Canon and Nikon have their new mid-level full frame DSLRs. How do the compare, what differences does resolution enforce, and more nitty gritty analysis. Not just what they changed, but why.
So Canon and Nikon have both announced their next full frame DSLRs and everyone is covering them everywhere – so if you aren’t up to speed, start here (Nikon) and here (Canon) and especially here (Nikon vs Canon). You back? Good. Interestingly, both have decided to keep their prior models around and just drop the price on those, keeping $800-$1300 difference between older and newer models. These new models have much in common – both shoot the same video sizes and video frame rates in H.264, both have new HDR modes, both offer myriad improvements over their prior models.
So what did they change? How do they compare? A lot of websites are diving into those details, but I’m more interested in another question – WHY did they change what they did? What paths did this force them down? And after that, what choices did they make that WEREN’T forced by other decisions? What does this show about their priorities, or more particularly the priorities for these particular models?
If you just want a quickie visual rundown of features side by side in a chart to begin with before diving deep with me, check out NikonRumors.com’s nice specs breakdown.
Canon had an excellent camera with the 5D Mark II. The first 1080p video capability in a full frame DSLR, and excellent well rounded stills performance – high resolution at 21MP, excellent low light performance (ISO 1815 on DXOmark Sports rating), but a mediocre 4 fps burst speed for stills.
Nikon’s D700 was essentially a cut down D3 – with the same 12MP sensor but it had a lower max fps of 5 (8 with the grip & big battery) than the D3, but better than the 5D Mark II, as well as amazing low light performance (DXOmark Sports rating of ISO 2303, 2nd best in the world) about the same buffer (17 RAW), but zero video capabilities and only a single CF slot (as compared to the D3).
When it came time to design a new model, Canon decided to update their camera, but keep targeting the same users – an evolution. Nikon decided to take the camera in an entirely new direction, and effectively obsoleting the previous top end high resolution (24MP) camera, the D3X, while losing some of the benefits of the previous model but adding new ones – a more radical approach, a revolution of product.
Nikon wanted a camera biased towards landscape, architecture, and studio work, but at the expense of maximum still shooting frame rates and low light performance – the strengths of its predecessor. Oh, and while adding competitive video capabilities as compared to the Canon 7D and 5D Mark II. Canon opted to evolve a more overall balanced camera, sacrificing extreme resolution for higher stills frame rates and low light performance, while refining their video performance.
Evolution tends to be slow, steady, predictable, but all aspects generally moving forward over time. Revolution is disruptive – some things may be massively changed for the better, but some aspects or features may suffer or even be lost. Canon tried to evolve, Nikon tried to revolutionize.
If you’ve been reading about cameras, the tradeoffs between resolution, maximum stills frame rates, low light performance, buffer size, and buffer clearing times should be aparent, and give some clues into the relative potential strengths, weaknesses, benefits and drawbacks of these core decisions.
If you haven’t been keeping up on all this nerdery, here’s how it works:
-more pixels means smaller pixels in the same size
-smaller pixels mean less light per pixel
-less light per pixel means less light sensitivity per pixel
-less light sensitivity per pixel more noise per pixel, and worse low light sensitivity
-more pixels also means bigger files that take longer to transfer off the sensor
-bigger files means a lower maximum stills shooting frame rate since so much data to process (AOTBE)
-bigger files also take longer to write to the card, means waiting longer for the buffer to clear (AOTBE)
So Nikon decided to push the megapixels REALLY high – all the way up to a never before achieved 36MP for a DSLR – into medium format resolution territory. Canon decided to leave theirs about the same – a tiny bump from 21.1 to 22.3MP. Here’s how that one choice affected the two cameras:
RESOLUTION:
Nikon: 36.3MP (7360×4912)
Canon: 22.3MP (5760×3840)
With nearly half again as many pixels, the Nikon should be able to generate markedly shaper images.
As mentioned above, the smaller the pixels for higher resolution, the less light pixels can receive, thus affecting noise levels. Thus the other side of the coin –
LOW LIGHT SENSIVITIY: Don’t know yet, but on paper at least, Canon’s bigger pixels are looking really good. I saw some very clean ISO 6400 sample shots from the 5D Mark III. Considering that Canon maxes out at over ISO 100,000 and Nikon maxes out around 25,000, Canon clearly seems more confident in their capabilities. (Edit since originally writing the above – I expect the Canon to exceed the Nikon’s low light performance, but I expect the D800 to at least match, if not exceed, the D700’s at a given print size).
ISO RANGES:
Nikon – 100-6400, extended range is 50-25,600.
Canon: 100-25,600, extended 50-102,400.
Canon expected to have better performance in low light due to bigger pixels.
Once you’ve shot that big frame, you have to record it – thus
FILE SIZES & TYPES:
RAW: Nikon uncompressed 14 bit RAW files are about 75MB apiece! Fortunately, Nikon offers lossless compression which brings these down to a bit over half that at 41MB a pop for the exact same quality, as well as offering a 12 bit option as well.
I couldn’t find an official size, but chugging the math the Canon’s 14 bit uncompressed RAW 22MP files should be about 30+ MB each. If you find some proper stats, please let me know and I’ll be happy to update this.
Nikon offers up to three different compression levels on the same number of pixels – uncompressed (75MB), Lossless compressed (41MB), and lossy compressed (36MB) – they vary the FILE size, while keeping the pixel dimensions the same. You can also choose between 12 and 14 bits, which on some other Nikon models affects shooting speeds, but I don’t know whether it affects the D800. All of these affect write speeds to card. Canon is ALWAYS 14 bit in RAW. Canon offers the same # of bits and seemingly the same amount of compression per pixel, but offers it at three different PIXEL sizes – sRAW, mRAW, and (full sized) RAW, scaling down the RAW pixels, but that is based on scaling down from full frame, NOT from cropping down the sensor or image. Nikon offers cropping (to DX & other sizes) but NOT scaling for the RAW. So if you want a smaller (in pixel dimensions) file, you MUST crop on the Nikon, thus changing your framing. For Canon, you can scale but not crop using mRAW and sRAW to cover the same image composition. but generate fewer pixels for it.
Adjust the compression vs adjusting the pixel sizes, cropping vs scaling – this is how their philosophies differ, and I can see advantages to both. Personally I’d prefer all the options combined – compression and file size options, scaling and/or cropping, but that isn’t going to happen. Both have their advantages, both achieve generally similar results in terms of trying to reduce file sizes to card, but with different crops.
JPEG: both offer different sizes & types of JPEGs – Nikon offers 3 sizes, Canon offers 5. Nikon also offers the ability to prioritize file size or optimize image quality – smaller files or better looking files, pick one.
…and since those frames have to be processed before you can shoot the next one, that affects the…
TOP SHOOTING SPEEDS:
FX: Nikon 4fps, Canon 6fps – Canon wins by a 50% margin
DX: Canon doesn’t, Nikon 5fps, but if you want 6fps it’ll do it, but only with the addition of a particular grip ($450), battery ($170), and port cover ($30), and you’ll need a charger for it ($350), adding $1000 to the camera – yowza! This starts to flip the price difference between them from $500 less to $500 MORE for the Nikon.
On a related note to shooting,
SILENT MODE: both offer them, don’t know how silent, but only Canon offers silent continuous mode (at 3fps) – great for weddings and other “pardon me” type events. Nature shooters? Could be handy there too. Nikon’s silent mode is one. button. press. at. a. time. Only.
…and once you shoot those frames, they have to be written to card while you’re still busy shooting. At some point that buffer will fill up and slow down your maximum shooting speed, possibly costing you a missed shot – which is why we’re concerned with the….
BUFFER CAPACITY:
Nikon: lots of options, but short version for apples to apples:
14 bit RAW lossless compressed FX format: 17 shots until the buffer fills (21 if 12 bit)
JPEG Fine Large: 56 until the buffer fills
Canon: best I can tell, 14 bit RAW – 18 shots
JPEG: I found a reference to 65 images, but without details beyond that
Speaking of which, Nikon has a big ol’ asterisk by their stats. What does the asterisk tell us?
“Maximum number of exposures that can be stored in memory buffer at ISO 100. Drops if optimal quality is selected for JPEG compression, ISO sensitivity is set to Hi 0.3 or higher, High ISO NR is on when auto ISO sensitivity control is on or ISO sensitivity is set to ISO 1600 or higher, or long exposure noise reduction, Active D-Lighting, or auto distortion control is on.” So yeah – pay attention to that if it matters for you.
..and it would be entirely foreseeable to have ISO over 1600, JPEG quality to optimal, Active D-Lighting and auto distortion control turned on. Long exposure noise reduction and high ISO noise reduction are less likely to be used regularly, but certainly plausible.
THINGS SO CLOSE IT DOESN’T MATTER, OR A LITERAL TIE:
OK, so those above are the things that are inherently affected by the choice of pixel size and thus resolution. Once they made a choice about resolution, that irretrievably defined some other characteristics of the camera – or at least were engineering challenges to be dealt with and managed if not overcome.
What aspects of the camera are NOT inherent to resolution? Where do we get into areas that are more Free Will if you will, not mandated by pixel size/density? Where is there some lattitude (pun?) to go in any direction desired? And having that will, it is interesting to see where they felt their customers needs were in accord – for the following things they both decided to settle in effectively identical ways.
Where do these two cameras come in at a tie? Lets look at some areas where they’ve made decisions that are the same, or so close to the same that it just won’t matter. These are factors I say to ignore if you’re trying to decide between the two, since they are functionally identical:
WEIGHT: 900 vs 860 grams – effectively a tie, – can you tell the difference between 40 grams? Canon edges out Nikon by wearing the skinner pants, but not by a significant amount. If I put one in each hand after addiing battery lens and card, and you were blindfolded, could you tell the difference? I’ll bet not. Three CF cards in cases weigh more than 40 grams.
STORAGE MEDIA: a perfect tie – both do CF & SD, both allow overflow, mirrored, and split writing (RAW to CF and JPEG to SD, for instance). I can’t think of any other option I’d want here, other than a test function to tell me which card was faster, and how fast they were. At least you can find out by just recording how long it takes to write a full buffer’s worth of frames to the card and stopwatching it. Quickest time wins (and don’t forget media matters on that one.)
BATTERY LIFE: Nikon is rated 950 CIPA, Canon 950/850 for viewfinder vs Live View. Effectively a tie in real life I’d say. In much the same way that a computer that is 10% faster than another – you can’t tell unless they are side by side, it won’t “feel” any different I say, although side by side the Nikon is likey to “win.” If you’re shooting an event, are you likely to go over 900 shots? If you are, how MANY times over? How many times would you need to change cards in a day? I see this as a non-differentiator.
VIEWFINDER: 100% coverage for both, .70 vs .71 magnification – effectively a tie
EXPOSURES: 1/8000 to 30sec for both – perfect tie. Both can also do bulb mode. Every once in a while I wish I could set to longer than 30 second exposures for astophotography. But by in large this is sufficent.
PICTURE CONTROLS & PRESETS – I don’t know enough about this one, but they both offer presets and customizability – what controls do you get in each? I can’t give an answer to this one yet, so I’ll call it a tie. Both allow you to create and save look presets. Any feel free to point out how wrong I am on this one and I’ll move it to appropriate other category.
HEADPHONE MONITORING – yep, they both have it with adjustable levels.
MINOR DIFFERENCES:
Here’s where there are similarities between the two, but small differences in functionality/quality/results, enough to start considering I’d say. These are features they both knew they wanted in the camera, but different philosophies on new features, or perhaps code legacies or hardware/programming challenges on existing features led them down different paths. I’d say that unless you’re super picky, these are damn near a tie for the casual user of these features.
VIDEO FORMATS: both do 1080p 24/25/30, and 720p50/60. The Nikon additionally offers 720p25 and 720p30, the Canon does not, so oops, my intro was wrong – my bad. I don’t know when you’d use those last ones, unless you were wanting an exact look match to slo-mo 720p50/60 footage. But only if you were delivering 25p or 30p. I’d rather shoot 1080p and scale it down to 720p, that would likely be sharper than shooting at 720p in the first place. So why didn’t they include 720p24 if they’re doing all the others? 720p25 and 720p30 are just half rate of NTSC and PAL rates – other than for web, when would you use them? Ah – I think I just answered my own question. Hey me – you’re welcome. Some Canon folks were hoping for 1080p60 for slomo on this camera but no dice – these are video frame rate cameras, not digital cinema cameras. Also, no offspeed for either camera – no shooting 27fps, for instance. I would THINK it should be possible to do project frame rates (“This is a 23.976 fps deliverable even though I’m shooting 30fps for overcrank.”) and offspeed shooting with these kinds of cameras, but they don’t. Yet. Where’s a HVX200 equivalent here?
LCDs: both are 3.2 inches, Nikon 920K pixels, Canon 1.04MP. In resolution, Canon edges out a win, but in terms of brightness, contrast, color etc, still TBD. Nikon has an ambient light sensor to adjust on the fly automatically. So far, I’d say roughly a tie with a possible nod to Nikon. The resolution difference is too small to matter in real life, and may be outweighed in real world usefulness by the ambient Nikon stuff. Or not.
CA CORRECTION: both offer it, Nikon is lateral only, Canon adds axial as well – catching up and superceding Nikon’s previous advantage. Win Canon, but how much does axial show up? Lateral is the biggie that I’ve noticed. Show me good examples and I’ll gladly move this to Meaningful Differences. In any case, a Canon win.
GRIP & BATTERY STYLE:
Nikon keeps the original battery in the body, lets you add one more – but the one more can be the same style second battery, or a different bigger one. But this requires a rubber covered socket on the bottom of the body, a potential point of weakness for exposure to harsh elements if the cover comes off, gets lost, or is breached. Lose the rubber flap and set it down on a wet surface? Potentially bad news. (or even if it just surface tension sucks it up in there)
Canon’s grip extends an arm up into the battery compartment (gotta stash that battery door!), keeping the bottom of the camera sealed, and holds two standard batteries – but no big battery option given. Nikon charges a fortune for their grips and big batteries, Canon’s pricing is as yet unknown for the BG-E11 grip (edit – actually is inbetween – is $350, $100 more than 5D Mk II’s, $100 less than D800’s). Overall, I’d say this is largely a wash for most folks unless you need MAXIMUM battery capacity, in which case Nikon gets the nod for flexibility – you can either maintain battery compatibilty, or if you need even more power, you have it as an option by going with the big battery and other charger, pricey as it is. Both offer the ability to use AA batteries as a punt/last ditch solution, a nice feature when out in the boonies with no AC socket to recharge from.
EXPOSURE COMPENSATION: Nikon +/-5EV, Canon +/-3EV. Both offer fine control. Nikon wins this one, but rare is the situation where your metering is 3 stops off – that is a 8x longer/shorter exposure than the camera thinks is right. How often does that happen? It does affect HDR bracketing, however (more later on that).
DON’T STOP! SEE NEXT PAGE FOR THE MEANINGFUL DIFFERENCES
MEANINGFUL DIFFERENCES:
Here is where there are meaningful differences in the capabilities of the two cameras in areas where it was a matter of design choice – they DECIDED to do it this way, rather than they HAD to do it this way. For some it is a matter of notably different implementations, for others a matter of having a feature vs not having a feature. Often this is a matter of legacy – this is how they’ve done it, so they’ll keep doing it this way – even if it would be easy to fix/change/add in software.
PRICE:
Nikon: $3000
Canon: $3500
Nikon is clearly less costly for the base camera. However, if you want to add the grip, big battery, battery cover, and battery charger that allows for 6fps DX crop shooting, add literally $1000 to the price. If you want to add the grip and extra battery to the Canon, it doesn’t change the fps, but gives you (surprise!) a grip and more battery life, it’ll add $410 (not $510 like Nikon). But since it can run two identical batteries, and you don’t need a different charger (you might want an extra, but we’ll save that for later), you’re only spending $60 for an extra battery in addition to the grip. Beyond the $450 Nikon grip, for the D800 you need a different battery for $170, a $30 battery port cover and a $350 (yowza!) charger (they only offer a dual battery charger for this model). So bare? Nikon costs less. With grip? Canon costs less. With grip and big battery? Nikon wins on feature flexibility, but loses on price.
MULTIPLE CROP MODES: Nikon offers a number of crop modes – FX, DX, 5:4, 1.2:1. Canon doesn’t – FX only. So why doesn’t Canon value crop values? 22MP is plenty to crop from – 1.6 crop would still be about 9MP, 3600×2400 – that’s 100 pixels per mm of sensor! It isn’t great, but still entirely valid. Hell, even my 12MP D3S allows me to shoot 5MP DX crop files, they’re only 2,784 x 1,848. Canon – why u no doo eet? I don’t get it. They offer sRAW and mRAW, but not this. I’d love to hear an explanation.
MAX AUTO FOCUS POINTS: Nikon has 51 points, Canon has 61. Canon offers a lot of control tweaks. Both are pulling systems from their new big brothers, so no firm data on this one. Both sound quite impressive. Nikon has face tracking, at least in Live View – not sure about TTL (I think so, don’t know so). But Canon has more points that seem to cover more of the viewfinder. As a practical matter, 51 vs 61 points – will it matter most of the time? Where is the point of diminishing returns? Since the 5D Mark II had only 9 points, this is a VAST improvement for Canon. I debated putting this isn Minor Differences, but figured I’d get shot over 10 points or face recognition. As a practical matter most of the time, just based on number of points, it’ll probably be a wash from the specs. Now, how well do those points work? Under what light levels? THAT could be significant! So I looked it up – Mark III supposedly works down to EV -2, and the D800 down to….EV -2. OK, so on paper at least, a tie on light levels, we’ll need to see how they do in the real world – it could go either way.
LOW LIGHT PERFORMANCE: I like the DXOmark Sports/Low Light rating, because it gives a definitive bar for low light performance: “low-light ISO is the highest ISO setting for a camera that allows it to achieve an SNR of 30dB while keeping a good dynamic range of 9 EVs and a color depth of 18bits.” You can argue with the seeming arbitrariness of that bar, but the bar is consistently applied to all comers. Unfortunately, we don’t have third party test results like this for these cameras yet. However, we do have the companies’ own statements about their cameras relative to their prior models. So with MANY grains of (noisy) salt, here’s what I think we can glean from what has been presentd to date:
Nikon: the prior model was outstanding, literally the second best low light performer ever, losing out only to the D3S. The D700 was a virtual tie with its big brother the D3 at about 2300. I don’t recall seeing anything official, but there seems to be a lot of hinting that the D800 is close to that level. My own napkin math with the D7000’s sensor (almost exactly the same pixel size/density) and forward projecting Nikon’s usual generation to generation improvements makes me think the performance should be in the ISO 1600-2000 range for the DXOmark score. Yep, totally guess based on many layers of assumptions, so a LARGE margin of error in there. Again, if they matched the D700’s performance, which had literally 1/3 as many pixels, that’d be 2300, and I’m not expecting (but I’d be delighted since I have one on order) the D800 to best the D700. EDIT – saw some samples – when scaled/printed to same size, D800 actually looks better than D700 – good news.
I’ve read from a couple of sources about a two stop increase in low light performance over the 5D Mark II, so starting from the 5D Mk II’s DXOmark Sports ISO of 1815 (fifth best of ANY camera), that’d be 7260 – more than double the prior champ, the D3S’ 3253. (The D4 is expected to have similar noise levels, but sharper images -so I don’t know how that’ll reflect in the score). I hope to see something at least over 4000, but I don’t expect it to score as high as 7000. Again, just my limited experience and extrapolation, I could well be wrong. In any case, I don’t see any possible way for the 5D Mark III to do anything but best the D800 in low light performance, and it could easily be over a stop of difference between the two. Time will tell. Or actually, DXOmark will. Get on it, lads.
But this opens the door for the 5D Mark III to actually be the new all time champion low light performer. Wow.
Follow up note – yeeeeeeah, there are JPEGs posted from both cameras now, but I’ve spent enough time on this article. Not gonna dive into it now, even though it is important. I’ll wait for DPReview.com to post some controlled test results. Have an informed opinion and some data or examples to back it up? Shoot me a link and I’ll include it!
CODECS: Video nerd that I am, I’m going to file this one under meaningful differences. For shooters handing over footage and don’t edit it, or stills only photogs, this is a minor, “Meh, who cares?” difference, but for those in post it is a BIGGIE. Either All-I or IPB for Canon, with either .mov or .avi file wrappers.(Canon also offers SMPTE timecode as well). All-I means bigger files but fewer “pops” in compression (that every half second pulse/twitch as the whole frame is updated). IPB means smaller files, but they can twitch on fine details or gradients in motion.
This will be Nikon’s first (or 2nd, depending on whether the D4 ships first) full frame DSLR with 1080p H.264 video recording to .mov files. Canon is clearly and literally a generation ahead with their timecode, wrapper and file format options. I’d love to know data rates for 1080p24 for both, but I don’t have those numbers handy.
In any case, for flexibility and options, Canon win. For quality of results? I don’t know yet.
VIDEO OUTPUT: The new Nikons (D800 and D4) are the ONLY DSLRs that can output full raster, clean output HDMI 1080p24 (with 3:2 pulldown? I suspect but don’t know, but the smarter recorders can remove it). But it’ll do it ONLY when NO CF or SD card is present – if you have a card in and recording to it, HDMI resolution drops BELOW 1280×720 – they don’t say what exactly though. So monitoring will NOT be full raster – same challenge the 5D model has. Nobody’s perfect…yet.
EDIT: FLASH Well duh – since I pretty much never use flash, I didn’t think to include this, but there is an marked difference – the D800 has a pop-up flash, the Canon does not. Sync speed is 1/250 on Nikon, 1/200 on Canon, and the Nikon has wireless flash control.
METERING – Nikon has 91K sensor, Canon has 63 “zones”
HDR STILLS: both cameras can now generate HDR stills. If you aren’t into it, then file this under Minor Differences. I’m not one of those people, however.
Nikon: In camera – 2 shots, up to 3 stops apart, no aligning. Don’t know about how it stores originals and resulting HDR. For DIY HDR in post, up to 9 bracketed shots 1 stop apart each, adding 8 stops of range to original.
Canon: 3 shots, aligns for you in camera, optionally keeps the HDR (non-RAW) and source frames. Conclusion: Sounds like Canon thought about this more. Canon wins. Up to 3 stops apart for HDR mode stills. Bracketing? Up to 7 shots, but unclear how much range you can give. 7 shots at 1/2 EV steps it seems? So that’d be a total 3 additional stops for the DIY in post crowd. Meh.
With the same total gap, I’d say 3 shots likely better than 2, and with alignment capabilities enabling hand holding, Canon seems to win this one for in camera capabilities. Oh, and Canon offers 5 different HDR looks in camera, too, compared to “like it or don’t” one setting for the Nikon. For post DIY HDR, however, Nikon clearly wins with wider bracketing capabilities.
EDIT Thom Hogan nails what I didn’t consciously state:
That said, as usual both Nikon and Canon have managed to include enough irritations in their new products that there’s definitely something to complain about from everyone. I’m still convinced that the camera designers don’t really know how people are shooting (or perhaps trying to dictate how people shoot). I’ll cite one example: HDR. On the Nikons, you have exactly one choice: create a two-shot HDR JPEG or not. If you choose not, the bracketing system will fight you in setting up exactly the sequence you want. Neither option (automated or on your own) gives you any help in defining where the “edges” of your scene’s brightness are. In other words, none of the engineers seem to actually have shot an optimized HDR sequence. They do know how to combine a dark and light image, though, thus the JPEG two-shot combo. I’ll reiterate my offer: I’m willing to take a dozen Japanese engineers out shooting for a few days and show them the way things are actually done in the field. My only goal would be to make sure that the camera engineers actually here real user demands.
AGREED! My own methodology is to Manually seek out what is the brightest detail I want to capture, by starting from auto exposure then switching to manual, dialing in those settings, then stopping down untill I’m getting detail in the bright thing (aka sun). Then I count stops on my fingers while I double the exposure time in my head four times – that is my baseline for a 9shot bracket. What the helll!!!! How about a mode to shoot 9 or more temp shots, and I pick the darkest and brightest I want to use in the range, and tell it how many steps I want between? Or better yet – hey mr camera, you have a metering tool, FIGURE IT OUT FOR ME!!! End rant.
AE/HDR BRACKETING: up to 9 shots 1 stop apart for Nikon, 7 shots 1/2 stop apart for Canon. Nikon easily and indisputably wins this one. Now – would YOU ever use that feature, and that much of it for AE purposes?
Then again….
Rant for both companies – HEY!!! This HDR thing has been around for a while! How about letting us space out more than 1/2 or 1 stop!!! If we are just exposure bracketing for safety, then sure, tight spacing makes sense. Why is it that some lesser cameras can shooter with wider spacing, and the higher end ones can’t? Insulting. My D90 and D7000 are limited to 3 shots, but at least they can space 2 stops apart! I would MUCH rather shot 5 shots 2 stops apart, only writing 5 shots to card and thus reducing my cycle time, rather than 9 shots 1 stop apart. It nearly doubles my storage and timelapse spacing times – obnoxious. YES, it would introduce a little complexity into the programming that you can’t do 9 shots 2 stops apart because that would exceed the maximum exposure bias, but That Is Fixable In Code – limit to the range available! OK, rant off. Again – why is it done this way? Because That Is How They Do It, and so often, Change=Bad in some eyes. So how do we change it? Get more people to complain. I hereby am doing my part.
RATING: Canon lets you rate from 1-5 stars straight from the camera – very nice user oriented, workflow think-ahead detail. Nikon doesn’t do this at all – and they should. OH, and as long as we’re thinking workflow….
MAC/PC SOFTWARE: Canon’s included software is better and more comprehensive. Canon gives away stuff you have buy from Nikon, and it ain’t cheap. Want to shoot tethered? Gotta pay for it with Nikon. BIG WIN – CANON.
TIMELAPSE/INTERVALOMETER CAPABILITIES: Nikon still builds in an intervalometer, Canon still doesn’t. You have to buy the Canon TC-80N3 for $135 (or metric equivalent), and it doesn’t offer as effective control for HDR timelapse. Then again, you can shoot indefinitely on the Canon, and can’t on the Nikon (although third party intervalometers can do it). Rant mode – OK Canon, how many years has it been? How many lines of code would it be to build in an intervalometer? Especially one capable of time efficient bracketing, which the TC-80N3 doesn’t do? This just seems stubborn, painful, old school and client punishing. Yeah, they get to sell a $135 piece of hardware, but when it would be an engineer work week or two to build one in? Stubborn, obnoxious legacy issue. Why do they do it this way? Because That’s How They Do It, apparently. Tautological frustration.
TIMELAPSE VIDEO:
Nikon can shoot timelapse and record as video, ready to go
Canon can’t. Same beef as above.
WHEN TO USE WHICH?
OK, so all that said, when would you use one versus the other? Imagine you’re a platform agnostic photographer renting for a job.
IT WON’T MATTER:
Comfortably lit exteriors that let you work no higher than, oh, maybe ISO 1250 that don’t require high shooting frame rates. No sports, just posed/posing/casual people. I’m confident you’ll get excellent results from either camera. I defy you to tell the difference at web & print sizes. Color and style differnces? Flavor to taste, dude, flavor to taste. Open question in my mind: is there any colorimetry difference between Canon and Nikon that can’t be accounted for with on camera picture styles and/or Lightroom tweaking? Not having shot enough of both, I have no informed opinion on that.
Studio, architectural, landscape work not printed ginormously huge and with enough light. Need a full page magazine shot? Either one will generate lovely results. You might be able to crop a bit tighter with the Nikon, but if you’re shooting in studio, while the hell are you notably cropping?
WHERE IT MIGHT MATTER:
If youre shooting in very low light levels without flash or other lighting assist, I expect (see above) that the 5D Mark III will generate notably cleaner results than the D800. How low? I dunno – there aren’t DXOmark scores, and I haven’t studied the samples scrupulously enough yet. Somewhere north of ISO 2000 I’m betting you’ll see meaningful/noticeable noise differences between the two, if not sooner.
Anything sporty that requires deep buffer and fast fps – if you need full frame, although buffer depth is a tie the Canon is literally 50% faster than the Nikon – so go Canon. However, if you need a bit more reach, and have $1000 to throw at grip/battery/cover/charger, you can shoot DX at 6fps, and up to 54 12 bit lossy compressed RAW images that are 4800×3200 pixels – not much smaller than the Canon’s. Likely less sharp than FX, however, since you’re cropping in 1.5x. Use SHARP glass.
Printing at absurdly huge sizes? Nikon, with primo glass, on sticks, mirror up, between f5.6 and f11, remote shutter release, PERFECTLY set focus. Shooting handheld at f2.8? Meh – your technique (and glass) are more important than camera, shoot whichever you’re more comfortable with**. Which leads me to….
ACTUAL REAL WORLD NON-THEORETICAL ANSWER:
Whichever system you are more familiar or comfortable with and know how to get good results with. Surprise! Or if that doesn’t matter to you whichever one you get better gear/a better deal on. The quality of your photos, with the exceptions noted above, will not meaningfully differ between these two cameras. While I’ve never touched either of them, I’ve worked with their parents enough, and seen enough good results, to know that excellent, world class, award winning results are possible with either of these cameras if you stand in front of pretty things, at just the right time, point it in the right direction, twiddle all the knobs Just So and gently press the button called, I think, “Shut Her.”
-mike
Curious about what it costs to kit out either of these cameras? Read this
Want to read more about why Nikon went for such high resolution in the D800? Read this.
** Seriously – all those extra pixels do you NO GOOD if you can’t resolve them. The way to think of those additional pixels is not as MORE pixels for you to look at on a screen, but rather as more detail you’d HAVE to resolve onto the same 36×24 slab of silicon onto those tiny pixels. If there’s any camera shake or motion blur, the smallest resolvable detail will start to smear over a number of pixels instead of just one. At that point, you might as well be recording onto bigger pixels. And since there are about 200 pixels for each linear millimeter of sensor, that isn’t much motion AT ALL. And to resolve the difference between 22 and 36 megapixels (that is, , you’ve got to take serious advantage of technique to get them all. Lenses are sharpest around f5.6-f8, but beyond f11 diffraction limitations hamper the D800. So literally, your sweet spot to take FULL advantage of the D800’s resolution is ONLY if on sticks, right lens, right f-stop, right ISO, PERFECT focus, right shutter setup and release. Otherwise, if there’s any vibration of motion blur, you could have caught that level of detail with 22, or 16, or 12MP. It doesn’t take much to make that level of 36MP perfection…imperfect.
Legally Required Ridiculous Disclaimer – I own 4 Nikon bodies and a bunch of lenses, and as much as I wish Nikon would send me new toys for free, they don’t. Drat them. But in any case, I have no affiliation/association with them other than I’ve bought their products with my own money. Canon makes lovely cameras and lenses that don’t fit my toys. Drat that as well. I have bought some of their Powershot cameras and continue to enjoy and use them. But I don’t have an affiliation or work relationship with either company. Which part of this is ridiculous? The legal requirement? That I own Too Many Cameras? That camera companies aren’t psychic and send me free gear out of the goodness of their hearts, hearing my psychic yearning? You decide.