Classic "brick wall" tests of Sigma 30mm f/1.4

http://www.pbase.com/fjp/ts_2005_07_jul_20

This is the second in a series of tests I will be making on the Sigma 30mm f/1.4. They are time-consuming and I have other things to do, so I'll only post no more than one series with the same theme, per day (at most).

Each test is a series of exposures at all eight full f-stops, i.e. from f/1.4 to f/16.

The second series is of the classic "brick wall." The wall is photographed from about ten feet away. I'm not going to repeat myself. Go to the gallery for the details.

--
FJP, Software Engineer

Thanks again for the test images Frank.

I didn't expect to see quite so much barrel distortion to be honest. I also noticed there is quite a bit of vignetting too, once stopped down to f/5.6 things seem to look more even across the frame though.

I wouldn't mind seeing some centre and edge 100% crops. It's quite hard to judge from the down-res images; acceptable sharpness at f/2.8 and maximum sharpness/contrast at f/8?

I have done a vignetting analysis of your shots fjp.



There is some serious vignetting at f/1.4. It clears up pretty well by f/5.6.
Some mathametician will tell me how many stops the L* values represent.
--
Geoff B
http://www.imageryhome.com

I stacked all your shots in PS as layers. They all pin registered (kudos to your technique).

I found two sections of brick in the centre and the corner that were the same luminance at f/8.0. I blurred those sections by 10 pixels and put a 5x5 pixel sampler over the area. Then i just read out the Luminance value as I switched on each layer.
--
Geoff B
http://www.imageryhome.com

I used a stereo auto-focusing/aperture analogue imaging system with built in VR and rapid lens cleaning system, linked to a 100 billion cell bilobal neural-net processor.... Or in other words, my eyes and brain.

One of the consequences of this much vignewtting is that the scene becomes progressively overexposed in the centre. I guess this could be a problem if the camera meter blows out any detail in the centre.

A good light meter such as in the D2X will tend to overcome the problem in many but not all situations.

I wouldn't let this vignetting put me off buying the lens however. For the type of portrait shots I would use it for, I often apply a vignetting action in PS anyway! This lens could save me some post processing!

However, for architecture - foreget it, or use f/8 and f/11; f/16 will not be sharp enough.
--
Geoff B
http://www.imageryhome.com

Lex Talionis said:

I used a stereo auto-focusing/aperture analogue imaging system with
built in VR and rapid lens cleaning system, linked to a 100 billion
cell bilobal neural-net processor.... Or in other words, my eyes
and brain.

Click to expand...

Lex, you've probably got far too many of those cells. My guess is that I've killed at least 50 billion with alcohol.
--
Geoff B
http://www.imageryhome.com

alabaster said:

I have done a vignetting analysis of your shots fjp.



There is some serious vignetting at f/1.4. It clears up pretty well
by f/5.6.
Some mathametician will tell me how many stops the L* values
represent.

Click to expand...

It's beginning to look like a return to B&H. Disappointing.

--
FJP, Software Engineer

alabaster said:

One of the consequences of this much vignewtting is that the scene
becomes progressively overexposed in the centre. I guess this could
be a problem if the camera meter blows out any detail in the centre.

A good light meter such as in the D2X will tend to overcome the
problem in many but not all situations.

I wouldn't let this vignetting put me off buying the lens however.
For the type of portrait shots I would use it for, I often apply a
vignetting action in PS anyway! This lens could save me some post
processing!

However, for architecture - foreget it, or use f/8 and f/11; f/16
will not be sharp enough.

Click to expand...

And if that's the case, what on earth is the point of getting a 1.4? Ths lens is beginning to look like selling hype and not much beyond that. I'll bet you wouldn't get this from either Tamron or Tokina.

alabaster said:

--
Geoff B
http://www.imageryhome.com

Click to expand...

--
FJP, Software Engineer

I kind of agree Frank. I really had my hopes up for this lens, but a prime with such limitations isn’t what I’d consider a sensible purchase.

Another one on my lens lust list is the Sigma 10-20mm, I hope it doesn’t turn out to be a turkey too.

Try going to shoot some street pix at sunset/night and see if it works for you before sending it back to B&H.

--
WW

Lex Talionis said:

I kind of agree Frank. I really had my hopes up for this lens, but
a prime with such limitations isn’t what I’d consider a sensible
purchase.

Another one on my lens lust list is the Sigma 10-20mm, I hope it
doesn’t turn out to be a turkey too.

Click to expand...

My hopes for this lens are going down the toilet. I've said all along this lens would be too difficult to build for the price they're offering it, unless they're pulling a miracle. With all the technology that went into the 30mm, they should have pulled this off. Of course I still haven't seen what pictures look like that you'd actually want to take.

--
FJP, Software Engineer

Wilfred Wong said:

Try going to shoot some street pix at sunset/night and see if it
works for you before sending it back to B&H.

Click to expand...

That's just what I was thinking. Who takes pictures of brick walls anyhow? And the flowers I photographed yesterday didn't look half bad. And the lens does seem to be sharp. Brick walls are tough on a camera. Still haven't done a bokeh test yet, although the flower pictures looked promising. Maybe I'm reacting too hard.

--
FJP, Software Engineer

--

max..i am glad you are going to try it out in real conditions. though this lens isn't what you expected it can be a very usefull tool applied in the right conditions such as your flowers or portrature or low light where all you want is the center sharpness.

I still think people are hoping for this lens to be more than it is. Don't get me wrong. I'm one of the people checking every day for new info on this lens. But the hard fact is, people who are subjective (like myself), who make money from their photography, etc. need more than it's giving us. I think it was a good first effort. Kind of like the Simga 12-24. People were reaching to justify it's limitations when it was the only thing besides the Nikon available and they didn't want to fork out 1K. Now that there are other more valuable options you don't see the same amount of trying to justify the drawbacks. I think when Tamron, Tokina, and Nikon catch up on this lens we'll see better offerings, the prices going down, quality going up, etc. Kudos to Sigma for being the first out of the box. I've waited this long I can wait a little longer and not have to resell something no one will want when there's better out.
Paul

The lens did really poor on the brick wall tests. To get rid of light falloff the lens has to be set to f/5.6. That's 4 stops from f/1.4, not including f/1.4. That is unacceptable. Then there is the distortion issue, although I suspect PTLens will have the variables soon enough to correct for this. Although difficult to see the extent of it, I did notice softness in the corners at the larger aperature values.

I would be curious to see how Nikon' 35mm holds up. Its focal range is similar (52.5mm instead of 45mm FOV after adjusting for the crop factor). Yes, you're losing 1 stop, but it is still a fast lens. Light falloff should be better controlled because the lens was designed for full frame cameras, so the majority of the light falloff should be outside the area of the lens needed by the sensor. The same can be said of any softness issues at the corners or edges. I also think distortion would be better controlled, although once again PTLens could be used to minimize the distortion effects. The only question mark I believe would be CA. I also can't see it being that much slower than the Sigma at focusing, although I could be wrong. Fast lenses tend to focus quicker than slower lenses.

The best thing about Nikon's prime - it's cheaper than the Sigma.

I really appreciate it, as I understand how much time and effort that goes into somehing seemingly simple like shooting these subjects at different apertures. So thank you for taking the time to share them with us.

One request: If you've got the time, I'd love to see some 100% crops of the brick wall from the center and the edge, especially at f/1.4 (the most interesting test, IMO), and maybe at f/5.6 or f/8 (for seeing the lens' optimal sharpness).

I would use this lens mostly for low light street/people photography, and don't mind the vignetting that much. I used to have the Sigma 20mm f/1.8, and it was great for street stuff on my film SLR, and the 30mm 1.4 might serve the same purpose (although more of a 'normal' lens on my DSLR) on my D70. I also agree on the bokeh - it doesn't look bad from the flower shot, so I'm looking forward to more shots from this. However, the sharpness and (micro)contrast would be very important for me, so I'm looking forward to maybe seeing some 100% crops

Thanks,

Thomas.

fjp said:

fjp said:

However, for architecture - foreget it, or use f/8 and f/11; f/16
will not be sharp enough.

Click to expand...

And if that's the case, what on earth is the point of getting a
1.4? This lens is beginning to look like selling hype and not much
beyond that.

Click to expand...

I haven't tested this lens, but reasoning by analogy (Nikkor 50/1.4 vs. 1.8: the faster lens is sharper than the other one until f/4, the slower lens excels beyond f/4), I would be very surprised if the Sigma 30/1.4 were better than, say, Nikkor 17-55/2.8 at 30mm, f/5.6 or f/8.

However, if it is acceptable in low light situations*, it might fill a niche as Nikkor 28/1.4 is considerably more expensive - it costs more than six (!) times as much here.

• e.g. http://forums.dpreview.com/forums/read.asp?forum=1030&message=14299084

jlmap said:

see how Nikon' 35mm holds up. Its focal range is similar
Yes, you're losing 1 stop, but it is still a fast lens.

Click to expand...

I agree. Nikon ain't stupid. If they really saw a market
for a 30/1.4, they'd have one.
Nikon already makes a wonderful 35/2 (sharp and fast
to focus) that many people already own. It covers full-frame
and can be had for $275 new with a 5-year Nikon warranty.
It can be found for used even less.
For those who really need more light and more coverage,
Nikon makes a 28/1.4. It is very expensive. That's the only
complaint I've ever heard about the lens.
Sigma is trying to hit a very specific market (people who
think they need a 'normal' lens to replace their 50mm) and
will probably be fairly successful. Camera stores are going to
push it hard as the new 'normal' lens. I'm afraid people are
going to be so focused on owning a 'normal' lens that they
will overlook the many shortcomings in this lens.
The rest of us, those of us who know better, will buy the
Nikon 35/2 or 28/1.4.

Matt

I guess I was hoping for this lens, against common sense ... like everybody else I suppose. When I first saw its MTF, I thought 'this lens just has to be a dog in the corners' ... Looks like MTFs don't lie.
thx again and best, mark