FF vs APS-C

I think it deserves a separate thread. Just got an A850 and was thinking how it stands against A700 both equipped with medium Q walk-around lenses. This is not scientifical test, just quick snapshots at certain focal length and at certain distance to object.

Used:

A850 + Minolta 50/f1.4
A850 + Minolta 24-105/f3.5-4.5
A700 + Sigma 17-70/f2.8-4.5

All set to FOV equal to FF50/f1.4, so 24-105 was set to 55mm, 17-70 to 35mm.
Shot in JPG, A850 set to 14mp mode, A750 - to full 12mp mode.

Conclusion is that FF camera even with medium Q lens still has an edge over APC camera with good lens.

all sources here:
http://www.betini.com/FFvsAPC/

FOV:











--
Hertz

Thank's for sharing. It will be also interesting to have this test with the "crop APS_C mode" on the A850, for to compare with a same image ratio (according to the size of photocell).

BTW, it's too early to give some "conclusion" according to the fact of that kind of test need a tripod for give an equal chance to every lenses.

Regards,

--
Michel J

Michel J said:

Thank's for sharing. It will be also interesting to have this test with the "crop APS_C mode" on the A850, for to compare with a same image ratio (according to the size of photocell).

BTW, it's too early to give some "conclusion" according to the fact of that kind of test need a tripod for give an equal chance to every lenses.

Click to expand...

Ceased to be a valid test anyway when different lenses were used on different cameras. That's more like a "make up whatever result you want" test.

Walt

You tried, but the results cannot be interpreted.

Even if you had used the same camera, you would get different results.

You see, even if the output of the A850 and the A700 was exactly the same, you get get results that wouldn't show this.

You must use the same lens on both bodies and compare those results.

No surprise that a FF sensor image shows more detail if you pixel peep, but the more important issue is how big of a monitor or print do you need to see the difference between the two when viewing the entire image? As a practical matter, I doubt many could see a difference when viewing the photo on a typical monitor or reasonably sized print. The fact is that high resolution full frame cameras will remain a small nice market for those few only interested in studio and landscape work.

If Sony reallly wants to make their FF cameras attractive to APS-C owners they need to replace the 24MP sensor with a 16MP sensor and keep the 5 FPS and sell it for less than $2000 as well. Now that would be a DSLR to get excited about!
-Phil

You see... This test even if it has its own flaws obviously, still reflects real life experience. I have ultimate quality prime lenses - 28/2, 50/1.4, 100/2.8 and could gain in IQ switching to FF just because of that. The same as I used to have it back in film era. Also I became somewhat spoiled with quality of A700 + Sigma Zoom lately. And was really concerned to loose in IQ going with general purpose zoom like Minolta 24-105 instead for not-so-critical walk-around purposes. Not at all.

I don’t see practical advantage of APC-S lens on FF camera against FF lens to justify this test.

For this particular test Elinchrom flash was used and cameras were mounted on a tripod at fixed distance to a subject. Every time the focus point was chosen to be most optimal to a particular shoot. In this regards this eliminates a variance in human error factor pretty much.

disclaimer
sorry guys for the way of words I put together

--
Hertz

Based on prints I made, I would say the difference will be noticeable starting of 18”x24" and larger. At least in terms of resolution. I’m not talking about tonal range and color response and DOF.

--
Hertz

I agree with the other people who say the test is invalidated by the use of different lenses. All this comparison shows me is that the 50mm lens is sharper than the others.

Until now about 50 experienced photographers (advanced amateurs and professionals) have judged the prints closely and carefully. How many was correct? About 50 percent of the asked!

http://forums.dpreview.com/forums/read.asp?forum=1003&message=31269570

There still seems to be a lot of people who are skeptical about the benefits of going to FF but the easiest way of convincing yourself is to try either an A850 or A900.

If you do all this talk of needing a large high grade monitor will fly out of the window - you will need to have a serious eyesight problem if you can't tell the difference. The same frame captured on a FF will have 24MP as opposed to 12MP & believe me that makes for a clear difference - the cropping ability is amazing & seems to go on for ever.
--
Keith-C

WaltKnapp said:

Michel J said:

Thank's for sharing. It will be also interesting to have this test with the "crop APS_C mode" on the A850, for to compare with a same image ratio (according to the size of photocell).

BTW, it's too early to give some "conclusion" according to the fact of that kind of test need a tripod for give an equal chance to every lenses.

Click to expand...

Ceased to be a valid test anyway when different lenses were used on different cameras. That's more like a "make up whatever result you want" test.

Walt

Click to expand...

Sure... then with the APS-C crop, it's possible to use the same lens
--
Michel J

I have the 24-105 myself and have wondered how well this would work on FF compared to the A700 + CZ16-80 combination that I currently use. The CZ is obviously a different lens to the 17-70, but I have seen enough samples from both to have a feel for how those two lenses compare.

In your shots I thought the 24-105 held up very well. Definitely better than I might have expected. Compared on the same camera, the CZ can clearly be shown to be sharper, but your test shots give an indication that it is sufficiently sharp for FF use.

In looking at the A850/A900, I am not really looking for an increase in fine detail - I have enlarged and sold 30"x20" landscapes from my 6MP KM 5D. I see the advantage of the A850/A900 as providing more data points in rendering tone gradients in difficult light.

I have seen some very good looking landscape shots taken with the A850/A900. One example was a middle of the day desert landscape with the sun fully in the frame of the shot and blue sky that rendered down smoothly to white in the immediate vicinity of the sun and still retained excellent shadow detail.

If you have the time and would be willing to share, I would love to see a FF shot from the 24-105 at 24mm and f8.
--
Richard B.
http://www.pbase.com/richard_b

Hi,

Pixel peeping vs. printing; always a subject to stoke the fires of opinion. I read your article and, by and large, agree with your conclusions but that does not invalidate hertz's observations, despite the obvious flaws it his test procedure. In short, on a screen the advantages of a FF sensor with a good lens compared to a Crop Sensor with a good lens are fairly obvious and on a print up to 16 x 12", not so obvious, in pure resolution terms.

In fact the whole FF vs. Crop Frame debate seems a little foolish as the advantages and disadvantages of both systems are well known and have been done to death. I use both. I use a Sony A900 (recently purchased) for work and a crop frame DSLR for my hobby. I use the Sony A900 not for the massive resolution but because it has amazing dynamic range and tonal separation, better than any other camera, up to 35mm format, that I have ever used. In fact I truly believe that the A900 would have benefited by using a slight lower resolution sensor, say 18mp. The viewfinder is a complete joy and I always feel a sense of disappointment when returning to a crop frame viewfinder. Not everything is rosy however; the A900's sensor is noisy and if you want to produce jpegs from the camera then you will find them disappointing. You will also need top notch glass to get the best out of the sensor. It is also slightly easier to throw the background out of focus with a FF vs. a Crop Frame camera at a given aperture.

I also use a Pentax K7 and I love its portability and the colour I get from the limited series lenses but if I am brutally honest the results from the A900 are marginally superior and contain more detail but K7 is no slouch. So, I think the idea of, "better" is misplaced and it simply comes down to which set of pros and cons is the best choice for the user.
Best regards, Howard

Hi Richard,

I recently purchased an A900 but the dealer only had a Sony 24-105 lens in stock. I knew that the lens was discontinued and I didn't want it as I had my mind set on the Zeiss and "G" series lenses. However, I bought it to test the camera out and I thought the focal range would make a good walk around lens if nothing else. Knowing how a good lens on crop sensor can be spectacularly poor on a FF sensor I didn't have too high a hopes for this lens. Having used it I can say I am pleasantly suprised and it is not a bad lens at all but not in the same league as the expensive CZ alternatives. Overall the lens gives pleasing results and is quite sharp but expect vignetting at the wide end, some fringing on high/low contrast borders and I found that some very fine detail gets smeared at wide apertures but with the exception of the vignetting non of this shows on a print upto 16 x 12". Its great for a walk about lens but I would probably look for something a little more special for serious work.
Best regards, Howard.

hertz said:

You see... This test even if it has its own flaws obviously, still reflects real life experience. I have ultimate quality prime lenses - 28/2, 50/1.4, 100/2.8 and could gain in IQ switching to FF just because of that. The same as I used to have it back in film era. Also I became somewhat spoiled with quality of A700 + Sigma Zoom lately. And was really concerned to loose in IQ going with general purpose zoom like Minolta 24-105 instead for not-so-critical walk-around purposes. Not at all.

Click to expand...

I see what you're saying, about going from what you are shooting now (a700 + 17-70) to what you will be shooting with (FF + others).

This makes the test only valid for yourself. It is in no way about FF vs APS-C, it is only about hertz old setup vs hertz new setup.

hertz said:

I don’t see practical advantage of APC-S lens on FF camera against FF lens to justify this test.

Click to expand...

I think you've missed the point here. It's not about justifying anything. The title made it seem like it is aps-c vs ff. In order to do that, you have to take the lens out of the equation. Use the same lens on both.

Howard,

Thanks for providing some more information on the FF performance of the 24-105. I agree, I would definitely not expect it to be as sharp as the CZ24-70.

I can't imagine I would like the performance on the 24-105 wide open, but for landscapes I would use the 24-105 in the f8-f11 area.

I know from my past experience with the mid range f-stops that the APS-C center from the lens is sufficient for my enlargement needs (up to 30"x20"). For me, it's just a case of how much the performance drops off at f8 once you go outside the center area.

--
Richard B.
http://www.pbase.com/richard_b

Phixel said:

If Sony reallly wants to make their FF cameras attractive to APS-C owners they need to replace the 24MP sensor with a 16MP sensor and keep the 5 FPS and sell it for less than $2000 as well. Now that would be a DSLR to get excited about!
-Phil

Click to expand...

To make full frame attractive to APS-C owners, it would be necessary to lower the price to around $1,500 or even lower if possible. If it takes a 16mp sensor to do that, then that is fine. In fact, a 16mp FF promises lower noise than the 24mp sensor. For now, most full frame DSLR cameras are simply beyond the reach of most APS-C owners. Even the the A850 is too expensive for most people.

--
Michel J

--
Richard B.
http://www.pbase.com/richard_b

Anastigmat said:

Phixel said:

If Sony reallly wants to make their FF cameras attractive to APS-C owners they need to replace the 24MP sensor with a 16MP sensor and keep the 5 FPS and sell it for less than $2000 as well. Now that would be a DSLR to get excited about!
-Phil

Click to expand...

To make full frame attractive to APS-C owners, it would be necessary to lower the price to around $1,500 or even lower if possible. If it takes a 16mp sensor to do that, then that is fine. In fact, a 16mp FF promises lower noise than the 24mp sensor. For now, most full frame DSLR cameras are simply beyond the reach of most APS-C owners. Even the the A850 is too expensive for most people.

Click to expand...

Yes, as I think about it you are correct that a FF DSLR must be priced at around $1500 or less in order to be a success among APS-C owners, and especially so if it is a Sony.

Years ago when I worked retail we had a term for a price point that was just a little too high for potential customers to make the produc a sales success. It was called "The threshold of pain" price. Most customers have a hard time getting past it.

-Phil