Ron Martinsen's Photography Blog: 5D

Showing posts with label 5D. Show all posts

Saturday, November 5, 2016

REVIEW: Canon 5D Mark IV - Better than expected

I'll admit that I was pretty disappointed when I heard that Canon was trying to push even more megapixels (30.4 or 6720 x 4480 - > 8k video) into the 5D Mark IV as that's not what users of this camera body needed. They needed better dynamic range and high ISO performance, not fewer pictures fitting on a memory card and concerns about blurry shots from a high megapixel camera. After all, the 5D series is the camera of choice for many event photographers (which I bundle as wedding, general events, and concert).

We have seen with the poor 7D and 70D that when you try too hard to win the megapixel marketing game that you end up with terrible image quality and poor high ISO performance with nothing really to show for it in return. We've also seen with the D800 and D810 that extra megapixels can also mean a greater chance of getting a blurry image from motion shake, so fearing the worst I ordered my 5D Mark IV with the 24-70 f/4L IS lens and the new 600EX-RT II flash.

Behold, this isn't a normal Canon

If you've followed my blog then you know that I often do shots in a cathedral not to push a religious agenda, but instead to test a camera in some very tough conditions. The shot below is one of my torture tests because to the human eye pretty much most of the ceiling appears black and to most cameras all but the highlights are lost. However, I hadn't seen results this good since the A7R II (image comparison) and the D750 (image comparison) so this camera got my attention.

f/5.6 @ 24mm for 1/125 sec at ISO 25600
I expected shots at this high ISO to suck, not rock like this

What's interesting here is that you can see all the way to the ceiling and the color tones are well preserved. This camera was supposed to suck at high ISO and because it's a Canon, it was supposed to be inferior to Nikon and Sony in terms of dynamic range (if you believe DxO). However, what do my wondering eyes should appear - something truly amazing! Is it a miracle?

So without reading further, it was at this point that I realized that this camera didn't suck. In fact, it got my attention that it might actually be quite good! However, would my frustrations with Canon that once had me thinking of switching camps be resolved? Read on to find out.

Body Thoughts

No, not my flabby body - the 5D Mark IV. In generally if you've used the 5D Mark III, 5Ds or 5DsR then you pretty much know what to expect here. Yes, there been some subtle changes, but beyond the higher resolution touch screen display and USB 3 instead of USB 2 port it didn't feel especially different in every day use.

On the geek side you end up with better burst mode performance, better video resolution 4k, alledgely better battery life with the same battery, built-in wifi (with FTP support) and GPS, dual pixel RAW AF (requires DPP),

In practice I just found it to have very good AF performance and when paired with the 24-70 f/4L IS lens I didn't have the need to over compensate on shutter speeds to get a sharp shot. Even if I did, they've finally added my favorite feature from Sony whereby you can specify a multiplier for the "auto" settings whereby Aperture priority will favor faster shutter speeds than what its traditional calculations would use. This is especially helpful for event photographers who deal with moving subjects in poor light.

In fact, during my review I found myself so pleased with the 5D Mark IV that I ended up not using the 600EX-RT II at all!

In the Studio

f/7.1 @ 200mm for 1/200 sec at ISO 100
Here's where you'll love the extra megapixels

In the studio the extra megapixels translate into more detail in web and small prints so that eyelashes and hair just feel more real. The shots in this article are 100% unedited and straight out of the camera from the in-camera JPEG, yet when you click the photos to see the original you can see this camera creates a impressive file.

Even with some quick mugshot shots, I realized that the improvements in dynamic range and high ISO performance of this camera might mean a worthy upgrade for 5Ds and 5DsR owners even if it meant sacrificing some megapixels. All of the other improvements definitely make it worth it!

Dual Pixel Raw AF Correction

Yes, Canon's Dual Pixel RAW AF Technology works well in DPP
NOTE: This is not the image from above

Being a geek, I had to try out the raw dual pixel feature to see how well I could correct a lens focusing problem. As you can see from the image above, once I apply the AF micro adjustment AFTER the image had originally been shot, I can recover quite a bit of detail that was lost in the original. Click here to learn more about Canon's Dual Pixel RAW AF technology.

Features like this and the anti-flicker technology aren't new to this Canon model, but they are game changers for dealing with age old problems that event photographers face making this yet another reason to buy this camera.

Bookshelf Shots

After 8 years of doing bookshelf shots for this blog, I want to puke whenever I have to do these for a review. However, I like my bookshelf shots a lot more than test charts and fake environments others set up because of the deep shadows and a seeing what happens to all of the books on a different plane from each other.

The following shots are unedited in-camera JPEG's shot on a tripod with mirror lockup, a timer and aperture priority with the 24-70 f/2.8L II at 70mm using Auto White Balance and the eyepiece taped over.

Click the photos to see the full-size originals.

f/5.6 @ 70mm for 6 sec at ISO 100
This is a A7R II caliber image when you look at the blue strip poker spine

While the shoe laces in the blue strip poker book show slightly lesser detail than the 5Ds and 5DsR, it's definitely on par with the A7R II and overall close enough that I probably wouldn't sacrifice the benefits of this camera for the limitations of the higher megapixels Canon's. I'd be annoyed if I bought those cameras and I'd definitely be excited to get my hands on the Mark II version of the 5Ds!

I was also blown away by the detail in the shadows as this better than what my eyes in real life, so at 100 ISO this is going to be an impressive camera for landscape photographers!

f/5.6 @ 70mm for 1/10 sec at ISO 6400

At 6400 ISO I expected things to go bad, and while I wasn't impressed with the Canon in-camera noise reduction, I can say that this image is totally useful for any purpose in my book. Shadow detail is still good and the noise is cleaned up nicely enough while still preserving a lot of the detail.

f/5.6 @ 70mm for 1/40 sec at ISO 25600
Are you kidding me?

In most of the last generation of Canon cameras, 25,600 was an ISO that was there for marketing purposes but nobody in their right mind would actually use it for pretty much anything. Now I found myself saying - heck yeah, I'd use that! Sure I wouldn't want to shoot a birds feathers or bushy tree in the fall with that ISO, but for most real world scenarios - and especially a concert - I'd totally use that file.

After 25,600 things do start to get ugly so I consider this the realistic upper limit of this camera, but it's a very practical upper limit.

Another thing to take note of here is that many social media sites don't take advantage of photos larger than the ones you see above. Can you see a huge difference between the ISO's when the images are scaled down to this size? - No, so if your output is primarily for the web versus prints > letter size - then you shouldn't stress much about ISO - they are all good enough for the web.

Real World Shots

These are all hand-held shots except the outdoors shots with moving water

f/4 @ 50mm for 1/500 sec at ISO 640
I liked how smooth the fog felt in the original even at ISO 640

f/22 @ 70mm for 1/125 sec at ISO 6400
At 100% I'm not impressed, but even full-screen at 4k the detail in the bottom leaf and water droplets was pretty good. I'm still not seeing magic from Canon's diffraction correction for f/22 like I saw on the Fujifilm x100s

f/5.6 @ 70mm for 1/125 sec at ISO 1000
Using macro on the 24-70 f/4L IS brought out a lot detail in this leaf that you can only fully appreciate at 100%

f/5.6 @ 115mm for 1/500 sec at ISO 125 (-1 EV)
The bird was an accident, but I did witness the improvements in the shadows of the 5DM4 that Canon claims has drastically improved

f/9 @ 66mm for 1/250 sec at ISO 160 (-2/3 EV)
Extra megapixels help to preserve a lot detail from this shot even when downsized

f/5.6 @ 59mm for 1/250 sec at ISO 320 (-2/3 EV)
I'm very satisfied with the dynamic range improvements - long overdue for Canon!

f/16 @ 70mm for 1/250 sec at ISO 10000
Improved dynamic range means fewer blown highlights in mixed condition shots like this

f/4 @ 50mm for 1/200 sec at ISO 100
This was a torture test to see just how wide the dynamic range was. The colors were preserved very well, detail in the shadows was better than expected and the highlights blew out about where the highest cameras typically do these days

f/4 @ 70mm for 1/320 sec at ISO 5000
Again the 24-70 f/4L IS macro mode brings out microscopic details of the leaf

f/4 @ 35mm for 1/125 sec at ISO 5000
Another torture test, this is one that blew me away on the D600 so I've made it a part of my regular shots since then. I much preferred this result over the D750 in the highlights and shadows of both the foreground and background. The detail was way better here than the D600 & D610 thanks to extra megapixels

f/7.1 @ 61mm for 1/250 sec at ISO 10000
Another great example of highlights and shadows both doing quite well under tough conditions and high ISO

f/5.6 @ 66mm for 1/250 sec at ISO 4000
I'm not sure any camera I've tested could better, but this was the first time I took this shot and I definitely see room for improvement in the stained glass in the background. This might be my new dynamic range torture test!

f/5.6 @ 70mm for 1/500 sec at ISO 8000
This is an in-camera HDR (vs single frame here) that showed that HDR is still very necessary and with the dynamic range of this camera it can be very impressive and natural - even with in-camera JPEG processing!

f/22 @ 26mm for 0.6 sec at ISO 100
Again, in-camera JPEG diffraction correction didn't seem to make up for lens softness at f/22, but at reasonable print and web size it is good enough to illustrate a wide depth of field

f/22 @ 37mm in-camera HDR at ISO 100 using Landscape Picture Style and Art Vivid mode
Choose your HDR effects carefully as this was pretty nasty but still much better than the base exposure (here)

f/22 @ 70mm for in-camera HDR at ISO 100 using Auto White Balance and Natural mode
Here the colors feel much more normal and the highlights are way better than the non-HDR base exposure (here)

f/5.6 @ 70mm for 1/200 sec at ISO 10000 (handheld in the wind)
Even at f/5.6 with the 24-70 f/4L IS the bokeh is buttery smooth

Click here for a full gallery of images, but please only use them for review purposes while you are reading this article. The photos should not be edited, redistributed, linked or shared in any way without written permission.

Conclusion

My 5D Mark III had grown so long in the tooth that I stopped using it well over a year ago - it had become a dust collector. When the 5Ds and 5DsR came out nothing impressed me enough about them to make the switch, so I held fast to my 1D X and 5D Mark III. With the release of the new Canon pro bodies I sold both cameras. After a comparison of both, I decided that the 1D X Mark II still made the most sense for my needs, but the 5D Mark IV impressed me way more than I expected.

30.4 megapixels with fast processors seems to be a good balance between more resolution and details without the need to have a significant shutter speed multiplier like I needed on the D800 / D810. The A7R II managed to outperform the Nikon too, so I'd put the 5D Mark IV and A7R II on my list of recommendations for people looking for more megapixels without the shutter speed multiplier penalty.

However, what impressed me most about this camera was the excellent high ISO performance that exceeded the 5DsR, 5Ds, D810 and even the A7R II. To my eyes, only the 1D X Mark II and D5 have out performed it, so I feel quite comfortable in saying that my fears about this camera before I had it were unfounded. It seems to be a great choice for event photographers as we as those looking for great quality with high resolution. The improvements in dynamic range also help rest the Nikon envy, so Canon users should rejoice that this is definitely a worthy upgrade.

I highly recommend the 5D Mark IV.

Learn more or buy

Click here to learn more or buy your own 5D Mark IV via B&H.

Disclaimer

If you make a purchase using links found in this article, I may make a commission. Thank you so much for supporting this blog by coming back here and using my links when making your purchase!

NOTE: This site requires cookies and uses affiliate linking to sites that use cookies.

If you enjoyed this article, please support future articles like this by making a donation or saving money by using my discount coupon codes. Either way, your support is greatly appreciated!

This blog is intended for freelance writing and sharing of opinions and is not a representative of any of the companies whose links are provided on this site.

The opinions provided are of Ron Martinsen alone and do not reflect the view of any other entity

Sunday, October 23, 2016

Comparison: Canon 5D Mark IV vs 1DX Mark II - CORRECTED

Canon 5D Mark IV

Canon 1D X Mark II (Review)

When the 5D Mark III and 1D X first came out, I bought them both. At the time I was doing quite a bit of commercial shoots where I needed a proper sports camera and a proper event camera that was lighter weight. Both cameras served me well, but honestly my biggest test with both camera bodies wasn't the paid work - that was easy because it was always a controlled situation with the light I needed. What was really hard were the shots I did for personal pleasure where I had the pressures of the family not wanting to wait for me to take pictures, horrible light - often indoors with lamps or outdoors in the rainy Northwest.

Over the years that I had these cameras, I favored the 1DX for the superior image quality over the 5D Mark III and it's better high ISO performance. I also loved the better control I had over the settings, despite missing creature comforts like the HDR mode found on the 5D Mark III and not the 1DX. In fact, the extra megapixels offered on the 5D Mark III would become a negative as its image was always softer and noisier than the 1D X.

I've sold both of my beloved cameras and have decided that I'm going to limit myself to one DSLR this time around, so the basis of this review is NOT about burst mode performance as the 1DX Mark II has no equal. It's more about every day use and which one would I want in my hands if I were shooting any situation where I wanted the highest probability of getting the shot with an image that will print beautifully at 24x36 - the largest size I often print.

This article isn't about camera price or features, as I was really most interested to see if the extra megapixels resulted in a better image or if the lower resolution would give me a more usable image file.

October 23, 2016 Correction - I made a mistake

In the original article that follows here, it turns out that I made a mistake by not being scientific enough in my comparison. I was hand held and thought I was close enough to make accurate determinations but I was wrong.

To prove to myself the differences between the two I ended up shooting with both cameras again in the exact same position using the same 70-200mm lens at 200mm using camera settings of 1/200 sec at f/4 and ISO 8000 (to show noise) and creating tungsten white balanced 16-bit ProPhoto RGB PSD files from the CR2 files of both that I loaded into Photoshop.

I normalized both files to be the same size as a 1DX Mark II image size (5472x3648 px) which would show the benefits of the greater resolution of the extra megapixels in the 5D Mark IV. With the files normalized and aligned, I created this animated GIF to compare the two images as precisely as possible:

Animated comparison of a 100% crop from a normalized 5472x3648 pixels image

The net result is that I still have a preference for the 1DX Mark II, BUT you can see that my original conclusion below about the lens distortion was inaccurate. This animation more closely represents what you might expect to get if you printed images from both cameras at the same size on the same printer. This type of comparison favors the 5D Mark IV simply for the finer details achieved by the higher resolution image.

It All Started With A Spider

5D Mark IV (Left) 1D X Mark II (Right)
1/200 sec at f/5.6
100% zoom to actual pixels

I was out in my back yard when I spotted a spider, so when I ran in to grab the nearest review camera, I decided that I better get outside again and shoot with the same camera.

These shots were taken about 26 minutes apart - both freehand using the same 24-70mm f/4L IS lens. You can view the original 5DM4 image on the left HERE and the 1DX2 image on the right HERE. Both were shot in manual mode with Auto ISO and a single AF point using the macro range of the 24-70mm f/4L IS lens.

The color difference is due to the auto white balance and auto picture style settings which can easily be normalized in RAW, but my favorite here was easily the 1D X Mark II image for the sharpness of the spider, despite the superior bokeh of the 5D Mark IV. This observation would begin a theme where naturally a higher megapixel camera is going to have less depth of field which results in superior bokeh - unless you normalize the files to the same size (or print them out).

5D Mark IV (Left) 1D X Mark II (Right)
1/200 sec at f/9, ISO 25,600
100% zoom to actual pixels

As I closed down the aperture the ISO naturally climbed resulting in an image where the 1D X Mark II was clearly the better image in my opinion. Click here to view the in-camera JPEG from the 5DM4 and click here (shown right above) to view the same from the 1DX Mark II.

Doing this of course begs the question, would the 1DX2 advantage carry over in a studio where the lighting was excellent?

Model with Studio Lights Test

5D Mark IV (Left) 1D X Mark II (Right)
1/200 sec at f/7.1, ISO 100, 200mm

In the studio, I once again shared the same 70-200mm f/2.8L IS II USM lens between the camera bodies with the model under identical light sitting in the same spot the entire time. While I did shoot freehand, as you can see from the image the models head and hair were my guide for her placement within the frame. In this case I used auto picture style and white balance just to see how both would perform in an informal test.

These images are 100% unedited - straight out of the camera in-camera JPEG files. You can view the 5DM4 JPEG or RAW and 1DX2 JPEG or RAW. All images are copyright Ron Martinsen - ALL RIGHTS RESERVED. You may view them while this browser session is open and this is your active page, but afterwards you must delete them. You may not save, edit, print, etc... these images without written on paper signed consent.

The net result was that the 5D Mark IV made the model look a little more heavy than the 1DX Mark II with the image quality being very good for both.

The heavier look wasn't real but rather an illusion created due to the model and lens not being an EXACT distance from each other and comparing two files that weren't the exact same size. If printed, or the files normalized to the same size, this issue would not exist.

I then zoomed in on the eyes and without normalization I favored the 1DX Mark II but in print or with normalization it'd probably be a draw:

As I looked at the lips, again without normalization or printing I felt that the 1DX2 had more detail but that both were equally good results.

Finally, here's a nose comparison: