[stevebrot]

Adam needs to make a sticky thread just for the topic of FF vs. APS-C and equivalency.

Adam did one better and created this FF sub-forum. Its intent is to be a dumping place for all such discussions and arguments. It operates sort of like the Hindu cycle of rebirth. Same content multiple times reborn as different "living things", each with different degrees of sophistication and social grace and flavor of trolling.


Steve

(...wondering what became of the OP...)

[jsherman999]

Adam needs to make a sticky thread just for the topic of FF vs. APS-C and equivalency.

Until he does, very good intro here: What is equivalence and why should I care?

Or, for a more thorough overview: Joseph James: Equivalence

[jatrax]

(...wondering what became of the OP...)

Might be out taking pictures....................

[cali92rs]

Might be out taking pictures....................

He bought a Samsung Galaxy because all that matters is "pixel pitch and technology"

[clackers]

I looked at DX0 'screen' results for K5, K50, A7S

The results did not show the K50 being better than the A7S
The results showed the K50 nearly equal to the K5 with the A7S better than both.
This is what I would expect to see.

I am afraid your arguments are over my head. I am not sure any further comments will help me understand your position. I will have to leave off conversation with you. Thanks for the help, though, and have a good morning.

You're not to blame, Tan, the Equivalency Fanatics have stared at those charts for years without understanding them.

Equivalency is either breathtakingly superficial ("Use a wider lens, you'll get the same PoV!") or makes non-scientific claims ("You don't get more noise with a full frame!").

For anyone else who wants to be gob-smacked, let's look at your example - the K50 and the A7S.

Sony A7S vs Pentax K-50

Compare the real 'Screen' measurements with the mathematical assumption on the 'Print' screen and you find:

Sony A7S at ISO 10587 improves its signal to noise ratio by 23.5 to 25.3 dB.

K-50 at 10843 improves by from 19 to 22 dB - that's the cropped sensor cleaning up its JPG or whatever you want to call it *better* than the full frame one.

Hands up all who realised that?

[clackers]

I'll stop you there ^

As Tan68 pointed out, you're misinterpreting DxO results - the 'screen' tab is for when you want to directly compare pixel performance between sensors. .

I'll stop you right there, Jay.

The screen tab *is* the DxOMark measurement, and tells you the contents of your RAW file.

The Print tab does not. Nobody did a measurement with any sensor for that.

Total light & sensor size have been taken out of that comparison ...

Oh, beautiful, you're the Venn diagram intersection of Total Light Fanatics and Equivalence Cultists!

Did you know 'Total light' and 'Sensor size' are not considered at all in the Print values, if that's what you were hoping for?

Methinks you need to go the DxOMark site and see how those values are actually generated.

[clackers]

He bought a Samsung Galaxy because all that matters is "pixel pitch and technology"

A phone actually has incredibly small pixel pitch, Cali, but I'm sure you knew that.

The large number of them compensates their size when downsampling (can you mention that to JSherman? He thinks that step in DXOMark is due to the size of the sensor, not the number of pixels) which is why phones actually produce decent pictures these days.

Don't be ashamed to use one!

[clackers]

Or, for a more thorough overview: Joseph James: Equivalence

AFAIK, he's a fashion photographer chucked out of the DPR forum in the past who did a diatribic blogpost some have bookmarked as a kind of holy relic.

The time I once spent reading it is time I won't get back, unfortunately. My training is in Physics so I had to read genuine Optics papers by people who actually knew what they were talking about.

Good "appeal to authority" - not!

[Sandy Hancock]

Perhaps I am not as knowledgable as the combatants here, and I hope I'm much less dogmatic.

For portraiture, I see the K-1 as giving an advantage because for the same depth of field you can stop the lens down more to getter a sharper subject for the same amount of background blur - in other words, better subject isolation. The photographer can compensate by adding more light with flashes, etc, or by bumping the ISO, which should be IQ neutral as the K-1 performs better in that regard.

For landscapes, I see the K-1 as providing advantages because of the higher resolution and improved dynamic range. Stopping down a little to maintain adequate depth of field should be fine as the subject doesn't move much and a tripod as almost de rigueur anyway. Diffraction limits are higher with bigger sensors so stopping down to f/16 or f/22 shouldn't compromise on resolution. If you really need to maintain shutter speed the ISO argument above still holds. Plus my beautiful FA31 becomes a wide angle again

For macros, the K-1 offers a very useful tilting screen and wonderful resolution, but the K-3 gives a deeper focal plane and more magnification due to the crop factor and pixel density. I'd say the jury's out on this one.

For wildlife and sports, I reckon the K-3 still offers advantages of more reach, higher frame rate, and deeper field, even traded against the supposedly better AF and SR of the K-1. I suspect the K-3 successor will have these anyway, so APS-C will always have a place in my heart for these applications.

Please educate me if I've got it all wrong, but play the ball, not the man. We should all be friends here.

[Ian Stuart Forsyth]

The screen tab *is* the DxOMark measurement, and tells you the contents of your RAW file

Understanding what that data represents is another story

take the 36mp FF that's one data point in 36mp and the 16mp cropped that's 1 data point in 16mp this may have an affect on how the image is going to look and more importantly how we interpret those findings.

Or lets look at it another way if I use Film A ( sensor) on my 35mm camera can I get the same or better image on my 6x7 with the same film A only using more of it ( Larger sensor) i am sure none here would argue that the IQ would be the same and many would say that using the larger format would give you better IQ

you have to understand the data to understanding why you will see a difference.

If we use your own interpretation of the DXO data you will find that the k3 is not to far behind the k10 but none here would say that data represents what we see in the field or in the digital darkroom



Also you don't see many people using the K10 for its better performance up to iso1600

[monochrome]

Here's a really simple take from a casual, aspirational photographer.

There has to be a reason Canon and Nikon sell all those FF cameras to all those professional photographers. I don't need to understand what the reason is. Those pros want whatever it is those FF cameras offer, especially - it seems - wedding photographers.

I know 35mm film better than I know APSc sensors. I think I will be very comfortable using the K-1. I don't need to understand why - I only need my final image to relate to my lenses the way they do on my film cameras.

[Ian Stuart Forsyth]

Perhaps I am not as knowledgable as the combatants here, and I hope I'm much less dogmatic.
For portraiture, I see the K-1 as giving an advantage because for the same depth of field you can stop the lens down more to getter a sharper subject for the same amount of background blur - in other words, better subject isolation. The photographer can compensate by adding more light with flashes, etc, or by bumping the ISO, which should be IQ neutral as the K-1 performs better in that regard.

yup

For landscapes, I see the K-1 as providing advantages because of the higher resolution and improved dynamic range. Stopping down a little to maintain adequate depth of field should be fine as the subject doesn't move much and a tripod as almost de rigueur anyway. Diffraction limits are higher with bigger sensors so stopping down to f/16 or f/22 shouldn't compromise on resolution. If you really need to maintain shutter speed the ISO argument above still holds. Plus my beautiful FA31 becomes a wide angle again

I will add that because of the greater pixel count you can trade that resolution for NR and have a cleaner image when compared to a smaller format shot at the same DOF and shutter speed

For wildlife and sports, I reckon the K-3 still offers advantages of more reach, higher frame rate, and deeper field, even traded against the supposedly better AF and SR of the K-1. I suspect the K-3 successor will have these anyway, so APS-C will always have a place in my heart for these applications.

This reach is only apparent if you are FL limited but if you have the selection of lenses that allows you to obtain the same FOV on FF as cropped then the FF would give you more reach at a given FOV
Some will argue that well the same lens will give you better reach on the K3 that is correct but my decision on buying a lens is based on the FOV I am going to use it the most.

[Sandy Hancock]

If you have the selection of lenses that allows you to obtain the same FOV on FF as cropped then the FF would give you more reach at a given FOV.

Sure, but at much greater size, weight and cost.

[Ian Stuart Forsyth]

Sure, but at much greater size, weight and cost.

Only if that lens you have selected has shallower DOF or you are forced into buying a faster lens than you need

200 F/2 on cropped compared to FF 300 F/2.8
200 F/2.8 to 300 F4
cropped 300 F/2.8 to FF 400 F4
all about the same size, cost and weight

[derekkite]

So my 500mm f4.5 Sigma, full frame lens has a depth of field at 75m of 6m on FF, 4.2 or so on apsc. The coyote that I saw today at 75m and got a decent shot with my k3 1250iso 1/1600 f4.5. I cropped it by about 1/3, 2000*1350. I won't print it, it isn't good enough.

Dof is not an issue, I had adequate shutter speed. A k1 would give me a couple stops improvement on noise. It wasn't bad to start with and cleaned up reasonably well. If be at 1600*1000 roughly pixels, probably pretty close to equivalent. No real gain or loss. An hour later that same shot would be 1/800 2400iso, and on the k3 it would be a mess. But on the k1 it potentially would be close to what I got with more light.

I get lots of shots like that noise, vibration, focus are the limiting factors.

Another one a few weeks ago, 1/1250 Iso 1600 f5.6 500mm. Dof is not a factor here. A shot of a pied bill grebe, cropped a bit tighter than the first one, 1800*1200. Noise decreases resolution and detail, if I wanted to show it I would apply two noise levels, subject and background. The half stop gives better tonality from the lens, worth giving up. I could drop shutter speed a bit but I run into vibration below 1/1000 on the k3.

I suspect some of these shots would be better, some worse. I get few keepers in any case. I'm curious about the shutter. I could get handheld 1/40 400mm shots with the k5, but I run into mirror slap with the k3.

Another shot I wish I could have printed was early light in October, watching bears. 1/320 Iso 1000 f4.5. At these light levels the k3 noise and color is not very good. They were close, the image was cropped in half. Again a minor loss of resolution but I think this one would have been on my calendar. 15 minutes later the shots were wonderful, warm morning light.

I shot with a 400 mm manual lens for a while and liked the length. 500 missed me some shots, being too close in.

At 10m the lens wide open has a depth of field of .07 and .1m. Guess which number matches which sensor.

Noise really matters. Any gain in resolution from the smaller sensor can easily be lost by noise.