Rethinking 4/3 Depth Of Field

Started Apr 11, 2013 | Discussions
Anders W
Forum ProPosts: 17,051Gear list
Like?
Re: Rethinking 4/3 Depth Of Field
In reply to exdeejjjaaaa, Apr 11, 2013

exdeejjjaaaa wrote:

Anders W wrote:

exdeejjjaaaa wrote:

al_in_philly wrote:
Even as good as FF images are, the OM-D at 1/4 the ISO will always look better.

it is a typical wishful thinking of those who like to compare deep shadows w/ Canon FF's @ low iso's (where Canon does not deliver)... but move to SNR @ midtones, real high ISO and start counting Sony/Nikon and suddenly everything is going back to there is no replacement for displacement (= how much light you can grab).

At high ISOs, which is what the OP was talking about, the E-M5 set at two stops lower ISO than the FF camera (for equal DoF), will go equal for SNR at midtones but do better with regard to shadow noise/DR in comparison with all current FF cameras except the D4 (where it will be a tie).

it is soundly beaten by DR @ high ISO even by Canon (1Dx).

Please read what both the OP and I were actually talking about. With the E-M5 at ISO 3200 and the 1Dx at ISO 12800 for equal DoF, the E-M5 will come out slightly ahead.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,051Gear list
Like?
Re: I can already predict a response.
In reply to exdeejjjaaaa, Apr 11, 2013

exdeejjjaaaa wrote:

danijel973 wrote:

RoelHendrickx wrote:

I can already predict a response by some people likely to chime in here.

They will say that having shallow DOF at one's disposal is never a disadvantage, because if deeper DOF is desired, the aperture can always be stopped down a bit (while it cannot be "stopped up" to make the DOF more shallow than the lens wide-open on the sensor will allow).

Except when you stop down to have similar DoF you're no longer at f/2.8, you're at f/5.6, and then you push ISO up two stops to compensate for speed loss and suddenly you wonder where your "two stop advantage at high ISO" went.

except that in many cases you do not really need exactly the same DOF as you are forced in w/ m43/43 and then you do not need to push ISO 2 stops... so what you are saying is an artificial condition.

Sometimes you do need the extra DoF and sometimes you don't. The fact that you don't always need it doesn't make the OP's comparison irrelevant or artificial.

FF isn't always better. It is sometimes better, i.e., in those cases where DoF can be traded, without any disadvantage, for less noise. Currently, it is also better when you hit the lower ISO limit and therefore use a higher shutter speed with MFT than you really need. I hope/think that will change in the future and that we will soon see cameras with arbitrarily low base ISO appear.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,368
Like?
Re: I'm not convinced...
In reply to Anders W, Apr 11, 2013

Anders W wrote:

Great Bustard wrote:

Anders W wrote:

Great Bustard wrote:

al_in_philly wrote:

I do the majority of my shooting in very low light situations without a flash.  I started out shooting at night with a (then brand new) Olympus e-510 and a Pan/Leica 25mm f1.4.  I know, great lens choice, not best choice for a low-light camera, but that's what I happened to own when what became a three year nightime photo-shoot opened up for me.  Those photos BTW have been published in a variety of newspapers & magazines, as well as having two gallery shows of them.  Now I'm shooting different very low-light subjects with an OM-D and the m4/3 version of the PL 25 1.4, and I couldn't be happier.  Why?  A big part is because of the depth of field that combo gives me.

Like many others, I too like to seperate my subjects from their background a bit by using the selective focus which wide apertures afford.  But often, I don't want to obliterate all the background, or foreground, definition, just soften it enough to tell my viewers wher to plant their gaze in the picture.  So, if I'm shooting in a dimly lit bar or a city street at midnight, my OM-D produces "just right" images at f1.4 and an ISO of 3200, typically yielding a shutter speed of between 1/30 and 1/80 sec, depending on how dark the lighting is.

Of course, I could shoot FF at 2.8, but then I'd also be shooting at an ISO of 12800 for the same shutter speed.  Even as good as FF images are, the OM-D at 1/4 the ISO will always look better.

...about your last sentence above.  It would be interesting to see a comparison between a modern FF DSLR with a 50 / 1.4 at f/2.8 and the same shutter speed, and see which comes out ahead.  I am far from convinced that it would be the EM5 (which is not to say that I am convinced it will be the FF system).

On what grounds are you not convinced?

Image quality.

As I know you are well aware, any current FF camera would lose out to the E-M5 for DR/shadow noise in this comparison (except the D4 which would break even).

I think the 6D comes out the same.

OK. You are right about that. But not the 5D3, the D800, or the D600, nor any older FF camera.

In terms of SNR at higher light levels, it would be a tie. So are you thinking about how the PL 25/1.4 at 1.4 compares to a 50 mm FF lens at 2.8 or what?

I'm thinking that a FF camera at 50mm f/2.8 would resolve more detail everywhere in the frame than an EM5 at 25mm f/1.4, and, if the EM5 did have less noise for equivalent settings (same DOF and shutter speed), the judicious application of NR (noise reduction) to the FF photo would result in a photo that was both more detailed and less noisy (albeit not significantly so, in my opinion).

Could be although at these ISOs, I am not at all sure. Even if the FF lens at f/2.8 would clearly outdo the 25/1.4 for MTF, little of that difference is likely to register in view of the noise levels we are talking about here.

Well, that is something I'd like to see put to the test.

Further, if we compare at apertures slightly smaller than these, e.g., f/2.8 versus f/5.6, the difference in MTF becomes close to insignificant in the first place.

Let's put that to the test, too.  I'm honestly curious to see the results from a competent test.  As I said, no matter which way it falls, I'm of the opinion that the differences are nothing to fret over.

In any case, if you are always going for the deeper DOF in low light, seems to me that the size and bulk of the FF system compared to the EM5 system makes the EM5 system the overall winner, although there may be operational advatanges that may substantially favor one over the other.

To that end, I think it would be interesting to see a comparison with competently processed photos.

P.S.:  I think the EM5 would come out significantly ahead in situations where it could use IBIS and a longer shutter speed and FF could not use a tripod.

And, yes, don't forget the awesome IBIS in the EM5.

Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,368
Like?
Re: I'm not convinced...
In reply to tt321, Apr 11, 2013

tt321 wrote:

Great Bustard wrote:

P.S.:  I think the EM5 would come out significantly ahead in situations where it could use IBIS and a longer shutter speed and FF could not use a tripod.

You should be able to get a similar degree of IS with an IS-capable 24-70/2.8.

Sure, and that's one of the nice advantages of FF -- where mFT would require a prime (25 / 1.4, in this case), FF could use a zoom (e.g. Tamron 24-70 / 2.8 VC).  The question, then, is how FF would fare with the zoom wide open compared to mFT with the prime wide open.

Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,368
Like?
Re: Lovely pix and excellent presentation of your argument
In reply to Hen3ry, Apr 11, 2013

Hen3ry wrote:

And my argument if it comes to that.

Razor thin depth of field is a by-product, an artifact of big apertures. They allow you to shoot in lower light but more often than not, the very shallow DoF is a damned nuisance.

For me, shallow DOF is the bomb, and I'm wide open, more often than not, regardless of the light.  That said, I don't expect others to have my tastes, but it's not like I never stop down:

http://www.dpreview.com/forums/post/39537128

Besides, and editor might use one or two thin depth of field shots in a publication, but s/he certainly didn’t want page after page of them -- they were showing what the world looked like, not some arty-farty extract of the world governed by a super large aperture.

Well, a modern compact, like the RX100 might be better still for many, and DPR has been showcasing many pros making good use of cell phones for published works.

Then the argument was switched around so that razor thin DoF was practically the raison d'etre for taking a picture. Of course, you had to have the most expensive lens to achieve it so ordinary mortals were excluded.

There are quite a few fast primes for FF that aren't particularly expensive:  35 / 2, 50 / 1.8, 50 / 1.4, 85 / 1.8, 100 / 2.

This was a fad for a while, then the world got back to the business of expecting photographs to show them what the world looked like.

Like I said, you can always stop down, if it results in a better photo.

Super thin depth of field is a choice. Those who want it can have it; I'm not interested and mostly I'm not interested in their pictures either. Very thin DoF moved from the innovative to the boringly banal in a remarkably short time 40+ years ago.

Ah.  Well, don't click on the link above then.

Reply   Reply with quote   Complain
Yohan Pamudji
Senior MemberPosts: 2,859
Like?
Re: I'm not convinced...
In reply to Great Bustard, Apr 11, 2013

Great Bustard wrote:

al_in_philly wrote:

I do the majority of my shooting in very low light situations without a flash.  I started out shooting at night with a (then brand new) Olympus e-510 and a Pan/Leica 25mm f1.4.  I know, great lens choice, not best choice for a low-light camera, but that's what I happened to own when what became a three year nightime photo-shoot opened up for me.  Those photos BTW have been published in a variety of newspapers & magazines, as well as having two gallery shows of them.  Now I'm shooting different very low-light subjects with an OM-D and the m4/3 version of the PL 25 1.4, and I couldn't be happier.  Why?  A big part is because of the depth of field that combo gives me.

Like many others, I too like to seperate my subjects from their background a bit by using the selective focus which wide apertures afford.  But often, I don't want to obliterate all the background, or foreground, definition, just soften it enough to tell my viewers wher to plant their gaze in the picture.  So, if I'm shooting in a dimly lit bar or a city street at midnight, my OM-D produces "just right" images at f1.4 and an ISO of 3200, typically yielding a shutter speed of between 1/30 and 1/80 sec, depending on how dark the lighting is.

Of course, I could shoot FF at 2.8, but then I'd also be shooting at an ISO of 12800 for the same shutter speed.  Even as good as FF images are, the OM-D at 1/4 the ISO will always look better.

...about your last sentence above.  It would be interesting to see a comparison between a modern FF DSLR with a 50 / 1.4 at f/2.8 and the same shutter speed, and see which comes out ahead.  I am far from convinced that it would be the EM5 (which is not to say that I am convinced it will be the FF system).

One day I might experience a brief moment of industriousness and do such a comparison with my Canon 5DIII and Olympus E-M5.  Anecdotally from using both cameras I have a feeling that when you start reaching into the really high ISOs the E-M5 will be better than the 5DIII at 2 stops higher ISO.  For instance E-M5 at 3200 will probably be better than 5DIII at 12800, and E-M5 at 6400 will almost definitely be better than 5DIII at 25600.  I have a feeling that E-M5 at 1600 might even be better than 5DIII at 6400.  At lower ISOs though the difference is small enough that it doesn't matter--for all intents and purposes both produce very good output for the same DOF and shutter speed (i.e. 2 stops different aperture and ISO).

Thus my gut feeling is that for low ISOs where one isn't DOF limited because it's easy enough to get enough DOF the 5DIII is better, but at high ISOs where you want to achieve a certain amount of DOF the E-M5 is better.  The wrinkle is that at high ISOs where you don't so much care about the DOF as you do getting a usable shot the 5DIII is better because you can use faster glass on it and/or shoot at a smaller DOF, sacrificing DOF for shutter speed or having less noise from lower ISO.

Just reinforces the point that there is no such thing as one perfect or best camera for all situations.

Reply   Reply with quote   Complain
KenBalbari
Regular MemberPosts: 275Gear list
Like?
Re: Rethinking 4/3 Depth Of Field
In reply to al_in_philly, Apr 11, 2013

al_in_philly wrote:

Of course, I could shoot FF at 2.8, but then I'd also be shooting at an ISO of 12800 for the same shutter speed.  Even as good as FF images are, the OM-D at 1/4 the ISO will always look better.

Is this true though?  At the moment, it appears to be true with the OM-D, but that's only because of the efficiency of that Sony sensor.  In the past, on other 4/3 and m4/3 models it has been otherwise, there has been more than a 2 stop ISO difference in some cases.  Even now though, is there much difference in these samples?

http://www.dpreview.com/reviews_data/oly_em5/boxshot/p1010017.jpg

http://www.dpreview.com/reviews_data/canon_eos6d/boxshot/img_0048.jpg

If the sensors were equally efficient, then there would pretty much be no case where m4/3 had an inherent advantage other than size, convenience, and cost.

On the other hand, I think we are also getting to where there are fewer and fewer situations in which full frame has any practical advantage that would be worth the difference in cost.

 KenBalbari's gear list:KenBalbari's gear list
Fujifilm FinePix F30 Zoom Olympus E-330 Olympus PEN E-PL2
Reply   Reply with quote   Complain
Martin.au
Senior MemberPosts: 5,408
Like?
Re: Rethinking 4/3 Depth Of Field
In reply to KenBalbari, Apr 11, 2013

There is quite a bit actually. The EM5 cleans house there.

Might be similar noise, but the OM-D isn't showing anywhere near as much "fuzziness".

Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,368
Like?
Re: I'm not convinced...
In reply to Yohan Pamudji, Apr 12, 2013

Yohan Pamudji wrote:

Great Bustard wrote:

al_in_philly wrote:

I do the majority of my shooting in very low light situations without a flash.  I started out shooting at night with a (then brand new) Olympus e-510 and a Pan/Leica 25mm f1.4.  I know, great lens choice, not best choice for a low-light camera, but that's what I happened to own when what became a three year nightime photo-shoot opened up for me.  Those photos BTW have been published in a variety of newspapers & magazines, as well as having two gallery shows of them.  Now I'm shooting different very low-light subjects with an OM-D and the m4/3 version of the PL 25 1.4, and I couldn't be happier.  Why?  A big part is because of the depth of field that combo gives me.

Like many others, I too like to seperate my subjects from their background a bit by using the selective focus which wide apertures afford.  But often, I don't want to obliterate all the background, or foreground, definition, just soften it enough to tell my viewers wher to plant their gaze in the picture.  So, if I'm shooting in a dimly lit bar or a city street at midnight, my OM-D produces "just right" images at f1.4 and an ISO of 3200, typically yielding a shutter speed of between 1/30 and 1/80 sec, depending on how dark the lighting is.

Of course, I could shoot FF at 2.8, but then I'd also be shooting at an ISO of 12800 for the same shutter speed.  Even as good as FF images are, the OM-D at 1/4 the ISO will always look better.

...about your last sentence above.  It would be interesting to see a comparison between a modern FF DSLR with a 50 / 1.4 at f/2.8 and the same shutter speed, and see which comes out ahead.  I am far from convinced that it would be the EM5 (which is not to say that I am convinced it will be the FF system).

One day I might experience a brief moment of industriousness and do such a comparison with my Canon 5DIII and Olympus E-M5.  Anecdotally from using both cameras I have a feeling that when you start reaching into the really high ISOs the E-M5 will be better than the 5DIII at 2 stops higher ISO.  For instance E-M5 at 3200 will probably be better than 5DIII at 12800, and E-M5 at 6400 will almost definitely be better than 5DIII at 25600.  I have a feeling that E-M5 at 1600 might even be better than 5DIII at 6400.  At lower ISOs though the difference is small enough that it doesn't matter--for all intents and purposes both produce very good output for the same DOF and shutter speed (i.e. 2 stops different aperture and ISO).

Well, if you ever get around to it, it would be great to see.  My guess is that a 5D3 at 50mm f/2.8 ISO 12800 will outperform an EM5 at 25mm f/1.4 ISO 3200 -- once NR has been judiciously applied to the 5D3 photo to normalize the detail in the resulting photos.

However, as Anders noted, if we are comparing 50mm f/5.6 ISO 12800 on the 5D3 to 25mm f/2.8 ISO 3200 on the EM5, perhaps the tables will turn.  In either case, I think the differences will be small enough to not be worth worrying over.

Also of interest to me would be comparing with the 5D3 using the 50 / 1.4 and comparing with the 5D3 using a 24-70 / 2.8 zoom.  I wonder how much of a difference that would make.

The advantage of the FF kit will be the option for a more shallow DOF when shallow DOF is desirable, or using a more shallow DOF when the lower noise that accompanies the more shallow DOF in lower light matters more than the deeper DOF.

The advantage of the mFT kit, of course, is size, weight, and cost, not to mention IBIS (although there is the Tamron 24-70 / 2.8 VC).

Thus my gut feeling is that for low ISOs where one isn't DOF limited because it's easy enough to get enough DOF the 5DIII is better, but at high ISOs where you want to achieve a certain amount of DOF the E-M5 is better.

Perhaps -- moreso with the 5D3 than with the 6D, though.  However, I'd be curious to see how the D800 fits in the mix, since it should resolve a lot more detail and thus be able to handle a lot more NR, possibly more than offsetting the higher read noise, even compared to the 6D.

The wrinkle is that at high ISOs where you don't so much care about the DOF as you do getting a usable shot the 5DIII is better because you can use faster glass on it and/or shoot at a smaller DOF, sacrificing DOF for shutter speed or having less noise from lower ISO.

When a deeper DOF is desirable in low light, there's always the balance between DOF and noise, and where that balance point lies will depend on the desired amount of DOF and the amount of available light.

Just reinforces the point that there is no such thing as one perfect or best camera for all situations.

Huh -- news to me!  I thought it was a Leica. 

Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,051Gear list
Like?
Re: Lovely pix and excellent presentation of your argument
In reply to Great Bustard, Apr 12, 2013

Great Bustard wrote:

Hen3ry wrote:

And my argument if it comes to that.

Razor thin depth of field is a by-product, an artifact of big apertures. They allow you to shoot in lower light but more often than not, the very shallow DoF is a damned nuisance.

For me, shallow DOF is the bomb, and I'm wide open, more often than not, regardless of the light.  That said, I don't expect others to have my tastes, but it's not like I never stop down:

http://www.dpreview.com/forums/post/39537128

Besides, and editor might use one or two thin depth of field shots in a publication, but s/he certainly didn’t want page after page of them -- they were showing what the world looked like, not some arty-farty extract of the world governed by a super large aperture.

Well, a modern compact, like the RX100 might be better still for many, and DPR has been showcasing many pros making good use of cell phones for published works.

Then the argument was switched around so that razor thin DoF was practically the raison d'etre for taking a picture. Of course, you had to have the most expensive lens to achieve it so ordinary mortals were excluded.

There are quite a few fast primes for FF that aren't particularly expensive:  35 / 2, 50 / 1.8, 50 / 1.4, 85 / 1.8, 100 / 2.

On the other hand, many of those that aren't particularly expensive aren't very good wide open and thereabout. I am not saying that exceptions do not exist, but they are hardly frequent. Most of the fast Oly & Pany MFT primes, by contrast, do very well already from the get-go, and at least some of them are quite affordable too (20/1.7, 45/1.8).

This was a fad for a while, then the world got back to the business of expecting photographs to show them what the world looked like.

Like I said, you can always stop down, if it results in a better photo.

Super thin depth of field is a choice. Those who want it can have it; I'm not interested and mostly I'm not interested in their pictures either. Very thin DoF moved from the innovative to the boringly banal in a remarkably short time 40+ years ago.

Ah.  Well, don't click on the link above then.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
KenBalbari
Regular MemberPosts: 275Gear list
Like?
Re: Rethinking 4/3 Depth Of Field
In reply to RoelHendrickx, Apr 12, 2013

RoelHendrickx wrote:

Alumna Gorp wrote:

Have a read at this.

http://admiringlight.com/blog/full-frame-equivalence-and-why-it-doesnt-matter/2/

I had seen that before. But did not know anymore where I had.

Thanks for bringing it up again.

-- hide signature --

Roel Hendrickx
lots of images: www.roelh.zenfolio.com
my E-3 user field report from Tunisian Sahara: http://www.biofos.com/ukpsg/roel.html

An interesting point there:

"Lower Noise. Full frame cameras of similar sensor technology to their smaller counterparts will yield lower noise images at its baseline, and continuing throughout the range. Generally, it’s about half to one stop advantage over APS-C and about 1-2/3 stop advantage over 4/3."

For "full frame euivalence" to really work well, you need the diffence to equal more like 2 stops.  I think it seemed to be close to that when we had Panasonic sensors, but maybe Panasonic just never was good at making low noise sensors.

Recently, with the OM-D compared to cameras like the 5DMIII or D800, the difference looks more like 1 stop.  I wonder if there isn't a technical reason why it's difficult for larger sensors to be as efficient as smaller ones. I guess we'll see how this relationship holds going forward.

 KenBalbari's gear list:KenBalbari's gear list
Fujifilm FinePix F30 Zoom Olympus E-330 Olympus PEN E-PL2
Reply   Reply with quote   Complain
Dave Sanders
Contributing MemberPosts: 749Gear list
Like?
Horses for courses...
In reply to Great Bustard, Apr 12, 2013

Great Bustard wrote:

Further, if we compare at apertures slightly smaller than these, e.g., f/2.8 versus f/5.6, the difference in MTF becomes close to insignificant in the first place.

Let's put that to the test, too.  I'm honestly curious to see the results from a competent test.  As I said, no matter which way it falls, I'm of the opinion that the differences are nothing to fret over.

I agree. I think there are probably more things getting in the way of optimal IQ other than lens sharpness or noise.

In any case, if you are always going for the deeper DOF in low light, seems to me that the size and bulk of the FF system compared to the EM5 system makes the EM5 system the overall winner, although there may be operational advatanges that may substantially favor one over the other.

This, I think, is the key. I see my Nikon and m4/3 kits as complimentary; I don't understand the urge to have one replace the other or be the end-all-be-all do everything camera. My main consideration when choosing to use my OMD is size and, as you point out, other operational benefits such as the tilt screen and touch-to-focus. I'm not gonna pick up my OMD and then make a convoluted argument as to why its greater DOF is superior. I'm gonna pick it up and use it when it is the best tool for the job. Same goes for my Nikon: when it's the tool I need, I'll use it. Kinda like I used to select film back in the dark ages. I didn't choose Velvia and talk about why its hideous reddish rendering of skin tones was perfect for a certain look. I'd just choose Astia if I was shooting portraits.

And, yes, don't forget the awesome IBIS in the EM5.

Again, one of the operational considerations when choosing which camera is best for a given situation. These days I'm finding that my main consideration is 'I don't mind carrying it' and, as such, my Nikon is rather lonely.

-- hide signature --

Dave Sanders

Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,051Gear list
Like?
Re: Rethinking 4/3 Depth Of Field
In reply to KenBalbari, Apr 12, 2013

KenBalbari wrote:

al_in_philly wrote:

Of course, I could shoot FF at 2.8, but then I'd also be shooting at an ISO of 12800 for the same shutter speed.  Even as good as FF images are, the OM-D at 1/4 the ISO will always look better.

Is this true though?

I think it is true as a statistical generalization when comparing sensors of different sizes, not just MFT versus FF. As long as you look at SNR from the midtones up, the difference will pretty much conform to the one we would expect based on the principles of equivalence. But read noise will tend to favor the smaller sensors so that the difference in DR/shadow noise will be slightly less than we would expect on the basis of the area difference.

Currently, the best MFT sensor will beat all but a few (D4 and 6D excepted) of the current FF cameras if DoF is held constant so that the FF camera has to use two stops higher ISO. The difference is not large but it's there. And it's not really that the E-M5 sensor is unusually good technically. It's rather that those we had before were not state of the art.

At the moment, it appears to be true with the OM-D, but that's only because of the efficiency of that Sony sensor.  In the past, on other 4/3 and m4/3 models it has been otherwise, there has been more than a 2 stop ISO difference in some cases.  Even now though, is there much difference in these samples?

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
olliess
Contributing MemberPosts: 874
Like?
Re: Rethinking 4/3 Depth Of Field
In reply to Anders W, Apr 12, 2013

Anders W wrote:

exdeejjjaaaa wrote:

Anders W wrote:

At high ISOs, which is what the OP was talking about, the E-M5 set at two stops lower ISO than the FF camera (for equal DoF), will go equal for SNR at midtones but do better with regard to shadow noise/DR in comparison with all current FF cameras except the D4 (where it will be a tie).

it is soundly beaten by DR @ high ISO even by Canon (1Dx).

Please read what both the OP and I were actually talking about. With the E-M5 at ISO 3200 and the 1Dx at ISO 12800 for equal DoF, the E-M5 will come out slightly ahead.

From the DxO measurements that doesn't look to be the case for DR between the 1Dx and the E-M5. Interestingly it looks as if the 1Dx beats out the D4 slightly in DR above ISO 2200 or so..

Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,051Gear list
Like?
Re: I'm not convinced...
In reply to Great Bustard, Apr 12, 2013

Great Bustard wrote:

Anders W wrote:

Great Bustard wrote:

Anders W wrote:

Great Bustard wrote:

al_in_philly wrote:

I do the majority of my shooting in very low light situations without a flash.  I started out shooting at night with a (then brand new) Olympus e-510 and a Pan/Leica 25mm f1.4.  I know, great lens choice, not best choice for a low-light camera, but that's what I happened to own when what became a three year nightime photo-shoot opened up for me.  Those photos BTW have been published in a variety of newspapers & magazines, as well as having two gallery shows of them.  Now I'm shooting different very low-light subjects with an OM-D and the m4/3 version of the PL 25 1.4, and I couldn't be happier.  Why?  A big part is because of the depth of field that combo gives me.

Like many others, I too like to seperate my subjects from their background a bit by using the selective focus which wide apertures afford.  But often, I don't want to obliterate all the background, or foreground, definition, just soften it enough to tell my viewers wher to plant their gaze in the picture.  So, if I'm shooting in a dimly lit bar or a city street at midnight, my OM-D produces "just right" images at f1.4 and an ISO of 3200, typically yielding a shutter speed of between 1/30 and 1/80 sec, depending on how dark the lighting is.

Of course, I could shoot FF at 2.8, but then I'd also be shooting at an ISO of 12800 for the same shutter speed.  Even as good as FF images are, the OM-D at 1/4 the ISO will always look better.

...about your last sentence above.  It would be interesting to see a comparison between a modern FF DSLR with a 50 / 1.4 at f/2.8 and the same shutter speed, and see which comes out ahead.  I am far from convinced that it would be the EM5 (which is not to say that I am convinced it will be the FF system).

On what grounds are you not convinced?

Image quality.

As I know you are well aware, any current FF camera would lose out to the E-M5 for DR/shadow noise in this comparison (except the D4 which would break even).

I think the 6D comes out the same.

OK. You are right about that. But not the 5D3, the D800, or the D600, nor any older FF camera.

In terms of SNR at higher light levels, it would be a tie. So are you thinking about how the PL 25/1.4 at 1.4 compares to a 50 mm FF lens at 2.8 or what?

I'm thinking that a FF camera at 50mm f/2.8 would resolve more detail everywhere in the frame than an EM5 at 25mm f/1.4, and, if the EM5 did have less noise for equivalent settings (same DOF and shutter speed), the judicious application of NR (noise reduction) to the FF photo would result in a photo that was both more detailed and less noisy (albeit not significantly so, in my opinion).

Could be although at these ISOs, I am not at all sure. Even if the FF lens at f/2.8 would clearly outdo the 25/1.4 for MTF, little of that difference is likely to register in view of the noise levels we are talking about here.

Well, that is something I'd like to see put to the test.

I wouldn't mind seeing that either.

Further, if we compare at apertures slightly smaller than these, e.g., f/2.8 versus f/5.6, the difference in MTF becomes close to insignificant in the first place.

Let's put that to the test, too.  I'm honestly curious to see the results from a competent test.  As I said, no matter which way it falls, I'm of the opinion that the differences are nothing to fret over.

I had a go at something like that here (click on "show signature" to see the entire post; I made a mistake of including a dashed line and ... ).

http://www.dpreview.com/forums/post/50462878

Not saying that's ideal in all regards but that was what I could manage.

In any case, if you are always going for the deeper DOF in low light, seems to me that the size and bulk of the FF system compared to the EM5 system makes the EM5 system the overall winner, although there may be operational advatanges that may substantially favor one over the other.

For my personal/wants needs, there aren't a whole lot of pictures that would benefit from (or "tolerate") much less DoF than I get when I shoot my fast MFT primes (12/2, 20/1.7, 45/1.8, 75/1.8) wide open, although as you know I am less fond of background blur with very fast WAs than you are. That said, exceptions do of course exist. See example below.

To that end, I think it would be interesting to see a comparison with competently processed photos.

P.S.:  I think the EM5 would come out significantly ahead in situations where it could use IBIS and a longer shutter speed and FF could not use a tripod.

And, yes, don't forget the awesome IBIS in the EM5.

I am not since I strongly appreciate it. In all fairness though, you can get IBIS with FF too if you get yourself an A99 (although I am not sure whether the efficiency matches that of the E-M5). And below is an example of an image that we have discussed before where an FF-camera with IBIS would beat the E-M5 with IBIS. As you can see, this is with the 12 mm at f/2 and 0.4 s. But the scene is sufficiently far away that one could have used a 24/1.4 on FF and still have sufficient DoF.

And no, tripod-friends, a tripod would not have helped here. I was standing on a staircase that was constantly vibrating due to the traffic on the bridge to which it was connected.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,368
Like?
Re: Lovely pix and excellent presentation of your argument
In reply to Anders W, Apr 12, 2013

Anders W wrote:

Great Bustard wrote:

Hen3ry wrote:

And my argument if it comes to that.

Razor thin depth of field is a by-product, an artifact of big apertures. They allow you to shoot in lower light but more often than not, the very shallow DoF is a damned nuisance.

For me, shallow DOF is the bomb, and I'm wide open, more often than not, regardless of the light.  That said, I don't expect others to have my tastes, but it's not like I never stop down:

http://www.dpreview.com/forums/post/39537128

Besides, and editor might use one or two thin depth of field shots in a publication, but s/he certainly didn’t want page after page of them -- they were showing what the world looked like, not some arty-farty extract of the world governed by a super large aperture.

Well, a modern compact, like the RX100 might be better still for many, and DPR has been showcasing many pros making good use of cell phones for published works.

Then the argument was switched around so that razor thin DoF was practically the raison d'etre for taking a picture. Of course, you had to have the most expensive lens to achieve it so ordinary mortals were excluded.

There are quite a few fast primes for FF that aren't particularly expensive:  35 / 2, 50 / 1.8, 50 / 1.4, 85 / 1.8, 100 / 2.

On the other hand, many of those that aren't particularly expensive aren't very good wide open and thereabout. I am not saying that exceptions do not exist, but they are hardly frequent. Most of the fast Oly & Pany MFT primes, by contrast, do very well already from the get-go, and at least some of them are quite affordable too (20/1.7, 45/1.8).

Well, I guess it depends on what you mean by "aren't very good wide open" -- I found the 50 / 1.4:

http://www.dpreview.com/forums/post/34489299

85 / 1.8:

http://www.dpreview.com/forums/post/34631434

and 100 / 2:

http://www.josephjamesphotography.com/lenses/100f2/

more than "good enough", although that's not to say that I would not be against paying more for better still.  Indeed, I paid a lot more for the 50 / 1.2L:

http://www.josephjamesphotography.com/lenses/50L/

but I have to wonder how many, if anyone, appreciates the difference.

Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,368
Like?
Re: I'm not convinced...
In reply to Anders W, Apr 12, 2013

Anders W wrote:

I had a go at something like that here (click on "show signature" to see the entire post; I made a mistake of including a dashed line and ... ).

http://www.dpreview.com/forums/post/50462878

Not saying that's ideal in all regards but that was what I could manage.

Well, based on those samples, the EM5 has the clear edge.

For my personal/wants needs, there aren't a whole lot of pictures that would benefit from (or "tolerate") much less DoF than I get when I shoot my fast MFT primes (12/2, 20/1.7, 45/1.8, 75/1.8) wide open, although as you know I am less fond of background blur with very fast WAs than you are. That said, exceptions do of course exist. See example below.

Ironically, I think shallow DOF with UWA is "da bomb":

http://www.dpreview.com/forums/post/38116367

To that end, I think it would be interesting to see a comparison with competently processed photos.

P.S.:  I think the EM5 would come out significantly ahead in situations where it could use IBIS and a longer shutter speed and FF could not use a tripod.

And, yes, don't forget the awesome IBIS in the EM5.

I am not since I strongly appreciate it. In all fairness though, you can get IBIS with FF too if you get yourself an A99 (although I am not sure whether the efficiency matches that of the E-M5). And below is an example of an image that we have discussed before where an FF-camera with IBIS would beat the E-M5 with IBIS. As you can see, this is with the 12 mm at f/2 and 0.4 s. But the scene is sufficiently far away that one could have used a 24/1.4 on FF and still have sufficient DoF.

Thing is, though, that at such wide apertures, the corners suffer.  So, for low light wide aperture photography where corners matter (e.g. astrophotography), FF still has ample room for improvement.

And no, tripod-friends, a tripod would not have helped here. I was standing on a staircase that was constantly vibrating due to the traffic on the bridge to which it was connected.

Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,368
Like?
I had read that...
In reply to Alumna Gorp, Apr 12, 2013

Alumna Gorp wrote:

Have a read at this.

http://admiringlight.com/blog/full-frame-equivalence-and-why-it-doesnt-matter/2/

...and made a few notes on some of his opinions:

http://www.dpreview.com/forums/post/50805880

Reply   Reply with quote   Complain
Alumna Gorp
Senior MemberPosts: 1,531Gear list
Like?
Re: I had read that...
In reply to Great Bustard, Apr 12, 2013

Great Bustard wrote:

Alumna Gorp wrote:

Have a read at this.

http://admiringlight.com/blog/full-frame-equivalence-and-why-it-doesnt-matter/2/

...and made a few notes on some of his opinions:

http://www.dpreview.com/forums/post/50805880

No bleeding you again

 Alumna Gorp's gear list:Alumna Gorp's gear list
Fujifilm X10 Panasonic Lumix DMC-G2 Panasonic Lumix DMC-GF2 Olympus OM-D E-M5 Panasonic Lumix G 14mm F2.5 ASPH +4 more
Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,368
Like?
Re: I had read that...
In reply to Alumna Gorp, Apr 12, 2013

Alumna Gorp wrote:

Great Bustard wrote:

Alumna Gorp wrote:

Have a read at this.

http://admiringlight.com/blog/full-frame-equivalence-and-why-it-doesnt-matter/2/

...and made a few notes on some of his opinions:

http://www.dpreview.com/forums/post/50805880

No bleeding you again

That's right -- I had forgotten we had crossed paths before:

http://www.dpreview.com/forums/post/51210089

By the way, here's a current case-in-point as to why an understanding of Equivalence is useful for many:

http://www.dpreview.com/forums/post/51275066

Seems that f/2.8 isn't f/2.8 after all, eh? 

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads