APSC vs MFT?

I keep wondering what is the attractiveness of an APSC DSLR vs an MFT camera

Am looking at Nikon D500 vs Olympus OMD-MKII and I really can't see why someone would want to shoot a DSLR today

If we put ports out of the mix for a second if I look a pricing a camera plus housing with Nauticam I get

D500 + NAD500 = $1,500 + $3600 = $5,100

OMD EM1 + NA EM1MKII = $1,399 + $2,010 = $3,400

The larger MFT like the G9 add another $700 bringing the total to $4,100

So $1000 to $1,700 price difference for maybe 1 Ev of dynamic range that on print or screen does not even show

As a former APSC user I do not see a reason but I am intrigued of why there is such an adoption in UW shooters. Is there something I am missing?

Sometimes the subject of focus comes about but the Olympus has phase detection if that was even a concern but I really can't see anything worth $1,000+ in the change

16 minutes ago, Interceptor121 said:

I keep wondering what is the attractiveness of an APSC DSLR vs an MFT camera

Am looking at Nikon D500 vs Olympus OMD-MKII and I really can't see why someone would want to shoot a DSLR today

If we put ports out of the mix for a second if I look a pricing a camera plus housing with Nauticam I get

D500 + NAD500 = $1,500 + $3600 = $5,100

OMD EM1 + NA EM1MKII = $1,399 + $2,010 = $3,400

The larger MFT like the G9 add another $700 bringing the total to $4,100

So $1000 to $1,700 price difference for maybe 1 Ev of dynamic range that on print or screen does not even show

As a former APSC user I do not see a reason but I am intrigued of why there is such an adoption in UW shooters. Is there something I am missing?

Sometimes the subject of focus comes about but the Olympus has phase detection if that was even a concern but I really can't see anything worth $1,000+ in the change

I guess it has to do with the lens and accessories someone might have invested in...  and of course marketing

Between MFT, APS-C and FF? Take a hard look at APS-C. The sensor is much better than the MFT, but falls a tad short of FF. The top, in my opinion, is the Sony a6400. I don't think you could go wrong.

16 minutes ago, bill1946 said:

Between MFT, APS-C and FF? Take a hard look at APS-C. The sensor is much better than the MFT, but falls a tad short of FF. The top, in my opinion, is the Sony a6400. I don't think you could go wrong.

Where have you got the data from? Best case there is 0.3 Ev benefit between a current 20.7 Megapixel APSC that goes into a Nikon and a 20.3 MP MFT sensor that goes into an Olympus

In fact independent data even shows that the benefit is at 'lab' level but not even useable

Best full frame sensor if we want to talk about sensor are far away from any cropped format with a gap of around 1.6 to 2 Ev

1 hour ago, Interceptor121 said:

Where have you got the data from? Best case there is 0.3 Ev benefit between a current 20.7 Megapixel APSC that goes into a Nikon and a 20.3 MP MFT sensor that goes into an Olympus

If you look at dynamic range, you have at least 2 EV difference, if not 3 EV between a Nikon D7500 and EM1 Mark II.

This dynamic range is very important for UW shots, as it greatly improves the ability to adjust WB.

of

Also important feature of Nikon camera, is "ISO invariance", meaning that you get equivalent results by underexposing image at ISO 100 and then correct exposure in post processing, or by pushing the ISO when you take the shot. This means that when you color correct your picture (and over expose red channel) the result will be less grainy.

You also have the issue that the EM1 Mark II has a minimum ISO of 200, where the APS-C starts at 100. You'll get a less grainy image to start with.

10 minutes ago, Algwyn said:

If you look at dynamic range, you have at least 2 EV difference, if not 3 EV between a Nikon D7500 and EM1 Mark II.

This dynamic range is very important for UW shots, as it greatly improves the ability to adjust WB.

of

Also important feature of Nikon camera, is "ISO invariance", meaning that you get equivalent results by underexposing image at ISO 100 and then correct exposure in post processing, or by pushing the ISO when you take the shot. This means that when you color correct your picture (and over expose red channel) the result will be less grainy.

 

You also have the issue that the EM1 Mark II has a minimum ISO of 200, where the APS-C starts at 100. You'll get a less grainy image to start with.

Those shots do not say anything about dynamic range and are not meant to compare cameras between themselves but just on its own and it shows if for a certain camera it makes sense to raise the ISO or not to recover shadows.

The graphic below is a measure of dynamic range instead comparing D500 G9 and Sony A6400. The only benefit limited to 0.3 Ev is in the range below ISO 100 however consider that you can actually rarely go there except for macro where DR is not a consideration. I have not included the OMD EM1 MKII as actually that camera is one of a kind and has a higher DR than anything else (though is not ISO invariant like instead Panasonic cameras and new olympus are)

Indeed "dynamic range" was an improper term for DPreviews measurements that I quoted, it's "exposure latitude" and "ISO invariance".

However I find them very relevant for UW photography. 

We often have to correct WB in our photos, to compensate the filtering of colors by water. This means that we have to push selectively the exposure of the Red channel when developing the raw file in post-processing. So what matters is not the classical dynamic range (difference between the lightest and darkest areas of a picture), but by how much we may push exposure.

Actually those graphics only tell you how the amplifier of your camera work. If you look closely when you move between 200 and 400 ISO the noise takes a hit at 200 this is because the camera has two gain circuits I have included here the same bit with the D7500 to include ISO 200 you can see that it looks noisy

In terms of ISO invariance actually some MFT cameras today are better than Nikon or Canon example GH5 here vs OMD EM1 (that is not so invariant) and D7500 that has a bump at 400. I think those charts are only good for marketing purposes and really say nothing useful except make you understand how your camera works in terms of gain and if there is a single or multiple amplifires. Nikon seem to have two as there is a bump in all models pretty much at 400

One metric that I have found of some use is the DxoMark Sport Value which is the highest ISO you still get SNR of 30 dB 9 Ev of DR and 18 bit color depth. It shows 1482 for the D7500 and 1312 for the OMD EM1 which is 0.177 Ev so nothing really too exciting anyway. 

So when I combine the ISO to get a decent image together with the actual usable dynamic range I get same value on ISO and DR delta of 0.77 Ev take into account that some MFTs like the G9 have a delta of 0.33 Ev on DR due to the area below 100 and in terms of usable ISO get to the same value which is why I started this post in the first place

In real life I shoot my GH5 at max ISO 800 and when I am on a boat with APSC and FF there is no difference with APSC on actual real life images, if there is it was not because of the camera but the operator

Hi Interceptor,
forget the lab measurements, and try it in the field...
I had the Oly OMD 5 mkI and have switched to a Nikon D500.
Simply because the sensor is better and allows me more things in low light.
For travelling the m43 setup was great, no question.
And i know you tell us that the OMD 1 mkII is much better.
But physics tell us a simple story.
The same sensor with more pixels can not be better than a bigger sensor ...
And with FF its just a step more..
Thats why i now try a Sony A6300 ( more or less the same sensor as D500, but smaller camera and much more smaller housing).
Simply because i wait for the second generation Nikon Z50 ( however that will be named ) and a Nauticam housing in the same design as the Sony A6xxx series. Had some Sony stuff, been using a Nex5, and 6 but then changed.
Light weighted, small and great camera / sensor inside. And not FF, that would limit it too much...
Sorry, Oly was great at that time, but what happens now? Macro was absolute top, and now?
The sensor size is keeping it on a based field, not much possibilities.
The other companies came later, but took their customers with them, and their lenses...
Now i will be trying a Sea&Sea 12mm fisheye, manual focus, but small and light weighted,
have a 15mm Nikonos as well and other lenses, Sony 90mm as macro lens....
If the next generation Nikon comes and Nauticam builds a small housing,
i know how to spend my money ;-))

Regards,

Wolfgang

1 hour ago, Interceptor121 said:

Actually those graphics only tell you how the amplifier of your camera work.

Actually, not the exposure latitude test: this test take pictures at the camera smallest ISO, and compare how decreased exposure when the picture is shot  can be compensated in post processing. This is what I believe is more relevant to reflect how well a camera will behave when processing significant WB correction.

Overall, smaller sensor means "Low light ISO". There is large variance within each category, due to the quality of the camera, and progress of technology (overall each camera generation improves on the previous). The top MFT camera (Lumix G9) is at the level of the top APS-C (D7500), but overall there is a significant difference. The gap is wider with FF: sensors cannot defy physics.

I agree with you that in most cases, it is the photographer which will make the difference, not the camera. Not least because the good photographer will know the limits and weaknesses of his camera. Also, for most "normal" shooting conditions, current cameras (MFT, APS-C, FF) will get great shots, with limited difference in quality from a practical point of view. But in challenging conditions, the camera will enable a better image "quality". This in itself is relative: "grain" is not always a defect, but can be used with an artistic intent.

The original premise of the post was m43 vs APS-C.  The surface area of the sensors tell a lot of the story, m43 = 225mm2,  APS-C 368mm2 compared to full frame at 860mm2 .  You are looking at 1-2 stops benefit for full frame and it has over double the surface area compared to APS-C.  it makes sense that the gap between APS-C and m43 would be significantly less than the gap between APS-C and full frame.

When designing sensors and their readout noise the manufacturers can trade off between noise, dynamic range and other parameters.   All of these parameters get smaller as ISO is increased.  Underwater you have more trade offs to make, for example bigger sensors with dome ports present problems so you need to stop down more and if you have ambient light involved you are probably going to increase ISO to compensate so the dynamic range you gained is at least partly traded off as you increase ISO to compensate for stopping down.  Macro with 100% strobe illumination is different of course but you are at base ISO more than likely and the Dynamic range/noise advantage while there is not going to provide to much benefit that is actually visible.

I did a similar analysis to Interceptor121 when I bought my EM-1 MkII system, I could get the housing, camera and one port for the price of the housing alone for a Canon system.  That along with the full range of excellent quality lenses and ports made the deal.

For me, I was definitely conflicted between M4:3 and APS-C, both in mirrorless format. I could certainly be happy with several of the excellent M4:3 offerings. But the Sony A6400 plus Nauticam is less than the original examples:

Sony A6400 w/15-50mm + Nauticam NA6400 = $998 ($898 body only) + $1730 = $2728

APS-C is a much larger sensor, not the huge jump to FF but certainly significantly larger than M4:3. Larger is good, so this forum says. And it is larger in the range of 60% vs. M4:3.

The price to performance to portability ratio is excellent. Even more portable than some M4:3 due to the high top housings in some Oly cameras will not sit vertical in my Pelican 1535, the NA6400 does easily.

I looked further into the companies, Sony makes money on it's consumer photo products, Olympus and Cankon do not. Sony has recently made it easier for other party access to certain data thus making it easier for lens makers to offer additional product. I believe that Sony mirrorless cameras will continue to excel and innovate. 

The A6400 arguably has the fastest autofocus of any camera, a number of tests show it faster or as fast as the newest A7 series. I intend to use my new A6400 for outdoor, wildlife and motorsports photography as well as UW so I wanted a system that had the fast autofocus and tracking AF and lens selections that can be used on both FF and APS-C formats. If I get a Sony A7X it may never see UW use, but I can share lenses with my A6400. 

Things that irk me with the Sony cameras are inability to select manual only flash with no-preflash. And then there is that menu! And the 1/160 sync speed. But I am using a TTL trigger and I lived with 1/60 to 1/125 for many years with Nikon SLRs and Nikonos. So I can make do. And, the Sony is HSS capable. 

I like the Sony camera better, I liked the APS-C format better than M4:3 in the end, I like the amazing AF. I liked the bigger sensor, it is bigger, no two ways about it. 

I have had APSC latest was Nikon D7200 and full frame (canon Film so would not really compare anything)
The image performance of the D7200 and GH5/G9 is identical in any land shot in daylight in low light GH5 not that good G9 same
I have also repeatedly been on boats with APSC last year D500 and in underwater shots there was no difference with my GH5 I doubt this was my mastery as the reference point has won a few categories at UPY and other competitions
In terms or Sony A6xxx they fall behind underwater in terms of real life shots not sure if this was operator issue or what exactly but consistently you get worse shots than an OMD EM1 MKII note am talking about a 20 megapixel sensor
For what concerns shooting at ISO 100 to raise exposure in post of 4,5 Ev this doesn’t really work underwater at best you have two ev even with a full frame camera to create an image that looks great, note am not saying an image that you can use as you don’t spend $5k to get Ok pictures in my view
Read Noise and DR traditionally go with the square of crop factor so the ration between APSC and MFT is (2x2)/(1.5X1.5)=1.78 translated in Ev this is 0.83 Ev. Full frame instead is 1:4 ratio so 2 Ev. This is obviously generic data that does not take into account how good is a specific camera implementation.Another point to consider is depth of field where MFT has 2 stops to FF and APSC 1.2 so the relative gap between APSC and MFT is 0.83 Ev as well
In practical terms if I shoot an MFT fisheye at f/5.6 and an APSC at f/8 so the 1 stop benefit (if it ever was there) of ISO and DR is sucked up by underwater port challenges and furthermore I need half the power in strobes. 
From the above reasoning the data I have collected and evidence taken on boats on similar shots I see zero benefit between a 20 MP APSC sensor and 20 MP MFT underwater despite this gap can be seen in certain conditions on land.

Full frame with a gap of 2 Ev is another story but even there the benefit in terms of real data are somewhat limited by depth of field vs field of curvature issue

So taking a fisheye lens if I have:

MFT f/5.6

APSC f/8

FF f/11

The Ev benefits among the formats are cancelled by underwater port issues and will only be visible when the sensor is outside the low noise zone which is that sport ISO parameter that I was mentioning if we compare Panasonic GH5 (ore typical than OMD) Nikon D7500 and D850 we have

D850 2660 ISO

D7500 1483 ISO

GH5 807ISO

Gap GH5 to D7500 0.8 Ev

Gap GH5 to D850 1.7 Ev

Gap D7500 to D850 0.843 Ev

So unless you manage to shoot an APSC or FF at the same aperture of an MFT on a fisheye lens the DR benefit and SNR benefit is eaten by underwater optics


Sent from my iPhone using Tapatalk

Hi,
but if you take different lenses, the game is wide open....
Have a Sea & Sea 12mm FE for Nikonos that i will use this trip attached to the Sony A6300.
No need for going to f8, its a water contact lens. And will bring a Nikonos 15mm as well.
So i can shot wider open as with in dome lenses...
Regards,
Wolfgang

19 minutes ago, trimix125 said:

Hi,
but if you take different lenses, the game is wide open....
Have a Sea & Sea 12mm FE for Nikonos that i will use this trip attached to the Sony A6300.
No need for going to f8, its a water contact lens. And will bring a Nikonos 15mm as well.
So i can shot wider open as with in dome lenses...
Regards,
Wolfgang

I am aware that some Nikonos lenses can be fitted to A6X series however those were designed for full frame once you add 1.5x crop they are less interesting. 15 becomes 22.5 and 12 becomes 18 mm ?

150 degrees no longer a fisheye on APSC

http://thesybersite.com/general/12mm-fisheye/

Are you not better off getting a Tokina at least it is a true fisheye?

I am using the WWL-1 and Inon UWL100/dome. I have not had the rig out on a real trip, just pool familiarity but the WWL-1 and Sigma 19 f2.8 looks really clean edge to edge at all apertures on the A6400.

I own a NIB Tokina 10mm-17mm lens I bought a few years back when I almost went DSLR. I will get the adapter and dome for it when I recover from all these purchases ;).

Regarding the Nikonos, the 15mm is not a fisheye, the S&S 12mm is. They are both still very wide and very clean even considering the crop factor and they are water corrected so they do not lose FOV when submerged. 

James

4 minutes ago, Captain Fathom said:

I am using the WWL-1 and Inon UWL100/dome. I have not had the rig out on a real trip, just pool familiarity but the WWL-1 and Sigma 19 f2.8 looks really clean edge to edge at all apertures on the A6400.

I own a NIB Tokina 10mm-17mm lens I bought a few years back when I almost went DSLR. I will get the adapter and dome for it when I recover from all these purchases ;).

 

James

OK but WWL-1 is also MFT so that changes nothing in the mix here as the sharpness will follow crop? If we go on the path of non standard configuration I shoot a canon 8-15 on my GH5 that has a zoom range that no camera including tokina on APSC has and is sharper than WWL-1

Hi,
the Tokina has the same problem, it gets longer, no question.
And some FF users even take a teleconverter to use it.
But the big benefit is that i dont need a dome!
So no reason to step down like f8, lightweighted and small if there is current...
And sharp corners...
And the 8mm m43 gets a 16mm, FF 16mm gets 24mm....
But next question is, what is the port doing to it?
How is the image changed before the lens throught the domeport glas / projection?
Would be interesting to see a side by side comparison.
Regards,
Wolfgang

Tokina is an APSC and as far as I can see 178 degrees diagonal on cropped

From comparisons posted by mustard with Nikonos the Nikkor fisheye looked better 

I know that some people rave about it but those are 40 years old products no longer produced and therefore not a mainstream consideration for someone going out and buying a camera

New wet contact optics exist and seem more flexible however they are expensive and at the end none of them fisheye as neither was the sea and sea that was 167 degrees on full frame so on yours probably 130 same as the WWL-1