1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

something wrong with my camera d3300

Discussion in 'Help Team' started by steve dude, May 11, 2018.

  1. PeteRob

    PeteRob Well-Known Member

    Spot metering has nothing to do with AF. You need to use a single focussing point and lock focus on the swan's head.
     
    EightBitTony likes this.
  2. EightBitTony

    EightBitTony Well-Known Member

    Centre-weighted isn't a kind of focus, it's a kind of metering. That was my point. Also, I've never heard single point AF be called spot focus, but I have heard of spot metering. Anyway, moving along.
     
  3. EightBitTony

    EightBitTony Well-Known Member

    Since we can't convince you the camera is probably fine you better return it.
     
  4. steve dude

    steve dude Active Member

    .... i should have said "single point focus" instead of spot meter, sorry, i used both and was thinking about the previous message ...
    you can choose the focus point, i kept the center one.

    the second image is completely out!
    The swan is soft.
     
  5. steve dude

    steve dude Active Member

    i did give the iso for each shot .... 400 and 100

    the lens definitely has image stabilisation (VR) and it obviously kicks in when you hold the shutter down half way. the image becomes still. This is well within the parameters
    of vr with focal length 240mm using 1/160 the second.

    i can assure you that the jpeg compression is not responsible. The images are still very similar at 50% compression. The images are clearly not sharp .... and that includes the raw files for each jpeg.
     
    Last edited: May 14, 2018
  6. Craig20264

    Craig20264 Well-Known Member

    I would be looking at a minimum of 1/500 sec at that focal length, but I don't think we're convincing you.
     
  7. steve dude

    steve dude Active Member

    no i am afraid not guys .... i am convinced that there is something seriously amiss.

    i think it could be 3 things

    a the mirror slap is creating shake. Perhaps the previous owner dropped the camera or something and dislodged the sensor or made it loose and prone to vibration.

    ...... it would be great to test this with mirror lock up and live view ... BUT the d3300 only uses mirror lock up for cleaning.
    ..... worse, although live view does lift the mirror, it is only right up until you take the shot ..... and then it drops the mirror and goes into normal mirror shoot mode.
    .... it also seems that timer delay does not lift the mirror and then delay and take the shot, but instead just delays ... and then normal mirror shoot mode.
    ...... and i suspect that if i set it to take 2 shots on timer delay .... that it will do the same as live view and take each shot in normal mirror shoot mode.

    apparently this is to do with auto focus operating between shots. SO ... it might be possible if i disengage auto focus and use 2 shot timer delay, Second shot no mirror?
    But i doubt it. It looks like the mirror always operates for every shot on the d3300. (else the shots per second could be made faster)

    b the VR is awry. Apparently image stabilisation can create odd effects if faulty. But i could at least test this by turning it off and using a tripod.

    c the auto focus is faulty. Again i could test this by turning it off
    .... BUT it might not isolate the problem. Auto focus is complex and involves the camera body, firmware as well as the lens.
    .... if the d3300 is designed such that focusing is the reason why the mirror cannot be bypassed, then it potentially could be a problem indistinguishable from mirror slap. Hopefully this is not the case, I would expect to be able to isolate auto focusing problems simply by switching to manual focus and use live view to zoom in on the display and manually focus.

    I will try to figure out what the problem is. But my initial gut reaction has returned and tells me the camera is a dud.
     
    Last edited: May 14, 2018
  8. EightBitTony

    EightBitTony Well-Known Member

    Then return it, and stop investigating.
     
  9. PeteRob

    PeteRob Well-Known Member

    I cannot see your pictures properly on my ipad so it is difficult to comment precisely. But there are a couple of things.to ask given the confusion over nomenclature, also not helped because Nikon and Canon (which I use) have different names for everything. I'll tell you how I set up my camera for those shots. If you set up the same then either it is technique or the lens is out of adjustment or you are expecting too much of it.

    I came from film to digital and I treat my digital camera like my film one. The default set up is to use aperture priority and set the iso to give a hand-holdable exposure time at F8. What I consider hand-holdable depends on the lens and its focal length. For my oldest Canon lens - a 24-105 F4 - the image stabilisation is an early version and I'd only trust it one stop at 105 mm, so that is a maximum 1/60s at 105 mm. I have a full frame camera, if it were a crop sensor, that would be 1/90s to take account that the results are effectively enlarged when viewed. I generally use what I think Nikon calls multi-segment metering. It makes a reasonable go at getting exposure OK. I'll use exposure compensation for cases where it doesn't or, alternately, exposure lock. This is where you meter off something of medium tone and lock the setting before recomposing the shot. I have the camera on single-shot mode. This enables me to lock the focus at the distance I want using a half-press of the shutter and recompose. A habit I trained myself out of is "snatching" where, especially with telephotos, there is a tendency to lower the camera as the shot is taken rather than wait until the mirror returns. I nearly always use the centre focussing point. The exception is when photographing an event when I may place the focussing point at head height to avoid cropping feet off. As an alternative to focus and recompose I often disable the focus function of the shutter release. My cameras have an "AF on" button so that focussing can be done independently - it is like using manual focus but letting the camera do the focussing for you. Most cameras allow this customisation but don't forget it has been done as, next day, the AF will appear broken.

    Other settings - I have the camera on auto white balance and standard development mode. I save the raw files and will make any adjustments needed in post-processing.

    For posting images to this site, and to the internet in general. I use the forum guidelines which is to resize the image to 800 pixels on the longest side and constrain the file size to be no more than 200 kb. This produces a result that displays correctly in the browser. As you posted them it is the browser that resizes them to appear in the window. On my ipad (which is an old one) I get repeated failures to display images of that size and they don't appear in the thread but on a different tab.

    I don't save jpgs but it is best to use minimum compression. Usually this is option large or similar on the camera.
     
    steve dude likes this.
  10. EightBitTony

    EightBitTony Well-Known Member

    So, the only way to know for certain. Do some indoor (i.e. controlled) tests.

    Put the camera on a solid surface, put it into aperture priority mode, choose f/7.1 or something close, switch to timer mode (2 or 10 seconds), focus on something static (and preferably well lit) in the room using the viewfinder, once focused, switch to manual focus (so the camera doesn't try and focus again), then take a shot.

    If the result is in focus, there's nothing wrong with the camera or lens.

    I usually use news print or posters / books with words on them for this kind of test. I put the camera about 2 meters from the target, and use shortest, middle and longest focal lengths to take 3 shots.

    The key is to _make sure_ the focus is correct each time, and to ensure the camera does not move.

    *You may find turning off VR helps with these tests, because it can't always tell when the camera is already on a steady surface (or tripod).
     
    steve dude likes this.
  11. Craig20264

    Craig20264 Well-Known Member

    My money is still on camera shake ;)
     
    steve dude likes this.
  12. steve dude

    steve dude Active Member

    thanks pete rob

    i have just got back from testing the camera out and i think i have found the problem.... me (sort of) lol
    1 ......................... AUTO FOCUS
    2 ...........................'critical focusing required generally'
    3 ........................... too high expectations for improvement.

    I am not entirely convinced i have ironed out what is happening but i have just done 3 test shots (3 different subject distances (3m, 5m and 12m)., each with 3 focal lengths (wide, mid, long)

    for each of those i set the delayed timer to take 2 shots (for potential reasons given above ... ie camera shake and/or mirror slap)
    all shots taken from a tripod

    then first pair ................. auto focus and VR on
    second pair ....................auto focus and VR off (at the lens)
    third pair .........................man focus using optical viewfinder zoom in and then recompose, VR off,
    fourth pair........................man focus using live viewfinder zoom in, VR off.

    with regards the comparison between first and second shots from the timer, i couldnt see any difference (**except in a couple of cases - later)

    with regards VR on or off, (nb camera on a tripod), there was sometimes 'blur/softness' at the telephoto setting. So VR is best off on a tripod.
    (this kinda makes sense, particularly for telephoto, since scene change may be interpreted as camera shake, and attempted VR compensation kicks in
    to actually blur the image)


    with regards auto focus, there was inconsistency in focus point. I used single point focus, and i noticed that the camera seemed to focus, then a few seconds later
    sort of maybe changed its mind. I have since realised that AF-A setting does not always lock the focus. (Choose af-s for that.)
    So AF-A is not a reliable auto focus setting. I have read that if anything changes in the scene the AF-A focus point might change (or not!.) I think also the
    occasional difference in sharpness between 1st and 2nd shot, was because the scene changed (windy leaves) and this changed the initial focus point in AF-A. It
    only happened when AF-A was engaged.

    Manual focus (all done without VR engaged) was the most reliable. BUT whether done with the optical viewfinder and zoom, then recompose OR live view zoom
    it is very difficult to make focusing precisely accurate. The smallest change will radically alter the focus of the scene, particularly at telephoto, but still significant even at mid (70mm)

    The easiest way to focus was to use the optical viewfinder, BUT i think the recomposing can sometimes change the focus point. (Maybe in zooming back out i touched the focus ring?)
    The live viewfinder method has the advantage that it doesnt touch the lens, but it is difficult to be as accurate, especially in sunlight.

    SO! ... lol ... it seems this camera and lens are not faulty.

    What i have discovered is that in moving to a larger sensor, focus is critical ... and difficult!
    The sx260 with its small sensor (larger dof) is much easier.

    I also compared the shots with the canon sx260. I was surprised how subtle the difference was between the 2 cameras..
    Yes, the images IF FOCUSED ACCURATELY, from the nikon were a bit 'better'. Slightly sharper, and smoother too. But the latter was due largely to the areas of the image
    that were out of focus. Which is much more so than with the sx260 of course! BUT at the telephoto end, there wasnt always much difference between them. I take this to show that the all in one 18-300mm lens, is better at the wide and mid range focal lengths. There the difference compared to the sx260 was most noticeable.

    So another aspect is that i was expecting too much image quality improvement from the larger sensor size and larger lens.

    I hope that all this hasnt exasperated you all too much. I think if i had gone from full frame film to aps-c digital, then i would have loved the difference immediately.
    But having spent years with a compact ..... going back to a dslr is challenging, and the improvement more subtle. Particularly with this lens.
    ie if i used prime lenses i guess the image quality would go up again (though with the huge loss of 24x zoom, servo convenience :)

    ..... does that sound about right?

    the previous swan shot was soft because focus is critical, subject scene changing (water ripples and swan movement) and this changed the AF-A focus point.

    the street scene may have been blurred because ironically my steady position threw the VR processing, and it engaged? ie it was equivalent to using a tripod with VR with a scene change like the car moving?

    thank you all for your comments, i tried not to get peed off.
     
    Last edited: May 14, 2018
  13. Chester AP

    Chester AP Well-Known Member

    I have read these entries again, and have had another thought. Has the previous owner been trying to adjust the autofocus settings?
    I'm not familiar with Nikon DSLRs, but expect somewhere in the menus there will be a 'reset all to factory defaults' (or similar). Once you have done this, try going outside on a bright day and see if the focus works any better.

    There will also be menu adjustments for noise reduction and sharpening - the swan image may be suffering from the wrong settings for these. It looks like a 2,000+ ASA/ISO image from my 16 megapixel Pentax K5 that has a previous generation Sony sensor to the one in your camera body.

    When I part-exchanged a Pentax K10 at a branch of London Camera Exchange I warned them that I had changed the control dial settings, and was told that they always reset everything to the original factory defaults when reselling camera bodies. This ensured that the new owner would be able to refer to the camera manual and find everything as expected.

    As a last resort, if you got the camera body and lens from a shop, ask them if you can try another lens on the camera body to see if you get correctly focused images. If yes, then your lens may be faulty. If still out of focus, ask to try the lens on another body in case your lens is fine but your camera body is faulty. A decent camera shop should humour you for a few minutes for this (as long as you can avoid a busy Saturday) because they will know that if the can resolve your problem they will get your custom in future.

    If the body and lens both appear to be in very good condition, my money is on the camera body's settings which is why I suggested resetting these to the original factory defaults.
     
    steve dude likes this.
  14. steve dude

    steve dude Active Member

    yes thanks chester, the factory reset could be a good idea .... along with avoiding AF-A! ... and accepting focus is critical and is a new skill :)
     
  15. PeteRob

    PeteRob Well-Known Member

    I'm glad that you are sorting it out!

    I don't know the Nikon jargon but indeed if you want to lock focus then there will be only one setting that allows that, Canon call it "one shot" and the shutter won't release unless focus is confirmed. I presume this is AF-S with S for single. I'd guess AF-A is a track focus mode aimed to improve hit rate for objects that are moving.

    Generally zooming will change the focus - especially with a super-zoom. I hadn't realised it was an 18-300. These all-in-one lenses are a compromise and the longest end is intended for occasional use and probably needs the lens stopped down to F8-F11 to give its best performance.

    If there is a difference live-view to ordinary AF then that indicates the lens calibration is a little bit off. A while back the manufacturers introduced for some cameras a feature called micro-focus adjust. I don't know if your camera has it. The manual will probably say "don't touch". Many people did, there were hundreds of articles about optimising the camera/lens combination, but truly it is a quick way to spoil the focussing. This is the control that Chester is talking sbout in post #53.

    Compared to a film camera manually focussing with the viewfinder is more difficult. The mirror is much smaller so the image is dimmer, there are no focussing aids, the camera most likely doesn't use a prism which again gives a dimmer view.
     
    steve dude likes this.
  16. steve dude

    steve dude Active Member

    thanks pete rob. ,,,, i am mightily relieved that the kit is ok.

    there are several af modes, including af-c for continuous tracking. The live view focus i used as manual focus to compare with the optical viewfinder :)
    i have yet to explore focus locking. I think there is the equivalent of micro focus too .... but the correct af mode must be used else it just overrides it.

    yes f8 to f11 often for long focal length .... which will often need either high iso or a tripod. But i am into landscapes rather than wildlife and sport. (i now have so much admiration for that kind of photographer with the super critical demands of camera shake AND fine focus on full frame sensors!) swans are difficult enough lol

    cheers :)
     
  17. steve dude

    steve dude Active Member

    i have just checked and my sons girlfriend has a nikon d90 ... which accepts dx lenses. So i can test it out on her camera and also check her lenses out on mine i guess :) I see from one of her images that she has a lens capable of 300mm, so that will be an interesting comparison.
     
  18. Chester AP

    Chester AP Well-Known Member

    No more posts for the last nine days - have you solved your problems yet?
     
  19. steve dude

    steve dude Active Member


    hi chester

    well sort of maybe ....

    first off i now have a load of dust bunnies on the sensor! I dont know how. so i am buying some kit to clean it. apparently nikon have a reputation for over oiling their shutters and newish cameras can get it on the sensor. or it may be dust .... perhaps sucked in through the lens. it is not weather proof and apparently creates suction when extending. I have been very careful. It happened after a very hot day in the woods.

    but i do think there is potentially still a problem with the auto focus. i have found a youtube on how to check this. it can be fixed by carefully using a allen key to adjust it through a nut in front of the sensor.

    i was comparing to ken rockwells site and he was reviewing and extensively using the same kit. He reported taking pin sharp images even at 300mm as low as 1/8th of a second! I didnt try anything like that.

    BUT i have also found reports of the af system not working consistently on a tripod. The blur looked similar to some of my results. So ironically if i hold the camera very steady on af it can create blur.

    so it looks like i have to turn off af and also use manual focus at the moment, with a tripod and smaller apertures. Even so i find that i still cannot get consistently sharp images. Sometimes they are and sometimes not. i have tried live view digital zoom focus and also manual zoom to 300mm focus and then pull back the lens for the shot.

    having said all that there is of course the much narrower depth of field anyway. Something isnt quite right yet, but i have to get the dust bunnies out first.
     
  20. PeteRob

    PeteRob Well-Known Member

    After this history it sounds to me like it needs a service.

    The focus adjustments must be measured in micrometers - allen key adjustments don't sound very likely to me. Oil on the sensor shouldn't happen. I think there was one Nikon body subject to a limited recall on this a few years ago but I don't think it is a general Nikon problem. Ordinary dust should be able to be removed with a few puffs of clean air from a hand operated bulb blower. Don't use compressed air from a tin.

    Sensor dust gets very visible using apertures of F16 and smaller. If you have been using F22 - F32 then images will tend to look softer than at F8-F11. Holding 300 mm consistently at 1/8 s isn't very credible on a crop sensor camera - that is 450 mm full frame equivalent. The very best combination of stabilisation gives 5 stops which would be 1/30 for most people. Most early lens-only systems gave 2 stops, canon latest mk ii L lenses give 4 stops. The in-body 5 stop systems in some mirrorless cameras give this for short focal lengths not telephoto - the amount of physical correction needed is too great. The early stabilisation systems needed to be turned off for tripod use. The later ones not. You shouldn't need to turn AF off for tripod use. The only circumstance I can imagine it having an effect is using a self-timer with the camera not prefocused. Maybe it will move the lens to its choice of focus as the shutter releases.
     
    Craig20264 likes this.

Share This Page