lorenzi
By Natan Lorenzi (Mar 2, 2012)

Well you need pay $ 500 more and, of course, D800 is a different type of camera. 5D3 will never be a 36mpx camera and never have the incredible detail.

Comment edited 3 times, last edit 2 minutes after posting
1 upvote
Gothmoth
By Gothmoth (Mar 2, 2012)

incredible detail?
well i see good detail in D800 images .... but not incredible.
and what is it worth for most users who print 13x18cm 90% of the time?

do i need 36 mp for the occasional 2-4 pictures a year i print with 120x60cm??

nope... 22mp are the sweet spot today and you can talk whatever you want.... the 5D MK3 images look cleaner.

as written this is a kind of "worst case scenario".

Comment edited 1 minute after posting
2 upvotes
Natan Lorenzi
By Natan Lorenzi (Mar 2, 2012)

Hahaha, Gothmoth, are you blind? The D800 has the DSLR best IQ. Tell me what camera is better in iso 100, 200 400 and 800? Canon? ahahahah.
I'll use it in a studio. Is this the best DSLR for studio and landscape ever made.

0 upvotes
Zoomstein
By Zoomstein (Mar 2, 2012)

Very very impressive. Up to 12800 the quality basically remains very stable and images are perfectly usable.

5 upvotes
JohnHBailey
By JohnHBailey (Mar 2, 2012)

Wlad,

ISO 102400 could also be there for the same reason as ISO 25600 is on older bodies - just in case you are in a situation where any shot is better than no shot. My venerable 30D goes up to ISO 3200 - it' s rough and I don't use it, but it's there if ever I DO find I need it !

0 upvotes
Earthlight
By Earthlight (Mar 2, 2012)

These seem better than the official samples. But the asking price is way, WAY too high. A local pro shop quotes EUR 3799. Imagine that! Laughable.

3 upvotes
ashwins
By ashwins (Mar 2, 2012)

At ISO3200 noise is pretty nicely under control, but at ISO6400 chroma noise starts to show up clearly (and it's at ISO6400 where a FF DSLR should be able to make low noise photos).

4 upvotes
Silat Shooter
By Silat Shooter (Mar 2, 2012)

These samples are impressive! Looks very usable all the way up to 25,600. That's honestly a HUGE improvement! I agree the asking price is high at $3,500. But no denying, Canon addressed the 5D II shortcomings. Question is it worth $1400-$1500 more?

4 upvotes
treepop
By treepop (Mar 2, 2012)

Thanks for the fast details. In the forum it was stated default NR was on. Wish we could see some raws from this rig.

Comment edited 1 minute after posting
0 upvotes
Jared Abrams
By Jared Abrams (Mar 2, 2012)

Is there a reason why DPreview chose to use Non-Native ISO's for this test? Has something changed with the 5D3. Typically Native ISO's are 160, 320, 640, 1250, Etc. Unless something has changed this test is not valid without them.

2 upvotes
Andy Westlake
By Andy Westlake (Mar 2, 2012)

Can you explain what you mean by 'Native' ISO here? And, indeed, how shooting at ISO settings that the camera offers could be 'not valid'?

0 upvotes
Andreas Frank
By Andreas Frank (Mar 2, 2012)

ISO 100, 200, 400, 800, ... is native to Canon.

0 upvotes
kimsch
By kimsch (Mar 2, 2012)

Jared is right. Native ISO's for Canon are 160, 320 etd

1 upvote
seta666
By seta666 (Mar 2, 2012)

I think jared is wrong; native iso is 100, 200, and so on
Iso 160 is an iso 200 pull, that is why look cleaner (shadow noise is hidden)
http://shootintheshot.joshsilfen.com/2010/05/13/canon-hd-dslr-native-iso/

0 upvotes
Andy Westlake
By Andy Westlake (Mar 2, 2012)

OK, note to videographers. Showing that 160, 320 etc have the lowest black noise in video means they're a good choice for video - nothing more. In fact it's very good circumstantial evidence that they're non-native ISOs, and pulled by 1/3 stop (which, technically, reduces the visual impact of late-stage read noise).

0 upvotes
txabi
By txabi (Mar 2, 2012)

Color rendition and noise performance seem good in these and specially the official samples (those reds!). However, I find the images we're getting quite underwhelming so far. Where is the fine detail? The samples - both DPR and official - look mushy to me. My 5N seems to show more detail than this.

0 upvotes
Will Patterson
By Will Patterson (Mar 2, 2012)

I think they are pretty impressive shots considering the conditions, shame about the price though.

1 upvote
windmillgolfer
By windmillgolfer (Mar 2, 2012)

Very impressive indeed. I look forward to seeing the formal test results using a production model.

0 upvotes
wlad
By wlad (Mar 2, 2012)

omg, why did Canon even bother with ISO 102400, when ISO 25600 is already absolutely unusable ? pure marketing I guess

9 upvotes
Andreas Frank
By Andreas Frank (Mar 2, 2012)

I disagree.
ISO 25600 is perfectly usable for a lot of applications escpecially for on-screen or small print usage.

ISO 102800 imags are ugly, but still better than a completely blurred image or no image at all.

3 upvotes
Henry M. Hertz
By Henry M. Hertz (Mar 2, 2012)

exactly... for some work you don´t need perfect image quality.
because it could not be reproduced in the endresult (newspaper for example) anyway.

but you have to get the shot.

2 upvotes
greenlander
By greenlander (Mar 2, 2012)

If you think 25600 is unusable then you must define usable in terms of pixel peeping for noise. That image is perfectly printable and usable for MANY applications. Add in post processing, and we might have something here.

The only question is how close to these results the RAW files will be. If this is all in-camera processing magic, then it is much less exciting.

0 upvotes
Marcin 3M
By Marcin 3M (Mar 2, 2012)

Look at the hairs of the rat. While quite well defined at low iso, turning into something bad as iso grows... Could it be a NR?

2 upvotes
ace63105
By ace63105 (Mar 2, 2012)

Yes, it appears that NR is on.

I agree, these look pretty good, but there's not a lot of fine texture to notice the NR blurring. The hairs on the rat and the wood grain are already visibly blurred by ISO 800 (but fine), and starting to look pretty smeary by ISO 1600.

I'd like to see the RAW performance before making any real conclusions.

2 upvotes
Henry M. Hertz
By Henry M. Hertz (Mar 2, 2012)

you are a genius!!

higher iso means less details..... you should write that to some magazines im sure they don´t know yet.

2 upvotes
ZAnton
By ZAnton (Mar 2, 2012)

The rat is not in the focus. It is blurred from the ISO 50.

1 upvote
Total comments: 344
123