Full frame vs. cropped sensors


Status
Not open for further replies.

Amfibius

Deregistered
Jan 26, 2004
508
0
0
52
Perth
Jed said:
On the contrary, I'm waiting for you to put everything in a nice list rather than my having to drag through about 10 threads and 30 posts to collate your thoughts. Let me preface everything by saying that I don't have a problem with full frame, I just don't think DX is any worse. So if you want to say that there is a problem with DX, then you need to state what the issues are before I can address them.

Before I start, pardon me if some of this seems obvious to you. I don't know how much you know, and I'm also writing so that less technically inclined CS members can still follow what i'm trying to say.

1. You already have a lens system which is optimized for 35mm. Every glass in your range (except (except for newly introduced glass, like the EF-S and DX) delivers an image circle for a 35mm frame. I would argue that EF-S and DX is a kludge, definition of a kludge being "an error introduced to fix another error". If Hasselblad wanted to produce a digital back for its MF camera, would it buy a 6.0MP APS sized sensor from Sony? The answer is "no" for many reasons, but one of the prime reasons is because too much of the imaging circle has been wasted. The argument remains true, in a smaller scale, for FF vs. APS sized sensors.

2. Expanding on my first point, the greatest expense in designing a lens is to ensure the lens performs consistently from corner to corner at wide apertures. This is one of the touted advantages of DX/EF-S/Four-thirds lenses. Paying a small fortune for a 600 F/4 and then throwing away the most expensive part of the image sounds wasteful to me. This would not be so bad if the manufacturer rationalised the entire lens range so that savings could be realized across all the lenses - something that the 4/3 consortium is trying to achieve, something Nikon is halfheartedly doing, and something that Canon will not do.

3. To continue along this line, where are the savings which reduced crop lenses have promised us? First 3 examples, price obtained from B&H, last example, Canon Australia website:

Nikon 10.5mm DX fisheye, USD599.95 - Nikon 16mm fisheye (full frame) USD$564.95
Nikon 17-55/2.8 DX, USD$1349.95 - Nikon 28-70/2.8 (full frame) USD$1329.95
Canon EF-S 10-22/3.5-4.5 USD$599 - Canon 17-40/4L (full frame) USD$674.95 (not strictly comparable, since the 17-40/4L is a constant aperture lens)
Canon EF-S 17-85/4.5-5.6 IS AUD$1099.00 - Canon 28-135/3.5-5.6 IS AUD$999

... I am sure you will agree there is no difference. We are still paying the same price for reduced crop lenses as we were for full frame lenses.

4. The next part of my argument concerns sensor design. Among other things, sensor designers are concerned about resolution, dynamic range, and noise suppression. Pixel pitch (the size of each individual photo pixel sensor, in micrometers) is just one out of many factors which influence these. People like to use the rainwater and bucket analogy which is a good one - a larger bucket allows more precise measurement of rainfall. A small bucket spills water too quickly. If there is very little rainfall, the small bucket may not gather any rainwater at all.

You can design a more efficient water gathering system (microlenses), you can design a better ruler (both tricks employed in the new 20D and 1DMk2) ... but in the end bigger buckets are still better.

A full frame sensor ultimately allows you to fit more, and larger pixels. Eventually you will hit a limit on a smaller sensor before the conflicting demands of resolution and larger sensors force you to look at other means of increasing dynamic range and reducing noise. You just hit the limit later on a FF sensor. If two manufacturers have exactly the same noise reduction and image processing technology, the manufacturer that can produce a larger sensor will have higher resolution, increased dynamic range, and less noise.

5. Viewfinders. Its not strictly necessary to have a larger viewfinder - it's just nice to have.

Um, I have typed enough now. Can't be bothered typing more. If anyone else has any more points to add, feel free to contribute.
 

Eh... don't flame me. :)


1. You already have a lens system which is optimized for 35mm. Every glass in your range (except (except for newly introduced glass, like the EF-S and DX) delivers an image circle for a 35mm frame. I would argue that EF-S and DX is a kludge, definition of a kludge being "an error introduced to fix another error". If Hasselblad wanted to produce a digital back for its MF camera, would it buy a 6.0MP APS sized sensor from Sony? The answer is "no" for many reasons, but one of the prime reasons is because too much of the imaging circle has been wasted. The argument remains true, in a smaller scale, for FF vs. APS sized sensors.

Hmmm so what has this got to do with FF and FLM sensors? It's a well known fact that FLM sensors are a smaller sized frames as compared to FF, so what's the vs here? Egg vs Rock? Why is a FLM sensor an error? Looking back at the only FF canon has and costs $14K, I think the bus advertisement probably won't come true on it's claim of 70% of DSLR users choose Canon. I would say it's the smartest thing yet.





2. Expanding on my first point, the greatest expense in designing a lens is to ensure the lens performs consistently from corner to corner at wide apertures. This is one of the touted advantages of DX/EF-S/Four-thirds lenses. Paying a small fortune for a 600 F/4 and then throwing away the most expensive part of the image sounds wasteful to me. This would not be so bad if the manufacturer rationalised the entire lens range so that savings could be realized across all the lenses - something that the 4/3 consortium is trying to achieve, something Nikon is halfheartedly doing, and something that Canon will not do.

Urm.. I would say Canon is doing the same now with the EF-S, and to make it worse, it would only be usable on newer bodies as compared to the DX series which is here to stay (irregardless what you say, it is HERE to STAY) can be used across multiple bodies, be it old or new (as of D2X). I seriously don't think major players will adapt to the 4/3s especially when they specialise so indepth to their manufacturing of glasses. And again, what has this got to do with FF vs APS?




3. To continue along this line, where are the savings which reduced crop lenses have promised us? First 3 examples, price obtained from B&H, last example, Canon Australia website:

Nikon 10.5mm DX fisheye, USD599.95 - Nikon 16mm fisheye (full frame) USD$564.95
Nikon 17-55/2.8 DX, USD$1349.95 - Nikon 28-70/2.8 (full frame) USD$1329.95
Canon EF-S 10-22/3.5-4.5 USD$599 - Canon 17-40/4L (full frame) USD$674.95 (not strictly comparable, since the 17-40/4L is a constant aperture lens)
Canon EF-S 17-85/4.5-5.6 IS AUD$1099.00 - Canon 28-135/3.5-5.6 IS AUD$999

... I am sure you will agree there is no difference. We are still paying the same price for reduced crop lenses as we were for full frame lenses.


12-24DX f/4 (18-36FLM) vs 17-35 f/2.8, $1600 vs $2950
17-55DX f/2.8 (25.2-82.5FLM) vs 28-70 f/2.8, $2200 vs $2950
18-70DX f/3.5-4.5 vs 28-105 f/3.5-4.5, $450 vs $700

Urm... who says no savings? Heehee.




4. The next part of my argument concerns sensor design. Among other things, sensor designers are concerned about resolution, dynamic range, and noise suppression. Pixel pitch (the size of each individual photo pixel sensor, in micrometers) is just one out of many factors which influence these. People like to use the rainwater and bucket analogy which is a good one - a larger bucket allows more precise measurement of rainfall. A small bucket spills water too quickly. If there is very little rainfall, the small bucket may not gather any rainwater at all.

You can design a more efficient water gathering system (microlenses), you can design a better ruler (both tricks employed in the new 20D and 1DMk2) ... but in the end bigger buckets are still better.

A full frame sensor ultimately allows you to fit more, and larger pixels. Eventually you will hit a limit on a smaller sensor before the conflicting demands of resolution and larger sensors force you to look at other means of increasing dynamic range and reducing noise. You just hit the limit later on a FF sensor. If two manufacturers have exactly the same noise reduction and image processing technology, the manufacturer that can produce a larger sensor will have higher resolution, increased dynamic range, and less noise.


Mmm I would want to know, perhaps you can enlighten me, how does a larger sensor with higher resolution can have increased dynamic range?

5. Viewfinders. Its not strictly necessary to have a larger viewfinder - it's just nice to have.

Actually, I would prefer a larger VF, but since I've never had one, I'm quite satisfied with what I have.

All in all, just replying for fun, make this a healthy one without flames and arguments.
 

You already have a lens system which is optimized for 35mm. Every glass in your range (except for newly introduced glass, like the EF-S and DX) delivers an image circle for a 35mm frame. I would argue that EF-S and DX is a kludge, definition of a kludge being "an error introduced to fix another error". If Hasselblad wanted to produce a digital back for its MF camera, would it buy a 6.0MP APS sized sensor from Sony? The answer is "no" for many reasons, but one of the prime reasons is because too much of the imaging circle has been wasted. The argument remains true, in a smaller scale, for FF vs. APS sized sensors.

Agree that glass is for 35mm format, but it's glass I already bought, so it's not an added expense.

Actually, medium format backs haven't until very recently been approaching full frame. For a long time, the most popular Phillips chip for medium format backs was actually 35mm sized (ie 36 x 24mm). So there was a lot of wastage, and it was even harder to get true wide angles given that it is far harder to get ultra wides to fit medium format cameras compared to 35mm systems. There is a lot more wastage in using a 36x24mm chip in even a 645 system, compared to an APS in a 35mm. So your point is flawed and actually does much to undermine your argument!

Expanding on my first point, the greatest expense in designing a lens is to ensure the lens performs consistently from corner to corner at wide apertures. This is one of the touted advantages of DX/EF-S/Four-thirds lenses. Paying a small fortune for a 600 F/4 and then throwing away the most expensive part of the image sounds wasteful to me.

At the same time, knowing that you have a DX size sensor could save you a bundle of money. If you knew you'd need a 600/4 on full frame, but now had a D2x, you could buy a 300/4 on the crop mode (assuming 6.8 mp is sufficient), and save about S$12000.

Furthermore, it doesn't matter how well you design a lens, it's corners and edges *always* perform worse than its middle. By using a full frame lens on a full frame camera, or a DX lens on a DX camera, your corners and edges suffer from quality deterioration. If anything, the only way of achieving a lens that performs consistently "from corner to corner at wide apertures" is to actually use an oversized lens. For instance a full frame lens on a DX sensor... take the difference between the Nikon and Sigma 12-24 lenses for instance.

So in response to your points, you might be throwing away the corners, but the corners are poorer relatively anyway. And also, if you've already bought the lens, great, it's no real loss to you since you bought it for what it was. But if you're contemplating a lens purchase, having a smaller sensor can save you plenty of money since you could realise you don't actually need to plough all that money into a 600/4.

Also in exchange, you get better resolution, less (if not any at all) light fall off at wide apertures, and less distortion and aberration.

To continue along this line, where are the savings which reduced crop lenses have promised us?

Actually, there is a lot of confusion in this area. At shorter focal lengths, you don't actually save much glass so there shouldn't necessarily be much cost savings. As you've demonstrated, the DX lenses are about the same prices as their non-DX counterparts. But then this isn't in any way a criticism of DX lenses... they are not any more expensive than their full frame counterparts either, so full frame isn't better either.

That's on a one to one comparison. But I've always held the savings come when you put together a basket of normal lenses, and price it up as a whole.

Say you want to get from 20mm through to 300mm with a constant f2.8 aperture.

With full frame, say you get 20/2.8, 24-70/2.8, 70-200/2.8 and 300/2.8. Total cost about S$13500 at a very rough approximation.

With DX, say you get a 14/2.8, 17-55/2.8, 70-200/2.8.
Total cost about S$7000 at a very rough approximation.

... I am sure you will agree there is no difference. We are still paying the same price for reduced crop lenses as we were for full frame lenses.

Nope, agree, the difference will come in longer lenses, if Nikon ever bother doing that. I cringed when the 200-400 and then the 200/2 were full frame, don't look at me. But as you yourself say, it's not any more expensive either, so this is not 1-0 to full frame, it's more a halve, to borrow Ryder Cup speak.

Among other things, sensor designers are concerned about resolution, dynamic range, and noise suppression. Pixel pitch (the size of each individual photo pixel sensor, in micrometers) is just one out of many factors which influence these.

But as you say, it is just one of *many* factors that influence these. Having a smaller sensor means a lens has to cover a smaller image circle, which in turn means that manufacturers can make higher-resolving lenses as they have to spend less effort correcting for aberration, which in turn eliminates any issues with resolution problems. With regards to noise suppression and dynamic range, with an equivalent lens, I agree completely. However in theory DX lenses *could* be designed a stop faster than their equivalent full frame lenses, for the same size, weight and price (excepting initial R&D considerations). That it isn't being done is a source of much grief to me.

But clearly we are not yet at the technology threshold yet either at the moment, which will also no doubt continue to improve. Any claims that Nikon cameras suffer from poor noise levels compared to their Canon counterparts due to the DX sensor is missing the point. Clearly chip technology is at the heart of any problems. The D1x for instance, has a smaller photosite than the 1Ds, yet has clearly better noise characteristics.

Canon have shown with the 20D that they can get 8 million pixels into a DX sized sensor to even greater effect than the 8 million in a 1.3x 1DII sensor.

Another factor is that microlenses really prefer light to be hitting them as close to perpendicular as possible. This has always been a problem with full frame chips, as obviously at the edges and corners light strikes at an angle. This results in light fall off and chromatic aberration around the edges and corners. And was one reason why Kodak decided to leave microlenses out of their Pro 14x series (the others we won't go into!). So a full frame chip, ceteris paribus, has more optical problems around the corners and edges as does a DX sensor.

A full frame sensor ultimately allows you to fit more, and larger pixels. Eventually you will hit a limit on a smaller sensor before the conflicting demands of resolution and larger sensors force you to look at other means of increasing dynamic range and reducing noise. You just hit the limit later on a FF sensor.

Agree that obviously you will be able to fit more larger pixels on. But as mentioned, in theory you should have a 300/2 on a DX at ISO 100 as opposed to a 300/2.8 at ISO 200 for instance, which is exactly the situation back at square one.

And clearly you can get 8 million pixels into a DX sensor (20D) with stunningly good results, as you yourself regularly attest. So clearly there aren't major issues yet. And we are really at the stage where the vast, vast majority of photographers do not need an ounce more resolution. In fact we were there with 6mp, never mind 8mp.

I am no in any way saying we don't need more resolution, but I'm talking in practical terms. There used to be the days when everyone sung the praises of Velvia. For that, you had to tolerate ISO 50 and no exposure latitude, and very contrasty (ie poor dynamic range) images. Colour aside, the previous generation of 6mp cameras gave you all that with a far greater ISO range.

I have images from a D1 done to 6m x 4m and they look great; images from a D2h done to 3' x 7' and they look terrific too. Critically examinging D1x files to 24" x 36" they firstly look sharper than tranny stock, and secondly look absolutely fine. Sure a 11, 12 or 16 mp file will look better, but if you don't put them side by side, you wouldn't in any way complain about the 6mp file (for most applications).

Viewfinders. Its not strictly necessary to have a larger viewfinder - it's just nice to have.

Agree completely. Full frame cameras have got bigger viewfinders.

Other factors you've forgotten:

You've gone on about DX lenses not saving money. I suggest they do. And you've completely forgotten than DX *cameras* are significantly cheaper than full frame cameras. At the anticipated street prices, you could nearly afford two D2x bodies for the price of a single 1DsII.

While it hasn't happened yet, primarily because Nikon are sadly still committed to full frame lenses on their digital cameras, a smaller sensor means the camera could be made smaller as well. A smaller, lighter system (including lighter lenses - look at the E series), is a welcome boon to *all*. I know a die hard Canon fan who keeps whinging about the size of cameras. Better on your back, and more likely that you will have your camera with you rather than sitting in your dry cabinet at home.

And I'm also too lazy to think, but I think there's just a lot of inertia in resisting the legacy that is full frame. It's like paper sizes. There is no real reason for the industry to stick to 4x6, 5x7, 6x8, 8x10, 11x14, etc, when there is no convention to the sizes and no relation, and the most common film format was in 3x2 dimensions, only really fitting the 4x6 paper size. The rest are leftover from when 4x5 was the neg size, or other formats. The A4 system is actually a much cleaner naming system, with a clear relationship between A0 through to A-whatever, and with much less paper wastage than an 8x10 sheet using conventional 3x2 dimensions.
 

Jed, I haven't forgotten. It's just that at the moment i'm too wined up and too full from dinner to bother typing any more. I got to a point, and then thought "AH FLUG IT!" and just hit the post button. I'm going to sleep soon. Will respond to yours and espn's post tomorrow.

Oh, and espn: :blah: :blah: :blah:

There, consider yourself flamed ;)
 

great stuff..

I'm personally pro 'DX sensors' cause well.. they're cheaper, I don't see the need for higher resolving power (yet), and its generally smaller (D70! :) )

I'm not too good in lengthy discussions, but I'm definitely keeping this thread for reads!

:thumbsup: :thumbsup: :thumbsup:
 

OK just a quick reply before I go to work :)

espn said:
If Hasselblad wanted to produce a digital back for its MF camera, would it buy a 6.0MP APS sized sensor from Sony? The answer is "no" for many reasons, but one of the prime reasons is because too much of the imaging circle has been wasted.
Hmmm so what has this got to do with FF and FLM sensors? It's a well known fact that FLM sensors are a smaller sized frames as compared to FF, so what's the vs here? Egg vs Rock? Why is a FLM sensor an error? Looking back at the only FF canon has and costs $14K, I think the bus advertisement probably won't come true on it's claim of 70% of DSLR users choose Canon. I would say it's the smartest thing yet.

The answer to your question is in the bit you quoted, in red. It is an error because most of the imaging circle has been wasted, and because the entire lens system has been optimized for a larger imaging circle.

[/quote]Urm.. I would say Canon is doing the same now with the EF-S, and to make it worse, it would only be usable on newer bodies as compared to the DX series which is here to stay (irregardless what you say, it is HERE to STAY) can be used across multiple bodies, be it old or new (as of D2X). I seriously don't think major players will adapt to the 4/3s especially when they specialise so indepth to their manufacturing of glasses. And again, what has this got to do with FF vs APS?[/quote]

I think that again, the answer was contained in the bit that you quoted. You are throwing away the most expensive part of the image if you use an APS sized sensor. And I am not suggesting that major players adapt to 4/3, I just quoted the 4/3 consortium as an example of manufacturers who are designing a whole new lens system, with a smaller imaging circle, to take advantage of the smaller chip (2x crop).

12-24DX f/4 (18-36FLM) vs 17-35 f/2.8, $1600 vs $2950
17-55DX f/2.8 (25.2-82.5FLM) vs 28-70 f/2.8, $2200 vs $2950
18-70DX f/3.5-4.5 vs 28-105 f/3.5-4.5, $450 vs $700
Urm... who says no savings? Heehee.

Well I don't know where you got your prices from, mine are from B&H. You could easily check.

And BTW, in your first example, did it escape your attention that the 12-24DX is F/4, and the 17-35 is F/2.8? Maybe that accounts for the difference in price?

Mmm I would want to know, perhaps you can enlighten me, how does a larger sensor with higher resolution can have increased dynamic range?

Read this response for a start. The answer was also partially contained in my post - larger sensors collect more light, so they can measure light more precisely. I also pointed out that the electronics play a part too - you can amplify signals from a dark region of an image to compensate, but then you run the risk of introducing noise.

Digital sensors basically are electronic buckets. When you dump photons on them, electrons get kicked out. This generates a charge on each pixel which is measured. If you dump a LOT of light on the sensor, a lot of electrons get kicked out. This will saturate neighbouring pixel sites, causing digital blooming - the unnatural looking blown highlights that are the hallmark of digital cameras. OTOH if these pixels don't get enough photons, they will hardly get excited and you will have to amplify the signal. Sensor designers try very hard to get rid of the excess electrons - they design "drains" in their sensors to remove them - but these drains have the effect of taking up die space and reducing pixel pitch further. Or, they could just have a bigger sensor in the first place.
 

Amfibius said:
Read this response for a start. The answer was also partially contained in my post - larger sensors collect more light, so they can measure light more precisely. I also pointed out that the electronics play a part too - you can amplify signals from a dark region of an image to compensate, but then you run the risk of introducing noise.
Frankly I simply don't any logic in that so called answer in the first place. For one, what is your definition of dynamic range?
 

Breaking News

The Last of 35mm Camera Stolen from British Museum

(Baker Street News Service):The last surviving 35mm camera body was stolen last night from the British Museum. The 35mm cameras were pioneered by the German giants Leica and Zeiss in the early part of the twentieth century. However, professionals preferred the FULL FRAME field cameras to the handy 35mm bodies.

Though handy, the 35mm cameras posed severe dynamic range problem. According to professionals, the FULL FRAME field cameras allow for enourmously huge pigments in the film, which can collect more light just like large buckets can collect more water. This in turn improves the dynamic range. The high dynamic range makes the FULL FRAME field cameras especially suitable for Amphibius (both underwater as well as surface) use.

Due to the severe problems posed by the 35mm technology, the professionals refused to budge from the FULL FRAME field cameras. Hence the 35mm technology died a natural death just a few years after its invention and 35mm camera bodies disappeared from the market. The 35mm camera exhibited at the British Museum was the last of surviving bodies.

Police are suspecting some link between the crime and the infamous Professor Moriarty. Scotland Yard detective Lestrade is investigating the case and a young Baker Street detective Mr. Holmes is reported to be assisting him.
 

Let me first say that this is a good opportunity for both views to be aired, especially when we remain fair to every brand here.

I'll cherry-pick the responses, I hope Jed and ESPN don't mind :p

amfibius said:
The answer to your question is in the bit you quoted, in red. It is an error because most of the imaging circle has been wasted, and because the entire lens system has been optimized for a larger imaging circle.
Wasting the worse part? Did you read Jed's comment on this? The lens maker try to make it better, but if you take a look at a FF digital shot using the Sigma 12-24 vs Nikon 12-24 (on a DX sensor), you will see significant quality difference. The corners are normally the worst part of the quality. I mean if you throw away the chicken feet, is it a waste? Or how about the chicken backside? (no intention of insulting those who love to eat chicken feet or backside :D ) Don't forget that most Western countries (including the one that you are current in Amfibius) throw away these parts.

amfibius said:
Well I don't know where you got your prices from, mine are from B&H. You could easily check.
And BTW, in your first example, did it escape your attention that the 12-24DX is F/4, and the 17-35 is F/2.8? Maybe that accounts for the difference in price?
Those are street prices in SG. You can easily do an equote to the various shops like MS Color and CP for them. As for the f/4 thing on 12-24, yes, but like someone gave the reason why they don't mind the f/4.5-f/5.6 on the Sigma 12-24: when you shoot landscape, you normally don't use it wide open at all. So for 1 f-stop, it is rather a minor quibble given that a press and a twist will make up the difference.

As for the dynamic part, like what was said by Zerstorer on this thread and the other recent thread.
 

Amfibius said:
The answer was also partially contained in my post - larger sensors collect more light, so they can measure light more precisely. I also pointed out that the electronics play a part too - you can amplify signals from a dark region of an image to compensate, but then you run the risk of introducing noise.

i would say that it has very much to do with how u calibrate the sys. a big bucket has 10cm of water & was calibrated to read rainfall with a rating of 1.8. another smaller bucket uses a different way of rating, if it has 1cm of water, that relates to rainfall with the same rating of 1.8. so both r accurate. the concern remaining is noise. how useable is the signal when noise is present? it seems like canon has taken care of that as well, so no problem with light reading accuracy. btw, 1Ds II & 1D II has 7.2 micron pitch & 8.2 micron pitch respectively. does that mean the 1Ds II FF DSLR is less accurate in its meter? ;p

it is tricky to make pixels that can take both very bright light & very low light. need to balance a lot of things, that explains the S3 pro approach. it doesn't mean having a FF sensor means it can handle bright light w/o bloom. u need technological break thru for that.

i think the 2 biggest problems regarding FF is the telecentric issue & yield.

personally i dun mind a FF DSLR if i can afford one. :cheers:
 

Zerstorer said:
Frankly I simply don't any logic in that so called answer in the first place. For one, what is your definition of dynamic range?


isn't it define by the range between the weakest useable signal before noise mask & the highest useable signal before clipping? :dunno:

relates to: can the camera sensor handle hi-lites w/o lossing shadow details (w/o blotchy noise)?
 

nightpiper said:
isn't it define by the range between the weakest useable signal before noise mask & the highest useable signal before clipping? :dunno:

relates to: can the camera sensor handle hi-lites w/o lossing shadow details (w/o blotchy noise)?
That query was meant for amfibius, unless you would like to answer his claim on why larger sensors WILL have a greater dynamic range.;)
 

nightpiper said:
i think the 2 biggest problems regarding FF is the telecentric issue & yield.

Telecentricity is a lens issue and affects both FF and DX sensors.
 

Amfibius,
I'm afraid you've been sidetracked. What started off as "Canon ahead of Nikon", which is true and easy to defend, has turned into "FF better than APS", which is less true and less easy to defend. Sometimes companies do high end stuff just to prove that they can, and are ahead of the game, i.e. Mercedes in F1, not necessarily make a profit from it. In making the 1DsMk2, Canon has thrown down the gauntlet to everybody else, "I can do it, how about you?". It must stick in Nikon's execs throats......
 

There is one related issue - resolving power of lenses. With the same lens and and sensors with the same resolution (e.g. 20MP), a cropped sensor might exceed the resloving power of the lens (as in the centre part of the lens is not able to resolve more details than could be captured by the sensor). But similarly, a full frame sensor might exceed the resolving power of the edges of the lens if the edges are significantly poorer than the centre.
 

Zerstorer said:
That query was meant for amfibius, unless you would like to answer his claim on why larger sensors WILL have a greater dynamic range.;)


oops, sorry. me & my big mouth. ;p
 

i dont' really care about 'Canon vs Nikon', but personally i do stand by 'FF is better than APS'.

No one's going to argue that a larger negative beats a smaller one, all things equal.

But first, the pixel size issue:

- a larger sensor does not mean larger pixels; it depends on how they cram it. i believe the D30 and 1D have the largest pixels around, despite not being FF? THEREFORE, the pixel size arguement is IRRELEVANT to the FF/APS discussion. Especially since Canon and Nikon are both going to up the mpix count as the days go by, and the pixels are just going to shrink, FF or APS.

- As the pixels get better and better, it becomes meaningless to compare the pixel sizes of older models and new ones. They don't relate directly anymore, ie a 10um two years back may not beat a 6um today.

- As an aside, larger pixels do potentially have better signal range than smaller ones, all things equal. This can translate to cleaner pixels or better dynamic range. However, larger pixels also mean lower resolution on the same size sensor. Dynamic range and resolution are balanced for the best image quality for any given sensor size. In the past, the optimal pixel size was 7-10 um; it seems to have shrank to 5-7 um these days, with better sensor technology. But like i said, pixel size is irrelevant to the FF issue.

Back to larger film/sensor being better than smaller ones - a larger negative requires much less enlargement than a small one to get to the same size print. The 36x24 (aka Full Frame) is the sweet spot, where the entire 35mm system is designed around, and where lots of lenses and accesories are available. Things that don't function the same when cropped include fisheyes and angle finders.

If (say) Nikon makes a D2X at 12.5mpix with 5.5um pixels using a 1.5x crop - with a full frame, they can get a 12.5mpix with bigger pixels for better d-range or cleaner images, or a higher resolution camera with the same characteristics. Either way, FF beats APS.

FF has more DOF control than APS, for the same framing. An 85/1.8 has shallower DOF than a 50/1.8 on a 1.6 for a picture framed the same way.

FF makes less demand on glass than APS, just like large and medium format lenses have more 'spare resolution' than 35mm stuff. To get to a required resolution, say 40 lp/mm on fullframe, across the entire frame, one would need the glass to deliver 60 lp/mm on an APS sensor for the same picture. 1.5 (or 1.6) times more resolution. The MTF at higher resolutions drop significantly.


***********************

... to be continued....


***********************
 

I am amazed at the some of the knowledge that is shared in this thread and the detail you guys look at your gear.

I am wondering for those in the cropped sensor camp, if you are offered FF body for the same price, would you still opt for a cropped sensor?

On a similar track - why would people pay top dollar for a 6MP MF digital back instead for shooting on a D60/10D or D70/D100? Surely it's a whole lot cheaper to buy 35mm glass (which need not be good towards the corners somemore!).

And how is it possible that these 6MP digital backs used in medium and large format capable of producing images at A0/A1 in size which a 10D or D100 can't really match visually at that size.

It has also been pointed somewhere that the 1D sensor is 12 times that of the G2 while both are 4MP, we all know the difference.

My take is that FF is better than APS and the proof is in the print IMO.

What's going on in today's technology is a whole lot of compromises - using one flaw to match another. It's not necessary a bad thing as it has resulted in cost effective interim solutions which as expanded our photography options.

I look forward to the day when the cost of manufacturing these large hi-res sensors drops to affordable levels. It wasn't too long ago when 1-3MP DSLRs almost cost as much as the FF cams today. I guess it's a matter of time and by then, I think the EF-S and DX lenses can be used as loupes. Then again, maybe slides will also be non-existent by then :)
 

I'm waiting for the day when they use 4, 9 or 16 small sensors in an array, combining the images into one large image, rather than manufacturing one large sensor. Parallel processing supercomputers beat the pants off Crays nowadays.

Medium format digital SLR, anyone?

Heck, take a leaf from insect eyes and have a honeycomb matrix of sensors.
 

Most of the people who are against full frame are nikon users. Is there some correlation? Correct me if i'm wrong. I'm not sure who uses which system anyway.

But if it's true.. something is really wrong.
 

Status
Not open for further replies.