"When pushed 5.5 stops" - I'm far more interested in what happens when pushed 2 stops. Do these artifacts still show up? Who is really trying to recover over 5 stops on a regular basis?
But just HOW incompetent? Is there a full-auto mode in existence which could be that far off? I'm pretty sure my Minolta from 1992 wouldn't have been that bad
Wanako Plays! Such a silly comment. Just making excuses for canon and there is definitely situations where having the option is great. Blending landscapes has been happening for decades and the new nikons and sonys really help cutdown on the need to blend. I'm sure it is also helpful for concert and other high DR situations. There is times a single shot will never get the desired effect but having great shadows really helps. It has nothing to do with incompetence.
Higgins2001 well being that it’s a Mirrorless camera and you can see the exact exposure and DOF before you take the shot. There really shouldn’t be any reason for anyone to be under or over by 5 stops.
Quentin Tarantino presents 'How The Test Was Won': Tony N: "I can compare it to whatever I want, I can compare it with this goddamn hand-canon, mutherflipper! Do you want me to compare it with this hand-canon? Well, do ya'?" Matt G: "Say "only one card slot" again. I dare ya, I double dare ya, mutherflipper!" Chelsea N: "You know what you guys look like? A coupla' dorks." Kai: "[censored]" Lok: "[inaudible]"
Download sample raw files (converted to DNG for your convenience): sdp.io/eosrsamples Rent cameras and lenses at lensrentals.com, coupon code 'northrup15' for 15% off! Canon EOS R: sdp.io/r Canon 5D Mk IV: sdp.io/5d4 Canon 6D Mk II: sdp.io/6d2 Sony a7 Mk III: sdp.io/a73 Sony a7R III: sdp.io/a7r3 The a7 III is in focus. As a control I have objects in front of and behind the focal plane which allows me to verify consistency. I reshot with both the Canon 70-200 f2.8 III (via adapter, the same lens used on the Canon bodies) and the Sony 70-200 f2.8 GM just to make sure the lens wasn't the issue. Also, the a7R III used those same lenses and it's far sharper. The a7 III just seems to have a heavy AA filter. FYI, I tested it with and without IBIS and with and without lens IS (there was no difference). You'll see a Post-It in one of the shots keeping track of which was which.
Tony, your simple, honest, but yet powerful review is always better than some other review that is so long but just turn round and round without any clear summary. Really glad to have you in TH-cam. Thank you so much.
I felt your 6D Mk II samples were clearer/sharper than the EOS R samples. And any slight (it'd have to be very slight) shadow recovery benefits of the EOS R over the 6D II is ruined by that old original digital camera era banding o.o
That's lateral chromatic aberration at 6:00 and 7:20, not moire. Moire happens in areas of high frequency color information, not where you have a single high contrast edge.
And software deals pretty well with this artefact. With a Canon lens mounted on a Canon body, I'd expect that the in body image converter does this too. Probably the option wasn't set ?
Even from this brief comparison it is evident that all modern cameras are pretty darn good. Really, who in good mind is going to salvage a 7 stops underexposed image and complain about the noise they were getting? There's simply no bad camera in production anymore. Now we've reached the point where the entire ecosystem should be compared, not body to body or lens to lens. So personally I'm staying with Nikon, but I'm not going to shout Canon or Fuji people crazy. Today the gear is not anymore important the way it used to be some 5 years ago. Now it's all an upper-OK class or better.
If I remember correctly during the 5d3 vs Nikon D800 era...the Canon handled Overexposed images better where the Nikon handled the Underexposed images better. I wonder how that applies to the R vs Sony world. But to me the most important deciding factor of the Nikon/Sony/Canon mirrorless battle...which body is the most durable and with the best weather seal? I live in Alaska. I read the R has better weather sealing than the 5d4. That would be very impressive. I've rented Sony Full Frames and found them excellent..but they were not durable enough for what I do...and that has held me back from purchasing one.
Yea. I own the A7r3 and it is not a camera you want to be careless or in the wild with. I haven't tested weather sealing (!) but what irks me is the battery hatch. *Edit* Otherwise it's a effing great camera if you have the lenses you need.
The R has "worse" weather sealing than the 5Dmk4 from what I've seen, but still better than the Sony cameras. Something like being equivalent to a 6Dmk2
Incredibly useful comparison, and concise too, as usual best on the net at least in my experience. You really earned that million subscribers. Having NO long introduction is SO WONDERFUL, as is just saying "thanks, bye" at the end. (Of course now my A73 goes into the toilet .... ;-) )
Art, I'll take that crappy A73 off your hands ;) But yes, great video. Tony, please compare to the X-T3. I realize there's a huge difference due to sensor size, but I'm curious what that means in the real world.
Thanks Tony! I'm really happy for Canon for putting such a good and well balanced camera out. *A small question*, how how did you process the EOS R raw images? does Lightroom / ACR / anything else supports them already? The banding I see in the images can come from image processing not from the sensor (see it all the time in my work). How much are you sure that the software processes the EOS R raws correctly? Maybe some sort of Adobe's patch might fix that?
As the laws of physics are non-optional the more moving parts any camera has the more it’s subject to failure in longevity. Hence Canon's reluctance for IBIS. PS don't shake any camera with IBIS.
I owned the 5D Mark IV for a couple years, then sold it and got the R and this represents my experience exactly. The 5D produced the best image of any camera I shot with in 15 years - even at high ISOs and even when files were pushed and pulled. The R was much more fragile, so any editing of files at moderate to high ISO's would produce shadow noise and banding and colour shifts. If you only shoot daylight exposures (and dont need to pull up shadows) or in studio, the IQ of the R is great! But beware as the IQ is not the same as the 5D Mark IV. Some mirrorless have this noise/banding issue because of the way the AF is attached to the sensor. The z6 has similar issues, and the A7III has an issue in backlit situations. It looks like in 2020, the R5 and R6 have fixed this with amazing image quality - arguably the best in their class. But, most youtubers only care about video specs and headlines.
They could have mounted the same lens (with settings fixed) on each body and kept each sensor at the same position--maybe allowing some slight position adjustment to account for cover glass differences.
so true ...thats y i didnt still buy that 24-70 g master and 70-200g master.. still not impressed with sony's 2.8... canon has the best 24-70 and 70-200 out there .. he should mount the same 24-70 canon to both cameras..if he wanted a fair review ..my a7III + 24-70 f2.8 or 70-200 f2.8 is effin rezor sharp... side by side with my 5ds ..they are the same even if i enlarge to 50mp the a7III shot
How does the EOS R compare with the 5DSR, purely in terms of image quality and dynamic range? I recollect you guys saying that purely from an image quality perspective, you thought that the 5DSR is better than 5D-IV. Which of these three - 5DSR, 5D-IV, Sony A7Riii and EOS-R do you think works best for a landscape shooter, shooting primarily off tripods, usually at ISO < 800, with image quality being the biggest criterion?
The a73 pictures are either out of focus, or there is an issue with the lens or the camera. The sensor in the EOS R is the 5dm4 sensor, you can compare that with the a73 on any site, DP Review being the best interface, and see they are comparable in sharpness and resolution. Image Resource also has a comparison against the 6d Mark II, which in your video the 6dm2 even blows the a73 away www.imaging-resource.com/PRODS/sony-a7-iii/sony-a7-iii-image-quality.htm
Wy Lee I agree, Canon has the best glass and I bet that 50mm is incredibly sharp. But that doesn’t explain why the 6dm2 is also noticeably sharper. There is no data to support the 6dm2 has a far superior sensor than the a73.
Tony - I was just wondering if these are the actual files they are going to release out of camera or is that just a pre production file as well? Thanks for the review, kinda unexpected that the quality was better than the A7III.
Hey Tony! What lenses you shot with? I'm kinda curious about eliminating glass as a variable between the EOS-R and the A7iii. Also when are the giveaway winners announced? And thank you so much for running that! UPDATE: I looked at the DNG files and I think the A7iii is using a different lens from the A7R3. The A7iii is using the Sony GM 70-200 which LensRentals found to be soft.
I trust Tony, and I trust the test to show what it shows when you account for variables. I just like to know what I'm looking at and I think Tony did a good job of explaining everything else.
I agree with the questions in regards to the variables of this experiment. When Canon is shot with native mounted lenses while Sony is using an adapter does not quite qualify as an apple to an apple comparison. Since many reviewers had already made direct comparisons between the 5D IV to the a7rIII, I don't see the reason for the EOS R vs a7III instead of strictly to a7rIII. The cost factor to me is very lame...
I wouldn't necessarily be concerned about a Metabones being used to make sure every camera uses one piece of glass...I would be more concerned if it was a prime vs zoom test, or a set of lenses that vary in objective quality by some degree (a cheaper 50mm vs a high end pro 50mm and so on). In a static, well lit, tripod-shot kind of situation, I'd expect the adapter to be more or less invisible for this kind of test--maybe not in all other tests, but certainly in this one.
Eos R now has Firmware update wich completly remove the banding issues,it will be interesting if you make another video comparing Eos R with the latest firmware vs the 5D mark iv dynamic range/shadow recovery.
Canon sensors of the past had not been great on dynamic range, which means that you often did end up with shadows that needed recovery unless you wanted to completely blow out the overall scene. In fact, the 5D MKII is the camera that really made me embrace silhouettes and dramatic contrast!
You are too young to have memory, I can guess. Canon sensors have been for many years the best in dynamic range, that make Canon a best seller in digital photo. It took sony several generations to beat them and surpase them. Last generation of Canon sensors have closed the gap in dynamic range and are much more close to sony's equivalents. When you elevate ISO the gap in dynamic range gets narrower. The dynamic range of canon sensors at high ISO is better than sony, as canon sensors loss less DR pero ISO step than Sony's. For that reason it is much more important not to underexpose in Canon than in Sony. If you get an image 1 step under exposed (you used ISO 800 instead of 1600) and you correct at developing stage, you will get more noise in shadows than if you do the same with sony.
The dynamic range of any camera degrades when you increase ISO, there is no camera where the DR keeps the same or gets better when you increase ISO. Canon sensors loss less DR per ISO step than sony sensors, in sony sensors you lose almost one step (a bit less) per ISO step. But canon sensors have less dynamic range at base ISO than sony ones. Sony sensors produce less electronic noise in reading pixel data, and that is why they have more DR at low ISO. But as they almost aproach and ideal sensor with no electronic noise, that implies that you gain almost nothing when you use an analog gain to amplify signal before reading and converting it to digital (that is what ISO is, an analog amplifier). Thus in Canon you get less dynamic base and more noise when you boost shadows, but the DR gap gets narrower when you boost ISO settings. In both cases the DR is less when you use a higher ISO. Several years ago, canon had the best sensors with more DR than its competitors. When sony introduced the sensors with backlight technology, they got a DR boost. A Canon sensor two generations old, had a DR more than two steps lower than a sony equivalent a base ISO. But at high ISO (1600 or 3200) they surpased the sony sensors and had a bit mor DR than sony. The ultimate canon sensors have less than a step lower than sony's equivalent, as they produce less reading noise. But the DR lost pero ISO step is now greater too than in previous versions. You can see all these in DXOmark in the curves you get of DR against ISO, where they carefully measure the behaviour or each sensor at different ISO.
Thanks for the honest unbiased review. As a project manager in my working life I can't help but agonize over the many technical issues that are involved when bringing a new camera to market. I'm sure there are times when the manager asks the designers to improve one feature only to be told that improving that one will make another aspect of the camera worse. There must be many trade-offs. The same then comes true for the photographer who must consider their own photographic product and how best to achieve that product by selecting the right camera.
Kinda ridiculous how it's even worse than the 5D mark IV, but am I the only one who wasn't that impressed by the A7III? Just seems very overhyped if I'm being honest.
Stewart Gilligan Griffin isn’t it more like the A7iii is (very) good at many things but not the best at any? It’s just the combined package at the price that builds the hype?
justin, ergonomics, yes... everything else, no. How important are those ergonomics when you need a feature and the Canon just doesn't have it? Serious question.
Would love to see this review updated with the 1.2.0 firmware to see if banding is better. Using a 16-35 2.8 III (with adapter, obviously), in Lightroom if you select the lens profile the banding gets really bad (at ISO 6400). Don't know if that's a Lightroom issue or camera
Does Lighroom fully support the RAW-Files of the EOS R? Because the test assumes that it does. What about using Canon Software for processing the RAW-Files?
One thing that holds me from jumping:- LCD display. Does it at any point show all three (or more) readings at once? (or glance) just like the originals do, or one has to literally search one by one?
tony.. someone has to tell you this. you're doing it wrong. you need to DOWNSIZE the larger image instead of ENLARGING the smaller image. you just make the smaller image look soft because its not viewed at 100%.. you need to stop doing that. it doesnt really indicate anything. a lens cant resolve more than 100% of the sensor can it???? so please think about it.. it just makes more sense to downzise the larger image and THEN see if you can actually see more detail..
No way. If you're comparing across the board, that's the truest way to compare. Why would anyone downsize their picture to lose detail because your camera's direct competitor has a lower resolute sensor? Insane for a comparison video. Should we also make the Sony recover 6.5 stops to the Canon's 5 because it has bigger pixels? WTF? Hahahahaah
i understand Tony comparing each camera body to a given 30Mpixel-standard (which is no standard anywhere except in this clip). it might be ok if you upsized the Rs result's by 44% to the a7RIIIs resolution as well. it's never apples to apples this way, i'd rather call it an apples to cider comparison ;)
Why would you try to equalize them out of their native form when you're trying to compare their native capabilities? *scratches head* Should we post process the R's low light recovery pictures to make them more closely match the ARIII's too?
Yes, he's comparing a 30mp photo to a 24 mp photo and yes, the 30 mp photo obviously is sharper and has more detail, but that's the point. One camera comes with a higher resolution sensor one doesn't. Advantage the one that does. At least as far as detail is concerned.
Hey Tony, could you do some long exposures with the cameras to see if that noise is evident? I'm guessing that low-light noise on the EOS R is very noisy since the 5 stop test was bad. A long exposure would compare to a 5 stop amplification in post I imagine. 👍
I just purchased an a7RIII and am not in the market for a new new camera but I always learn something new from watching one of your videos.Thanks Tony!
It could be super interesting to see comparison on colors. How do you differ. Shooting trees, shooting faces, shooting cars etc. So it's not only about sharpness since that will never actually be seen for professional work unless you do heavy cropping. It's more up to the general visual perception of how the camera performs.
Tnx Tony, Canon is killing me; I want dynamic range that is better that D850/7RIII sensor. Come on Canon you've been milking this sensor since Sept 2016 and it wasn't that great to start with. I'm a Canon shooter for 20 years but this is now getting silly. Lets hope the pro-R version comes with a better sensor or for a 5D mark V, a 7D mIII, or a 5DsR m II. I don't care about the mirror I just want a better sensor.
I know I'm tempted but it means switching all my lenses to the Nikon house and it will not be cheap. also if I'm to jump ship I may as well do it with SONY. I've been impressed with some of their lenses; much more so than the Nikon equivalent offering. And I think the A7RIV is not far away in the horizon that will decimate whatever Nikon offers plus it will make the RIII much cheaper.
If I was a Canon user I would stick with my mature and dependable DSLRs and rich lens ecosystem and probably wait a couple of years for Canon's mirrorless line to mature.
I agree. I had the MK2 for many years which served me well. Till I upgraded to the 4 this year. Probably wise to wait till the R series is on par or better than Sony's offerings
I have a question: I’m thinking of buying the EOS R BUT I want to use my lenes from my canon 80d... I know the R is a full frame and the 80D is not! It’s possible that I will need some sort of adapter, if so, will I lose any resolution or clarity when shooting video if I use a adapter? Thanks in advance!
Is it really? I mean how often are you recovering 5 stops of exposure? 1-2 is a lot for most of us... I find the lack of IBIS and dual card slots to be a much bigger deal.
Don Schiffer that’s the thing. Without he extra 3 stops you can get from the ibis, you might end up shooting faster shutter speeds and having to recover more...
Maybe I missed it...Is there a test list, and a sentence or 2 to what parameters used? I couldn't tell if you used the same lenses, or if the format was RAW or something else(Of course I would imagine raw, as that is what you have to download). But still worth clarifying and adding to the description..
Concerning your video quality: even if you film in 1080p, try to sample the video in 4k and upload as such to TH-cam. It will look much better because firstly the Lightroom examples will be in 4k and secondly TH-cam won't nearly compress as much as in 1080p. So even normal 1080p footage will look better on TH-cam if uploaded in "4k."
EnormousSmartass its so funny, that most people don’t know that. In fact, there is almost no difference to a 1080p video uploaded in 4K as compares to the native 4k upload. This means: all youtubers should should in 1080p and upload in 4k in my opinion
I guess I was too early watching it. :) I'd like to use this chance to ask you what lens you used for the A7III and especially why you did not use the same glass on both, the Canon and the Sony. The A7III looks weirdly blurry, as if the lens just isn't that great.
Thanks for this in-depth look! These are the types of reviews that really help in making purchasing decisions. Love the entire channel, you and your partner have really helped jump start my adventure in terrestrial photography!
My thought as well. I have a feeling that if we look closely, the words "Coca Cola" may be imprinted on the front of the lens - the images sure looks like it...
Appears to be the 24 70 gmaster. The issue is that he’s going so deep into the images (400% or so) that you actually see the low pass filter results. In a real scenario I’d say they would be indistinguishable.
David Neto, that makes sense - I forgot that the A7 III had a low pass filter, and you are probably right, it really takes a lot of the bite out of the Sony at that kind of magnification.
For the color banding, wouldn't you want to test it in different programs than only Lightroom? Adobe may need to update their software to be able to account for the new camera, right? Maybe I'm wrong.
It's a total failure in my book. This now proves how far Sony is in front. Canon are a total failure now and I wouldn't be surprised if in 10 years they are like Pentax of today.
I, being a canon person, believe they will catch up somehow, the Nikon Z7 isnt ready to take on sony either but those guys make the fkg D850! Canon cant keep up but they know how to make the 1dx2. Hopefully, this becomes a battle of giants that we can all profit from.
I respectfully challenge that the A73 image used for comparison at 4:54 was captured with an "out-of-calibration" lens (internal optical elements not properly aligned) and NOT a representation of 25mp vs 30mp. Notice the ghosting/doubling of the lines. Did you delete the "pinned comment" that you referenced? I cannot find it.
Chelsea & Tony, do you by any chance have an ETA for the full review of EOS R? Preorders are open and shipping begins in a couple of weeks, yet it is as if there is some sort of standing gag order out there...
Thanks for featuring the Argus. Used nearly that exact model in college and still have it--but mine had the conical front on the lens, perhaps a year or so newer. Had to dismantle and clean the range finder system in the basement of the library once...thought for sure I'd lose a spring somehow. Darkroom was under the staircase in the gym. Good times...thanks for the memories; keep it on the shelf! That said, if they're worth $2,300, I might trade for the R. (Now, where is that top-mount light meter...)
The title says "Image Quality" guys, so lets not go off-topic about the card slot, no IBIS, weak 4K, $200 dollar USB charging option, etc etc lol etc, lets just stay on image quality for this video. So with that yes the Canon does have a 6mp advantage over the A7iii, as well as um um umm...um....yeah.
Yeah, I agree.... I think this is great news for Canon shooters from an IQ perspective. It's everything else that's a disappointment. But hey, not everyone cares about all of those features. So if you're a 6D-type customer... then this looks like a great camera for you.
Interesting review, however I would like to say that many of your shots of the a7-III appear improperly focused or have a lens with sharpness issues. The 100% views are very unsharp and some even exhibit typical 'micro bokeh' out-of-focus-ness (4:59) on the lettering. I own the a7-III and am extremely picky with sharpness and have never seen any images so blurry as these - not close-up and not portraits. There is an obvious resolution advantage for the Eos R but the A7-III is by far not as blurry as yours. Consider re-shooting with a different lens?
Which lenses dit you use? Would be good to mention. I did not expect the EOS R to outperform the Sony a7iii like that, interesting. Are you also going to do a video comparison?
Amazing the brains of Canon photographers... if you don’t know how DR works for others then you should trade yours in, because DR and recovery in post isn’t about correcting mistakes SMH!
Nice Made. Im Surpised the EOS R has more banding problems than the EOS 5D MK IV. Is it shot with a 2.8 Lens? If it is, wouldn't you get more sharpness out of the Lens ar f 4.0 oder f5.6 to test the Camera. It deosn't matter so much, just wanted to mention it
Please do a 160 vs 200 base iso when you get the XT3 vs XT2. I want to see what difference the revised base ISO makes. I also want to see the XT3 vs a comparable full frame camera at base ISO (say 6d2). Thanks Tony
I am that kind of people. I shoot landscape and astro. When facing camera directly to the sun at sunrise/sunset , to make sky correct exposure, all the foreground and subject will be very dark (-2 to -3 stops) Believe me, you dont need sky to be over-exposed because it is really harder to be recovered than under-exposed foreground. Then in Lightroom/Photoshop I normally add an extra local adjustment like radial adjustment. It adds up 1-2 stops. Then dodging and burning for detail adjustment adds another 0.5 to 1 stop. For the astro work, unlike stars and milky-way, all foreground elements seem to be dark. Sure we can work on light painting foreground, but not all the time. Recovering dark shadow from night shot is always happen.
Tony, question. What was the lens on A7III. I see in Canon L lens but on Sony looks like 50/1.8 basic lens which is soft. Honestly i have never seen before such a soft output on A7III before. Was not the difference due to cheap lens on A7III?
Ethan Deshaies I’m so damn tired of people who can’t shoot but can read a spec sheet and think they’re any good. So by your theory, there has never been any real art or good photographs before 2018’s dynamic range ??? Go out and make something great
What lenses did you use on Sony a7iii/Riii? I believe that sharpness can be defined by the lenses you use not only from the sensor's capabilities. I would want to believe that the a7iii can produce sharper images with Zeiss or G Master lenses.... any thoughts?
Hello T&C, I enjoy your videos. My question is about the EOR R and lens compatability. I have a 5DMk2, and I have a Tamrom 70-200 lens, a Sigma 85mm 1.4 ART, and two Canon L lenses. I know that the L lenses should work fime with one of the EOS R adapter rings, but what about third party lenses with the Canon EF mount on them? Can I expect third party (Canon mount) lenses to work on the EOS R just as if they were on a Canon EF mount camera? If I buy an EOR R, I don't want to lose use of third party lenses, especially since I will keep my Mk2 as a second body. Any thoughts?
Tony, please could you tell us how the R performs when shadows are pushed only a couple of stops. I'm assuming there's a fix for the banding you showed. Sorry wasn't clear. I mean I'm assuming the banding can be fixed in post processing software. I know this would be a pain but surely there's only so many images you'd cause this by extreme post in-camera exposure adjustment.
The sharpness tests are useless because you don't tell us which lenses are used and whether they are comparable in sharpness. I suspect that it was an unfair prime lens on the Canon vs. zoom lens on the Sonys.
It was the same lenses. You can download the raw files. Here's a stock comment I've been repeating: The a7 III is in focus. As a control I have objects in front of and behind the focal plane which allows me to verify consistency. I reshot with both the Canon 70-200 f2.8 III (via adapter, the same lens used on the Canon bodies) and the Sony 70-200 f2.8 GM just to make sure the lens wasn't the issue. Also, the a7R III used those same lenses and it's far sharper. The a7 III just seems to have a heavy AA filter.
30 Mp can't be THAT more sharp than 24 mp like in the test. It's not conform with my experiences and reviews from other photography websites. It looks more like 10-12 a megapixel difference.The issue is maybe: sharpness and noise reduction settings (on camera or lightroom, how does Lightroom open the raw-files of the Eos R by the way), dirt or fingerprints on the lens, motion blur because of shake, lens adapting issues... you have a whole video on this issue called "flaws in technical camera reviews". Something is wrong and you have to find out what it is.
The fact that you have you usually have to do 5 stop shadow recovery and zoom in to tell the difference means that if you already have a Full Frame camera, you probably don't need to go mirrorless. Unless you just enjoy spending $2,000+ on camera junk LOL.
Personally, I think that looking at the trajectory is more important, Sony will be the best for the most people getting into photography today. Unless you are on a budget.
Tony and Celsea literally no one else is doing what you do on youtube. You guys are bar none the best. Never mind the haters. Keep doing your business and making the best content. Thank you.
Sony underexposed images looks so soft looks like they're using noise reduction or something for underexposed images to make it look better. I hope sony don't play games like Volkswagen with emission test :)) Canon only have extra 6Mpx so sharpness should be roughly equal..
But the point is not Canon being sharper. That can be true. The point is that the A7 mkiii image seems taken with a Bencini Comet (Google if you don't know what it is). I mean, it's not possible.
Tomas Ramoska - Nah, it's not soft. 6Mpx is huge. Compare Sony A7R (36Mpx) with Sony A7R III (42Mpx), only 6Mpx difference but A7R III has more detail and noticeably sharper image when you pixel peep.
The a7 III is in focus. As a control I have objects in front of and behind the focal plane which allows me to verify consistency. I reshot with both the Canon 70-200 f2.8 III (via adapter, the same lens used on the Canon bodies) and the Sony 70-200 f2.8 GM just to make sure the lens wasn't the issue. Also, the a7R III used those same lenses and it's far sharper. The a7 III just seems to have a heavy AA filter.
I dunno man, I've played with an A7Riii and own an a7iii, and these photos seem consistent with my experience with the cameras. The a7iii is definitely a tiny bit soft, and it's really my only complaint with the camera; it's otherwise amazing. I don't think I'd even care if I hadn't played with the a7Riii. Whenever I need to make a significant crop, I'm always a little sad I didn't pony up for the a7Riii. I do think absolute sharpness is overrated(most of the time), though, and, in cases with only minor cropping, the a7iii is more than sharp enough.
Could there be some chromatic lens flaws ? Canon glass does have a reputation of very high quality , but could there be something wrong with this copy? to me the R system is not ready yet for birds in flight/wildlife, 3 fps at RAW, 30Mp a bit low if you need cropping, the z7 can do 8fps, 14 Bits RAW , 45,7 Mp (D850 like DR? / pixel quality?) no viewfinder blackout , IBIS ,looks more like it... but what will the autofocus be of the z7 and a 500mm?
Umm, what? "terrible" sensor?? The only thing bad about it is the banding, but who is going to be recovering close to 5 stops of shadows regularly? Maybe it's a technical issue that's causing it since it's still "pre-production" after all. It's literally Canon's flagship sensor, far from terrible. What's your idea of a fantastic sensor then?
@@wesleylee99 the glass makes a huge difference, i always said that canon L glass is on of the best besides Fujinon lenses. Would have been interesting to see a camera without AA filter in this comparison too.
Tony, You show 5.5 stops recovered is bad. How many stops can you recover on the R w/o noise banding? I almost never need to recover shadows on my mark 4 but when I do its maybe only 3 stops. Good job on the quick update! Thanks! -Mike.
@Tony & Chelsea Northrup - two questions, one I asked before but didn't get a response (totally understand, btw). 1) did you use the same lens on all the cameras to eliminate that variable. 2) did you convert the R raws to DNG and then adjust exposure or just open the raw files and begin working with them? Thanks!
Should shut Sony fanboys up a bit, sensor, sensor, sensor blabla. And you didn't even mention the far supérior lenses and the fact the Canon isn't cheaply built in China!
Dont make us laugh, Sony glass are way more expensive than Canon L glasses, and if they was that great for the money, Sony user would sell their Canon lenses to buy native Sony lenses. But what do I see most of the time??Yes, adapted Canon L lense. Why ? Simple Canon EF system is just the best for what it cost and Sony lenses are just not competive. you are just lucky that Sigma now make sony FE lense
@@cyrildu972 Look at dxo Lens performance reviews and ratings chart. Also some L glass is more expensive than the equivalent G master Sony lens with the sony being better in some cases. People have adapted glass because they most likely came from a cannon ecosystem. People new to cameras also aren't buying canon lenses to put on their Sony bodies very often.
Hi Tony, I was wondering if there will ever be a mirrorless alternative to the D500 or 7D's. Basically a fast outdoors crop sensor body that allows you to get closer to wildlife, much more than a full frame, is that something we can expect from the mirrorless world??
Why Tony has to apologize any tiny flaws in these excellent and informative videos? I really appreciate and respect the hard work your team put into these videos.
HELP after watching this I am not sure what to do!!! I am committed to Canon (I have plenty of glass). Tony and Chelsea I am a landscape, nightscape photographer. Would you recommend the 5dM4 or the EOS R. I concerned about DR which is very important in my landscape work. Thanks
Hi, great video Tony. I was just wondering, could the difference in the low noise color bands between the EOS-R and the 5D Mark IV be due to possibly having different default Picture Style Settings or Custom Functions (C.Fn) or even both internally?. Meaning that even though they might be the same for, lets say Contrast, usually a default value of 0, if it is changed to -1 or +1, if they match better then that means that the camera internally has different values in the processor? In the processor as they both according to Canon have the exact same sensor, but different processors. Just a guess, if so, then maybe a Custom Picture Style or Custom Function (C.Fn) could match both cameras if a person shot using both of them during some event or something. If so it would be great if Canon put out some info on matching these settings with current DSLR's :-) Again, just an idea or maybe a guess, no need for anyone back lashing in the comments LOL
Seems to me the Eos R is a good still camera but I’d be interested if you did the same test for video. All those other cameras would leave it for dead in terms of video. Also you mentioned comparison to the A7riii and how people might take issue with this. The A7riii is probably about the same cost as a5d3 here in nz so it’s a fair comparison.
"When pushed 5.5 stops" - I'm far more interested in what happens when pushed 2 stops. Do these artifacts still show up? Who is really trying to recover over 5 stops on a regular basis?
Incompetent photographers that don't know how to expose properly and want the camera to do all the work for them.
But just HOW incompetent? Is there a full-auto mode in existence which could be that far off? I'm pretty sure my Minolta from 1992 wouldn't have been that bad
Wanako Plays! Such a silly comment. Just making excuses for canon and there is definitely situations where having the option is great.
Blending landscapes has been happening for decades and the new nikons and sonys really help cutdown on the need to blend. I'm sure it is also helpful for concert and other high DR situations.
There is times a single shot will never get the desired effect but having great shadows really helps.
It has nothing to do with incompetence.
Higgins2001 well being that it’s a Mirrorless camera and you can see the exact exposure and DOF before you take the shot. There really shouldn’t be any reason for anyone to be under or over by 5 stops.
Actually in mirrorless is soooo hard to miss an exposure for more than 2/3 of stop due the real time exposure in EVF
“I CAN, I can do whatever I WANT” lol love it 😅
"Don't Tell me what to do!". I had to rewind and watch again. Hahaha
It was great.
Tony's version of "Come at me, bro!"
Whenever the dweebs in forums goes "Oh you cant compare this to this because of blah blah." I read it in my head with the Comic Book Guy's voice.
Quentin Tarantino presents 'How The Test Was Won':
Tony N: "I can compare it to whatever I want, I can compare it with this goddamn hand-canon, mutherflipper! Do you want me to compare it with this hand-canon? Well, do ya'?"
Matt G: "Say "only one card slot" again. I dare ya, I double dare ya, mutherflipper!"
Chelsea N: "You know what you guys look like? A coupla' dorks."
Kai: "[censored]"
Lok: "[inaudible]"
Download sample raw files (converted to DNG for your convenience): sdp.io/eosrsamples
Rent cameras and lenses at lensrentals.com, coupon code 'northrup15' for 15% off!
Canon EOS R: sdp.io/r
Canon 5D Mk IV: sdp.io/5d4
Canon 6D Mk II: sdp.io/6d2
Sony a7 Mk III: sdp.io/a73
Sony a7R III: sdp.io/a7r3
The a7 III is in focus. As a control I have objects in front of and behind the focal plane which allows me to verify consistency. I reshot with both the Canon 70-200 f2.8 III (via adapter, the same lens used on the Canon bodies) and the Sony 70-200 f2.8 GM just to make sure the lens wasn't the issue. Also, the a7R III used those same lenses and it's far sharper. The a7 III just seems to have a heavy AA filter.
FYI, I tested it with and without IBIS and with and without lens IS (there was no difference). You'll see a Post-It in one of the shots keeping track of which was which.
Tony, your simple, honest, but yet powerful review is always better than some other review that is so long but just turn round and round without any clear summary. Really glad to have you in TH-cam. Thank you so much.
I felt your 6D Mk II samples were clearer/sharper than the EOS R samples. And any slight (it'd have to be very slight) shadow recovery benefits of the EOS R over the 6D II is ruined by that old original digital camera era banding o.o
That's lateral chromatic aberration at 6:00 and 7:20, not moire. Moire happens in areas of high frequency color information, not where you have a single high contrast edge.
And software deals pretty well with this artefact. With a Canon lens mounted on a Canon body, I'd expect that the in body image converter does this too. Probably the option wasn't set ?
if that is ca why does pixel shift clear it up?
By the way, what is the impact of using pixel shifting when it comes to chromatic aberration?
@@shang-hsienyang1284 i think i still dn get what is benifit of pixel shift we also can stack images to increase image quality
none as far as I know
Hey Tony, the A7III is out of focus or something wrong with the lens.should be super sharp
Even from this brief comparison it is evident that all modern cameras are pretty darn good. Really, who in good mind is going to salvage a 7 stops underexposed image and complain about the noise they were getting? There's simply no bad camera in production anymore. Now we've reached the point where the entire ecosystem should be compared, not body to body or lens to lens.
So personally I'm staying with Nikon, but I'm not going to shout Canon or Fuji people crazy. Today the gear is not anymore important the way it used to be some 5 years ago.
Now it's all an upper-OK class or better.
I hope I win a Canon EOS R.
Then I can sell it and buy the A7III.
Haha, we have the same though process when it comes to canon camera competitions XD
Thats what I thought 😃
Except nobody is gonna buy it
I want you to win.
Lmfaoooo
If I remember correctly during the 5d3 vs Nikon D800 era...the Canon handled Overexposed images better where the Nikon handled the Underexposed images better. I wonder how that applies to the R vs Sony world. But to me the most important deciding factor of the Nikon/Sony/Canon mirrorless battle...which body is the most durable and with the best weather seal? I live in Alaska. I read the R has better weather sealing than the 5d4. That would be very impressive. I've rented Sony Full Frames and found them excellent..but they were not durable enough for what I do...and that has held me back from purchasing one.
Yea. I own the A7r3 and it is not a camera you want to be careless or in the wild with. I haven't tested weather sealing (!) but what irks me is the battery hatch.
*Edit* Otherwise it's a effing great camera if you have the lenses you need.
The R has "worse" weather sealing than the 5Dmk4 from what I've seen, but still better than the Sony cameras. Something like being equivalent to a 6Dmk2
I agree Cannon recovers highlights better. I always expose to the right
Incredibly useful comparison, and concise too, as usual best on the net at least in my experience. You really earned that million subscribers. Having NO long introduction is SO WONDERFUL, as is just saying "thanks, bye" at the end. (Of course now my A73 goes into the toilet .... ;-) )
Art, I'll take that crappy A73 off your hands ;)
But yes, great video. Tony, please compare to the X-T3. I realize there's a huge difference due to sensor size, but I'm curious what that means in the real world.
Art Altman q
@SwitchRich Except, actual users are questioning the so-called banding. See www.dpreview.com/forums/post/61778113
Thanks Tony! I'm really happy for Canon for putting such a good and well balanced camera out. *A small question*, how how did you process the EOS R raw images? does Lightroom / ACR / anything else supports them already? The banding I see in the images can come from image processing not from the sensor (see it all the time in my work). How much are you sure that the software processes the EOS R raws correctly? Maybe some sort of Adobe's patch might fix that?
IBIS on the a73 though probably more than makes up for the 6mp outside the lab.
Barnabas Mackay and also means in a real world it can use much lower ISO cos you can get a slower shutter speed. Why would they not have IBIS in 2018
As the laws of physics are non-optional the more moving parts any camera has the more it’s subject to failure in longevity. Hence Canon's reluctance for IBIS. PS don't shake any camera with IBIS.
what happens?
thefailingstudent the sensor shakes. The sensor actually needs to move in order for IBIS to work.
Say z6 will beat a7iii so what excuse u will put then :P with IS lens its a non issue
I owned the 5D Mark IV for a couple years, then sold it and got the R and this represents my experience exactly. The 5D produced the best image of any camera I shot with in 15 years - even at high ISOs and even when files were pushed and pulled. The R was much more fragile, so any editing of files at moderate to high ISO's would produce shadow noise and banding and colour shifts.
If you only shoot daylight exposures (and dont need to pull up shadows) or in studio, the IQ of the R is great! But beware as the IQ is not the same as the 5D Mark IV.
Some mirrorless have this noise/banding issue because of the way the AF is attached to the sensor. The z6 has similar issues, and the A7III has an issue in backlit situations.
It looks like in 2020, the R5 and R6 have fixed this with amazing image quality - arguably the best in their class. But, most youtubers only care about video specs and headlines.
The A7 III images looked more like they were shot with a poor lens (or out of focus) rather than being less sharp due to lower MP.
a7iii and the g-master 24-70 2.8 www.ninirockxphotography.com/911Memorial/i-kTNshKh/A.
They could have mounted the same lens (with settings fixed) on each body and kept each sensor at the same position--maybe allowing some slight position adjustment to account for cover glass differences.
Nini RockX some mint shots there bruh...
Fantastic set of images, Nini!
so true ...thats y i didnt still buy that 24-70 g master and 70-200g master.. still not impressed with sony's 2.8... canon has the best 24-70 and 70-200 out there .. he should mount the same 24-70 canon to both cameras..if he wanted a fair review ..my a7III + 24-70 f2.8 or 70-200 f2.8 is effin rezor sharp... side by side with my 5ds ..they are the same even if i enlarge to 50mp the a7III shot
How does the EOS R compare with the 5DSR, purely in terms of image quality and dynamic range? I recollect you guys saying that purely from an image quality perspective, you thought that the 5DSR is better than 5D-IV. Which of these three - 5DSR, 5D-IV, Sony A7Riii and EOS-R do you think works best for a landscape shooter, shooting primarily off tripods, usually at ISO < 800, with image quality being the biggest criterion?
The a73 pictures are either out of focus, or there is an issue with the lens or the camera. The sensor in the EOS R is the 5dm4 sensor, you can compare that with the a73 on any site, DP Review being the best interface, and see they are comparable in sharpness and resolution. Image Resource also has a comparison against the 6d Mark II, which in your video the 6dm2 even blows the a73 away www.imaging-resource.com/PRODS/sony-a7-iii/sony-a7-iii-image-quality.htm
It's because of the Canon glass he's using the 50 1.2 is probably the best and sharpest 50mm out right now other then maybe a few manual 50mm
Canon's got better glass than Sony.
Wy Lee I agree, Canon has the best glass and I bet that 50mm is incredibly sharp. But that doesn’t explain why the 6dm2 is also noticeably sharper. There is no data to support the 6dm2 has a far superior sensor than the a73.
Tony - I was just wondering if these are the actual files they are going to release out of camera or is that just a pre production file as well? Thanks for the review, kinda unexpected that the quality was better than the A7III.
Hey Tony! What lenses you shot with? I'm kinda curious about eliminating glass as a variable between the EOS-R and the A7iii.
Also when are the giveaway winners announced? And thank you so much for running that!
UPDATE: I looked at the DNG files and I think the A7iii is using a different lens from the A7R3. The A7iii is using the Sony GM 70-200 which LensRentals found to be soft.
maybe he used the metabone or sigma mc11 adapter and simply used the very same lens
Vince Ha - maybe - we don't know that lenses were used and we can not trust these videos to show what the cameras actually can do.
I trust Tony, and I trust the test to show what it shows when you account for variables. I just like to know what I'm looking at and I think Tony did a good job of explaining everything else.
I agree with the questions in regards to the variables of this experiment. When Canon is shot with native mounted lenses while Sony is using an adapter does not quite qualify as an apple to an apple comparison. Since many reviewers had already made direct comparisons between the 5D IV to the a7rIII, I don't see the reason for the EOS R vs a7III instead of strictly to a7rIII. The cost factor to me is very lame...
I wouldn't necessarily be concerned about a Metabones being used to make sure every camera uses one piece of glass...I would be more concerned if it was a prime vs zoom test, or a set of lenses that vary in objective quality by some degree (a cheaper 50mm vs a high end pro 50mm and so on). In a static, well lit, tripod-shot kind of situation, I'd expect the adapter to be more or less invisible for this kind of test--maybe not in all other tests, but certainly in this one.
Eos R now has Firmware update wich completly remove the banding issues,it will be interesting if you make another video comparing Eos R with the latest firmware vs the 5D mark iv dynamic range/shadow recovery.
Looking forward to an Ultimate Showdown..
A7iii vs Eos R vs Z6 vs X-t3
😵🤯
I want to see that too.
Same! Im excited to see how the x-t3 holds up!
Nikon z6 will smoke each n every one
@@senaritradutta Probably, at least in terms of image quality. But theres a lot more to it then that
...vs GFX 50R :)
Guys, GREAT stuff here! What a nice well rounded initial review mate.
Canons sensors have always been rough on shadow recovery.
Esp when folks way under expose their photos.
Canon sensors of the past had not been great on dynamic range, which means that you often did end up with shadows that needed recovery unless you wanted to completely blow out the overall scene. In fact, the 5D MKII is the camera that really made me embrace silhouettes and dramatic contrast!
You are too young to have memory, I can guess. Canon sensors have been for many years the best in dynamic range, that make Canon a best seller in digital photo. It took sony several generations to beat them and surpase them. Last generation of Canon sensors have closed the gap in dynamic range and are much more close to sony's equivalents. When you elevate ISO the gap in dynamic range gets narrower. The dynamic range of canon sensors at high ISO is better than sony, as canon sensors loss less DR pero ISO step than Sony's.
For that reason it is much more important not to underexpose in Canon than in Sony. If you get an image 1 step under exposed (you used ISO 800 instead of 1600) and you correct at developing stage, you will get more noise in shadows than if you do the same with sony.
Canon sensors have always been rough. Full stop. Their lenses are rubbish too.
The dynamic range of any camera degrades when you increase ISO, there is no camera where the DR keeps the same or gets better when you increase ISO.
Canon sensors loss less DR per ISO step than sony sensors, in sony sensors you lose almost one step (a bit less) per ISO step.
But canon sensors have less dynamic range at base ISO than sony ones.
Sony sensors produce less electronic noise in reading pixel data, and that is why they have more DR at low ISO.
But as they almost aproach and ideal sensor with no electronic noise, that implies that you gain almost nothing when you use an analog gain to amplify signal before reading and converting it to digital (that is what ISO is, an analog amplifier).
Thus in Canon you get less dynamic base and more noise when you boost shadows, but the DR gap gets narrower when you boost ISO settings.
In both cases the DR is less when you use a higher ISO.
Several years ago, canon had the best sensors with more DR than its competitors.
When sony introduced the sensors with backlight technology, they got a DR boost.
A Canon sensor two generations old, had a DR more than two steps lower than a sony equivalent a base ISO.
But at high ISO (1600 or 3200) they surpased the sony sensors and had a bit mor DR than sony.
The ultimate canon sensors have less than a step lower than sony's equivalent, as they produce less reading noise.
But the DR lost pero ISO step is now greater too than in previous versions.
You can see all these in DXOmark in the curves you get of DR against ISO, where they carefully measure the behaviour or each sensor at different ISO.
Thanks for the honest unbiased review. As a project manager in my working life I can't help but agonize over the many technical issues that are involved when bringing a new camera to market. I'm sure there are times when the manager asks the designers to improve one feature only to be told that improving that one will make another aspect of the camera worse. There must be many trade-offs. The same then comes true for the photographer who must consider their own photographic product and how best to achieve that product by selecting the right camera.
Kinda ridiculous how it's even worse than the 5D mark IV, but am I the only one who wasn't that impressed by the A7III? Just seems very overhyped if I'm being honest.
Stewart Gilligan Griffin isn’t it more like the A7iii is (very) good at many things but not the best at any? It’s just the combined package at the price that builds the hype?
the fact is in real life all modern sensors are very very close!
@@DriveCancelDC you are forgetting ergonomics, build quality, made in Japan vs China, weather sealing, resell value, lens quality.......
justin, ergonomics, yes... everything else, no. How important are those ergonomics when you need a feature and the Canon just doesn't have it? Serious question.
@@brois841 I can't honestly think of a single feature that my 5d4 and lenses can't fulfill for my needs
Would love to see this review updated with the 1.2.0 firmware to see if banding is better. Using a 16-35 2.8 III (with adapter, obviously), in Lightroom if you select the lens profile the banding gets really bad (at ISO 6400). Don't know if that's a Lightroom issue or camera
Does Lighroom fully support the RAW-Files of the EOS R? Because the test assumes that it does. What about using Canon Software for processing the RAW-Files?
One thing that holds me from jumping:- LCD display. Does it at any point show all three (or more) readings at once? (or glance) just like the originals do, or one has to literally search one by one?
tony.. someone has to tell you this.
you're doing it wrong. you need to DOWNSIZE the larger image instead of ENLARGING the smaller image.
you just make the smaller image look soft because its not viewed at 100%.. you need to stop doing that. it doesnt really indicate anything. a lens cant resolve more than 100% of the sensor can it???? so please think about it.. it just makes more sense to downzise the larger image and THEN see if you can actually see more detail..
my thoughts exactly. megapixels count for cropping & printsize, not for upsized comparison in resolution. nice vid anyway :)
No way. If you're comparing across the board, that's the truest way to compare. Why would anyone downsize their picture to lose detail because your camera's direct competitor has a lower resolute sensor? Insane for a comparison video. Should we also make the Sony recover 6.5 stops to the Canon's 5 because it has bigger pixels? WTF? Hahahahaah
i understand Tony comparing each camera body to a given 30Mpixel-standard (which is no standard anywhere except in this clip). it might be ok if you upsized the Rs result's by 44% to the a7RIIIs resolution as well. it's never apples to apples this way, i'd rather call it an apples to cider comparison ;)
Why would you try to equalize them out of their native form when you're trying to compare their native capabilities? *scratches head* Should we post process the R's low light recovery pictures to make them more closely match the ARIII's too?
Yes, he's comparing a 30mp photo to a 24 mp photo and yes, the 30 mp photo obviously is sharper and has more detail, but that's the point. One camera comes with a higher resolution sensor one doesn't. Advantage the one that does. At least as far as detail is concerned.
Hey Tony, could you do some long exposures with the cameras to see if that noise is evident? I'm guessing that low-light noise on the EOS R is very noisy since the 5 stop test was bad. A long exposure would compare to a 5 stop amplification in post I imagine. 👍
No one comments on how the 1.2.0 firmware fixed the green banding? Typical...
I just purchased an a7RIII and am not in the market for a new new camera but I always learn something new from watching one of your videos.Thanks Tony!
Excellent comparative video, Tony! Yours is the gold standard against which all reviews should be judged!
It could be super interesting to see comparison on colors. How do you differ. Shooting trees, shooting faces, shooting cars etc. So it's not only about sharpness since that will never actually be seen for professional work unless you do heavy cropping. It's more up to the general visual perception of how the camera performs.
Tnx Tony, Canon is killing me; I want dynamic range that is better that D850/7RIII sensor. Come on Canon you've been milking this sensor since Sept 2016 and it wasn't that great to start with. I'm a Canon shooter for 20 years but this is now getting silly. Lets hope the pro-R version comes with a better sensor or for a 5D mark V, a 7D mIII, or a 5DsR m II. I don't care about the mirror I just want a better sensor.
Get the Z7 then. Latest Z7 evaluation shows it actually outperforms the D850 in texture and DR
I know I'm tempted but it means switching all my lenses to the Nikon house and it will not be cheap. also if I'm to jump ship I may as well do it with SONY. I've been impressed with some of their lenses; much more so than the Nikon equivalent offering. And I think the A7RIV is not far away in the horizon that will decimate whatever Nikon offers plus it will make the RIII much cheaper.
When is the 90d full review coming Chelsea and Tony!!??
I am so happy with my Canon 5D Mark IV now...
David Law Yep! It’s not time to switch to mirrorless yet.
If I was a Canon user I would stick with my mature and dependable DSLRs and rich lens ecosystem and probably wait a couple of years for Canon's mirrorless line to mature.
Sold my 5D4 and preorder R today
How much did you sell it for?
I agree. I had the MK2 for many years which served me well. Till I upgraded to the 4 this year. Probably wise to wait till the R series is on par or better than Sony's offerings
I have a question: I’m thinking of buying the EOS R BUT I want to use my lenes from my canon 80d... I know the R is a full frame and the 80D is not! It’s possible that I will need some sort of adapter, if so, will I lose any resolution or clarity when shooting video if I use a adapter? Thanks in advance!
That banding in recovered shadows actually is a really big deal...
Is it really? I mean how often are you recovering 5 stops of exposure? 1-2 is a lot for most of us...
I find the lack of IBIS and dual card slots to be a much bigger deal.
@@SketchTurnerZero it means that banding always is there, it only shows up if pushed so much.
May be try to properly expose the shot ...
Don Schiffer that’s the thing. Without he extra 3 stops you can get from the ibis, you might end up shooting faster shutter speeds and having to recover more...
Mr Hist so what? If I cannot see it in the final image all is good.
Maybe I missed it...Is there a test list, and a sentence or 2 to what parameters used? I couldn't tell if you used the same lenses, or if the format was RAW or something else(Of course I would imagine raw, as that is what you have to download). But still worth clarifying and adding to the description..
Concerning your video quality: even if you film in 1080p, try to sample the video in 4k and upload as such to TH-cam. It will look much better because firstly the Lightroom examples will be in 4k and secondly TH-cam won't nearly compress as much as in 1080p. So even normal 1080p footage will look better on TH-cam if uploaded in "4k."
This is uploaded in 4k (the screencast is 4k). YT is still processing it.
EnormousSmartass its so funny, that most people don’t know that. In fact, there is almost no difference to a 1080p video uploaded in 4K as compares to the native 4k upload. This means: all youtubers should should in 1080p and upload in 4k in my opinion
I guess I was too early watching it. :) I'd like to use this chance to ask you what lens you used for the A7III and especially why you did not use the same glass on both, the Canon and the Sony. The A7III looks weirdly blurry, as if the lens just isn't that great.
Thanks for this in-depth look! These are the types of reviews that really help in making purchasing decisions. Love the entire channel, you and your partner have really helped jump start my adventure in terrestrial photography!
Video 9:34min, posted 4 min ago, 2 downvotes. Classic!
maybe for sound quality? Or maybe just Jared Polin😂
Most likely for the intro regarding the review not being out
Damn Tony... What lens did you use for the A7III samples? And why in gods name jpeg?
What lens was attached to A7III?
My thought as well. I have a feeling that if we look closely, the words "Coca Cola" may be imprinted on the front of the lens - the images sure looks like it...
Appears to be the 24 70 gmaster. The issue is that he’s going so deep into the images (400% or so) that you actually see the low pass filter results. In a real scenario I’d say they would be indistinguishable.
Even at 400% that is too soft, I think he rented that for the test, just like a rental car it was probably dropped 10 times.
David Neto, that makes sense - I forgot that the A7 III had a low pass filter, and you are probably right, it really takes a lot of the bite out of the Sony at that kind of magnification.
That a7iii looks really bad if Canon can fix that bending issue
For the color banding, wouldn't you want to test it in different programs than only Lightroom? Adobe may need to update their software to be able to account for the new camera, right? Maybe I'm wrong.
"it's a 5D but bad. Please clap"
-Canon
It's also a thousand dollars cheaper
yeah people don't see the price point,
They want a 1DxII for 1000 dollars.
It's a total failure in my book. This now proves how far Sony is in front. Canon are a total failure now and I wouldn't be surprised if in 10 years they are like Pentax of today.
I, being a canon person, believe they will catch up somehow, the Nikon Z7 isnt ready to take on sony either but those guys make the fkg D850! Canon cant keep up but they know how to make the 1dx2.
Hopefully, this becomes a battle of giants that we can all profit from.
I respectfully challenge that the A73 image used for comparison at 4:54 was captured with an "out-of-calibration" lens (internal optical elements not properly aligned) and NOT a representation of 25mp vs 30mp. Notice the ghosting/doubling of the lines. Did you delete the "pinned comment" that you referenced? I cannot find it.
You can find the A7RIII for $2,500. So, it’s a wiser choice. I got mine from the official Sony store in Mexico City. For $2,400 USD.
Lame. In a month or two you can find the R used for $1900, then what? Let's compare apples to apples, people.
@@derekv6479 in one month you'll sell all the canon gear to jump Sony so...
@@hectorrivera2785 Yeah right. I like actually having lenses. Haha Hope you like your metabones. 🤢🤢🤢
@@derekv6479 Because Zeiss are not good lenses lol
Because I can buy a Zeiss 600mm, put a 2x on in and autofocus at f/11 on my A9. :| Bwahahahahahahah
Chelsea & Tony, do you by any chance have an ETA for the full review of EOS R? Preorders are open and shipping begins in a couple of weeks, yet it is as if there is some sort of standing gag order out there...
Canon simply forgot the DR issue.....and the IBIS....and the dual card slot....and the...and the...
BUT THERE'S A FLIPPY SCREEEEEN! XD
whoopty doo
True...
...and the fact they have strong competition
they didn't forgot to remove 1000$ from the price too.
Thanks for featuring the Argus. Used nearly that exact model in college and still have it--but mine had the conical front on the lens, perhaps a year or so newer. Had to dismantle and clean the range finder system in the basement of the library once...thought for sure I'd lose a spring somehow. Darkroom was under the staircase in the gym. Good times...thanks for the memories; keep it on the shelf! That said, if they're worth $2,300, I might trade for the R. (Now, where is that top-mount light meter...)
The title says "Image Quality" guys, so lets not go off-topic about the card slot, no IBIS, weak 4K, $200 dollar USB charging option, etc etc lol etc, lets just stay on image quality for this video.
So with that yes the Canon does have a 6mp advantage over the A7iii, as well as um um umm...um....yeah.
Native, adaptable glass...
I wasn't convinced. I would still take my A73 and 5D3 over the EOSR. The banding alone would disqualify it.
Yeah, I agree.... I think this is great news for Canon shooters from an IQ perspective. It's everything else that's a disappointment. But hey, not everyone cares about all of those features. So if you're a 6D-type customer... then this looks like a great camera for you.
Lul there's an option to get USB charging for 200 dollars? Kek
The best choice of glass with a fully articulating screen. Yeah i know it bothers you, ever wonder why your watching canon reviews.. Yeah ummmm
Interesting review, however I would like to say that many of your shots of the a7-III appear improperly focused or have a lens with sharpness issues. The 100% views are very unsharp and some even exhibit typical 'micro bokeh' out-of-focus-ness (4:59) on the lettering. I own the a7-III and am extremely picky with sharpness and have never seen any images so blurry as these - not close-up and not portraits. There is an obvious resolution advantage for the Eos R but the A7-III is by far not as blurry as yours. Consider re-shooting with a different lens?
glad i got the best choice.. a7r3
Which lenses dit you use? Would be good to mention. I did not expect the EOS R to outperform the Sony a7iii like that, interesting. Are you also going to do a video comparison?
5 1/2 stops of recovery in post., that's asking a bit much. Time to trade in the photographer, not the camera.
The blurry images from the Sony a7iii is pretty bad, No idea how Tony could say they were good.
Amazing the brains of Canon photographers... if you don’t know how DR works for others then you should trade yours in, because DR and recovery in post isn’t about correcting mistakes SMH!
highlight recovery bro.
Professor Hines' Choice bracketing bro
Yep, the banding is pretty bad for sure.
Nice Made. Im Surpised the EOS R has more banding problems than the EOS 5D MK IV. Is it shot with a 2.8 Lens? If it is, wouldn't you get more sharpness out of the Lens ar f 4.0 oder f5.6 to test the Camera. It deosn't matter so much, just wanted to mention it
"I do whatever i want"
WHAT A MAD LAD
DeathCoreGuitar you're the only one having feelings right now.
Please do a 160 vs 200 base iso when you get the XT3 vs XT2. I want to see what difference the revised base ISO makes. I also want to see the XT3 vs a comparable full frame camera at base ISO (say 6d2). Thanks Tony
That a7iii shot is never ever in focus.
Check the pinned comment
he got some money from canon
where is the pinned comment? I can't see it on to top of the comment section!
@@merlinalfonso6374 only a bad photographer needs 5 stops for editing...i never pushed over 1.5 🤣 usualy i go till 1 stop and thats it. So whatever...
Out of curiosity, what lens did you use on the A7III? Did you adapt whatever lens you used on the EOS R/6DII/5D Mark IV?
Interesting but sounds like it was filmed in a toilet
He did apologize for the bad sound quality. .
Oh right didn't hear .. lol
Dee it was a black screen with text at the start, how are u so imperceptive
Maybe Dee is blind.
So he didn't SAY - he put text - well I don't look at the screen every seconds - no need to at the beginning is there???
Great comparison. As a 5DIV user, wonder how close the 5D4 compares to the a7RIII?
Cmon, who recovers 5 stops?
Some people in some scenarios. If you don't, then DR doesn't matter to you and you can disregard that part of the test.
I am that kind of people.
I shoot landscape and astro.
When facing camera directly to the sun at sunrise/sunset , to make sky correct exposure, all the foreground and subject will be very dark (-2 to -3 stops)
Believe me, you dont need sky to be over-exposed because it is really harder to be recovered than under-exposed foreground.
Then in Lightroom/Photoshop I normally add an extra local adjustment like radial adjustment. It adds up 1-2 stops.
Then dodging and burning for detail adjustment adds another 0.5 to 1 stop.
For the astro work, unlike stars and milky-way, all foreground elements seem to be dark.
Sure we can work on light painting foreground, but not all the time.
Recovering dark shadow from night shot is always happen.
Stress test; it's indicative.
to protect highlights. and you don't need to pull up 5 stops to see the difference.
How do you think you do not need it? Daylight in the sky background To express all the clouds and buildings it is too more need
Tony you slipping men, no review on the pre release/rumors Panasonic full frame mirrorless?
6:20 Best part of the video. Laughed so hard at that. xD
Tony, question.
What was the lens on A7III. I see in Canon L lens but on Sony looks like 50/1.8 basic lens which is soft. Honestly i have never seen before such a soft output on A7III before. Was not the difference due to cheap lens on A7III?
If you have to recover 5 stops, there’s a bigger problem than a sensor. :)
I would love to be able to recover 50 stops for astrophoto!
Ethan Deshaies I’m so damn tired of people who can’t shoot but can read a spec sheet and think they’re any good. So by your theory, there has never been any real art or good photographs before 2018’s dynamic range ??? Go out and make something great
telemaq76 Hahahhaa me too
What lenses did you use on Sony a7iii/Riii? I believe that sharpness can be defined by the lenses you use not only from the sensor's capabilities. I would want to believe that the a7iii can produce sharper images with Zeiss or G Master lenses.... any thoughts?
Canon Sensor cant compete with Sony Sensor
are you blind? it looks far better than the A73
What?? It just smoked Sony.
justin holding no, canon lens are better, sony sensor is better
Have a look at 1dxmk 2 best in the world
Hello T&C, I enjoy your videos. My question is about the EOR R and lens compatability. I have a 5DMk2, and I have a Tamrom 70-200 lens, a Sigma 85mm 1.4 ART, and two Canon L lenses. I know that the L lenses should work fime with one of the EOS R adapter rings, but what about third party lenses with the Canon EF mount on them? Can I expect third party (Canon mount) lenses to work on the EOS R just as if they were on a Canon EF mount camera? If I buy an EOR R, I don't want to lose use of third party lenses, especially since I will keep my Mk2 as a second body. Any thoughts?
A7iii photos look really soft, softer even than 6Dmkii
That's because Canon's glass is sharper in this example...
Indeed, to the point I question if the focus was completely off
I’m pretty sure he’s zooming into more than 100%. My a7iii images are super sharp.
@@BenTroxell yeah he kind of explained that. Less megapixels so to make it same size you get poorer image quality.
take with a grain of salt :P A7III is zoomed in to 150%
IMO all sensors paired with good glass produce sharp images even on APS-C at 1:1.
Tony, please could you tell us how the R performs when shadows are pushed only a couple of stops. I'm assuming there's a fix for the banding you showed. Sorry wasn't clear. I mean I'm assuming the banding can be fixed in post processing software. I know this would be a pain but surely there's only so many images you'd cause this by extreme post in-camera exposure adjustment.
The sharpness tests are useless because you don't tell us which lenses are used and whether they are comparable in sharpness. I suspect that it was an unfair prime lens on the Canon vs. zoom lens on the Sonys.
It was the same lenses. You can download the raw files. Here's a stock comment I've been repeating:
The a7 III is in focus. As a control I have objects in front of and behind the focal plane which allows me to verify consistency. I reshot with both the Canon 70-200 f2.8 III (via adapter, the same lens used on the Canon bodies) and the Sony 70-200 f2.8 GM just to make sure the lens wasn't the issue. Also, the a7R III used those same lenses and it's far sharper. The a7 III just seems to have a heavy AA filter.
30 Mp can't be THAT more sharp than 24 mp like in the test. It's not conform with my experiences and reviews from other photography websites. It looks more like 10-12 a megapixel difference.The issue is maybe: sharpness and noise reduction settings (on camera or lightroom, how does Lightroom open the raw-files of the Eos R by the way), dirt or fingerprints on the lens, motion blur because of shake, lens adapting issues... you have a whole video on this issue called "flaws in technical camera reviews". Something is wrong and you have to find out what it is.
This was awesome. Great review! Looking forward to watching more reviews in the future.
The fact that you have you usually have to do 5 stop shadow recovery and zoom in to tell the difference means that if you already have a Full Frame camera, you probably don't need to go mirrorless. Unless you just enjoy spending $2,000+ on camera junk LOL.
Personally, I think that looking at the trajectory is more important, Sony will be the best for the most people getting into photography today. Unless you are on a budget.
Tony and Celsea literally no one else is doing what you do on youtube. You guys are bar none the best. Never mind the haters. Keep doing your business and making the best content. Thank you.
Sony underexposed images looks so soft looks like they're using noise reduction or something for underexposed images to make it look better. I hope sony don't play games like Volkswagen with emission test :)) Canon only have extra 6Mpx so sharpness should be roughly equal..
But the point is not Canon being sharper. That can be true. The point is that the A7 mkiii image seems taken with a Bencini Comet (Google if you don't know what it is). I mean, it's not possible.
Tomas Ramoska - Nah, it's not soft. 6Mpx is huge. Compare Sony A7R (36Mpx) with Sony A7R III (42Mpx), only 6Mpx difference but A7R III has more detail and noticeably sharper image when you pixel peep.
Hey Tony, Love the video are you going to be doing the video quality version of the same cameras?
Yeah
preeetttyy sure that a7iii was out of focus, that or that lens was terrible
The a7 III is in focus. As a control I have objects in front of and behind the focal plane which allows me to verify consistency. I reshot with both the Canon 70-200 f2.8 III (via adapter, the same lens used on the Canon bodies) and the Sony 70-200 f2.8 GM just to make sure the lens wasn't the issue. Also, the a7R III used those same lenses and it's far sharper. The a7 III just seems to have a heavy AA filter.
guess its actually softer...kinda shocked
_Here we see a wild Sony fan unable to cope with the fact that something might be better than their beloved Sony_
I dunno man, I've played with an A7Riii and own an a7iii, and these photos seem consistent with my experience with the cameras. The a7iii is definitely a tiny bit soft, and it's really my only complaint with the camera; it's otherwise amazing. I don't think I'd even care if I hadn't played with the a7Riii. Whenever I need to make a significant crop, I'm always a little sad I didn't pony up for the a7Riii. I do think absolute sharpness is overrated(most of the time), though, and, in cases with only minor cropping, the a7iii is more than sharp enough.
Results are results, buddy. Looks like it does. Time to upgrade or buy an R.
Could there be some chromatic lens flaws ? Canon glass does have a reputation of very high quality , but could there be something wrong with this copy? to me the R system is not ready yet for birds in flight/wildlife, 3 fps at RAW, 30Mp a bit low if you need cropping, the z7 can do 8fps, 14 Bits RAW , 45,7 Mp (D850 like DR? / pixel quality?) no viewfinder blackout , IBIS ,looks more like it... but what will the autofocus be of the z7 and a 500mm?
even worse than jared polins results, just a terrible sensor
I wouldn't say it's terrible. It's just not competitive.
Umm, what? "terrible" sensor?? The only thing bad about it is the banding, but who is going to be recovering close to 5 stops of shadows regularly? Maybe it's a technical issue that's causing it since it's still "pre-production" after all. It's literally Canon's flagship sensor, far from terrible. What's your idea of a fantastic sensor then?
@@FairlyUnknown Sony sensors are far better and more competetiive in late 2018. Even Samsung makes better sensors :-)
But wait, the Canon image is SHARPER.
@@wesleylee99 the glass makes a huge difference, i always said that canon L glass is on of the best besides Fujinon lenses. Would have been interesting to see a camera without AA filter in this comparison too.
Tony,
You show 5.5 stops recovered is bad. How many stops can you recover on the R w/o noise banding? I almost never need to recover shadows on my mark 4 but when I do its maybe only 3 stops.
Good job on the quick update! Thanks! -Mike.
Banding not noise.
The fanboys will go crazy.
ViveLa France Not Really, still old drama. Sony body is better, canon lens are better
@Tony & Chelsea Northrup - two questions, one I asked before but didn't get a response (totally understand, btw). 1) did you use the same lens on all the cameras to eliminate that variable. 2) did you convert the R raws to DNG and then adjust exposure or just open the raw files and begin working with them?
Thanks!
A7Riii. Wish I could have it.
I got sick of waiting for the A7iii to be available and end up finding an A7Riii open box for $2500
Why, will the A7R III make your photos better?
@@ChillWithMe_musics define better. I shoot landscape.. So the extra megapixels and sharpness is great to have with the A7Rlll.
why are the pics from the Sony A7R3 smaller, than the ones from the Canon? 42MP vs. 30MP
Should shut Sony fanboys up a bit, sensor, sensor, sensor blabla. And you didn't even mention the far supérior lenses and the fact the Canon isn't cheaply built in China!
justin holding agree, Sony camera cheaply built with cheap low res touch screen, cheap EVF and bad ergonomie
www.dxomark.com/category/lens-reviews/
Sony have great glass, unlike Nikon and expensive Canon L. Check the numbers.
Dont make us laugh, Sony glass are way more expensive than Canon L glasses, and if they was that great for the money, Sony user would sell their Canon lenses to buy native Sony lenses. But what do I see most of the time??Yes, adapted Canon L lense. Why ? Simple Canon EF system is just the best for what it cost and Sony lenses are just not competive. you are just lucky that Sigma now make sony FE lense
@@cyrildu972 Look at dxo Lens performance reviews and ratings chart. Also some L glass is more expensive than the equivalent G master Sony lens with the sony being better in some cases. People have adapted glass because they most likely came from a cannon ecosystem. People new to cameras also aren't buying canon lenses to put on their Sony bodies very often.
Did you get any feedback from Canon about the banding? A lot of people have put money down on these cameras shipping in October like myself.
But but but canon color science,!!!!!
Tony can you please try other raw converters? I really want to like this camera 😊
Hey Tony, just an fyi there are denoisers like izotope RXdenoiser that can get rid of clipping distortion in audio very easily.
Hi Tony, I was wondering if there will ever be a mirrorless alternative to the D500 or 7D's. Basically a fast outdoors crop sensor body that allows you to get closer to wildlife, much more than a full frame, is that something we can expect from the mirrorless world??
Yeah, I'm sure. You'll probably see an a6700-type from Sony that does that.
Why Tony has to apologize any tiny flaws in these excellent and informative videos? I really appreciate and respect the hard work your team put into these videos.
HELP after watching this I am not sure what to do!!! I am committed to Canon (I have plenty of glass). Tony and Chelsea I am a landscape, nightscape photographer. Would you recommend the 5dM4 or the EOS R. I concerned about DR which is very important in my landscape work. Thanks
Hi, great video Tony. I was just wondering, could the difference in the low noise color bands between the EOS-R and the 5D Mark IV be due to possibly having different default Picture Style Settings or Custom Functions (C.Fn) or even both internally?. Meaning that even though they might be the same for, lets say Contrast, usually a default value of 0, if it is changed to -1 or +1, if they match better then that means that the camera internally has different values in the processor? In the processor as they both according to Canon have the exact same sensor, but different processors. Just a guess, if so, then maybe a Custom Picture Style or Custom Function (C.Fn) could match both cameras if a person shot using both of them during some event or something. If so it would be great if Canon put out some info on matching these settings with current DSLR's :-) Again, just an idea or maybe a guess, no need for anyone back lashing in the comments LOL
These are raw photos (so picture styles don't matter) and the cameras have the default settings.
Seems to me the Eos R is a good still camera but I’d be interested if you did the same test for video. All those other cameras would leave it for dead in terms of video. Also you mentioned comparison to the A7riii and how people might take issue with this. The A7riii is probably about the same cost as a5d3 here in nz so it’s a fair comparison.
Will do video soon
Tony & Chelsea Northrup cool
Thank you... Helped me a lot in deciding between EOS-R and 1Dx Mark II
Looks like there is a slight defocus/shutter vibration visible in the first pictures from the sony a7III, or not!? at 4:04
Checked the pinned comment
i dont see any.