The full frame look is a myth. I did some tests with the A7S and Canon 70-200 F2.8L II- you can too with any camera. The math is easy. Matching settings is a little trickier, but it's clear there's nothing special or magical happening with a full frame sensor vs. crop sensor. Just like there's nothing special happening with medium format vs. full frame.
The Full Frame Look is a Myth. Here's how to Prove it to Yourself
- Thread starter jcs
- Start date
Ryan Lightbourn
Well-known member
Alex, I'll take full frame over crop for $400.
HeldvomFeld
Active member
Don't get it! What myth are you talking about? Everybody knows that it's simple math, nothing to do with magic. Which doesn't mean that there isn't a full frame look, there is!
Ryan Patrick O'Hara
Veteran
Interesting write up! Nice work getting the images to match so well!
Well, inherently there are some differences in the realm of optical physics and with sensor variables depending on what you are discussing.
When discussing optics, clearly the FoV of a lens should be taken into account, as should the difference of depth of field when matching focal length FoV between two sensor sizes. You can match DoF with aperture adjustment, in many cases, but that also changes the performance of a lens in regards to sharpness, uneven field illumination and chromatic aberration.
I'm a S35mm man, myself. I don't prefer to work with full frame, but either way, I do know there are differences and I do take them into account when shooting.
It's true that full frame doesn't have any sort of magic that makes it definitively *better* than S35mm, but it does have the potential to do things that smaller sensors cannot. For one, it can acheive shallower DoF than smaller sensors. It has the potential to be more light sensitive, but not guarenteed. The list goes on, but you certainly have the cons too!
Well, inherently there are some differences in the realm of optical physics and with sensor variables depending on what you are discussing.
When discussing optics, clearly the FoV of a lens should be taken into account, as should the difference of depth of field when matching focal length FoV between two sensor sizes. You can match DoF with aperture adjustment, in many cases, but that also changes the performance of a lens in regards to sharpness, uneven field illumination and chromatic aberration.
I'm a S35mm man, myself. I don't prefer to work with full frame, but either way, I do know there are differences and I do take them into account when shooting.
It's true that full frame doesn't have any sort of magic that makes it definitively *better* than S35mm, but it does have the potential to do things that smaller sensors cannot. For one, it can acheive shallower DoF than smaller sensors. It has the potential to be more light sensitive, but not guarenteed. The list goes on, but you certainly have the cons too!
Last edited:
Lol Ryan. Full frame has advantages in affordable lens choices- however full frame doesn't do anything special with the image when set up with equivalent settings with a Super 35 / crop setup. By equivalent settings- the settings aren't the same- they are different (scaled by the cropfactor). The DOF and FOV are equivalent. When this is done, there isn't any significant different between S35 and FF. The images I posted demonstrate this- did you check them out?
Baddgsx- I have a Speedbooster and FS700- that's true if you don't use the equivalence math to set up Super 35 vs full frame (SB Ultra is actually a 1.5*.71 = 1.07 crop*). You can't compare the same focal length and aperture- that's why full frame appears to have shallower DOF. To compare the FS700 with & without the Speedbooster, you'd need to change the settings as follows when using the Speedbooster: after shooting without the Speedbooster (Super 35) you'd add the Speedbooster, then change the focal length by *1.5 (the crop factor), e.g 70mm*1.5 = 105mm, change the aperture from e.g 2.8*1.5 to 4.2 (or 4.0 if closest), and up the ISO from e.g. 800*(1.5*1.5) to 1800 (or 1600 if closest). As noted the Speedbooster Ultra isn't actually full frame, it's a 1.07 crop, so to get a better match, you'd use a crop factor of 1.5/1.07 = 1.4 instead of 1.5.x.
Did you look at the settings and images from the link? Can you tell the difference between the full frame and Super 35 shots when the camera is set up for equivalent DOF and FOV?
* on the MB site, they had listed the gen 1 SB EF to E-mount with a 1.09 crop. I had never done the math, and like many others quoted a 1.09 crop for the SB + FS700. The gen 1 SB also has a .71x focal reducer. The correct crop factor is 1.5*.71 = 1.07. These guys reported what MB stated (1.09) and measured 1.07: http://www.dpreview.com/articles/2667195592/first-impressions-metabones-speed-booster .
Baddgsx- I have a Speedbooster and FS700- that's true if you don't use the equivalence math to set up Super 35 vs full frame (SB Ultra is actually a 1.5*.71 = 1.07 crop*). You can't compare the same focal length and aperture- that's why full frame appears to have shallower DOF. To compare the FS700 with & without the Speedbooster, you'd need to change the settings as follows when using the Speedbooster: after shooting without the Speedbooster (Super 35) you'd add the Speedbooster, then change the focal length by *1.5 (the crop factor), e.g 70mm*1.5 = 105mm, change the aperture from e.g 2.8*1.5 to 4.2 (or 4.0 if closest), and up the ISO from e.g. 800*(1.5*1.5) to 1800 (or 1600 if closest). As noted the Speedbooster Ultra isn't actually full frame, it's a 1.07 crop, so to get a better match, you'd use a crop factor of 1.5/1.07 = 1.4 instead of 1.5.x.
Did you look at the settings and images from the link? Can you tell the difference between the full frame and Super 35 shots when the camera is set up for equivalent DOF and FOV?
* on the MB site, they had listed the gen 1 SB EF to E-mount with a 1.09 crop. I had never done the math, and like many others quoted a 1.09 crop for the SB + FS700. The gen 1 SB also has a .71x focal reducer. The correct crop factor is 1.5*.71 = 1.07. These guys reported what MB stated (1.09) and measured 1.07: http://www.dpreview.com/articles/2667195592/first-impressions-metabones-speed-booster .
Ryan Patrick O'Hara
Veteran
I certainly did read the article. You did do a nice and thorough job matching the two with settings. However, you did have to forsake ISO and stop down the full frame lens in order to match the two. By doing this you inherently change the quality. Maybe not so noticeable in a photograph, but I know with almost any camera I typically use, doubling ISO changes the noise floor noticeably. I also can tell you that almost any vintage lens will show considerable differences in performance from wide open to a stop-and-a-half down. Shooting SuperSpeeds or StandardSpeeds at T/1.3 or T/2.1 is a much different image than T/2.5 or T/3, etc.
You certainly prove that the camera formats can be matched, but it ignores the advantages full frame can have with FoV and extremely shallow DoF, which S35mm cannot match and it also ignores the sacrifices made to match the formats. Small, but significant depending on the camera and lenses being used.
But it was a very good job matching.
Well as you mention in your article, Full frame has the potential to do some things s35mm cannot, and vice versa.
You have to admit that despite FoV and DoF matching, the resulting lens being shot wide open or 1.5 stops closed down will have different optical qualities. That's also ignoring the blatant effect of having increased noise from pumping ISO in order to stop down.
Your test is pretty good. i suppose the amount of wiggle room depends on the equipment and settings. i can think of some situations where the compensation will greatly effect image quality and some where it will not effect the image nearly as much
There is nothing special about a full frame sensor. We can set up a Super 35 sensor to look equivalent to the full frame sensor. Can you tell the difference between the 4 images in the link I posted? Super 35 vs full frame. Same lens, different settings to match FOV and DOF. We match a full frame sensor to Super 35 by scaling the focal length and aperture by the crop factor (ISO is scaled by the crop factor squared). Going the other way we divide. What is the full frame look when we can get the same look on Super 35, micro four thirds, medium format, etc? Full frame has the most affordable lens options for shallow DOF, however shallow DOF can be achieved with other sensor sizes at varying costs / available selections. So there's no such thing as a full frame look if all a full frame look implies is shallow DOF. If the full frame look implies something other than shallow DOF that other sensor sizes cannot do, let's see the examples which demonstrate an exclusive full frame look. Otherwise we can simply say shallow DOF (which will have different lens requirements based on sensor size) and perhaps even develop a vocabulary to describe specific shallow DOF which is not related to any specific sensor or lens.
[EDIT- the first Ryan was for Ryan L! Sorry about that!]
Ryan O- I wasn't able to find any examples of the full frame look other than implied shallow DOF. Do you know of any? Shallow DOF can be achieved with the right lenses. The only reason a smaller sensor can't match a full frame sensor for shallow DOF would be lack of lens availability. I can get decent shallow DOF with the 2x (and 2.3x in 4K) GH4 crop sensor and the Voigtlander 25mm F.95. This works out to F.95*2 = F1.9 and F.95*2.3 = F2.2 (4K). Speedbooster is another option. In both cases, still a small sensor, where we're increasing the shallow DOF capability. Your points regarding ISO and lens performance at different stops are right on, however for the A7S higher ISO isn't much of an issue until it's really high- 12,800 for some, for others much higher. My point is that full frame doesn't have any inherent special optical qualities in and of itself. It's really the limitations of the available lenses for the sensor sizes.
I've simplified this as much as I can and provided example images which match what the math predicted. This guy goes into far more detail (it's where I found the math): http://www.josephjamesphotography.com/equivalence/
Last edited:
HeldvomFeld
Active member
Well, yes but all of this is old news. Its like saying there isnt a difference between a Ferrari and a jetta when they both go 50mph. You need way more light with a smaller Sensor, faster aperture, higher ISO. And even then, you don't take into account lens compression, distortion and lesser image quality when opening up the aperture all the way. Not saying FF is always better, but all this stuff is old and it seems like it has been coming up since people are trying to justify buying the GH4.
Did you have a chance to look at the matched images between full frame and Super 35? Perhaps you're more skilled at googling than I- can you post similar images (old news) which do the same type of comparison? The idea is there's no such thing as a full frame look, because there's nothing inherently special about a full frame sensor. Smaller sensors can look the same as full frame, as demonstrated in the 4 images. Medium and large format cameras can of course also do shallow DOF. Thus, we're talking about shallow DOF, not anything specific provided only by a full frame sensor. If there was, we could call it the full frame look. This debate repeats over and over. Only recently did I try the equivalence math to match S35 to full frame and it became clear that there's no such thing as a full frame look. Aren't we're better off describing it as the shallow DOF look, since many different sensor sizes can achieve that look, some easier and lower cost than others?
HeldvomFeld
Active member
Although I did look at the pictures, I Didn't have to since, as I have said twice now, it is old news. There are many articles and videos on the math and how to achieve the same look with crop sensors. If this whole threat is really only about your dissatisfaction with the term "full frame look", then this is a waste of time. Call it what you like. I can't imagine anybody ever thought that there is mystery or magic behind full frame. It is a sensor with a specific size, that's it. Although, by stating that you can replicate the look with a small sensor, you pretty much imply that there is a look which you want to emulate.
alaskacameradude
Veteran
I think maybe the FF 'look' people refer to is shallow depth of field
at wider field of views. A 50mm F1.4 on Super35 is plenty shallow.
But it's harder to match the look of a 24mm F1.4 on a FF camera if you
are using a GH4. That would require a 12mm faster than F1 lens.
But personally I'm a Super35 fan....
Samuel H
Major Contributor
I made a very similar test and post a while ago, here, with mouseover images too. Basically I aggree with most of what has been said here: there's nothing magical about a bigger sensor, but may get better options in terms of what levels of sharpness, field of view and depth of field you can afford.
Edit:
Also, I'm glad that everyone here is so knowledgeable and thinks this is old news, but that's not the case everywhere. Plenty of people insist there's something magic about FF, other than the FoV/DoF that you can get with relatively cheap lenses.
Edit:
Also, I'm glad that everyone here is so knowledgeable and thinks this is old news, but that's not the case everywhere. Plenty of people insist there's something magic about FF, other than the FoV/DoF that you can get with relatively cheap lenses.
Last edited:
HeldvomFeld
Active member
This is a debate which just wont end, since people just want to justify whatever their camera choice is. Fact is, Sensor and lens both play a big role in creating an image but with each sensor size some pictures are easier created and some more difficult to create. As said above, is it easy to get a 24mm 1.4 equivalent picture with a m4/3 sensor? No it isnt! Is it easy to replicate a m4/3 200mm 2.8 equivalent with a FF-sensor? Well, you would need a 400mm lens for that, which most people dont have. We can spin this around forever. Would I buy a FF camera if I was shooting 90% wildlife or sports? Nope! Would I buy a m4/3 if 90% of my shots are supposed to have shallow DOF or I shoot in lowlight a lot? No!
morgan_moore
Major Contributor
Technically/Optically there is no debate.
Practically.. well FF does tend to 'win' if cheap lenses are your thing - a 24 or 28 2.8 or 50 1.4 ($150 or less) are hard to match on a smaller sensor with a litmited lens budget.
The last three years have even changed that tradeoff somewhat with the minor manufacs making some lenses that 'work' well on S35 or smaller - Samyang 16, SLRM12, voigtlander, et al.
CaNikon have not really made good glass for sub FF as they seem to consider it an 'amateur' format..
S
Practically.. well FF does tend to 'win' if cheap lenses are your thing - a 24 or 28 2.8 or 50 1.4 ($150 or less) are hard to match on a smaller sensor with a litmited lens budget.
The last three years have even changed that tradeoff somewhat with the minor manufacs making some lenses that 'work' well on S35 or smaller - Samyang 16, SLRM12, voigtlander, et al.
CaNikon have not really made good glass for sub FF as they seem to consider it an 'amateur' format..
S
Darren Levine
Veteran
Indeed old news, but im typically happy whenever there is talk against full frame, it needs persepective because there are still people obsessed with it.
By far the most Oscar nominations for this year, and likely every year before that have been films shot on a s35 format.
Regardless of pros an con's, there are the facts of the industry, and full frame just isn't a determining one.
By far the most Oscar nominations for this year, and likely every year before that have been films shot on a s35 format.
Regardless of pros an con's, there are the facts of the industry, and full frame just isn't a determining one.
HeldvomFeld
Active member
Well, for Film, Full Frame has never been "the" sensor size, and there are reasons for it. With photography its very different!
photochris
Well-known member
Like others have said, "the FF look" is FOV and DOF - which in many cases can be replicated with a smaller sensor when you do the math. Fast lenses wide open, well you're not going to get that extra stop of DOF and the FOV without a Booster. Same goes for UWA's. FF is just the most efficient way to take advantage of the fast glass or get the actual FOV of a lens since Speedboosters lack native AF capabilities and occasionally have compatibility issues. Its really not a big deal when shooting video because f1.2 is not an ideal aperture with anything moving. But shooting stills is a different animal - full length portraits shot with the 85/1.2 or the magical 200/2 are unreal. Given the fact that its pretty easy to get a sub-$1000 FF body, its a great time to be in the image-making game. That new Canon 11-24 is another one that's going to be something special and difficult to replicate with a crop sensor. Its great to have options.
Erik Naso
Veteran
S35 is magical. LOL! The only issue I have with smaller sensors like MFT is having to open the lens up to get shallow depth of field that I'm used to getting with APS-C and full frame. In doing so you're hitting the lenses weakest IQ. Also need really fast lenses to achieve it or shoot long but it's not always practical to use a long lens.
Larger sensor also seem to handle low light and noise much better than MFT. S35 is the sweet spot.
Larger sensor also seem to handle low light and noise much better than MFT. S35 is the sweet spot.