this question and your replies made me think about snow pictures... I know that the camera will be fooled into thinking it’s much lighter than it is because of the snow so one needs to compensate for that. I thought that is best done by using the exposure compensation dial? You suggest it’s done by prolonging the shutter speed?
Its the same thing.... its a question of 'Metering' and the way you translate that meter-reading to camera settings.
The SI unit of light intensity is the Lumin. If you ever chance across a scientific or industrial light-meter, that's how it will be 'calibrated', with a scale marked in Lux... and a bit more peculiar in that it's a normally logarithmic scale... and almost utterly useless to help any-one trying to take a photo.
If you are trying to take a photo, what you want to know, is what 'settings', of Aperture, Shutter-Speed and ISO you need to use to get a picture that's not too dark or too bright..... See
Exposure Exposed because there is actually no such thing as a 'correct' exposure, let alone one that you can get from a super-scientific light-meter 'reading'.. its mostly down to intent and personal preference, and trying to force a set of one-size-fits all rules onto metering, is a bit like telling people how much salt they should have on thier crisps.... some might actually prefer prawn cocktail!
So, if you pick up an old fashioned hand held photographic light meter.... it has a scale marked in EV or 'Exposure Values'.... still not quite what you want to know... but a little more convenient than Lux! I conveniently have my old Leningrad Selenium cell light-meter to hand. Incidental aside; its a clever bit of Russian cold-war era technology, the selenium cell makes electric from the light falling on it. The swing needle is then actually a simple volt-meter, that moves the needle more, the more volts the cell makes, cos the more light that's falling on it... means it don't need batteries! But.... the needle swings across the scale, and rather than being marked in Volts which actually whats making the needle move, or Lux whats making the volts to make the needle move... its Marked in these EV's.... from EV8 to EV14, because thats the 'high' or 'daylight' range. There's actually a switch that changes a resistor in the Wheatstone bridge circuit behind the galvanometer so that it can react to lower light, the 'low' or twi-light range, and actually moving that switch pulls another scale down behind the needle to read from EV1 to EV9... like I said, wonderful Cold-War era Russian technology!
Anyway, beneath the EV scale is a little mechanical Computer! Basically two discs on a common axle, and a few arrows! It was designed for 'film' and we tended to be pretty much stuck with whatever ASA film speed or 'ISO' sensitivity that was, so the first thing I do is set the 'ISO' setting in a window of one of the discs. Then I line up the EV I read off the scale, on the outer-disc with the arrow... and that brings shutter-speeds into line with corresponding Apertures... and I get perhaps 8-pairs of aperture and shutter-speed settings I can choose from that all offer the same 'exposure' for that EV at that film-speed... sounds more complicated than it is... BUT... significant thing is that I get a RANGE of settings I might use, not just one.
More... I can take a light meter reading in one of two ways. I can go stand in the same light as my subject, and take an 'incident' reading of the light actually falling '
on' the subject.... or I can stand back, where I am holding the camera, point the meter at my subject and take a 'reflected' reading of the light bouncing '
off' the subject.
Oft said that an incident reading is more 'accurate' than a reflected one, but pedant in me says no, the accuracy is how close to target the result is... both can be as accurate or inaccurate as they may be... what they mean is that an incident reading 'tends' to be more reliable... and give you a reading that more often results in a good or pleasant exposure... but as said, see the tutorial, there's no such thing as a 'correct' exposure.....
And in this instance, with a predominantly white snowie scene, an 'incident' reading is probably more 'appropriate' than a reflected one, because all that snow will reflect a lot of the light falling on it.... and as has been mentioned, most reflected metering schemes have an inbuilt flaw in the presumption that the scene will reflect about 18% of the light that falls on it..... so if you take a reflected metering of a black-cat in a coal hole, its going to give a lot lower reading than if you took a reflected reading of a white rabbit on a ski-slope, BUT, assuming that 18% of the light falling is reflected... when you translate the EV to 'settings'... you would tend to make settings that under-expose or make dim, the white rabbit on a ski-slope, or over-expose, make bright, the black-cat in a coal-hole.... if you used an incident reading.... you measure the light falling on the subject, not reflected off it, so you should get the same 'reading' for both scenes, and not over-expose your cat or under-expose your rabbit.... whether you could see either in the resultant photo, and if you could, whether you 'LIKED' that exposure, is an altogether different matter.... like I keep stressing... there's no such thing a a 'correct' exposure!
BUT this brings us to the matter of exposure compensation.... which is essentially NOT taking the numbers as gospel, and applying a bit of judgement WHEN you turn an Exposure Value into 'settings' or in picking the settings you actually use.
In the case study presented, the 'essence' is that the predominantly bright white scene has come out 'under exposed' and a bit grey. Hence to brighten it up a bit, adding a bit of compensation.. and letting in a little more light than the meter suggests... which you could do by either manually setting a longer shutter speed, OR by telling the camera to do that 'automatically' using an exposure-compensation dial.
NOW! Using Exposure-Compensation, on a dial, on the camera.... what that does, is tell the camera to pick settings to use, that will make the final photo brighter or dimmer than it 'thinks' it aught to be, just from the meter-reading it got.
Doing it manually, you pick an alternative shutter speed or aperture or ISO, to what the camera 'suggests' to make the photo brighter or dimmer.... and here we have to consider how the camera's meter is 'coupled' to the settings..... because if you want to brighten up the picture, you could just select a lower shutter speed.... but the cameras electrickey could just as easily over ride YOU, and as you add say one stop of shutter, reduce one stop of aperture or one stop of ISO, to keep the exposure what IT thinks it should be, going by the meter!
So its all ways to skin that black cat in the coal-hole.... and the important thing is to note that there are two ways to meter a scene, reflected and incident, which may be more of less expropriate or convenient in different circumstances. Having got your meter reading... you then have to translate that it to more or less appropriate 'settings' by way of aperture, shutter and ISO.... and you can pick many many combinations of the three that ALL result in the same exposure.... BUT is that 'exposure' the one most pleasing to you?
If you see tutorial, you can get an exposure that makes an 'acceptable' picture, across a range of perhaps six stops of Exposure Value... in fact, in the tut, I think that I showed more or less 'acceptable' photo's six stops either side, so twelve stops range... any of which might be more to your preference.... BUT the camera only takes ONE meter reading.... and assumes you want an exposure slap in the middle of that range....
Interesting 'point;' here, is that the OP 'prefers' the exposure that his smart-phone chose for him, to the one that his Digital SLR did...... neither are 'wrong'... both are photo's, BUT one is more pleasant than the other.... in one opinion....
There's also some small revelation in the OP's comment about the fact that it was too cold, he didn't feel inclined to faff about to much, and 'just' left most of the job and decision making to the machine. And there in lies the 'nugget'. The camera-phone picked 'settings' the OP 'liked better' pretty much by happy-chance. The DSLR didn't, similarly, mostly by chance....
If you leave it to chance? You get what the camera gives you. The 'lesson' here is that to get what you want... you need to know what the camera is doing, and how its going to give you what you want..... or not.. and if not... what to do to MAKE the camera give you what you hope for.... which means YOU cant just leave it to chance, you have to do some work, which probably means faff, and putting up with the cold, or inconvenience... and its probably THAT 'diligence' or often shear bludy mindedness! that most often separates the OK, from the pretty good, from the excellent, as much as applying the small extra know-how, and faffability of not just leaving it to chance and the programming mushrooms that made the decissions you cant be bothered with.