Canon must hurry up on FF mirrorless, Sony's new A9 is killer

Mikehit said:
Canon has closed the gap significantly in the last 2 years

And - let's be clear - it's only a "gap" (if you wish to acknowledge it as such - some of us don't, as it impinges on our photography in no meaningful way) of a very specific nature; and Real-World relevant only to a relatively small subset of the photographic community...
 
Upvote 0
Darkly said:
Many arguments here are missing a major point, and that is Sony's proven ability to enter a given field and dominate it before the well-established players can play catchup. Look at the PlayStation.

Look at Vaio. How'd that work out for Sony?


Darkly said:
Now, while mirrorless may represent a small proportion of camera sales right now, this will change as Sony get their act together with user interface design and handling.

Will it? With no evidence to back up such a claim, it's merely an opinion.


Darkly said:
However, if you're new to photography what reasons are there for choosing Canon? Or Nikon? Or Sony? They all have merits but Sony's reputation is in the ascendant, particularly with relation to imaging sensor technologies.

Yet, in spite of that 'ascendent' reputation, Canon continues to gain market share while Sony loses it.


Darkly said:
Canon need to up their game significantly before they lose their reputation as one of the big two go-to, no-brainer camera manufacturers. Anyone citing figures showing mirrorless sytems as having a small market share, let alone Sony's modest market share would do well to look at other incumbent giants who lost their way through complacency. Remember Nokia? They could do no wrong until a lean and hungry Samsung came along and innovated their way to the top (even if the they were blatantly taking design cues from Jony Ives).

Anyone citing Nokia would do well to look at the many examples of incumbent giants who remain the leaders of their markets. Nokia is an example of what can happen, not a typical example of what usually does happen. In particular, Nokia is an example of a company failing to anticipate and react to a paradigm shift. The important difference in the situation under discussion here is that no paradigm shift is occurring for the ILC market. Film to digital was a paradigm shift. How did Canon handle that one?


Darkly said:
So, I think Canon either need to follow the same technological direction as Sony/Nikon to achieve similar noise and dynamic range figures

So, Canon should follow the same technological direction as Sony'a a9 and Nikon's D5, both of which have more low ISO noise and thus less DR than their predecessor? The a9 is billed as a 'pro' camera and yet it has over a stop less low ISO DR than the 1D X II. Why, exactly, should Canon follow suit?


TL:DR — YAPODFC.
 
Upvote 0
Mikehit said:
Darkly said:
Nikon made a very smart move by partnering with Sony, leveraging Sony's R&D and their quick turnaround on silicon fabrication.
So why did Nikon stop using Sony sensors?

??

They didn't. Sony may not be the exclusive supplier of sensors for Nikon SLRs (Toshiba for example supplied them for the D7100 and D7200, but of course Sony owns Toshiba's sensor business now), but they remain a (the?) primary one.
 
Upvote 0
neuroanatomist said:
Darkly said:
Many arguments here are missing a major point, and that is Sony's proven ability to enter a given field and dominate it before the well-established players can play catchup. Look at the PlayStation.

Look at Vaio. How'd that work out for Sony?


I can't speak for this user, but the way I read that says it is very possible Sony can take over this market and become the leader, not that it is guaranteed to. I didn't see an argument that Sony is unstoppable. So not sure why you pointed out a failure, when all companies have failures at times.

Darkly said:
Now, while mirrorless may represent a small proportion of camera sales right now, this will change as Sony get their act together with user interface design and handling.

Will it? With no evidence to back up such a claim, it's merely an opinion.


Why are you expecting conclusive evidence to prove a prediction? You're very oppositional I think for the sake of just arguing with people.

Perhaps one can look at a trend and speculate. Sony is looking to make money. They keep building better and better mirrorless bodies and introducing more lenses in the FF realm. They are obviously trying to move into this market. If they are trying to do that, it isn't unreasonable to predict that they'll make improvements to appeal to more users and gain more market share.




Darkly said:
However, if you're new to photography what reasons are there for choosing Canon? Or Nikon? Or Sony? They all have merits but Sony's reputation is in the ascendant, particularly with relation to imaging sensor technologies.

New or not, there's more to photography than a sensor. Sony can have the greatest sensor on Earth, the camera handles like a piece of crap. It literally has no grip. You are holding on to the end of a minimalist body.

Yes, there will be many thousands of pros who will skip a superior sensor that offers them no practical advantage, for a camera that actually has handling that allows them to get work done.

Grip > Sensor.

As I've written about in the past, try holding and wielding a camera for 8 hours or more in a day - active shooting. Being able to comfortably wrap your hand around a grip is of big value. And given that Canon and Nikon offer up plenty of IQ for what many need, the priority then shifts to other factors.

For mirrorless to really move into DSLR, I said in another post that there will need to be mirrorless bodies the size of 5D4 w/ battery grip, or 1DX2 sized.

Think of it this way, you could build a Honda Civic into the greatest car in the World - it won't matter, as there will always be a market for a larger, more comfortable vehicle even though it might be inferior in features.

Battery is another factor. Before mirrorless can do anything serious - they have to get the battery life issue under control. Right now, their battery life even at the best of the best is pathetic compared to DSLR.


Darkly said:
Canon need to up their game significantly before they lose their reputation as one of the big two go-to, no-brainer camera manufacturers. Anyone citing figures showing mirrorless sytems as having a small market share, let alone Sony's modest market share would do well to look at other incumbent giants who lost their way through complacency. Remember Nokia? They could do no wrong until a lean and hungry Samsung came along and innovated their way to the top (even if the they were blatantly taking design cues from Jony Ives).

Reputation.

Canon's reputation among pros is still the best. Canon's reputation among hobbyists and amateurs is fading. One group does work with cameras. The other doesn't mind the various shortcomings of Sony, and values all-out sensor performance.


Darkly said:
So, I think Canon either need to follow the same technological direction as Sony/Nikon to achieve similar noise and dynamic range figures

So, Canon should follow the same technological direction as Sony'a a9 and Nikon's D5, both of which have more low ISO noise and thus less DR than their predecessor? The a9 is billed as a 'pro' camera and yet it has over a stop less low ISO DR than the 1D X II. Why, exactly, should Canon follow suit?


TL:DR — YAPODFC.


Nikon took a step back on DR for sure. They realized, and rightly so, that high ISO performance is of more value in the these bodies than the silliness of maxing out the DR at ISO 100.

Speaking of which. I remember countless, endless threads where people told us that DR doesn't come at the expense of ISO. Oh really? What are the SoNikonian pseudo-electrical engineers saying now about the D5? Shouldn't the D5 be a high ISO beast and still crank out the big DR?

Obviously, sensor design can be TUNED for an expected result.

Looking at the various tests and charts, the D5 gives up low ISO performance, to gain it at high ISO's. It follow more of a Canon style approach of having a flatter graph to give the best results across a wide spectrum of ISO.

The mighty D810 is a turd at high ISO. But all we see and hear about from the Sony crew is how amazing underexposing a shot at ISO 64 by 5 stops allows excellent recovery.

New school of photography. The ISO 64 underexpose by 5 stops method and recover in post. Look ma! I saved the highlights!

Even though my photos look flat, have no punch, with weaker colors and have an HDR'ish character to them.
 
Upvote 0
The numbers don't really align with your theories, but what's new?

The loss of DR in the Nikon is much higher than the gain in ISO performance, the Sony A9 beats the 'tuned' D5 everywhere, the 1DX MkII does most places on the DR and ISO range too.

The simple fact is the 'tuning' went arwy and the D5 sensor is worse than the D4 sensor pretty much everywhere but for a 1/2 a stop at unusable ISO IQ ranges. I don't know anybody, even for newsprint, who is shooting regularly above 25,600 ISO for publication and actual image sales, that a camera has an advantage over that number that still isn't usable is pretty much irrelevant.

Besides, I don't believe it is the high ISO that creates the noise, I believe it is the fast frame readout that is the wall the high ISO and frame rate cameras are hitting.

Personally, as a primarily 100-1,600 ISO shooter, I am pretty happy that the Canon DR is the best of the bunch in that range, but ithe 1DX MkII, D5 and A9 are all within a stop of each other so nothing to get bent out of shape over nor to swap systems over.
 
Upvote 0
The Bible foresaw the mirrorless-mirror debate and has the answer to darkly most specifically:

Corinthians 13:12 King James Version (KJV)

For now we see through a glass, darkly; but then face to face: now I know in part; but then shall I know even as also I am known.
 
Upvote 0
AlanF said:
The Bible foresaw the mirrorless-mirror debate and has the answer to darkly most specifically:

Corinthians 13:12 King James Version (KJV)

For now we see through a glass, darkly; but then face to face: now I know in part; but then shall I know even as also I am known.

Sounds like a scathing smite of Sony's SLT cameras and Canon's Pellix series. 8)
 
Upvote 0
neuroanatomist said:
AlanF said:
The Bible foresaw the mirrorless-mirror debate and has the answer to darkly most specifically:

Corinthians 13:12 King James Version (KJV)

For now we see through a glass, darkly; but then face to face: now I know in part; but then shall I know even as also I am known.

Sounds like a scathing smite of Sony's SLT cameras and Canon's Pellix series. 8)

Oh, bible pellicle snap. Nice.

- A
 
Upvote 0
Sony A9 is not ready for prime time.

It's a very positive step forward for mirrorless, but mirrorless is still in its infancy.

Sony A9 is no threat to the 1DX2 because of -

Ergonomics, grip, controls
Canon's vast array of great white lenses.
Superior battery life
Ruggedness and surviveability in the environments where such pros actually operate
Extensive Canon network support of Pros

All this doesn't even begin to touch any matters of IQ, metering and how photos are taken. Nor any of the other pro features built into the 1DX2.

The Sony A9 though will be popular for those needing a completely silent shutter. I can see it being a good side camera for a wedding pro. Might be OK for some indoor sports where noise matters and distances are close. Maybe like billiards or something LOL.

High volume, all day shooters will not appreciate the lack of a grip and the awful battery life. Basically, serious pros.

Most of the people praising the Sony A9, have never held a 1DX2 in their hands. Let alone used one. And certainly have never been in a scenario where they can appreciate/need it.

This is why they spew the nonsense that the A9 is a 1DX2 competitor. It's laughable to say the least.
 
Upvote 0
privatebydesign said:
The numbers don't really align with your theories, but what's new?

The loss of DR in the Nikon is much higher than the gain in ISO performance, the Sony A9 beats the 'tuned' D5 everywhere, the 1DX MkII does most places on the DR and ISO range too.

The simple fact is the 'tuning' went arwy and the D5 sensor is worse than the D4 sensor pretty much everywhere but for a 1/2 a stop at unusable ISO IQ ranges. I don't know anybody, even for newsprint, who is shooting regularly above 25,600 ISO for publication and actual image sales, that a camera has an advantage over that number that still isn't usable is pretty much irrelevant.

Besides, I don't believe it is the high ISO that creates the noise, I believe it is the fast frame readout that is the wall the high ISO and frame rate cameras are hitting.

Personally, as a primarily 100-1,600 ISO shooter, I am pretty happy that the Canon DR is the best of the bunch in that range, but ithe 1DX MkII, D5 and A9 are all within a stop of each other so nothing to get bent out of shape over nor to swap systems over.

Can you explain pls. I want to learn this.
 
Upvote 0
privatebydesign said:
The numbers don't really align with your theories, but what's new?

The loss of DR in the Nikon is much higher than the gain in ISO performance, the Sony A9 beats the 'tuned' D5 everywhere, the 1DX MkII does most places on the DR and ISO range too.

The simple fact is the 'tuning' went arwy and the D5 sensor is worse than the D4 sensor pretty much everywhere but for a 1/2 a stop at unusable ISO IQ ranges. I don't know anybody, even for newsprint, who is shooting regularly above 25,600 ISO for publication and actual image sales, that a camera has an advantage over that number that still isn't usable is pretty much irrelevant.

Besides, I don't believe it is the high ISO that creates the noise, I believe it is the fast frame readout that is the wall the high ISO and frame rate cameras are hitting.

Personally, as a primarily 100-1,600 ISO shooter, I am pretty happy that the Canon DR is the best of the bunch in that range, but ithe 1DX MkII, D5 and A9 are all within a stop of each other so nothing to get bent out of shape over nor to swap systems over.

I thought the A9 is a stop worse than the d5 in terms of DR?
I had no intention to buy an a9, but I was very curious to check out at least its AF. After reading the overheating reports, I'm not sure now if it's even worth trying? I'm in Texas, and 110F ambient in summer is nothing extraordinary here. I doubt the camera will survive it... The jello effect is a very minor concern, I'm not using panning often.
 
Upvote 0
Jopa said:
I thought the A9 is a stop worse than the d5 in terms of DR?

It "holds its own" (which for the Sony fanboys will be damning with faint praise):

https://petapixel.com/2017/05/29/sony-a9-versus-nikon-d5-canon-1d-x-ii-dynamic-range-tests/

As an aside, I'm enjoying the fact that at low ISO, the 80D bests the D5 for DR!
 
Upvote 0
K said:
Grip > Sensor.

Or in longhand: a comfortable, effective grip is more important than the last few percent of sensor performance - yes, yes, and yes again.

K said:
As I've written about in the past, try holding and wielding a camera for 8 hours or more in a day - active shooting. Being able to comfortably wrap your hand around a grip is of big value. And given that Canon and Nikon offer up plenty of IQ for what many need, the priority then shifts to other factors.

For mirrorless to really move into DSLR, I said in another post that there will need to be mirrorless bodies the size of 5D4 w/ battery grip, or 1DX2 sized.

Film DSLRs had slimmer, lighter bodies, and I'd be very pleased to see a return to that, but there is more need than ever for the body to be large enough in the other dimensions.
 
Upvote 0
Steve Balcombe said:
Film DSLRs had slimmer, lighter bodies, and I'd be very pleased to see a return to that, but there is more need than ever for the body to be large enough in the other dimensions.

A film DSLR? Surely you typed that in haste.

In defense of the weight, film cameras didn't exactly have to have a Fotomat booth on-board. ::) But I hear you, any reduction in weight should be welcomed. And Canon is doing that. Latest gen L lenses are lighter than the prior revisions, the 5D4 shed 150g from the weight of the 5D3, etc.

But I don't think we'll ever see a 'Canon 5D Air' for a host of reasons:

  • Pros like the weight as a counterbalance to the big pro glass, which cannot reasonably get that much lighter given how much glass is required to support wide aperture use.

  • Such a product (an ultralight FF DSLR) would require a top to bottom redesign -- it's more than just swapping out steel/magnesium for carbon fiber outer components. That's a huge delta to their manufacturing processes, tooling, assembly, etc., let alone a huge step away from the reliability of how the cameras are currently being put together today.

  • I'm sure they've run the numbers on this, and either the amount of extra money / new customers they'd get for offering this would pale in comparison to the cost it would take to deliver.

So I could see a one-foot-in-front-of-the-other approach to weight in which they shave 10% or so each generation. Maybe we'll see a carbon fiber body someday, but not a magically paper-weight sort of rig.

- A
 
Upvote 0
Steve Balcombe said:
Film DSLRs had slimmer, lighter bodies, and I'd be very pleased to see a return to that, but there is more need than ever for the body to be large enough in the other dimensions.
no they didn't. film cameras with AF/AE. modern ergonomics and built to support larger f2.8 glass weren't small.
the EOS-3 is bigger than the 5D Mark IV

EOS-3: 161 x 119.2 x 70.8mm
5D IV: 150.7 x 116.4 x 75.9 mm

if you remove the top plate LCD, touch buttons,etc,etc and turn the camera into the ergonomic nightmare of a rebel T6, then sure, you got a smaller film (or even digital camera) but seriously who the hell wants that anyways?
 
Upvote 0
ahsanford said:
Steve Balcombe said:
Film DSLRs had slimmer, lighter bodies, and I'd be very pleased to see a return to that, but there is more need than ever for the body to be large enough in the other dimensions.

A film DSLR? Surely you typed that in haste.

In defense of the weight, film cameras didn't exactly have to have a Fotomat booth on-board. ::) But I hear you, any reduction in weight should be welcomed. And Canon is doing that. Latest gen L lenses are lighter than the prior revisions, the 5D4 shed 150g from the weight of the 5D3, etc.

But I don't think we'll ever see a 'Canon 5D Air' for a host of reasons:

  • Pros like the weight as a counterbalance to the big pro glass, which cannot reasonably get that much lighter given how much glass is required to support wide aperture use.

  • Such a product (an ultralight FF DSLR) would require a top to bottom redesign -- it's more than just swapping out steel/magnesium for carbon fiber outer components. That's a huge delta to their manufacturing processes, tooling, assembly, etc., let alone a huge step away from the reliability of how the cameras are currently being put together today.

  • I'm sure they've run the numbers on this, and either the amount of extra money / new customers they'd get for offering this would pale in comparison to the cost it would take to deliver.

So I could see a one-foot-in-front-of-the-other approach to weight in which they shave 10% or so each generation. Maybe we'll see a carbon fiber body someday, but not a magically paper-weight sort of rig.

- A

One of the problems with going to lighter composite materials is heat dissipation.... With high burst rates and particularly with video, we have a "killer ap" :)
 
Upvote 0
ahsanford said:
Steve Balcombe said:
Film DSLRs had slimmer, lighter bodies, and I'd be very pleased to see a return to that, but there is more need than ever for the body to be large enough in the other dimensions.

A film DSLR? Surely you typed that in haste.

Yes :-)

ahsanford said:
In defense of the weight, film cameras didn't exactly have to have a Fotomat booth on-board. ::) But I hear you, any reduction in weight should be welcomed. And Canon is doing that. Latest gen L lenses are lighter than the prior revisions, the 5D4 shed 150g from the weight of the 5D3, etc.

But I don't think we'll ever see a 'Canon 5D Air' for a host of reasons:

  • Pros like the weight as a counterbalance to the big pro glass, which cannot reasonably get that much lighter given how much glass is required to support wide aperture use.

Not sure about term "big pro glass" - two of my most frequently used lenses are a 500/4L IS II and a 100-400L II, which has nothing to do with being professional. The lenses I use for my professional work are a fraction of the size.

ahsanford said:
  • Such a product (an ultralight FF DSLR) would require a top to bottom redesign -- it's more than just swapping out steel/magnesium for carbon fiber outer components. That's a huge delta to their manufacturing processes, tooling, assembly, etc., let alone a huge step away from the reliability of how the cameras are currently being put together today.

I think you've lost the context here - we're talking about a FF mirrorless to replace a DSLR, which would obviously be a new design from the ground up. Perhaps you wrote in haste? ;-)
 
Upvote 0
Mt Spokane Photography said:
Sony is Sony. They excell at making products smaller, and have had success doing that. They seem slow to understand the camera market, first, they thought putting floppy disks into huge (relatively) point and shoot cameras would start a revolution, and a lot of them sold to consumers based on the Sony name. There were two big flaws.

1. File compression resulted in poor images
2. The floppy disk drives were horribly failure prone, and usually died in a year or two, Sony wanted $300 to replace one.

Then, they replaced the Floppy Drive with Memory Sticks, even though SD cards were already the standard. I guess they thought buyers would go for that, and die hard Sony fans did.

Now, they use SD cards, and are putting out tiny but very expensive cameras but lenses are still huge, and they have yet to release some super telephotos, their best are very expensive and do not match the competition. Sure enough, they will sell some, but they still see their market as those who buy anything Sony and are willing to pay a 30% premium. The Sage continues.
You may want to test the G Master lenses. They are expensive but the are also very good. I'm just saying that because as renters we have to provide what the client wants and in motion picture the Sony A7S II is widely used in specialist situations.
 
Upvote 0
neuroanatomist said:
Sporgon said:
I had a good laugh at this given how the Sony sensors had been the Holy Grail of low ISO DR, and how incredibly important to everyone this was. And if you didn't know how important this was to you then you were clearly in need of enlightenment, and of course there were quite a few Evangelists out there who were prepared to do that for you.

12 stops of low ISO DR for the a9, 13.5 stops for the 1D X II. I wonder why Rishi isn't comparing the a9 to it's "best-performing peers" (which is how he referred to the a7RII in the 1D X II testing)? ::)

I did, 6 days before your post in fact:

https://www.dpreview.com/articles/7266455439/sony-a9-real-world-iso-invariance-and-dynamic-range

But I'm sure you'll find some way to suggest this piece was actually an ode to Sony in disguise. I look forward to seeing the convoluted logic you'll use to pull that one off.

Also, I see you've cleverly / deceivingly compared my pixel-level DR estimate (24 MP) of the a9 to the print normalized DR (8 MP) of the 1D X II. You wouldn't be trying to mislead our own friends here, would you?

You really think that's fair? Back when I published papers in scientific journals, I didn't plot comparisons on vastly differently scaled axes on the same graph without indicating the different scales, but maybe that's something your lab is used to? Does it help you get funding by faking your data? I'm genuinely curious.

The theoretical max pixel-level DR of a 12-bit ADC (what I suspected the a9 is run at, much to Sony's chagrin btw) is 12.5 EV, which normalized to 8MP is 13.3 EV... which, incidentally, is precisely what DXO measured.

So that's actually only 0.2 EV behind the 1D X II. Though in our review of the a9 we mentioned we preferred the look of the 1D X II's deep shadows b/c the a9 shows some pattern noise.

In other words, my estimate was correct all along, and I even compared the a9 to the a7R II from the beginning. Your implication I didn't in a public post can be classified as fake news, but that could be said about so many of your posts, like your misleading comparison of pixel vs. print DR.

Frankly, I have no idea what you're actually complaining about, other than simply trying to mount another unprovoked, specious, unfounded personal attack.

-Rishi
 
Upvote 0