Lens porn: Sigma 50-100mm F1.8 Art

[Cinegain]

39 minutes ago, TheRenaissanceMan said:

[..]

Of course, the practicality of using a lens this huge, expensive, and heavy on a GH4 or BMPCC leaves something to be desired...

Yes. I wonder, though. Who is this for?

The professionals would most likely shoot fullframe. If they're into sports and wildlife photography, the crop does help with the tele reach. But how much tele can you get out of 100mm? Besides, my understanding is that like the 18-35mm f/1.8 this does not have optical image stabilization. So I wouldn't really see this fit in with the pro stills shooter? Unless, they are dedicated to their APS-C camera and wouldn't mind to shoot something else... like portraits with it. Seems like a great option for that last application.

People who already have the 18-35mm f/1.8. With which they target the mass consumer market for a serious up in quality from any kitlens with great sensitivity. So you kinda got your semi-wide 18mm to the 25mm, 35mm, 50mm, 85mm and 100mm f/1.8 all in two lenses should you get the new one as well. My problem with this. Is this is for the mass consumer market. Maybe someone who picked up the T6s/760D... are they really going to drop 1499 USD on a lens 2x the cost of their camera kit or almost 2.5x the 18-35mm f/1.8? One without image stabilization? I don't know. I don't really see it.

For us video shooters it would be way nice. The range, the sensitivity, the consitency with the 18-35mm f/1.8. Having only 2 lenses to shoot 85% of more of what you shoot. We're used to rigging things up, we don't mind the lack of IS neccessarily. What we perhaps do as well mind, is the price. I mean, I would've said that at 1.5x the 18-35mm f/1.8, coming in at just below 1000 USD, they would've sold me on this. 1499 USD?? I don't know. It seems a little stretch. But yeah, it's one of a kind. Still as sexy as I first thought, but I did gain some perspective reading and thinking about it. Not sure.

[The Chris]

11 minutes ago, DayRaven said:

#delurk

my brain is melting trying to follow this, mostly because that video makes perfect sense - but you all are saying it's wrong! The Cris, can you post your pics, you just took, do they have the same amount of blur in the background? I only have a 16-35 on my bmpcc with the dedicated speed booster and am really interested in this lens, but it will or won't give me the same performance as a theoretical lens with the same mm range, the same f stop but covers a full frame - and would that be the case if I put that theroetical lens on a theoretical ff camera it would perform differently to the sigma on a crop frame....

ARGH! I'm in the market for a new camera and some that I'm looking at include the A7r II, Ursa Mini, so while I could just buy the lens and be happy, I do want to make an informed choice.

#relurk

Bokeh and exposure are two different things, and that's what the Grim dude is mixing together. If you want to compare bokeh between different lenses and sensor sizes there are website's that allow you to select parameters and see the difference. 

[DayRaven]

5 minutes ago, The Chris said:

Bokeh and exposure are two different things, and that's what the Grim dude is mixing together. If you want to compare bokeh between different lenses and sensor sizes there are website's that allow you to select parameters and see the difference. 

Ok, great, thanks, but one more question - in that video, at the start he puts three different cameras on the same lens, and doesn't change anything on the lens and they all expose differently, the smaller the sensor, the darker the image - what causes that, did he cheat the test, because I watch him every week and he doesn't seem the type!

[Nikkor]

41 minutes ago, DayRaven said:

Ok, great, thanks, but one more question - in that video, at the start he puts three different cameras on the same lens, and doesn't change anything on the lens and they all expose differently, the smaller the sensor, the darker the image - what causes that, did he cheat the test, because I watch him every week and he doesn't seem the type!

But the tits are big on all three, aren't they?

Anyway, imagine you crop a fullframe image, voilá, there you have your aps-c, micro four thirds or whatever you want. If the image is darker it's only because he either hasn't set the same iso and shutter speed or because the iso has some variation (iso is a standart, but it's never spot on -see dxo-).

The F number, the ISO, and the shutter speed, are the three parameters that will let you expose correctly no matter what sensor ,film or whatever you are using.

Another story is noise, dynamic range,etc...

[Cinegain]

6 minutes ago, DayRaven said:

#delurk

my brain is melting trying to follow this, mostly because that video makes perfect sense - but you all are saying it's wrong! The Cris, can you post your pics, you just took, do they have the same amount of blur in the background? I only have a 16-35 on my bmpcc with the dedicated speed booster and am really interested in this lens, but it will or won't give me the same performance as a theoretical lens with the same mm range, the same f stop but covers a full frame - and would that be the case if I put that theroetical lens on a theoretical ff camera it would perform differently to the sigma on a crop frame....

ARGH! I'm in the market for a new camera and some that I'm looking at include the A7r II, Ursa Mini, so while I could just buy the lens and be happy, I do want to make an informed choice.

#relurk

That's because people are mixing depth of field and exposure and making false statements about them. Also, they want crop cameras to adhere to fullframe camera behaviour, although you could just as easily state you can conform fullframe to crop cameras if all you want is make them identical. Just bear with me:

[> At the exact same settings: ISO, shutterspeed and aperture, regardless of sensor, you will get an evenly lit picture on either of two cameras

Next, we will define two cameras to compare. We'll be comparing a fullframe sensor camera (1.0x, no crop) against camera with Micro Four Thirds sensor (has 2x crop). For ease of comparison we give these two imaginary cameras the same megapixel count and shoot the same resolution video with it without additional crops.

[> At the exact same settings: ISO, shutterspeed and aperture, to get the same framing between the two cameras, you'd have to use a lens half the focal length on the 2x crop body

Illustration intermezzo:

Say, we use two identical 70-200mm f/2.8 lenses at 200mm f/2.8 on both cameras as our starting point. The image circle is the same for both: 200mm f/2.8. For this example we just assume that in the illustration above the circular image on the left is what the lens gives off at 200mm f/2.8. The 35mm fullframe sensor uses as much width as a squared sensor can get out of that circular image, capturing most of the scene. The cropped sensor crops in on that image circle (which, as we can see, doesn't affect the image's brightness in any way). So, we said that for our example we use a 2x crop camera to compare to. Looking at the pictures at the same physical dimensions, we notice that the image of the cropped sensor looks zoomed in. So in order to get back to similar framing we will have to use the lens at a wider position for the 2x crop body (or move the entire camera back in opposite direction of shooting until the framing is the same, but this is inconvenient and may not always be possible). Luckily we're using a 70-200mm zoom lens, so for the crop sensor we'll adjust the focal length to half of that we've set on the fullframe (200mm f/2.8), at 100mm f/2.8. Let's continue:

[> At the exact same settings: ISO, shutterspeed and aperture, with the same framing (two different focal lengths), the 2x crop body will have a deeper depth of field compared to a fullframe camera

We're now using the lens on the 2x crop body less zoomed in, meaning we are decreasing the subject isolation and increasing the depth of field

[> At the exact same ISO and shutterspeed, with the same framing (two different focal lenghts), the 2x crop body will match the depth of field with the fullframe camera having its lens selected to a full stop darker

Pinching the aperture on the lens mounted on the fullframe camera to 200mm f/4 will result in a deeper depth of field to match that of the depth of field with the 2x crop sensor camera that's at 100mm f/2.8. However. Pinching the aperture on the the lens will make the image on the fullframe camera darker... leaving only the one on the 2x crop sensor exposed properly

[> At the exact same shutterspeed, with the same framing (two different focal lengths), with matching depth of field and the fullframe body with lens one stop darker than the lens on the 2x crop body, you'd have to increase the ISO on the fullframe camera with one stop to even out the exposure again

Luckily earlier on we would assume both sensors were made up out of an equal amount of megapixels. However, the fullframe sensor is physically larger than the cropped sensor, giving each individual pixel better light gathering capability. Kinda of like solar panels and their better efficiency when they're bigger. So upping the ISO doesn't have to result is terrible noise per se

THE LESSONS LEARNED (hopefully)

• A lens is a lens. F-stop = f-stop. Cropping in on the image circle a lens gives off does nothing, absolutely nothing to the brightness. A crop sensor camera doesn't record f/4 brightness with a lens set at f/2.8, that's just bollocks!
• For the same focal length used, the crop sensor will have framing that appears 'zoomed in'. To counter this you have to either step back and create more distance between you and your subject, or the more convenient solution: use a wider lens or a zoomlens at a wider position
• When you want to equal the framing with the different sensor cameras at the same place using different focal lenghts you will create a difference in perceived depth of field
• To equal the depth of field you either have to pinch the aperture on the lens that's attached to the fullframe camera, or brighten the other one (this might not be possible)
• To even out exposure you have to work with the forbidden love triangle that is: ISO, shutterspeed and aperture (additional influences: external lighting, (variable) ND filters)

In the end I think we're all on the same page. Just some people have poor wording or might just misunderstand the concept a tiny bit. Which is no problem as long as they get it right within their own world and others in theirs. Now, it's well past midnight and I've been up for way too long, so I hope I've jotted down everything the way I meant. If you feel I'm in the wrong somewhere, I'm open-minded enough to accept essays in my EOSHD Inbox and rectify this comment according my newly acquired knowledge. But you really got to bring it to make me a believer...

Now let's get back to that Sigma, ey?

[DayRaven]

That's amazing, Thanks

EDIT: Just to check - you mention moving the camera towards or away from the subject or zooming, but physically moving the camera away from the subject causes the scene to looked more compressed, right?

[Andrew Reid]

I love what Sigma are doing and am excited for the 50-100 F1.8 which I am sure will be mega sharp

But a Speed Booster solution is definitely better value for money.

The XL on Micro Four Thirds for instance is a 0.64x so an F2.8 lens will be F1.8.

The E-mount Speed Booster will make F2.8 a F2.0

The Tokina zoom mentioned earlier or the Sigma 50-150mm F2.8 II EX DC APO for APS-C would be the ones to go for. I have the Sigma, it's great.

[Timotheus]

This thread has gone quite of the rails. Still interesting luckily :-) Was working on an explanation also but Cinegain beat me to it, cheers haha. Very concise indeed...btw, this and what Tony Northrup is explaining is exactly the same.

As for Andrew's recommendation of the Sigma 50-150: it has my vote too. Great, portable lens well below 1kg (the one without OS that is). Probably very comparable to the aforementioned Tokina 50-135.

[Nikkor]

9 minutes ago, Andrew Reid said:

I love what Sigma are doing and am excited for the 50-100 F1.8 which I am sure will be mega sharp

But a Speed Booster solution is definitely better value for money.

The XL on Micro Four Thirds for instance is a 0.64x so an F2.8 lens will be F1.8.

The E-mount Speed Booster will make F2.8 a F2.0

The Tokina zoom mentioned earlier or the Sigma 50-150mm F2.8 II EX DC APO for APS-C would be the ones to go for. I have the Sigma, it's great.

But the XL will make the 1.8 f1.15

The problem is that such a long lens without stabilization is kind of useless right now. 

[Cinegain]

10 minutes ago, DayRaven said:

That's amazing, Thanks

EDIT: Just to check - you mention moving the camera towards or away from the subject or zooming, but physically moving the camera away from the subject causes the scene to looked more compressed, right?

Yes, that's true! That only works for the framing. You're now comparing the same lens set-up, but the sensors are not aligned in the same plane, resulting in different compression indeed.

[Guest Ebrahim Saadawi]

Not really interested in a standard tele with no IS.

This is a HUGE limitation on an otherwise good concept. No IS on the 18-35mm is acceptable, but not at 50-100mm.

Also 50-100mm is a very small range for me to use as my sole standard tele.

For APS-C S35 I'd much rather use (a super sharp) 70-200mm f/2.8 IS (even without a SB), compared to this sigma: you have extra IS, 100-200mm, and only lose f1.8 vs f2.8, both pretty damn fast. 

Or as others mentioned if you'll use it on a Sony s35, with a SB it gives you a 50-140mm f/2 IS. Also a much better proposition than this lens for APS-C/S35 shooters. 
 

There are 3 potential users:

-Canon/Nikon S35 bodies:   70-200mm f/2.8 IS  vs   50-100mm f/1.8. Which would you get? 

-Sony s35 bodies: + SB:    50-140mm f/2 IS   vs   50-100mm f/1.8. Which would you get? 

-M43s 4k bodies: + SB XL:     35-70mm f/1.2 m43s lens. Which is a cool lens I guess. 

So I only see this new lens a use in speed-boosting it for a m43s 4k body on which up until this lens came out they couldn't get the FF 70-200mm F/2.8 look equivalent, now they can get it (but with a shorter reach at about 150 not 200, and no IS)

 

For me as a whole proposition, no, there are many extremely sharp stabilized fast teles out there covering APS-C to sacrifice IS. 

 

What's that price anyway?

[cantsin]

The problem is: both arguing parties in this debate are right. ;-)

Yes, in terms of exposure measurement, f2.8 is the same for every camera, no matter the sensor size. But: f-stops are relative, not absolute values for light gathering. They describe the factor between the incoming and the outgoing image circle of a lens. In absolute terms, however, the image circle/sensor surface of full frame is twice the size of APS-C and four times the size of MFT (if you factor vertical and horizontal dimensions). That means that a full frame lens at f2.8 captures four times as much light as an MFT lens at f2.8, because it has to project the image onto a four times bigger surface.

This is why, for example, a f2.8 full frame lens can be speedboosted to an f2.0 APS-C lens - essentially, by taking the superior light gathering quality of the FF lens and focusing all the light/image it gathers, in the manner of a burning glass, onto the smaller image circle of an APS-C sensor. With a conventional lens adapter, you would lose 50% of the light the FF lens gathers since this light will hit areas outside the APS-C sensor's smaller surface and hence not be captured.

This mathematics also explains why, in the old times, almost all Super 8 cameras, with their mini image size, had f1.8 superzooms while it was impossible to build such lenses for 35mm stills cameras. Indeed, those lenses only gathered about 6.25% of the light gathered by an f1.8 lens for a 35mm photo camera.

In practical terms, the superior light gathering of larger sensors can be used in two ways:

1. increasing pixel resolution (the recipe for cameras like the Sony A7R, Nikon D800/D810, Canon 5DS): compared to an MFT sensor, you can have 4 times as many pixels on the sensor yet still with the same signal/noise ratio (=dynamic range/noise level/image quality) per pixel. So, taking the GH4 with its 16MP MFT sensor as an example, you could theoretically cut a 64MP FF sensor from its die and have the same per-pixel photo quality, but four times the image resolution, and effectively four times the light on each picture thanks to the superior light gathering of both FF lens and FF sensor.
2. increasing pixel size (the recipe for Cameras like the Sony A7s and Nikon D4); when a FF sensor as the same amount of pixels as an MFT sensor, and if the FF lens has the same aperture as its MFT equivalent, each pixel will gather 4 times as much light, resulting in better dynamic range and ISO sensitivity.

Tony Northrop is right suggesting that camera manufacturers play an unfair marketing game when they advertise cameras with FF-equivalent focal lengths but native f-stops. For example, the lens of the Sony RX10 does not have - as the product advertising page says - an "expansive 24–200 mm zoom range" with "F2.8 aperture over the entire zoom range", but really is a 8,8-73,3mm/f2.8 lens for a 1" sensor which, in absolute terms - but also in terms of visual depth of field - gathers as much light as an f8 lens for a full frame sensor.

 

 

 

[Nikkor]

This is a thread about the 50-150 1.8, and some people compare it to fullframe 70-200 2.8 lenses. So the whole discussion, if relevant, is about 1.8 crop vs 2.8 fullframe (or boosted).

[TheRenaissanceMan]

F/stop is a ratio between focal length and pupil size. It has nothing to do with image circle size.

[Timotheus]

A lot of Sigma announcements today, in addition to the 50-100mm F1.8:

• EF to E-mount adapter, promising fast AF and good price ($249). We'll see about AF speed, but surely some competition for Metabones.
• 30mm F1.4 for E-mount and MFT
• A new (ugly) mirrorless camera (two versions), unfortunately no video capabilities it seems

[Nikkor]

1 hour ago, Timotheus said:

A lot of Sigma announcements today, in addition to the 50-100mm F1.8:

• EF to E-mount adapter, promising fast AF and good price ($249). We'll see about AF speed, but surely some competition for Metabones.
• 30mm F1.4 for E-mount and MFT
• A new (ugly) mirrorless camera (two versions), unfortunately no video capabilities it seems

I think the mirrorless is totally awesome (for stills obviously) and the estetics gave me a gear-kind of hardon.

[AndrewM]

On 2/20/2016 at 5:39 PM, The Chris said:

Dude, you're mixing light gathering and bokeh with your long winded wikipedia cut and paste.

I just put my 16-35 on the a5100 and the A7rII, set the aperture to F/4 and left everything else to the camera - they both picked the exact same exposure - 1/60, ISO 3200. F/4 is F/4 no matter what the sensor size is, and a F/1.8 lens is faster than a F/2.8 on a focal reducer since you only gain one stop.

Why is this so difficult for you to understand?

I hate to be the one to do this, because there is more than enough contention on this thread already, but...

The 16-35 is an APS-C lens. So when you put it on your A7rII, it automatically went into crop mode. So of course the pictures look the same...

 

I think one of the things that is confusing things here is ISO. Aperture and shutter speed are real physical measures, but ISO - not so much. Or rather, it is a calibrated/normed number. When you set your ISO to 3200 for a particular f stop and shutter speed, you are setting the amount of gain that is being applied to the image, and if the camera maker has done their job properly, you should end up with a roughly similarly-exposed (same perceived brightness) resulting image as you would with any camera with the same settings. So of course different camera/lens combinations should come out the same if set the same - ISO is calibrated so they should.

BUT (and this is the big thing) the same ISO can mean a completely different things in terms of processing and amplification. So if all other things are equal (same sensor technology, same processing sophistication), ISO 3200 on a small sensor is going to require more boosting to get the same output compared to a larger sensor, which means potentially more noise.

[Phil A]

29 minutes ago, AndrewM said:

I hate to be the one to do this, because there is more than enough contention on this thread already, but...

The 16-35 is an APS-C lens. So when you put it on your A7rII, it automatically went into crop mode. So of course the pictures look the same...

 

No it's not... I'm not sure which 16-35mm he means but the Sony FE 16-35mm f/4, Sony ZA 16-35mm f/2.8, the Canon 16-35 2.8 L, Canon 16-35mm f/4, etc. are all Fullframe lenses. I don't think there exist any 16-35mm APS-C lenses.

[WyattMaurer]

http://nofilmschool.com/2016/02/sigma-50-100mm-f18-30mm-f14-mc-11-e-mount-lens-adapter

Looks like it will be $1100 which seems very reasonable.  The leak had it at 167400 yen (aprox $1500) I'm glad the western market price is lower. I can't see this being big in the stills community because of the narrow zoom range, no IS and crop sensor size, but I think this will be huge in the film/videography set as a complement to the 18-35 as other people have said.

[AndrewM]

2 hours ago, Phil A said:

No it's not... I'm not sure which 16-35mm he means but the Sony FE 16-35mm f/4, Sony ZA 16-35mm f/2.8, the Canon 16-35 2.8 L, Canon 16-35mm f/4, etc. are all Fullframe lenses. I don't think there exist any 16-35mm APS-C lenses.

Sigma 18-35, the one we have been talking about, is APS-C.  I made the mistake, didn't notice 16-35. My apologies!