Canon 5D Mark IV -- Time to Discuss!

sploo wrote in post #18115809
I think the gripe is, for example, that the D500 has been out for a while, has a UHS-II slot, and a 200 capacity raw buffer. Yet, in a $3500 body we have UHS-II and 21 shots. What's frustrating is that there's likely enough memory to do more, but it's a firmware limitation.

I'm less annoyed about the 4K crop because there are significant processing issues, so despite it being a pain I can understand why it works that way.

Well, one thing's for sure: it won't be giving aliased video at 4K.

gabebalazs wrote in post #18116055
And with the improved DR of the mark IV, now even dynamic range differences are minor (except maybe in the case of the new D5 which is a low 12.3 EV, ironically). And I'm saying all that as a pixel peeper, analyzer sort of guy.

The D5 doesn't even do that well in reality; it has very strong banding noise.

gabebalazs wrote in post #18116055
I do like Tony Northrup and his videos, nothing against him, but all those viewers/posters drawing IQ conclusions from a compressed 1080p video is just insane. I spend hours staring at DPreview studio samples on my calibrated monitor to decide which camera handles noise better etc. And I bet many of these "expert" commenters viewed this video on their cell phones

Tony Northrup certainly has his own opinions, and his fairly recent predictions video really highlighted his own desires (IE Nikon and Canon would be introducing 75+MP cameras). For real world performance and IQ comparisons with the 5D4, I'm finding he and Fro are about the same. They both rush right out with the paper specs and then draw their conclusions. It's not much different then what we can do in these forums. At least the initial hands on reviews of the pre-release 5D4 seems that the dual pixel RAW is a nifty feature (that doesn't seem to diminish performance).

Since DR has improved with the latest Canon models, I'm seeing less of an all out advantage with Nikon/Sony. The Nikon D810 RAWs are still larger then a 5D4 RAW, as they still offer slightly more MP. My coworker who was a Canon shooter/ now works with a D810 has liked the D810 for pulling shadows in challenging situations (especially harsh lighting). He's not printing large format for most portraits, and likes the ability to crop in heavily. I'm not as optimistic as Tony Northrup, who has stated the successor to the D810 will be 75MP. My speculation is that it's 42MP like the A7Rii.

The all out stills IQ is pretty much negligible between Canon, Nikon, and Sony now....each one offering a better feature somewhere, but all good offerings. Since I've been invested in Canon glass, I never saw the need in switching camps to Nikon, and I didn't want to take my chances with the lens adapters for Sony. Even with the successor to the D810, I don't see myself switching camps. I like my Canon glass too much, and now the DR differences are minuscule.

John Sheehy wrote in post #18116063
Well, one thing's for sure: it won't be giving aliased video at 4K.

I just saw a video of a Canon rep responding to the criticism of why the 5D4 uses motion jpeg compression. Their response is that even though it's high bandwidth, it's tried and true. And that it can yield a high quality still retrieval. That makes sense to me. Since multithreading is always coming down in price, the memory requirements don't seem as bad to me. Though I would have thought it would also use CFast for video....guess Canon wanted to make it and 4K 60fps exclusive to the 1DX2

I find Tony Northrup annoying.

Not saying he is wrong or knows nothing, but just plain annoying.

Do not watch his videos anymore.

Regarding video.. if you have 5 minutes to spare and a sense of humor

https://youtu.be/FJYXr​ilGk_8

This made me lol..

It was funny because its like they read my mind ..especiallu towards thr end when makes the sony and m43 comments

I watched Tony's tldr, it's not the best but close. Don't see anything wrong with that.

I do agree with Gabe on the dpreview tool for IQ, mostly. I do avoid the corners and edges in case of weak lenses.

In use, I'm guessing it'll be like the 5d3 refined, it's not like camera manufacturers make huge ergonomic changes between generations.

don1163 wrote in post #18115615
I saw this one as well...He seems to be saying it is a shift in sharpness within the available depth of field rather than an actual shift in focus.

That would seem to make sense, as the sensor (i.e. each pixel in a pair) will be "seeing" light from either side of the lens (both for AF - wide open, and when the shot is taken - possibly stopped down).

As such, each pair of pixels will contain a signal that's come from the sides of the lens with the aperture that was used to take the shot - thus they contain a "range" of data that's related to the depth of field at shooting time.

I'm aware my use of "sides" is probably incorrect in optical terms, but that's how phase AF diagrams tend to be drawn.

FEChariot wrote in post #18115660
I would imagine that a firmware dual pixel for the 1dx2 is a possibility but so could have many other technologies over the years that didn't. I was very please with the 7D increased buffer capability upgrade but I don't hold my breath for these type of things.

I would actually be quite surprised if the 1Dx2 (and maybe 80D) hardware couldn't make the individual pixel pairs available to the firmware for writing out. Perhaps even some older bodies may have that capability too. Whether we'll see them get it in a firmware update is another matter (though I suspect it's likely with at least the 1Dx2).

gabebalazs wrote in post #18116055
When comparing today's modern FF cameras, say D810, D750, 5DIV, K-1, A7RII, image quality is so close that other factors (MP, fps, ergonomics, AF, weather sealing, card slots, lens selections, etc.) are much more important than 1/3 stop of noise difference at ISO 25000. And with the improved DR of the mark IV, now even dynamic range differences are minor (except maybe in the case of the new D5 which is a low 12.3 EV, ironically). And I'm saying all that as a pixel peeper, analyzer sort of guy.

I think it was in an earlier post on this thread that John made a comment on the lines that Canon are out of the dog house - that is, the low ISO DR is sufficient that, whilst not quite matching that of rival sensors, it's now close enough that it shouldn't be a major factor in choosing a system (unless perhaps that was a very primary consideration... which means: buy a D810 and stick it on ISO 64).

John Sheehy wrote in post #18116063
Well, one thing's for sure: it won't be giving aliased video at 4K.

True. I'm just a bit sad they weren't able to offer a quality 1.5x subsampling of the sensor, as that would have resulted in a tiny crop factor. IRC The 5D3 resolution (width) is exactly 3x1920, so I assume that was deliberate (to achieve a high quality 1920 result with much less moire than the 5D2).

Indeed if they had been able to offer the 1.5x scaling on the 5D4 they could have even marketed it as having full frame and Super 35 modes. Ok, actually about a 1.09x and 1.64x crop respectively, but close enough for marketing

John Sheehy wrote in post #18116069
The D5 doesn't even do that well in reality; it has very strong banding noise.

Like old Canons I guess?

Charlie wrote in post #18116177
I watched Tony's tldr, it's not the best but close. Don't see anything wrong with that.

I do agree with Gabe on the dpreview tool for IQ, mostly. I do avoid the corners and edges in case of weak lenses.

In use, I'm guessing it'll be like the 5d3 refined, it's not like camera manufacturers make huge ergonomic changes between generations.

His love of the 5Ds is interesting. I guess that if you have good light (not pushing the ISO) and a scene that's relatively low DR (not pushing Canon's historic low ISO DR issues) then with a good lens you're going to have 50MP of superb quality. If you're a landscaper or architectural guy who can bracket shots for DR, or a fine art studio shooter (who can control light) I can see it'd be an ideal body. I suppose those types of shooters probably also don't tend to spray - so you don't end up having to work your way through that many lots of 50MP.

As soon as you're having to trade MP for noise though (with aggressive noise reduction), you will still be physically dealing with a 50MP file (i.e. pushing your computer's processing) but you'll have lost some of the resolution benefits.

sploo wrote in post #18116359
I think it was in an earlier post on this thread that John made a comment on the lines that Canon are out of the dog house - that is, the low ISO DR is sufficient that, whilst not quite matching that of rival sensors, it's now close enough that it shouldn't be a major factor in choosing a system (unless perhaps that was a very primary consideration... which means: buy a D810 and stick it on ISO 64).

ISO 64 should be possible with "normal" RAW headroom with dual-pixel RAW, if the headroom claims are true. Canon might not support it directly, but you can DIY with ETTR or maybe Magic Lantern might do something that can make it transparent to the user if they figure out the firmware.

That's computation-intensive. Now you're resampling, rather than binning like in the 5D3. Resampling can give the least aliasing, but 1.5->1 resampling is rough on details. They probably figured line skipping was not used in the 5D3, so that is no longer an option in the 5D series. Line skipping is the worst method, as far as aliasing goes. The best video from a Bayer sensor would come from a very high sensor resolution, and a heavy downsampling, with full color at every pixel and no aliasing, but that would require lots of processor power and battery power. Binning with overlapping source tiles could avoid aliasing, too, but that needs precise ratios of pixels. It's really a big struggle to resize to close numbers; the options are not ideal.

Indeed if they had been able to offer the 1.5x scaling on the 5D4 they could have even marketed it as having full frame and Super 35 modes. Ok, actually about a 1.09x and 1.64x crop respectively, but close enough for marketing

Like old Canons I guess?

I downloaded a set of D5 blackframes someone provided. I haven't done a direct comparison, but by sight and memory, the D5 seemed to be a bit worse than the 5D2 at ISO 100. That stuff shows up, too, at extremely high ISOs, when the pre-gain noise is very low and you downsample so that the random noise mostly disappears and only the banding noise is still strong. Banding noise and blotchy low-frequency noise do not disappear as fast when downsampling as random noise does. I suspect that even though the D5 outperforms the 5D4 and 1Dx at high ISOs in DxOMark, that in practice, if you pushed the Canons to ISO 3.28 million, and the final output was small 450*300web images processed by someone who knew how to handle the data with the least destruction, that the Canons may be a lot closer than DxO would suggest, or maybe even a little better. We don't see many tests like that, though. People tend to judge the usefulness of very high ISOs by 100% crops or large images sizes, where they are likely to be disappointing. Yet, people ooh and aah at small web images at medium-high ISOs, which are not very challenging at all if the the processor knows how noise, filtering, and resizing work.

Preordered mine from Canon. I hope I get the mini 5D4 with 24-70 replica. I believe it's a USB like in the past.

Levina de Ruijter wrote in post #18115658
Yes, that's how I understood it too.

Right. I hadn't seen the other video. Will watch that one too.

I do wonder where all this new technology will take photography. Dual Pixel Raw and what it can do is starting to look more like holography than photography. Are we moving into that direction? Or is that a very silly remark?

Same, there used to be skill in being a photographer..

cwphoto wrote in post #18116515
Same, there used to be skill in being a photographer..

Much of it being unnecessary, and only a struggle against technological limitations. You still have to know where to be and when to be an interesting photographer. We can only sample an infinitesimal percentage of all time slices and perspectives, and knowing which ones to position ourselves in is an art in itself. Picking frames from a burst or HQ video, and picking focal planes after the fact are still artistic decisions.

Good riddance to boasting about what one can do with their hands tied behind their backs, useful only as a competitive novelty for the bored.

John Sheehy wrote in post #18116069
The D5 doesn't even do that well in reality; it has very strong banding noise.

How do you know this

Use it or something you read on the net ?

Anybody have a guess at how long will it take B&H to fulfill all their preorders? I ordered mine last weekend, just hoping to get it before the end of the month. I should have ordered immediately, I have had my 5d II for 4-5 years and I knew I was upgrading when this camera was released.

umphotography wrote in post #18116560
How do you know this

Use it or something you read on the net ?

Neither are necessary. This criterion of "using" cameras is bogus. The electronics and their noise do not change in the hands of users. They are in the RAW files like a watermark.

I have RAW files of the lenscap at all ISOs from the D5. They are ugly as hell, at all ISOs. When you boost the levels to see the noise, the banding is just as strong as the random noise dynamics, visually, even at 100%.

So, while the D5 has relatively low noise quantity at high ISOs, the quality or character is not very good. It could be a lot better, without changing the kind of score it would get at DxO, by having finer, more random noise.