Sigma 18-35mm 1.8 lens price announced!!!

[pdirestajr]

And it's 799! (USD). That's a pretty sweet price if the lens is as good as the hype. I'm definitely interested now.

 

[verysimplejason]

Really??? Wow! I want an FF equivalent (with constant F1.8 aperture, not F2.8) at sub-1K. Hopefully it's possible.

 

[untitled10]

Really??? Wow! I want an FF equivalent (with constant F1.8 aperture, not F2.8) at sub-1K. Hopefully it's possible.

Most likely not, if this were stretched to full frame coverage it would be 2.8 due to light magnification and such like, so unfortunately this is just a full frame 2.8 zoom with a smaller image circle

 

[Wildfire]

Really??? Wow! I want an FF equivalent (with constant F1.8 aperture, not F2.8) at sub-1K. Hopefully it's possible.

Most likely not, if this were stretched to full frame coverage it would be 2.8 due to light magnification and such like, so unfortunately this is just a full frame 2.8 zoom with a smaller image circle

What? That doesn't make any sense.

This is not a f/2.8 zoom, it's f/1.8, regardless of which body it's mounted to. Someone has already mounted this lens to a 5D2 and the images exposed at f/1.8 like all other f/1.8 lenses (of course, there was heavy vignetting around the edges because of the smaller image circle).

If they kept the design the same and scaled it up it would be a 1.8 zoom for full frame... the problem is that it would most likely be too large and heavy after being scaled up to fit a FF image circle.

 

[BruinBear]

Is it just me or does it look like this lens zooms the wrong (Nikon) way?

 

[risc32]

And it's 799! (USD). That's a pretty sweet price if the lens is as good as the hype. I'm definitely interested now.

looks pretty damn good to me.
http://www.lenstip.com/index.php?test=obiektywu&test_ob=374

 

[rs]

Really??? Wow! I want an FF equivalent (with constant F1.8 aperture, not F2.8) at sub-1K. Hopefully it's possible.

Most likely not, if this were stretched to full frame coverage it would be 2.8 due to light magnification and such like, so unfortunately this is just a full frame 2.8 zoom with a smaller image circle

What? That doesn't make any sense.

This is not a f/2.8 zoom, it's f/1.8, regardless of which body it's mounted to. Someone has already mounted this lens to a 5D2 and the images exposed at f/1.8 like all other f/1.8 lenses (of course, there was heavy vignetting around the edges because of the smaller image circle).

If they kept the design the same and scaled it up it would be a 1.8 zoom for full frame... the problem is that it would most likely be too large and heavy after being scaled up to fit a FF image circle.

To get FF coverage from this design of lens (throughout the zoom range without vignetting or possibly painfully soft corners), you'd need to fit a mythical 1.6x TC to it. That same quantity of light from the small projection of a crop sensor is spread a bit dimmer over a larger area. So with a 1.6x conversion, you multiply focal length and aperture by 1.6, and you get a 29-56/2.88 FF lens.

If the 'same quantity of light from the small projection of a crop sensor is spread a bit dimmer over a larger area' is confusing, imagine a projector with a zoom option - due to compression, a small image is brighter than a large image, yet both have the same quantity of light.

And yes, the FF sensor with a slower lens will need a higher ISO to get the same exposure.

 

[Wildfire]

Really??? Wow! I want an FF equivalent (with constant F1.8 aperture, not F2.8) at sub-1K. Hopefully it's possible.

Most likely not, if this were stretched to full frame coverage it would be 2.8 due to light magnification and such like, so unfortunately this is just a full frame 2.8 zoom with a smaller image circle

What? That doesn't make any sense.

This is not a f/2.8 zoom, it's f/1.8, regardless of which body it's mounted to. Someone has already mounted this lens to a 5D2 and the images exposed at f/1.8 like all other f/1.8 lenses (of course, there was heavy vignetting around the edges because of the smaller image circle).

If they kept the design the same and scaled it up it would be a 1.8 zoom for full frame... the problem is that it would most likely be too large and heavy after being scaled up to fit a FF image circle.

To get FF coverage from this design of lens (throughout the zoom range without vignetting or possibly painfully soft corners), you'd need to fit a mythical 1.6x TC to it. That same quantity of light from the small projection of a crop sensor is spread a bit dimmer over a larger area. So with a 1.6x conversion, you multiply focal length and aperture by 1.6, and you get a 29-56/2.88 FF lens.

If the 'same quantity of light from the small projection of a crop sensor is spread a bit dimmer over a larger area' is confusing, imagine a projector with a zoom option - due to compression, a small image is brighter than a large image, yet both have the same quantity of light.

And yes, the FF sensor with a slower lens will need a higher ISO to get the same exposure.

Well that's not an FF equivalent to a 18-35 f/1.8, that's a 18-35 f/1.8 mounted to a teleconverter.

If you scale the design of the lens up by a factor of 1.6, then you'll have a 28-55 f/1.8 lens for full frame, which is what we want. The question is whether or not such a lens would be reasonable in size, weight, and price.

EDIT: I suppose we could try to calculate the answer to that question by scaling the size, weight, and price up by a factor of 1.6 as well. In which case, our full frame 28-55 f/1.8 zoom lens would be 125mm diameter x 194mm length, and weigh 1296g, and cost $1280. That's larger in diameter, nearly equal in length, and slightly less weight than a Canon 70-200 f2.8 IS II.

I'm guessing that rather than simply upscaling the design and leaving it completely unchanged, Sigma would want to modify this new full frame lens design to be more efficient, increasing the cost but reducing the size and weight. Would the size/weight reduction be significant enough? Would the cost increase be too high? Perhaps Sigma engineers have already answered these questions with prototype lenses.

 

[rs]

Well that's not an FF equivalent to a 18-35 f/1.8, that's a 18-35 f/1.8 mounted to a teleconverter.

If you scale the design of the lens up by a factor of 1.6, then you'll have a 28-55 f/1.8 lens for full frame, which is what we want. The question is whether or not such a lens would be reasonable in size, weight, and price.

I get your point. From a scale point of view, the Sigma 18-35/1.8 with a built in 1.6x TC is a 29-56/2.88, so to go from that to a 29-56/1.8 lens is an aperture increase of 1.6x. Not too far out from the 1.4x difference between an f4 70-200 and an f2.8 70-200. Therefore the differences need to be scaled up by a small factor to get from 1.4x to 1.6x, and apply that to the sigma.

70-200/2.8 IS II vs 70-200/4 IS:
Price: 2x
Weight: 2x
Length: 1.15x
Width: 1.17x

Scale those up from 1.4x to 1.6x, and you end up with a 29-56/1.8 FF lens which is 2.3x the price, 2.3x the weight, 1.31x the length and 1.34x the width of the Sigma. Or in other words, a £1840 1.9kg lens which is 160mm long and 104mm wide. Very heavy for a normal zoom, but similar price to the Canon 24-70 II.

 

[Wildfire]

So I guess we can all agree that a 28-55 f/1.8 with a design similar to Sigma's 18-35 f/1.8 would be big, heavy, and expensive.

However, I still believe a full frame faster-than-2.8 zoom is possible with some additional compromises in aperture and focal range. Maybe with an all new design (external zoom instead?), something like a 28-50 f/2 or 35-85 f/2 lens could be possible. Or even a variable aperture 24-70 f/1.8-2.8 lens.