Home · Register · Search · View Winners · Software · Hosting · Software · Join Upload & Sell

Moderated by: Fred Miranda
Username   Password

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Canon Forum | Join Upload & Sell

1      
2
       3       end
  

Archive 2013 · Sigma 18-35 f/1.8 hands-on and samples online
  
 
myron lee
Offline
• •
Upload & Sell: Off
p.2 #1 · Sigma 18-35 f/1.8 hands-on and samples online


andyjaggy82 wrote:
I understand and agree with everything posted above. I think what really annoys me about the term is that some people talk like a full frame 2.8 zoom is better than a 2.8 crop zoom because it has "more light gathering" ability. I guess that's technically true, but really kind of pointless in the real world, where your crop sensor can't use that light anyway.

So some people discredit this lens claiming that's it's really just the equivalent of a full frame 2.8 zoom, because it "gathers the same amount of light" as a full frame 2.8 zoom. As if
...Show more

+ 1. Lets get back to basics here. A crop camera captures the image from the center portion of the lens onto a small sensor. Same light/ sq mm light capture. Same DOF. Same bokeh. Only difference is the image composition is different. Shoot at a person with a ff and with a crop camera. You will see it's the same everything, except the image on the crop is ...well a "cropped" image of the ff image.
So with that said, f1.8 has a huge advantage on any camera than a f2.8. It will capture more light on the sensor per square feet. One just needs to know how to take advantage of it. With current technology, I prefer ff because of the issue with noise on densely packed sensors which is the crop camera disadvantage. The other is framing of object and background blur. Huge for me, but not others.




Apr 23, 2013 at 11:02 PM
super35
Offline
• •
Upload & Sell: Off
p.2 #2 · Sigma 18-35 f/1.8 hands-on and samples online


Equivalence, light gathering ability, and total amount of light captured...

These terms are literally blowing the minds of photographers all over the internet. Why is that? What is so hard to understand about it?

Part of it is ignorance and the other part is ego. Some people, though they can comprehend, refuse this knowledge because their ego is tied up in the equipment purchases they have made.

...a Hasselblad H4D (40.2 x 53.7mm) has more light gathering ability than FF 135 (24 x 36mm), yet you never hear anyone cry about that.



Apr 23, 2013 at 11:28 PM
bemyzeke
Offline
• •
Upload & Sell: Off
p.2 #3 · Sigma 18-35 f/1.8 hands-on and samples online


myron is right. This is becoming another myth. Lens characteristics, such as aperture and focal length, are physical constants that do not change over time. Noise, however is not constant, not universal, and also not straight forward to measure. Different testing groups will come with different figures (sometime contradicting each other).

Noise is not a characteristic of the whole sensor. It is the characteristic of individual pixel site (and the A/D converter associated to that pixel). When the pixel gets more light the S/N ratio improves and the apparent noise decreases. Due to bayer interpolation, the noise gets averaged between adjacent pixels, so the noise we see is a cumulative function of a bayer group. However the signal (and noise) output of a pixel is not a function of another pixel 20-30 mm away which may or may not be there.

A bright lens will shine more light on a pixel, increasing its exposure. With more exposure, you can either increase your shutter speed or reduce ISO. A bright lens just gives you more light, how you make use of that light is upto you. You may want to use it to compensate for the small size of your sensor (your choice), or increase shutter, but f/1.8 is f/1.8 whether it is on full frame or on a cell phone.

That said, why do people assume that FF is one stop better in ISO over APS-C. This is a gross over simplification. Noise levels vary from camera to camera, vendor to vendor, and generation to generation. Is D4/D600/D800 only one stop better than D70 or even D300? Is D7100 one stop worse than Canon 1Ds Mark I (it is actually better) or the original 5D ( it is about the same)?

[For the record, I own FF cameras and have done so for last several years. So no ego is involved.]






Apr 23, 2013 at 11:30 PM
super35
Offline
• •
Upload & Sell: Off
p.2 #4 · Sigma 18-35 f/1.8 hands-on and samples online


The myth is that f/1.8 = f/1.8 = f/1.8 on different size sensors or films. This is the thing that trips up so many people.

Understanding the concept of equivalence can only help photographers. They will be able to understand why an 18-35mm f/1.8 zoom on a Canon 7D is approximately equivalent to a 29-56mm f/2.8 zoom on a 5D Mark III. Knowledge of the physics and math behind photography is no way harmful or detrimental to the art of photography. The two can and SHOULD compliment each other. Photographic Equivalence explained here.



Apr 24, 2013 at 12:40 AM
bemyzeke
Offline
• •
Upload & Sell: Off
p.2 #5 · Sigma 18-35 f/1.8 hands-on and samples online


super35 wrote:
The myth is that f/1.8 = f/1.8 = f/1.8 on different size sensors or films. This is the thing that trips up so many people.

Knowledge of the physics and math behind photography is no way harmful or detrimental to the art of photography. The two can and SHOULD compliment each other.


That is right. Which is why f/1.8 is f/1.8 because it is based on the physical characteristic of the lens, not of the sensor. An f/1.8 will give you same exposure (in terms of EV) no matter which camera you choose.

The equivalence certainly has its place, but where people get it wrong is assuming that f/1.8 equals to f/2.8 because there is one stop of ISO difference between sensor. This is incorrect. Noise level difference between sensors can be one stop, could be more, or could be less depending on sensor and ISO chosen (see examples above in my previous post). If I shoot my D800 in DX mode, the noise level is the same as in FX mode. However with f/1.8 I can increase my shutter speed and capture fast action or just keep the shutter same and increase image brightness.

Still all of this is how I choose to use the aperture. How I use the aperture DOES not change the value of aperture.





Apr 24, 2013 at 01:28 AM
super35
Offline
• •
Upload & Sell: Off
p.2 #6 · Sigma 18-35 f/1.8 hands-on and samples online


bemyzeke, you are right there in the forest but you have missed the trees. I have a question for you.

The premise:
I make a photo in FX mode and then another DX mode with a D800 of the same subject. I use the same aperture, shutter speed, angle of view, and ISO. Now, I print both images at the same size of 24x36 inches.

The million dollar question:
Which print has more noise?

Bonus question:
Which image has more resolution?

Edited on Apr 24, 2013 at 02:19 AM · View previous versions



Apr 24, 2013 at 02:15 AM
bemyzeke
Offline
• •
Upload & Sell: Off
p.2 #7 · Sigma 18-35 f/1.8 hands-on and samples online


What if I want to only print a 4x6 photo?


Apr 24, 2013 at 02:19 AM
cputeq
Offline
• • • •
Upload & Sell: On
p.2 #8 · Sigma 18-35 f/1.8 hands-on and samples online


super35 wrote:
The myth is that f/1.8 = f/1.8 = f/1.8 on different size sensors or films. This is the thing that trips up so many people.

Understanding the concept of equivalence can only help photographers. They will be able to understand why an 18-35mm f/1.8 zoom on a Canon 7D is approximately equivalent to a 29-56mm f/2.8 zoom on a 5D Mark III. Knowledge of the physics and math behind photography is no way harmful or detrimental to the art of photography. The two can and SHOULD compliment each other. Photographic Equivalence explained here.



It's no myth at all. The problem stems from people arguing two different facets of the same thing (really probably three facets), and being either too stupid or proud to realize it. They either don't comprehend the nuances of the arguments (to realize they're both right, conditionally), or they're too proud to admit it.

There is an operational argument (how the camera acts when determining shutter speeds, how many photons are collected per sensor-mm^2 per second, etc) and there is a results argument (how the image actually looks on screen or paper, but let's keep to screen). Lastly, there's a common-denominator physics argument facet that is almost completely f-ing pointless (in at least two ways), but people keep bringing it up.

The operational "camp" is correct - assuming 15 camera bodies from various brands will meter the same scene the same way, assuming the T-stops of various lenses are in fact "true" to the f-stop (which hardly happens, if ever), and assuming these 15 bodies aren't playing games with their indicated-vs-true ISO settings, then f/1.8 = f/1.8 = f/1.8 in the sense that the attached cameras will give exactly the same shutter speeds. Because, ultimately, everyone on planet Earth is concerned about shutter speeds when taking a photo. So, they're worried about the operational characteristic of COMBO A vs COMBO B and they find they are, in fact, equivalent in that sense.


The results camp begs to differ. They make the argument that the f/2.8 lens on FF is equivalent to the f/2 lens on the APS-C camera, so that f/1.8 != f/1.8 != f/1.8. They are conditionally correct, just like the operational people. The results camp seems to assume the FF camera is always at least 1 stop better than the APS-C camera, therefore the end results are equivalent -- the FF camera can just bump up ISO to gain an equal shutter speed and equal noise level as the "faster" APS-C combo. In their mind, the f/2 lens has to be f/2 on APS-C just to "catch up" to the FF camera.

This is also a flawed argument, but conditionally correct - Again, this assumes the FF camera is ALWAYS at least 1 stop better than the APS-C camera. This argument fails/will fail when considering old FF vs modern APS-C. It also fails when considering cameras react differently to increased ISO levels. I once compared my Panasonic GX1 vs NEX5n vs Canon 5D3 all at ISO400 (I think) and was completely unimpressed with the 5D3 - I literally had a hard time telling them apart (dark shot, not brightly lit stuff). I then performed the same comparison at ISO 3200. The 5D3 was more than 1 stop ahead of the other two cameras, but you didn't hear me shouting from the rooftops that my 40/2.8 pancake was actually a badass 40mm/1.4 in "light gathering" for only $150! (and yes, this took resampling into effect, but let's not go there)


Then you have the "common-denominator physics" arguments that are completely f-ing pointless, even if they ARE, technically true. YES Captain of the Obvious, we KNOW that a PHYSICALLY LARGER LENS (f/2.8 vs f/2.8), such as a FF lens vs an APS-C lens, will actually capture more photons/second. No-freaking-duh.

This argument is pointless, but people keep bringing it up as "light-gathering". First, while YES the FF lens has "scooped" more photons/second, it has to distribute them over a larger area in order to get the equivalent shutter speed. This is true as you go up and down the sensor size. Nobody really gives a damn about how many photos the unmounted, physical pieces of glass and plastic could collect in 1/100 of a second - they only care about where those photons actually go (the results camp) or how the aperture affects the metering of the camera (the operational camp).








Edited on Apr 24, 2013 at 02:23 AM · View previous versions



Apr 24, 2013 at 02:19 AM
super35
Offline
• •
Upload & Sell: Off
p.2 #9 · Sigma 18-35 f/1.8 hands-on and samples online


bemyzeke wrote:
What if I want to only print a 4x6 photo?



, thats what the D800 is made for.



Apr 24, 2013 at 02:19 AM
bemyzeke
Offline
• •
Upload & Sell: Off
p.2 #10 · Sigma 18-35 f/1.8 hands-on and samples online


That is beside the point. I have a camera and I make a picture from it.

How I use the camera does not change the specs of the camera, nor does it change the specs of Lens. Is this that hard to understand?



Apr 24, 2013 at 02:21 AM
 

Search in Used Dept. 



super35
Offline
• •
Upload & Sell: Off
p.2 #11 · Sigma 18-35 f/1.8 hands-on and samples online


cputeq, Please read this before commenting further. Everything is explained clearly.

http://www.josephjamesphotography.com/equivalence/

"Saying "f/2 = f/2 = f/2" is like saying "50mm = 50mm = 50mm". Just as the effect of 50mm is not the same on different formats, the effect of f/2 is not the same on different formats."




Apr 24, 2013 at 02:30 AM
bemyzeke
Offline
• •
Upload & Sell: Off
p.2 #12 · Sigma 18-35 f/1.8 hands-on and samples online


cputeq wrote:


Congratulations! You have managed to insult both sides (or three sides) equally well.



Apr 24, 2013 at 02:31 AM
super35
Offline
• •
Upload & Sell: Off
p.2 #13 · Sigma 18-35 f/1.8 hands-on and samples online


, I shoulda known better. Enjoy the bliss of ignorance.

bemyzeke wrote:
That is beside the point. I have a camera and I make a picture from it.

How I use the camera does not change the specs of the camera, nor does it change the specs of Lens. Is this that hard to understand?




Apr 24, 2013 at 02:33 AM
cputeq
Offline
• • • •
Upload & Sell: On
p.2 #14 · Sigma 18-35 f/1.8 hands-on and samples online


bemyzeke wrote:
Congratulations. You have managed to insult both sides (or three sides) equally well.




If it'll somehow lessen these ludicrous pseudo-correct arguments, so be it. One shouldn't be insulted when told that you're not right all of the time, especially when balance with you are right some of the time




Apr 24, 2013 at 02:33 AM
bemyzeke
Offline
• •
Upload & Sell: Off
p.2 #15 · Sigma 18-35 f/1.8 hands-on and samples online


super35 wrote:
, I shoulda known better. Enjoy the bliss of ignorance.



I do, I do. There is certainly a lot I don't know about (and dont want to know about), but this is not one of those things. You are arguing for equivalence, which is not the same as equal. In physics we used the sign of double tilda to indicate equivalent. Equivalence is conditional, subjective, and will change based on circumstances.



Apr 24, 2013 at 02:42 AM
myron lee
Offline
• •
Upload & Sell: Off
p.2 #16 · Sigma 18-35 f/1.8 hands-on and samples online


super35 wrote:
bemyzeke, you are right there in the forest but you have missed the trees. I have a question for you.

The premise:
I make a photo in FX mode and then another DX mode with a D800 of the same subject. I use the same aperture, shutter speed, angle of view, and ISO. Now, I print both images at the same size of 24x36 inches.

The million dollar question:
Which print has more noise?

Bonus question:
Which image has more resolution?


It depends on alot of things. Do both FX and DX mode uses the full area of the sensor? Probally not, Explain to me "same angle of view". Are you moving in closer when you are in FF mode? If you can explain how the image is processed in the dx800 to us, I bet I can make a pretty good guess.


Edited on Apr 24, 2013 at 04:59 AM · View previous versions



Apr 24, 2013 at 04:46 AM
Yohan Pamudji
Offline
• • •
Upload & Sell: On
p.2 #17 · Sigma 18-35 f/1.8 hands-on and samples online


bemyzeke wrote:
That is right. Which is why f/1.8 is f/1.8 because it is based on the physical characteristic of the lens, not of the sensor. An f/1.8 will give you same exposure (in terms of EV) no matter which camera you choose.

The equivalence certainly has its place, but where people get it wrong is assuming that f/1.8 equals to f/2.8 because there is one stop of ISO difference between sensor. This is incorrect. Noise level difference between sensors can be one stop, could be more, or could be less depending on sensor and ISO chosen (see examples above in my
...Show more

Bolded the text in question. Actually no, you don't get the same noise level. Per pixel yes but per image no, assuming you view or print both images at the same size. The DX image will be noisier than the FX image because of enlargement. If there is any doubt of the effect of enlargement on noise, print the same photo at 4x6 then 20x30, view from the same distance away and tell me which one is noisier.

The equivalence argument assumes all other things being equal, which is basically impossible when comparing different cameras but your example of using FX and DX mode in the D800 actually works well for this.

To help guide the discussion let's agree on a common goal: to take the same exact picture with APS-C and FF. Same picture means same everything--angle of view, DOF, motion blur or camera shake (or lack thereof), noise, viewed at the same size (screen or print). As photographers that's what it's all about after all right--the final image? Does that sound like a reasonable assumption? If so the question boils down to this: how can this be achieved on APS-C vs. FF?

Angle of view:
Shoot both from the same distance from subject. Take the DX focal length and multiply by 1.5x (easier number than Canon's 1.6x). So from the same distance shoot DX at 50mm and FF at 75mm. Not so easy with the dearth of 75mm lenses, but bear with me. You could do 24mm and 36mm if you like since 35mm lenses are abundant and close enough to 36mm in practical terms.

DOF:
There's roughly a 1 stop difference in DOF control between APS-C and FF due to enlargement. So shoot the DX at f/2 and the FF at f/2.8. Technically the DX aperture would need to be slightly larger than f/2 but close enough anyway. You would also view/print both images at the same size, for instance at the same size on the same monitor. DOF is a function of viewing size and distance after all, so it makes sense to ensure viewing conditions are identical.

Motion blur/camera shake:
Easy--shoot with the same shutter speed.

Noise:
Because FF is a stop better than DX, shoot FF at 1 stop higher ISO. Example: DX at 1600 and FF at 3200.

This gives you the same exact photo from both DX and FX. This is what I think of when I think "equivalence". There are certainly different facets to this and different ways of approaching the issue, which is what causes a lot of misunderstanding and frustration when the topic comes up. If you want to discuss or contest any of this I'll be happy to follow up. We're all adults. No reason why we can't have a discussion without shouting each other down.

FYI I'm not saying any of this to put down the new Sigma 18-35mm f/1.8. I think it's a fantastic looking lens that evens the playing field with FF in many ways. I'm not a FF zealot either since I happily shoot with both FF and m4/3 (2x crop), and if we want to get into a thin DOF shouting match there's always large format to put everybody in their place or better yet the van camera (http://vimeo.com/39578584). But it always pays off to keep in mind the final result and the various ways that said result can be achieved.



Apr 24, 2013 at 04:49 AM
myron lee
Offline
• •
Upload & Sell: Off
p.2 #18 · Sigma 18-35 f/1.8 hands-on and samples online


Yohan Pamudji wrote:
Bolded the text in question. Actually no, you don't get the same noise level. Per pixel yes but per image no, assuming you view or print both images at the same size. The DX image will be noisier than the FX image because of enlargement. If there is any doubt of the effect of enlargement on noise, print the same photo at 4x6 then 20x30, view from the same distance away and tell me which one is noisier.

The equivalence argument assumes all other things being equal, which is basically impossible when comparing different cameras but your example of using
...Show more

Noise comes from the circuits being to close together in both the sensor and camera. It has nothing to do with it the camera being a crop or full frame. All that dense wiring needs to be insulated from one another. It is the same principal with our wiring for our home speakers and TV sets. Pack them up close together with poor insulation, you will get signal noise. The crop camera has a smaller sensor, they have more of a challange with noise issues if they need to maintain the same megapixel count as the FF. Same happens with the sensor being so dense, It is harder to separate the light photons as they to create interference. Newer cameras have less noise because they found better ways to insulate between the curcuity between photons. Also if there is very little photons around, say at night, then these photons do not provide enough image details so the noise from the curcuits being so close together shows up more. Especially when the camera turns up the sensitivity to get as much light as possible. Same thing happens with your eyes in the dark. You start to see crap floating in your eyes when in the dark.



Apr 24, 2013 at 05:10 AM
bemyzeke
Offline
• •
Upload & Sell: Off
p.2 #19 · Sigma 18-35 f/1.8 hands-on and samples online


Yohan Pamudji wrote:
DOF:
There's roughly a 1 stop difference in DOF control between APS-C and FF due to enlargement. So shoot the DX at f/2 and the FF at f/2.8. Technically the DX aperture would need to be slightly larger than f/2 but close enough anyway. You would also view/print both images at the same size, for instance at the same size on the same monitor. DOF is a function of viewing size and distance after all, so it makes sense to ensure viewing conditions are identical.

Actually DOF works the other way when enlarging. Since crop sensor images have to be enlarged more, the circle of confusion gets bigger and therefore the depth of field is reduced. However the angle of view comes to rescue (like you said) because instead of 75mm I have to shoot with 50mm to get the same frame, and 50mm has better DoF, so the end result is that we get more DOF.


Motion blur/camera shake:
Easy--shoot with the same shutter speed.

Why shoot at the same shutter speed. Why not faster?

This is why I am personally interested in this lens. In poorly lit family rooms, dance halls, sport gyms, where action is happening, I can use all the light that is available to keep my shutter speed high and eliminate subject blur. As long as the focal length works for me, I have no qualms in shooting DX to get the faster shutter speeds. True, if printed at 40x60 the images will be noisier, but noisier images are a hell of a lot better then blurry images. Noise can be removed, details can be recovered, but so far there is no commercial working algorithm to remove motion blur.


Noise:
Because FF is a stop better than DX, shoot FF at 1 stop higher ISO. Example: DX at 1600 and FF at 3200.

Says who?

First of all, in my given scenario, of D800 DX/FX, noise is the same.

Secondly

D800 - D300 = way more than one stop difference
D7100 - Canon 5D = Same level of noise
D7100 - Canon 1Ds Mark I = D7100 beats the pants of this first generation full frame
Fuji X-Trans (what is noise?)
Olympus OM-D - Canon 5D = about same






Apr 24, 2013 at 05:38 AM
Snopchenko
Offline
• • • •
Upload & Sell: Off
p.2 #20 · Sigma 18-35 f/1.8 hands-on and samples online


I was hoping for some discussion of the new and exciting lens (and probably its usability on APS-H bodies, courtesy of Ian), but instead it's all more of the same "equivalence" stuff...







Apr 24, 2013 at 08:49 AM
1      
2
       3       end




FM Forums | Canon Forum | Join Upload & Sell

1      
2
       3       end
    
 

You are not logged in. Login or Register

Username   Password    Retrive password