PDA

View Full Version : f/1.8 and flouro light globes



Takoma
27-10-2009, 1:46am
My fiancee bought me a Nikkor AF f/1.8D yesterday, and I'm finding it a bit...odd. first port of call after walking out of the shop yesterday was the dimly lit Fremantle Markets, and I was getting hand-holdable shutter speeds on Aperture Priority, which is to say 1/100 and upwards. Later that day at home, after the sun went down, I was getting 1/250 inside with the residual light from the well-set sun.

Tonight though, in my brightly-lit office, I'm lucky to get 1/4. I've got three of those flouro light globes burning, and it's almost like daylight in here. It's SERIOUSLY bright. But even so, my D40 just doesn't want to play ball.

To put it to the test, I walked out into my almost pitch black backyard just now, and pointed it at the streetlight 100m away over the back fence, and it gave me 1/25. Not exactly hand holdable in my mitts, but a heck of a lot better than I'm doing in my blazing bright office.

The only thing I can think of is maybe it's the globes, or perhaps the colour of them. They're those warm yellowy ones, and images taken in here with my other lenses + tripod all come out in a tone surpassing warm and diving headfirst into HOT. Seriously, shots in here are almost sepia in tone. Is it possible that my camera's getting confused? I've tried all the available metering modes, though in my inexperienced state I still don't fully understand their respective purposes. The shots I've done in here to test the new lens don't seem to be overexposed, so I suppose the metering's working like it should. But why does it take such a slow shutter speed to capture what is otherwise a bright almost sunny office, when it can go out the back into the darkness and perform (by comparison) brilliantly?

For the record:

Nikon D40
Nikkor AF f/1.8D

Thanks in advance :)

Colourised
27-10-2009, 2:59am
i think probably the issue youre having here is with White Balance. i have those same yellowy warm globes in my aptmnt too and whenever i take the shots it does get a bit too warm for my liking. id say to try and change the WB to experiment. hopefully you would get better color reproduction than before!

sometimes when its in Auto it does tend to not be accurate. hopefully tht helps!

I @ M
27-10-2009, 6:41am
Sounds a bit like "auto ISO" being switched on. Inside it decides that there is a heap of light and drops the ISO to 200 but leaves you with a slow shutter speed and smallish aperture. Outside in the darker areas it is probably setting the ISO to 1600 or so and opening the aperture to allow more light in.
Apart from that, inside lighting is very deceptive, what appears to be plenty of light to see by is very often interpreted by the camera as being quite dark.

ricktas
27-10-2009, 7:55am
Agree with Andrew. ISO is the issue I think. From memory, the D40 has two different ISO settings, you need to turn both OFF, to get full manual ISO control - check your user manual.

arthurking83
27-10-2009, 8:40am
post a few test, low res if need be, images with exif data intact.

How are the exposures looking?

Takoma
27-10-2009, 9:36am
ISO is most definitely under my control. I hate ISO noise, and if I can't get a shot at ISO 400 then I can't get the shot :p

The images, as mentioned, look incredibly warm, but otherwise ok. Free of noise, if that's what you're pointing at :)

Will post some images tonight.

ving
27-10-2009, 11:31am
ok just some stoopid questions... you arent closing the aperture are you? what metering are you using? thats about all i can come up with.

if you spot of center meter on something dark you will get a long exposure even if the sun is out

Takoma
27-10-2009, 8:25pm
Yeah, have noticed that with spot and center metering, and it stands to reason, especially with a D40 having only three AF points. I've fooled around with all available metering modes, and ISO increases yield linear results as far as shutter speed goes, so I'm confident it's metering correctly.

I've also asked this question over on Atomic, and there was some talk of flouro globes and the way they illuminate. Apparently (and this is new to me), they alternate full brightness and a 'cooling' cycle at about 60Hz, meaning that with shutter speeds faster than 1/60th you've got a chance of catching them in a cool cycle, which is to say significantly less than maximum brightness. It would be fair to say that my camera's noticing this, and decides that it's really only seeing half to two thirds of the light that we can see with the naked eye, since the eye isn't fast enough to detect this cycling.

Now whether that's true or not, I don't know. But if it is, it would stand to reason that under flouros it WOULD decide to use a slower speed, as opposed to incandescent globes which illuminate on both sides of the power alternation.

I happen to work at Bunnings, and we happen to carry a metric buttload of different light globes, so I'm going to pick some up either later tonight or tomorrow and have a good play around with this. Will let you guys know how I get on.

arthurking83
27-10-2009, 10:18pm
if that were the case, then you'd see this situation if you tried this:

set a manual exposure at high ISO, to keep a constant shutter speed, while 'trying to maintain a reasonable exposure.(hope that makes sense.. it should it's basic photography!)

What you need to do is to just try to get a good exposure under fluoro light at more than 1/60s.. and I think somewhere along the lines of 1/125 or 1/250 should do it OK.
You have to be in a full manual exposure mode tho.. but the aperture is only important in that it has to stay constant too.. the idea is to try to catch the 'cool' or dark side of the fluoro frequency.
I dare say you won't, but not having tried it myself I really have no idea.

The basic principle is that: As the shutter speed is faster than the cycle time, it should catch at least one (substantially)darker frame than the norm, basically along the lines of capturing a standard CRT screen.
if you try to shoot any CRT screen at a speed faster than the refresh rate(say 1/60Hz) the faster shutter speed will catch a 'dark' frame. usually on a CRT screen that means a screen that wil probably have a half dark/half image on it.
Very rare to catch it fully dark(black) unless you are using a very fast shutter speed, and then aperture, ISO, and available light will be the limiting factor anyhow.

Could be a weirdness with the cameras metering system, but I doubt it.

I've shot under fluoro light many times(in my kitchen) and never noticed any anomalies.

Just for reference!
@ 2m from my 60W incandescent globe, with my 50/1.4 @ 1.8 :

If I point the spot meter directly at the globe, I need to stop down the lens to f/3.5-f/5 to get 1/8000s on the shutter.

If I point the spot meter just to the side of the globe, say the distance through the viewfinder to where one of the side spot points is then superimposed onto the globe itself(ie. sideways by approx 1cm) the same exposure then drops the shutter speed down to 1/200s both @ ISO200.

By the sounds of it, from your office/backyard comparo, you've gone from pointing at reflected light(in the office) to pointing directly at the light source.
light is not linear in the way it shines/reflects. That is, using my example, even though that area 1cm from the globe is super bright, it's reflected light for starters, and the power is significantly reduced.
In my dimly lit office, pointing at any light reflecting surface that's bright(off white, not stark white) away from the ceiling(and remember I only have a single 60W globe!) I then get 1/13s @ f/1.8(ISO200) in matrix metering, and 1/25s at the same aperture and ISO.

The difference may be in the metering modes(and the scene you are pointed at!!)
In spot metering the only point that matters to the cameras metering system is the small 2-6mm area directly in the line of sight of the focus point.
In matrix, or center weighted more of the area is then taken into consideration to get a more balanced exposure. if the scene is all white(or the same colour) the required EV will be roughly the same(dependent on lens), but if there are other colours in the scene.. especially black or blue, then the EV required will be less, to expose the scene at the same overall mid tone, and you need a slower shutter speed.... even though the camera may over expose the brighter areas.

So in all the tests are you using the same metering mode?

the metering modes are quite easy to understand:

spot: a very small area directly on the focus point used, is used to calculate an exposure.. you control how much under/over exposure you need(best metering mode to use.. try to learn it)

center weighted: similar to spot metering, but a larger area is used to calculate the required exposure. Usually there's a setting in the camera to alter the size of the center weighted area size.
handy for portraits sometimes.

Matrix: basically what most people prefer to use, but not as simple as the name suggests. You;d think that the camera is going to try to achieve a properly balanced exposure(when it can) without blowing highlights or losing shadows, but it doesn't. A major portion of the scene is directed at the center, or spot area, so that in a scene with roughly the same toning, if you point the focus point at a part of the image that is darker or lighter, the exposure value changes.
eg. a half black half white wall, where you use a side focus point, once on the black half, the next on the white half.. with each different focus point the exposure requirement will change.. by up to about 1EV. the more stark the contrast, the greater the difference will be.

side note too: each different camera will also try to expose differently!
Digital sensors are not film, they each usually have their own requirement for exposing correctly.
What you D40 requires is almost certainly different to what my D70s and D300 require for the same exposure.. using the same lens. Some sensors need under exposure more than others for a better exposure. That's the difference between the D70s and the D300 as I've noticed. But I haven't done any 'formal' tests with notes to refer back too.. only that I know that for each of my lenses the D70s looks better slightly underexposed than the D300 does.. and that it also underexposes by about 1/3 to 1EV anyhow(at the same manual settings).. and the variance varies with each lens.

hope that helps in some way....

Takoma
27-10-2009, 11:22pm
I'm managing to get around it at the moment with some trickery in Lightroom, mostly a small bump up on the Exposure slider, around 75-85 on the Luminance noise reduction, and some white balance stuff. More fiddling than I like to do, but I'll live.

Thanks guys :)

Takoma
29-10-2009, 4:26pm
Just an update, this thread is now entirely irrelevant as I went down the shop this morning and let them talk me into this:

http://i73.photobucket.com/albums/i204/takoma808/Developed-5307.jpg

Geez, it weighs a ton :)