It's both, actually. But that said, Nokia uses very good glass.
It's only once you get up to P&Ss and DSLRs that the glass can really hold back the sensors -- phone sensors are always struggling to make a compromise between package size and image quality, and usually package size wins.
>> The problem with phone cameras (mostly) ins't the sensor, it's the glass, or lack there-of.
Is that a problem with the camera or the user's expectations of what that camera can do?
You can only push the laws of physics (with respect to optics, especially) so far, and if you want a camera that can fit in the pocket of a pair of skinny jeans, compromises must be made.
There's still a lot of room to push the laws of physics, although I'm not sure glass is the answer. I think manufacturing tolerances will be the biggest limit for inexpensive handsets.
If I play (hah!) stupid, I look at the flange distance and lens size for an SLR of any sort and compare it to a mirrorless camera such as the M9 or even NEX-7, I can get much higher quality on the same or smaller sensor with smaller lenses, especially in the normal field of view. This is purely based on that big empty space in the SLR. The main thing we'll have to give up is an expectation of a shallow depth of field, but "f/8 and be there" is why camera phones are useful anyway.
I've been impressed with the sapphire lens on the iPhone 5, all things considered. It's not what I would call a good camera, but perhaps coupled with a high-density sensor with an alternative pixel structure[1] and microlenses, it would not be unreasonable to expect 2013 "full frame" image quality in a mobile device in 2-5 years.
I don't want to burn my fingers trying to capture 4k raw video on my phone. Heat dissipation will become an issue, as it is on HDSLRs and MILCs.
>> My point is this is the _worst_ kind of spec padding.
How so? This is actually the first officially announced smartphone (I also have a keen interest in the upcoming Sony i1, but it's not a reality yet) that I'd consider buying just for the camera.
Personally, I think Nokia has taken an interesting approach with the sensor, because the added resolution gives them leeway in overcoming the optics problem (from a focal length perspective, at least) in that the "digital zoom" will produce better results than any other phone.
The upcoming Sony i1 should please many people. It's rumored to have a 1/2.3" sensor - while smaller than the Nokia's sensor, it's still quite large for a phone.
It's also worth noting that this and the 38 MP Nokia 808 phone have a 1/1.2" sensor, larger than all but one fixed lens compact camera, including cellphones. The only compact with a larger sensor is the Sony RX100/RX100 II. Not quite DSLR quality, but higher quality than you'll get out of most compacts.
The high MP counts allows you do downsample/pixel bin for higher quality images and also allows room to crop, providing a digital zoom which doesn't suck.
If you are a big DxOMark fan, you'll find the RX100mk2 scores a 67, while the RX100 scores a 66, which is the same score as a Canon 7D's sensor.
And while you may argue that the 7D's sensor is old, Canon's been pumping variations of the 7D sensor into all of their recent APS cameras short of the freshly announced 70D.
First, my statement had to do with the 1/1.2" sensor that Nokia is using. Close in size to the RX100 and Nikon 1, but we can't assume that means the same level of image quality- there's a pretty big gap between the RX100 and Nikon 1 IQ scores, for instance.
To your point, I agree. I've seen comparable (and sometimes higher) IQ from my RX100 than I do from various MFT and APS-C DSLR/MILC cams with their kit lenses including the Canon T3i and T4i, GX1, a few NEXes, and Nikon D3000-D3200. The RX100 fares well against those cameras at the wide end of the zoom range when it comes to low light performance and thin DOF as well. Of course, you could spend more and get better lenses for those cameras and beat the RX100 in low light.
I do hope that the new sensor in the 70D provides an actual improvement in ISO score if not IQ and isn't just the older sensor with OSPDAF added.
I didn't mean to sound like I was rebutting your original point about the 1/1.2" sensor -- Sorry about that.
I was just adding to the point about 1" sensors actually being as good as an SLR sensor, which they are.
Of course the RX100's optics limit some of its performance, but Sony's 1" sensor and the M43 sensor they sell to Olympus punch well above their weight, equalling or besting some existing APS sensors on the market today.
What a lot of people forget is that doubling sensor size gets you only one stop in high ISO performance. For raw quality in decent light, 1" seems to be a sweet spot.
It's been a while since I've read the paper, but technical aspects like runtime size of binaries led to a bytecode compiled language like Smalltalk, leading to NewtonScript.
They're used extensively in a lot of industry- just because it isn't used for logging onto Gmail doesn't mean it isn't used elsewhere. I work in utilities and client certificates are used for all or nearly all web clients and APIs.
I'm on Windows and Linux, but my reasoning is similar. I also occasionally use the mouse for text selection. On Windows I especially like using GVIM over VIM because of the extremely limited font selection for the Windows terminal.
I think if you're happy with what you're using, stick with it.
I switched from pico/emacs/TextMate/Notepad++ to VIM about three years ago. I always hated vi. I love it now, and really wish I would have switched earlier. But the advantages don't become obvious until you make yourself to live with it for a while. The initial learning curve is pretty steep, but once you get over that hump it's pretty easy going.
This is very different from emacs, which was very easy for me to do simple tasks but very hard for me to do anything complicated. It's also different from pico, TextMate, and Notepad++, which are easy to pick up but plateau, sometimes leaving to write a script to do what you need.
Using a vi-like mode for another editor doesn't really cut it, not unlike an emacs-workalike that isn't built atop a powerful scripting language. For me, it's about the way commands stack and compose rather than using hjkl instead of arrow keys.
Thanks for the good luck! I've definitely been having fun fiddling with MacVim and it's array of plugins in my spare time. The vi-mode is definitely a limiting factor... I find it great for the getting into basics and getting comfortable, especially once you add a block cursor to help switching modes.
Having used MVC for many years, and in the environment where it was invented (Smalltalk-80), I couldn't help but wonder why he was reinventing MVC while thinking he was criticizing it.
I don't see why what the first part (before the hyphen) has to do with the rest. I'd agree with the first bit, but not because anyone who dinks around a bit online at work is "lazy and unproductive."
I've probably overstated myself a bit, whilst failing to explain my opinion well because I was writing from an emotional response.
Dinking around online is lunch-time behaviour. If you really have nothing to do, why not take the initiative and do something productive for your employer instead of fooling around on the net? The occasional 5 minutes probably doesn't hurt, but that doesn't make it right either... an hour or two is totally unacceptable.
I'm probably bitter because I spent a fair amount of time in my younger years working low-pay bottom end jobs with lots of manual labour and "physical" engineering type work. There is no dinking around on the internet, and the amount of slacking that is typical in most office roles would result in prompt warnings and a firing, yet somehow easy-peasy office roles are paid more - even if they are similarly low-skill (data-entry, receptionist, office manager etc).
Even so, my "bad" youthful experiences are nothing compared to how, to a very good approximation, the entire human race earns their living.
Hopefully that can explain why my response was so driven by emotion rather than reason, and hence did not make perfect sense?
(I am a programmer now and I work in an office earning £25k - and yes, I do feel guilty about how much I get paid for such an easy job)
I guess it depends on what work you do. If I worked a 100% of the day, except for lunch break, my brain would be fried after a few days. 5 minutes an hour doing something else actually makes me more productive.
Sure, when I'm in the zone and are lucky enough not to get interrupted I can program for longer periods of time, but I still need a short break now and then.
I don't know where you worked, and since I'm from a different country, the standards may also be different. But where I live, I can't think of a physical job where you are required to work at 100% capacity for a whole day, except for a lunch break. Short rest periods are required.
5 minute breaks are effective for refreshing yourself and I do the very same probably about once an hour but this is a luxury of my workplace and is generally not allowed in manual work - in some situations - e.g. assembly line, it is crucial that no one take breaks for very long - asking to go to the toilet can get you fired (I've had this happen to the guy working right next to me once). I think we've probably tested methods to "perfection" over the last 5k years or so. This is part of why there is an invisible barrier between office and shop floor so often - because office guys get a lot of slack and have easier jobs by many measures. Just being allowed to sit whilst you work can seem like an incredible luxury to some of these guys stuck standing at an assembly line or running around a warehouse filling orders. And I'm sorry, but I'm a programmer - you can train someone to be a programmer, and most office type jobs on the fly just as easily as an assembly line worker - they won't be great but neither is the assembly line worker when he starts out - in both cases it takes years of practice. The difference, to me at least, seems to be artificially imposed by a combination of broken supply/demand models and general misconceptions - programmers are in huge supply, we just pretend they aren't. Most people rely on programming skills in everyday life - they just aren't told it that way.
I live in the UK there are two standard formats for most grunt work - either 2 hrs work 15 mins break 2hrs work 1hr break UNPAID (i suspect not allowed - i think this depends on a legal loophole against the spirit of the law) 2hrs work 15 mins break 2hrs work - or 8 hours with just 30 minutes paid break to be taken in one consecutive lump. i've worked both - its still nothing compared to what most of the world (i.e. China, India and Africa) deal with. I'm one of the very most privileged...
Whilst you work discipline can be illegally strict because it is accepted and nobody questions things if "80% of places I worked were like this - they can't all be illegal" or "its been this way for years" or "big companies know better than to break the law" etc. I'm one of the fortunates with sufficient balls to stand up for myself - even then I don't do it every time that I probably should, but I've seen so many people (the vast majority) treated illegally because it is accepted by their peers, and they just tolerate it. When you explain how they can protect themselves they don't want to because they need the money and don't want to tick anyone off, job security really matters when people tell you you are unskilled and jobs are hard to find - it doesn't matter how many times I tell them I've never been fired for standing up for myself.
Its one of my personal pet peeves - hence the irrationally strong response to, at best, a vaguely related link.
Am I the only one who didn't see this? I never go straight to http://google.com, but use the search box or vimperator's ":open google search terms." To think, I had to lose productivity in traditional ways...
All the really matters is that you use what you like. If you like gedit or notepad++ that's just fine. FYI, emacs also comes with an extensive tutorial.
Personally, I used emacs and clones (mg and JASSPA MicroEmacs) for a long time- though I never really went very deep into the features. Then, I started a job where I'd be doing some sysadmin work on AIX boxes- nothing but vi and ed installed- so I figured I should sharpen my rusty vi/vim skills. I've found myself using more of the features available in VIM, as well as writing plugins myself. After you get past the initial weirdness (for folks coming from emacs and other modeless editors) I've found it easier. Less of a learning/feature curve, far fewer keystrokes, and almost no "Meta-Shift-Control-Z Ctrl-X abc" acrobatics. UI feels more modular in that I'm stringing together simple commands rather than using a command that does something specific.
Not advocating so much as trying to explain where I came from when I started using vim...
The problem with the "use what you like" approach is that the emacs/vim learning curve are too steep. By the time I am actually using one of them I'll have spent so much that the fear of change starts to strike and I'm stuck with that editor.
