This topic is in the 2012 archive. You can read it and like its posts here. It is read only.
Archive 2012Canon 5D Mark III and Nikon D800 Review
p.3 #1Canon 5D Mark III and Nikon D800 Review
Reply to: utildayael
Can you say more? It's not clear to me how that would confirm focus using hyperfocal technique, which seems to require a stopped down view to assure accuracy. Am I missing something.
Apr 22, 2012 at 07:01 PM
p.3 #2Canon 5D Mark III and Nikon D800 Review
Reply to Ardea:
Exactly the conclusion I reached. Stick with the 5D II.
I had been pretty interested in switching, but I shoot in low light, sharp everywhere, hyperfocal, with manual focus lenses (hence not much excitement for me about 5D III) Live view performance is critical for all of that. I will give the Nikon a test, to be sure I understand the problem, but it sounds like it may be unusable for me. Too bad, because I also need the largest prints I can get away with, and improved dynamic range would be a big help.
Probably have to wait another generation. Hope the old 5D II makes it over the finish line.
Exactly the conclusion I reached. Stick with the 5D II.
I had been pretty interested in switching, but I shoot in low light, sharp everywhere, hyperfocal, with manual focus lenses (hence not much excitement for me about 5D III) Live view performance is critical for all of that. I will give the Nikon a test, to be sure I understand the problem, but it sounds like it may be unusable for me. Too bad, because I also need the largest prints I can get away with, and improved dynamic range would be a big help.
Probably have to wait another generation. Hope the old 5D II makes it over the finish line.
Apr 22, 2012 at 07:11 PM
p.3 #3Canon 5D Mark III and Nikon D800 Review
Reply to Parsons:
I shoot manual focus lenses about 90%. The problem with the focus-wide-open method you describe is it probably handicaps the advantage of shooting hyperfocally for maximum depth of field. Someone with hands-on D800 experience may wish to correct me, but I surmise from the comments here that you can't use live view with that camera stopped down to smaller apertures and still see critical sharpness.
The view you get wide open won't show the stopped-down sharpness, typically in the foreground area. If you use lenses that offer tilts, that makes it even more critical, because you MUST see the actual stopped down focus to optimize your tilt setting. That would be a major problem for me, and in fact a deal breaker.
Some here say that even the 5D III is less proficient with live view than the 5D II. They surmise that optimizing design for video capability is driving live view degradation. I hope that isn't happening. When good live view arrived, I thought it was a major step up in capability, one that would even drive lens design (I expected many more new TS-Es from Canon than they have so far offered). I don't want to see live view neglected.
One marketing-centered concern is that Canon cameras with live view are a natural for Zeiss manual focus lenses, which may not delight Canon lens marketers. Degrading live view would be one way to advantage Canon's auto-focus lens lineup in the marketplace, if not in the field.
I shoot manual focus lenses about 90%. The problem with the focus-wide-open method you describe is it probably handicaps the advantage of shooting hyperfocally for maximum depth of field. Someone with hands-on D800 experience may wish to correct me, but I surmise from the comments here that you can't use live view with that camera stopped down to smaller apertures and still see critical sharpness.
The view you get wide open won't show the stopped-down sharpness, typically in the foreground area. If you use lenses that offer tilts, that makes it even more critical, because you MUST see the actual stopped down focus to optimize your tilt setting. That would be a major problem for me, and in fact a deal breaker.
Some here say that even the 5D III is less proficient with live view than the 5D II. They surmise that optimizing design for video capability is driving live view degradation. I hope that isn't happening. When good live view arrived, I thought it was a major step up in capability, one that would even drive lens design (I expected many more new TS-Es from Canon than they have so far offered). I don't want to see live view neglected.
One marketing-centered concern is that Canon cameras with live view are a natural for Zeiss manual focus lenses, which may not delight Canon lens marketers. Degrading live view would be one way to advantage Canon's auto-focus lens lineup in the marketplace, if not in the field.
Apr 22, 2012 at 07:38 PM
p.3 #4Canon 5D Mark III and Nikon D800 Review
I understand that Canon and Nikon may have engineered their Live View systems very differently, and I've used the Canon's LV (5D mk2, mk3, 7D...) quite a few times.
I don't know anything about the science and engineering, however here are a few thoughts I have on Nikon Live View. I bet that if Fred tries even just the first tip, he would see a night-and-day difference in the ability to use LV focusing...
* You're sensible landscape photographers, we shoot RAW....so crank your in-camera sharpening all the way up. It won't hurt your images in LR at all, (or, you can un-do it in Capture NX etc.) ...and it will give you INCREDIBLE acuity in live view. Honestly as long as LV is using my base ISO, details are TACK SHARP.
* I actually think that Nikon's LV is more accurate to the final image, compared to Canons. Canon's LV looks more like an over-sharpened HD video screencap, and yeah its easy to confirm focus, but all you gotta do on Nikon to get the same results is crank your in-camera shaprening way up, and find the perfect mid-range magnification to use during LV. Zooming in ALL the way on a Nikon in LV is worthless, it's just pixels. True 100% is actually what Nikon calls "Medium Magnification" during playback. (Usually option f2 in the control customization menu, I LOVE that option and am very glad they've finally done something similar on the 5D mk3 with the "SET" button magnification...)
* Keep in mind that your picture style affects the live view image a little bit too; something neutral with the contrast turned down should give you a more accurate histogram, and might give you better results when trying to focus on shadow detail during LV.
* Use manual focus lenses, or just enter live view wide open on your G lenses for a second, and focus that way so that LV is using the lowest possible ISO... Live view is gonna be crap at f/16 in dim light. Again, we're all sensible landscape photographers; it's not that hard to pick the right hyperfocal distance if you know your lenses well. Focus there wide open, stop down, shoot. Works for me! I've never "missed the light" while shooting on a D300 or D700, and I don't think they've dramatically changed the way LV works on the D800...
* I actually use Nikon Live View quite effectively even to focus on stars at night; it works great!
* When all else fails, a V-Finder (Z-Finder) works wonders. Again, crank your in-camera sharpening all the way up, and you get an eye-popping visualization of focus; it almost harkens back to the days of film and TRUE "ground glass" focus sharpness.
I can see how maybe someone who has only ever shot Canon may find Nikon's Live View to be inadequate, or at least too different to get used to any time soon, but I just wanted to say that as someone who has been shooting Nikon since the first generation of LV cameras, ...I don't find it to be an issue at all, in fact I LOVE using Live View for manual focusing on landscapes...
Take care,
=Matt=
I don't know anything about the science and engineering, however here are a few thoughts I have on Nikon Live View. I bet that if Fred tries even just the first tip, he would see a night-and-day difference in the ability to use LV focusing...
* You're sensible landscape photographers, we shoot RAW....so crank your in-camera sharpening all the way up. It won't hurt your images in LR at all, (or, you can un-do it in Capture NX etc.) ...and it will give you INCREDIBLE acuity in live view. Honestly as long as LV is using my base ISO, details are TACK SHARP.
* I actually think that Nikon's LV is more accurate to the final image, compared to Canons. Canon's LV looks more like an over-sharpened HD video screencap, and yeah its easy to confirm focus, but all you gotta do on Nikon to get the same results is crank your in-camera shaprening way up, and find the perfect mid-range magnification to use during LV. Zooming in ALL the way on a Nikon in LV is worthless, it's just pixels. True 100% is actually what Nikon calls "Medium Magnification" during playback. (Usually option f2 in the control customization menu, I LOVE that option and am very glad they've finally done something similar on the 5D mk3 with the "SET" button magnification...)
* Keep in mind that your picture style affects the live view image a little bit too; something neutral with the contrast turned down should give you a more accurate histogram, and might give you better results when trying to focus on shadow detail during LV.
* Use manual focus lenses, or just enter live view wide open on your G lenses for a second, and focus that way so that LV is using the lowest possible ISO... Live view is gonna be crap at f/16 in dim light. Again, we're all sensible landscape photographers; it's not that hard to pick the right hyperfocal distance if you know your lenses well. Focus there wide open, stop down, shoot. Works for me! I've never "missed the light" while shooting on a D300 or D700, and I don't think they've dramatically changed the way LV works on the D800...
* I actually use Nikon Live View quite effectively even to focus on stars at night; it works great!
* When all else fails, a V-Finder (Z-Finder) works wonders. Again, crank your in-camera sharpening all the way up, and you get an eye-popping visualization of focus; it almost harkens back to the days of film and TRUE "ground glass" focus sharpness.
I can see how maybe someone who has only ever shot Canon may find Nikon's Live View to be inadequate, or at least too different to get used to any time soon, but I just wanted to say that as someone who has been shooting Nikon since the first generation of LV cameras, ...I don't find it to be an issue at all, in fact I LOVE using Live View for manual focusing on landscapes...
Take care,
=Matt=
Apr 22, 2012 at 11:11 PM
p.3 #5Canon 5D Mark III and Nikon D800 Review
huddy wrote:
I'd say Fred was pretty clear on it. He likes T/S lenses and the Nikon 24 doesn't have independently adjustable axes and also suffered from CA so he didn't use it as much. It didn't really feel biased at all what with the raving praise for the D800 sensor.
Using these two as an explanation for having hardly any D800 images is strange. CA is correctable in software. If this was really a comparison it would have been nice to have a side-by-side shot of the 24mm T/S from Nikon and Canon showing the differences. He did that with the sensors, why not the lenses.
As for independently adjustable T/S axes, I looked through the images he posted and did not see a single one where anything other than a "conventional" (top/down) tilt would have been needed.
I don't think it is fair to accuse any pro dedicated to a system of bias. Yes, they are biased towards what they are using, otherwise they would not be using it. However, this does not make for good comparison reviews, and this certainly was not one.
I'd say Fred was pretty clear on it. He likes T/S lenses and the Nikon 24 doesn't have independently adjustable axes and also suffered from CA so he didn't use it as much. It didn't really feel biased at all what with the raving praise for the D800 sensor.
Using these two as an explanation for having hardly any D800 images is strange. CA is correctable in software. If this was really a comparison it would have been nice to have a side-by-side shot of the 24mm T/S from Nikon and Canon showing the differences. He did that with the sensors, why not the lenses.
As for independently adjustable T/S axes, I looked through the images he posted and did not see a single one where anything other than a "conventional" (top/down) tilt would have been needed.
I don't think it is fair to accuse any pro dedicated to a system of bias. Yes, they are biased towards what they are using, otherwise they would not be using it. However, this does not make for good comparison reviews, and this certainly was not one.
Apr 23, 2012 at 12:02 AM
p.3 #6Canon 5D Mark III and Nikon D800 Review
Hooray for the D800 but I really don't need it. 
My Flickr Gallery
My Nikonians gallery
My interesting photos on Flickriver
My Flickr Gallery
My Nikonians gallery
My interesting photos on Flickriver
Apr 23, 2012 at 01:14 AM
p.3 #7Canon 5D Mark III and Nikon D800 Review
GroovyGeek wrote:
Using these two as an explanation for having hardly any D800 images is strange. CA is correctable in software. If this was really a comparison it would have been nice to have a side-by-side shot of the 24mm T/S from Nikon and Canon showing the differences. He did that with the sensors, why not the lenses.
As for independently adjustable T/S axes, I looked through the images he posted and did not see a single one where anything other than a "conventional" (top/down) tilt would have been needed.
I don't think it is fair to accuse any pro dedicated to a system...Show more →
You do realise the 'default' tilts and shifts for landscape (up or down shift combined with up or down tilt) are impossible to have together in the default PC-E layout?
In any case, Fred's already said he's getting another copy of the PC-E to try out. The TS-E II and the PC-E have been comparable in IQ IME, it's the practicality in which the TS-E smacks the PC-E (and the Schneiders smack both).
huddy wrote
I'd say Fred was pretty clear on it. He likes T/S lenses and the Nikon 24 doesn't have independently adjustable axes and also suffered from CA so he didn't use it as much. It didn't really feel biased at all what with the raving praise for the D800 sensor.As for independently adjustable T/S axes, I looked through the images he posted and did not see a single one where anything other than a "conventional" (top/down) tilt would have been needed.
I don't think it is fair to accuse any pro dedicated to a system...Show more →
You do realise the 'default' tilts and shifts for landscape (up or down shift combined with up or down tilt) are impossible to have together in the default PC-E layout?
In any case, Fred's already said he's getting another copy of the PC-E to try out. The TS-E II and the PC-E have been comparable in IQ IME, it's the practicality in which the TS-E smacks the PC-E (and the Schneiders smack both).
Apr 23, 2012 at 02:15 AM
Guest
Offline
Guest
Details
p.3 #8Canon 5D Mark III and Nikon D800 Review
ok,
if you focus hyperfocally you will know exactly where to set your lens for any given aperture. so nothing will change.
i do exactly that on my 1dsmk2, ap wide open, compose etc, then stop down to my prefered ap of F11 check exp and then shoot.
that is also how i teach it on the workshops.
this image below was taken that exact same way as i have worked out and examined many images to find the best point for my three zeiss and zuiko lens, also as seen in the image below.
http://www.sussexlandscapephotography.co.uk/VU2R6852_1.jpg
lens chart which i use in the field as reference as i have have not had the lenes long.
http://www.sussexlandscapephotography.co.uk/lens%20test.jpg
you will also notcice that the lenses do not match the markerss for their given hyperfocal distance on the lens, hence why you should check images to find you best point for you.
s
if you focus hyperfocally you will know exactly where to set your lens for any given aperture. so nothing will change.
i do exactly that on my 1dsmk2, ap wide open, compose etc, then stop down to my prefered ap of F11 check exp and then shoot.
that is also how i teach it on the workshops.
this image below was taken that exact same way as i have worked out and examined many images to find the best point for my three zeiss and zuiko lens, also as seen in the image below.
http://www.sussexlandscapephotography.co.uk/VU2R6852_1.jpg
lens chart which i use in the field as reference as i have have not had the lenes long.
http://www.sussexlandscapephotography.co.uk/lens%20test.jpg
you will also notcice that the lenses do not match the markerss for their given hyperfocal distance on the lens, hence why you should check images to find you best point for you.
s
Apr 23, 2012 at 05:42 AM
p.3 #9Canon 5D Mark III and Nikon D800 Review
Parsons, would it get any harder to focus hyperfocally if you were using f/5.6 on a 100mm lens? Or are you so good at distance estimation that you can reliably distinguish a focal target 290 feet away from one 343 feet away? And then twist your focus just right to match. By all means, understand the optimal hyperfocal values. Experiment. Build a table. Commit them to memory. But a lot of the time, especially when you are not using wide angles, you still have to see it to focus it, don't you?
Or maybe I am still missing something. I know my Zeiss 100 performs at it's absolute best, tack sharp right into the corners, at f5.6—provided my focal target is sufficiently distant. Too close, and I can't get critical focus in the center and corners at that aperture. However, that doesn't mean I don't often want the foreground as close as possible, and I am not a human rangefinder. Seems like with longer lenses the combination of focus capability, composition, and depth of field is a dynamic problem to be solved in the field. How do you pre-solve it with a table?
Or maybe I am still missing something. I know my Zeiss 100 performs at it's absolute best, tack sharp right into the corners, at f5.6—provided my focal target is sufficiently distant. Too close, and I can't get critical focus in the center and corners at that aperture. However, that doesn't mean I don't often want the foreground as close as possible, and I am not a human rangefinder. Seems like with longer lenses the combination of focus capability, composition, and depth of field is a dynamic problem to be solved in the field. How do you pre-solve it with a table?
Apr 23, 2012 at 08:52 AM
Guest
Offline
Guest
Details
p.3 #10Canon 5D Mark III and Nikon D800 Review
if i am using the 50mm for instance, the first part of the image visible with the camera level is around 5m so i will focus as above.
once my object of choice is further say 12m i know it will be covered in the dof of that aperture.
if for example i use your 100mm and your subject is 290 ft focusing at infinity the object will be covered by the dof of 5.6 as it starts at 116ft as your closet point viewable in the frame so i woulnt need to worry about hyperfocal focusing as the longer the lens the less dof you have depending obviously on your subject distance.
put it this way, when ever i use a lens longer than 85mm at f8 i will always focus on infinity regardless of the main object ( landscapes is my thing).
with any object that is closer, on a day out at the musem i will only focus on the object itself, letting the corners do there thing as its the staue that is most important.
sombody once said focus on your main subject, if you get blurry corners dont worry your main point of focus (statue) is always in focus.
just the way i work i guess
simon
once my object of choice is further say 12m i know it will be covered in the dof of that aperture.
if for example i use your 100mm and your subject is 290 ft focusing at infinity the object will be covered by the dof of 5.6 as it starts at 116ft as your closet point viewable in the frame so i woulnt need to worry about hyperfocal focusing as the longer the lens the less dof you have depending obviously on your subject distance.
put it this way, when ever i use a lens longer than 85mm at f8 i will always focus on infinity regardless of the main object ( landscapes is my thing).
with any object that is closer, on a day out at the musem i will only focus on the object itself, letting the corners do there thing as its the staue that is most important.
sombody once said focus on your main subject, if you get blurry corners dont worry your main point of focus (statue) is always in focus.
just the way i work i guess
simon
Apr 23, 2012 at 01:22 PM
p.3 #11Canon 5D Mark III and Nikon D800 Review
Hi Fred,
I enjoyed your review and it was interesting to see you had some of the same issues that I had.
In my review I had a hard time getting the camera to white balance as you can see in this video:
http://www.learningdslrvideo.com/nikon-d800-greenish-lcd-tint/
I read the manual and the 3 or 4 ways to custom white balance but I was never sure I got it right. Since you have a lot more experience with the Nikon's than I do I was wondering if you could watch the video above and tell me if I did it wrong or not?
Thanks I appreciate it.
Dave Dugdale
ps. If you want to watch my full review you can view it here:
http://www.learningdslrvideo.com/5d-mark-iii-vs-nikon-d800/
I enjoyed your review and it was interesting to see you had some of the same issues that I had.
In my review I had a hard time getting the camera to white balance as you can see in this video:
http://www.learningdslrvideo.com/nikon-d800-greenish-lcd-tint/
I read the manual and the 3 or 4 ways to custom white balance but I was never sure I got it right. Since you have a lot more experience with the Nikon's than I do I was wondering if you could watch the video above and tell me if I did it wrong or not?
Thanks I appreciate it.
Dave Dugdale
ps. If you want to watch my full review you can view it here:
http://www.learningdslrvideo.com/5d-mark-iii-vs-nikon-d800/
May 16, 2012 at 09:04 AM
p.3 #12Canon 5D Mark III and Nikon D800 Review
I do agree that Fred is a fan of the TSEs, as I am. After watching the "peak your JPEGS-shoot RAW) you tube I am convinced that I can get sharp live view pictures with a D800. Now for a modern Nikkor series of PC lenses (especially that 17mm they just patented).
Jun 02, 2012 at 12:54 PM