[falconeye]

Because the "how is FF different?" keeps coming back, I developed a "short answer" to the question which condenses wisdom and many pages of discussion, some of them I was part of. So, the summary is:

• Properties as found with FF are a true superset of those with APS-C (as long as the bayonet is the same).
• The superset narrows to identity for lenses used with FF and APS-C, respectively, which have the exact same physical diameter of aperture, i.e., in mm.
• A 100 MPixel camera would require FF.
• Properties as mentioned above include: Feasible resolution, Diffraction limits, Noise, DoF, Weight and Size (of both body and lens).

I know, this is a cryptic message. If I would have to condense it even further, it would read:

• FF gives more choice.

---

Last edited by falconeye; 10-29-2008 at 09:53 AM. Reason: added diffraction limit

[Class A]

FF gives more choice.

Have you read Joseph James' essay on comparing sensor formats.?
In general his conclusion is the same as yours, but he notes that you need better high ISO performance from FF sensors in order to have the same kind of DOF choice. Otherwise, FF limits you to comparatively shallower DOF.

Also, assuming FF implies more pixels corresponding to the increase in area, I'm not sure if the need for more storage capacity and faster processors is a real added option, but I agree that this is a secondary issue which will become obsolete as technology advances.

[Class A]

Pentax is not following the FF trend just because they are just unable to get a ticket for the FF train!

How do you know they are "unable" as oppose to "unwilling" (because of marketing considerations)?

DPR found that the noise performance of the A900 is worse at ISO 800 just for that particular body, not in general case - just because the pixel density is too high and thus physical pixel pitch size is too small (it has 24.6MP!).

Noise is mainly affected by sensor size, not pixel size.
Given the same area, a sensor with smaller pixels has more noise per pixel but also has more pixels. Just perform pixel binning to get the same result as the sensor with the larger pixels. Also, the noise of the higher pixel density sensor will be more fine grained, i.e., more pleasing to the eye. There might be a point when the ratio between light sensitive area and light insensitive area gets too unfavourable or read out noise of too small signals from too small pixels becomes a problem, but I don't think you have any evidence that this is the case with any 24MP sensor.

[FF] DSLRs have more freedom on choosing shallower DoF

Freedom = "forced to" unless they have better ISO performance.

and have better 3d feel.

Surely that must be a subjective observation based on existing models. How would you substantiate this claim based on physics?

[Class A]

a) you loose the x1.5 crop factor. That's great for those with FA wide lenses and sucks for those who are into wildlife and sports.

Contrary to what others have said, you don't loose reach (telephoto power) with a FF sensor, if it has more pixels (corresponding the the increased sensor area), since you can simply crop an APS-C format from the FF sensor image, giving the same performance as if the shot had been made with an APS-C camera.

Yes, you don't get the full FF resolution for the same FOV but you get the same performance as with an APS-C sensor.

This assumes that your FF lens is as sharp w.r.t. to the APS-C crop area as the corresponding DA lens, which it may not be.

[jeffkrol]

Nope. This is not correct. DPR found that the noise performance of the A900 is worse at ISO 800 just for that particular body, not in general case - just because the pixel density is too high and thus physical pixel pitch size is too small (it has 24.6MP!). So, same case applies for APS-C 1.5X DSLRs when the pixel count just at 12MP (not even more! AS APS-C is just of 42.4% of the area of a FF sensor and so does the physical size of a pixel for receiving light if the pixel count is the same)

So you want to spend 2x as much and yet have the same or LESS resolution as an APS-C camera????
Think you should read this thread and learn what pixel density really means.. Be sure to follow all the links.
Pixel density revisited: News Discussion Forum: Digital Photography Review
As for the evidence, Emil (http://theory.uchicago.edu/~ejm/) has pointed you to a post of his, there's a long series of posts which you didn't follow, but did discuss and present the evidence in some detail, there's John Sheehy's demonstration under the title 'the joy of pixel density' , and finally there have been extensive discussions of the physics behind it, which back up the position that in theory there is no causal link between pixel density and final image noise content at any given image size (with the caveat that there are noise effects such as random telegraph noise, which come into play at very small geometries). These discussions included a number of people who are research physicists (not me, I hasten to add), and included Eric Fossum.
Your urban legends are showing again.
Re: No really, it isn't: News Discussion Forum: Digital Photography Review
When one does this exercise, it becomes apparent that the main factor in image noise is sensor size. The result is largely independent of MP count for a fixed sensor size. Pixel density, which is sensor area divided by MP count, is poorly correlated to noise because both MP count and sensor size will vary from camera to camera, but only one of those factors is tied to noise level.
Read carefully and you may yet learn weedhopper......

[Blue]

The "Shit" is you are being so rude in making a totally useless and unhelpful post in responding to an FAQ which is generally interested and asked by many beginners (and I think they need to know too).

Because I knew you would be in here promoting your anti-pentax blog and talking abut cameras that you don't even own and have limited experience with. If you have used a k20d and Canon Mark III or similar for 4-6 months, I retract my statement.

Also, as pointed out there are a plethora of threads on here about the FF debate.

[falconeye]

he notes that you need better high ISO performance from FF sensors in order to have the same kind of DOF choice.

I've been thru all arguments (as most other members, probably). Which is why I wanted to contribute a single, condensed statement, ... only.

But since you ask this specifically... A certain increase in ISO performance is required for equivalent DoF with a larger sensor, correct, and exactly this required increase is delivered by the larger sensor. It mathematically cancels out.

You may try it out: Use f/11 or f/16 and you'll emulate a P&S, both in DoF and noise behavior. Have fun


Of course, in my statements in the post above, I assumed constant technology, read pixel size in µm and read-out noise per pixel.

---

Last edited by falconeye; 10-29-2008 at 09:52 AM.

[falconeye]

Read carefully and you may yet learn weedhopper

All this can be summarized as follows:

• In the first order, noise is a function of sensor surface only.
• Second order effect in favor of a high pixel count is: The read-out noise is minimized.
• Second order effect in favor of a low pixel count is: The area sensitive to light is maximized.
• So, the combined second order effects favor a specific pixel density which is technology dependent (and may currently be at about 3-4 µm).

[Class A]

A certain increase in ISO performance is required for equivalent DoF with a larger sensor, correct, and exactly this required increase is delivered by the larger sensor. It mathematically cancels out.

I had to think about this for a moment, but of course, if the same shutter speed and DOF (aperture) is used, the same amount of light will hit both big and small sensor. While the ratio light/area is worse for the big sensor, it also has more area to make up for it.

So Joseph James is wrong when he writes "... if we were to try to achieve an equivalent image from, say, a Fuji f31 at f/2.8, 1/50, ISO 1600 using a Canon 5D, we would have to use f/13, 1/50, ISO 33000." or is ISO sensitivity somehow specified in terms of a light/area ratio? Even though, the higher ISO number for a FF sensor would be kind of misleading then and comparing ISO numbers directly would be as invalid as comparing f-stops (f-ratios) directly.

Thanks a lot for pointing out the forces regarding secondary pixel size effects. Do you have a pointer to something that explains why read-out noise is lower with higher pixel density? In principle, the lower the signal, the higher any electronic noise floor should become in proportion.

[Gooshin]

the FF advantage is very simple

for any given focal length, the larger the sensor (or film) the more you see (FOV)

55mm on FF will allow you to see more than on a Cropped sensor without having to take a step back.

tele lenses (from what i undrstand) are fundementaly easier to make than wide angle (especially extreme wide angle).

there is alot of glass that was CONVINIENT to use in the past, take canon for instance. If you had teh 16-35 2.8 L and the 28(?)-105 2.8L, you're done, there is your collection!

you might want to pick up either a 35, 50 or 85 1.4 for "artistic" purposes but really, there you have it.

with APS-C new lenses had to be developed (like the DA* line) only its bullshit, because they are sticking to the old mount and making unsesseraly large lenses (which is why i'm keeping a close eye on the micro 3/4's system)

so you have these inbetween lenses... for no other reason than to supply more FOV, rather than just 2k'ing the old formula.

and the only way to get down to 10mm is from specialized lenses! at 500-600 dollars a pop

so the advantage is two fold, you get to see more from what you have, and older glass gets you more utility



that, and larger viewfinders.

thats about the only advantage of FF over cropped there is WITHOUT going into the specifics of silicone sensor development and production.

[k100d]

add more control of depth of field, about a 1 stop advantage
e.g. the depth of field of a 35mm lens at 1.4 on APS-C > 50mm at 1.4 on FF

[Big Dave]

the fact that "full-frame" lenses work on APS-C cameras is merely a happy coincidence.

Actually the mounts and registration distance were retained with the APS-C cameras intentionally for lens compatibility. The RD could easily be reduced, because of the smaller mirror.

Dave

[Gooshin]

Actually the mounts and registration distance were retained with the APS-C cameras intentionally for lens compatibility. The RD could easily be reduced, because of the smaller mirror.

Dave

or no mirror at all

[Blue]

or no mirror at all

But then would it be a dslr?

[Venturi]

Actually the mounts and registration distance were retained with the APS-C cameras intentionally for lens compatibility. The RD could easily be reduced, because of the smaller mirror.

Dave

I fully realize that it was intentional and planned from an engineering (AND marketing) standpoint by Pentax.
But, for you and I (consumers) and thousands of eBay sellers - it's luck.