[cwood]

FF is not overrated. It serves a purpose and does so very well. Last weekend I was shooting mountain biking next to a guy using a 1D Mark III. To achieve good results in the evening light we were both shooting ISO 800. Although that particular canon is not full frame (it is APS-H) the sensor is larger and the resolution is significantly better at higher ISO's.

[ghelary]

IMO signal-to-noise ratio is the principal argument for FF, but that said, there is still plenty of room for improvement on the APSC format. K10D sensor was much more noisy than K20D's, and we can expect better performance for the next APSC sensor.

I've read some article saying that Nikon's D300 have to have no shame against D700 regarding noise. (having higher ISO is just a conséquence of getting lower noise)

APSC have a future, FF sensor will remain expensive for the next years. My personnal mark is to upgrade on 1000€/$ bodies, I don't care that the new one would be FF or APSC, I just want it to be significantly better that my old one. I will probably buy a K20D next month, in particular if a newer body is announced (and then the K20D will be sold at bargain prices) and then I'll buy a new body every 2-3 years.

I'm an amateur, so the question I shall ask myself is how much money I am willing to put in this hobby and then try to have the best buying policy.

If I were a professionnal, the reasonning would completly different. The question would be : does it worth it ? Often the answer is : I need the best available.

In that context the D300/D700/D3 debate have a meaning, the D700 is currently the best reporter body available on Nikon lineup. It is more discete/lighter than D3 and have better performance than D300 for nearly same shape. But D300 is still a good body and can be found new for 1400€ (in France) half the price of a D700.

For someone with a limited budget, for example 3000€ every 4 years, he would have the choice of buying a D700 and wait 4years or buy a D300 and then a newer camera 2 years laters (D400 or D500) We can expect the D400 or D500 (or whatever it is called) to be at least as much performant as the D700. This is a very simplified reasonning of course, but digital photography is very similar to the rest of consumer electronics on that matter.

Regards,
Guillaume

[kenyee]

K10D sensor was much more noisy than K20D's, and we can expect better performance for the next APSC sensor.
I've read some article saying that Nikon's D300 have to have no shame against D700 regarding noise.

Not sure they can make the same big leap over the K20D's sensor. The main reason for the jump in performance from the K10D was going from CCD to CMOS...you get 1 stop.

As for the D300, it has a lot stronger NR than the D700. To put the photosite size in perspective, the D700's photosites are slightly larger than the 6 Mpix K100D's CCD sensor and that had 1 stop better performance than the K10D. That's why it has much better high-ISO performance...it's all physics.

With the K1D, assuming they use the Samsung FF, the photosites seem to be about the same size as the K20D's. You'll be able to downsample (resize images smaller) to get "better" ISO performance, but it'll be about the same if you do 100% crops. You can see this behavior in Sony's big 24mpix FF sensor...photosites are the same as their APS-C sensor and the ISO1600 images look pretty noisy

[Miserere]

I'm glad you guys liked it. As I said, I'm not bashing FF or saying it's useless, only that it's not necessary most of the time for most photographers. Like Axl/Peter said, let's stop moaning and go take some good pics.

As for viewfinder sizes in FF cameras, I had the chance to handle a Nikon D700 the other day, and also had a D80 at hand. I put each of them to my eyes at the same time so I could compare the viewfinders. The D700's VF wasn't significantly larger, and certainly wasn't any brighter. But if a FF DSLR could have the VF of my SuperProgram, then that would be a different story altogether...

[falconeye]

I'm glad you guys liked it.

Yes, I liked it. But such an educated article asks for comments. Not that I disagree with anything you said. I just feel that you miss a point which personally, I consider to be the most essential one.

You write:

We’ve lost the wide angle.
The focal lengths are all wrong.
We’ve lost control over the depth of field.
Cropped sensors have more noise.
Cropped sensors don’t have enough resolution.
Professional photographers use FF.
The viewfinder is so small.

Actually, you write how none of this is an argument against APS-C. And I agree.

However, assume for a second that FF and APS-C sensors would both have zero cost of manufacture (a situation we are approaching over time).

And assume a camera to be a K mount camera.

Then the missing point is that an FF camera won't be bigger, nor heavier, nor more expensive. It just would discard part of the image circle which, for longer than normal focal lengths, is always larger than even FF. What you give up with APS-C this way is freedom in digital post-processing. For no reason.

For this reason, the question must read "when", not "if". And then we turn into protection of investments. And now you got the only point you missed to address.


Great article otherwise.

[Miserere]

Thanks for your thoughtful comment, falconeye. And yes, I do welcome comments and discussion. All articles are opinions based on facts, and different people can reach different conclusions from the same set of facts.

Regarding the point you say I missed, the truth is it is not the scenario we're currently living, and it's unlikely that sensors will become so cheap that their price will be negligible compared to the rest of the camera, at least not in the short/mid-term. I focused on the complaints put forth now, and my answers are based on the current state of affairs.

Some day, FF sensors will become affordable, but even then APS-C will still be cheaper, so while a FF camera will be accessible to more people than it is now, APS-C cameras will be at an even lower price and still very attractive because of this.

But affordable FF cameras (say, $1000 after price has stabilised) are still far away. I think the best case scenario is we're 4 years away, but I would bet on 6.

So yes, if cost, size and everything else externally were the same, then maybe more people would choose FF, but this is not the case. Furthermore, APS-C lenses are smaller and lighter than FF ones, and the bodies have the potential to also be smaller and lighter. For some people these are desirable qualities, so even if price were the same, they would still choose APS-C.

Feel free to tell me I'm wrong! It's what we're here for, to chat.

[ghelary]

Not sure they can make the same big leap over the K20D's sensor. The main reason for the jump in performance from the K10D was going from CCD to CMOS...you get 1 stop.

Well going from 1600 to 6400 means 2 stops ;-)

As for the D300, it has a lot stronger NR than the D700. To put the photosite size in perspective, the D700's photosites are slightly larger than the 6 Mpix K100D's CCD sensor and that had 1 stop better performance than the K10D. That's why it has much better high-ISO performance...it's all physics.

Nope, it not all about physics, it's all about engineering. Physics gives limits that can't be outperformed, but when applying physics to technologies and in particular young technologies like CMOS / CCD sensors, there is still a sensible difference between performance that can be achieved as per physics and performance that can achieve engineers.

FF sensor have their own specific problems that are due to their size, unloading the photosites from their charge is for example more complicated, leaving a SNR degraded as compared on an APSC (and smaller sensors as on P&S)

So, about the D300 vs D700, I'm just reporting a blog that I read (and of course didn't remember the address ) But it wouldn't surprise me that the difference between the 2 is smaller that it should be.

Don't misinterpret, I'm not saying that FF is crap. Like Misere, I think it has a lot of quality (and yes it will be the future of SLRs) But the point is there is still plenty of room for progress for Full Frame and APSC sensors. For example, Sony just applied for a patent for a techno that would allow to put all the circuits behind the sensor, allowing the photosites to occupy 100% of the surface of the sensor. Signal processing techniques and low pass filters will surely make a lot of progress.

At the end, I'm nearly 100% sure that future APSC sensor will outperform of FF sensors.

[bigben91682]

"The only quality that never decreases is our capacity to complain about minutia."

I think I'm going to have this quote engraved on a nice piece of wood and hang it on the wall at work, so that when people complain I can just point at it.

Great article.

[Miserere]

"The only quality that never decreases is our capacity to complain about minutia."

I think I'm going to have this quote engraved on a nice piece of wood and hang it on the wall at work, so that when people complain I can just point at it.

Great article.

Thanks Ben, and make sure you pay the royalties every time you use that quote

[Gooshin]

i used to think FF was overrated

but as i mentioned in another thread, after using a Pentax Super Program, i changed my perception.

a well executed FF camera (there ARE bad film cameras) is more pleasing to use than a cropped camera

lens choices, particulary in the wide end, become much more usefull and accessable (example is the Canon 16-35 F2.8 L lens)

and you can squeeze out shallower depth of field.


the last two, really, are technicalities

but again, as i mentioned in another thread, its near impossible to explain to a person what 100% FF viewfinder LOOKS AND FEELS like, you have to go out and try it for yourself to really understand.

[kenyee]

Well going from 1600 to 6400 means 2 stops ;-)

AFAIK, the ISO6400 has issues (banding most notably) so it shouldn't be used all the time.
For the D80 vs. D300, and also the K10D vs. K20D, you get 2 stops in very bright light (so you can get faster shutter speeds), but in dim light, it's really only 1 stop. For the bigger D700/D3 photosites, you get an extra stop. Just IMHO.

And yes, you're right about engineering helping performance, but the base reason you get good ISO performance is photosite size (because physics dictates how many photons you can catch in an area/well). The new Sony technique also increases photosite size

[deejjjaaaa]

and it's unlikely that sensors will become so cheap that their price will be negligible compared to the rest of the camera

btw, just curiuos - what is the current wholesale price of Sony 10mp sensor to buy in thousands (or whatever quantity they use for quotes) for a 3rd party ? anybody knows ?

[Miserere]

i used to think FF was overrated

but as i mentioned in another thread, after using a Pentax Super Program, i changed my perception.

a well executed FF camera (there ARE bad film cameras) is more pleasing to use than a cropped camera

lens choices, particulary in the wide end, become much more usefull and accessable (example is the Canon 16-35 F2.8 L lens)

and you can squeeze out shallower depth of field.


the last two, really, are technicalities

but again, as i mentioned in another thread, its near impossible to explain to a person what 100% FF viewfinder LOOKS AND FEELS like, you have to go out and try it for yourself to really understand.

Gooshin, I too am using a SuperProgram for film, and it pretty much does everything I want for the type of shooting I do with it (which is different than what I do with digital). I agree that it would be nice to have a digital FF back for it. I also agree that if you're someone that likes shooting really wide really often, then FF might be something worthwhile considering and changing systems for ...or maybe just shoot film. Luckily for me, I'm not one of those wide people (but I know there are some out there!).

As for the view finder, like I mentioned further up, I looked through a D700 and was unimpressed. It didn't remind me of the SuperProgram at all. When FF DSLRs have VFs like the SuperProgram or the ME Super, I'll delete the VF paragraph from my article

[deejjjaaaa]

As for the view finder, like I mentioned further up, I looked through a D700 and was unimpressed. It didn't remind me of the SuperProgram at all. When FF DSLRs have VFs like the SuperProgram or the ME Super, I'll delete the VF paragraph from my article

but it is not a technical limitation of FF dSLR cameras in principle - it is just a cost cutting measure, so it will be better not to generalize, but to say that so far all FF dSLR cameras do not have as big viewfinders as it is possible.

[falconeye]

btw, just curiuos - what is the current wholesale price of Sony 10mp sensor to buy in thousands (or whatever quantity they use for quotes) for a 3rd party ? anybody knows ?

This is difficult to say as most sources are closed. One source I found is this:

from Cypress, a 9MP APS-C sized CMOS sensor which they project will sell for around $90 and which should be available in quantity in early 2006

So, APS-C was 90$ two years ago.

Another source from last year is this:

Sony [...] US $493.3 million to expand its Kumamoto Technology Center (TEC) in Kyushu over the next three years. [...] Production capacity will be ramped by about 20% at the site, run by subsidiary Sony Semiconductor Kyushu Corp., to the equivalent of >20,000 wafers/month for both CCD and CMOS sensors

Assuming they earn 250 million per year (750 for 3 years) just on the 20% increase, a wafer would sell for about 5000 $. A 300mm wafer holds about 150 APS-C chips assuming a yield of 80%. I don't buy the bad yield argument because CMOS sensor chips, unlike CPU chips, are allowed to have faulty sensor cells. Those numbers therefore end up to 35$ per APS-C sensor.

Taking both sources and a certain price curce into account, I would say that an APS-C chip today has manufacturing costs of about 50$.

From this, I would say that an FF chip can be produced for 150$ (**).

Disclaimer: All this is deduced from public sources. I don't know what camera makers really pay. And in the case of Sony they may not even know themselves

(**) The (cost of FF) = (2/y)*(cost of APS-C) , where y is the yield for APS-C production.