D5100: Nikon still needs to get a clue...

I just picked up a D7000 recently and am pretty happy with it. Just a couple weeks ago I was shooting a documentary and the director had me use a 5dmkii for a some of the shots (indoor, low/natural light situations.) I am quite familiar with and like the Canons but the d7000 made more sense for me because of a few reasons.
1. I have lots of good Nikon glass. From old manual primes to modern fast zooms I have some really contrasty, sharp glass.
2. I shoot weddings on the side and more than paid for the camera in 1 job.
3. I really like the Nikon image. This is personal taste to a degree... but a lot of the video I've pulled off the Nikon just seems to have killer dynamic range and rich, emotional imagery.

The only big frustration with me has been the limited frame rates. I would be ecstatic if it had 720p60. I'm not ecstatic, but I am still really pleased.

These comments about Canon being better are stupid. Pick the camera that suits YOUR needs best, based on what glass YOU have and what work YOU get that pays YOUR bills.

Compared to the D5100, I'm glad I picked up the d7000. It meters with AI-S lenses, has dual card slots for instant backups, has an alloy shell and feels solid, etc... It's a bummer Nikon seems to have made video more difficult on the D5100, especially since some of the concerns seem to be firmware related that would make manufacturing costs identical... Not that they're the only company that seems to handicap lower end models to the benefit of higher priced options...
 
Efexone said:
heh

you do have a problem with G lenses because by not offering certain functionality in their current crop of camera bodies, nikon has sudo-crippled their video users as well as making it difficult and half-assed to adapt G lenses to other camera bodies. As much as I wish Nikon would get more serious with their offerings in video, I know that at the end of the day, camera bodies come and go and what matters will always be the lenses. This is why I stick AF-D and AI-S glass. G lenses are great but you need to consider what you sacrificing by buying into that technology. Everything from the abandonment of the aperture ring, the long term reliability of the VR mechanism, the adaptability to other lens mounts, should be considered before buying into the very proprietary G nikkors.
Click to expand...

I do not have a problem with G lenses because I never want to mount them on anything which can't control them. I used to have a 550D and never wanted to mount my Nikon lenses on it because of all the issues I hear about with people using lens adaptors. Even if I had all manual lenses, I just don't like those adaptors. My only complaint is that Nikon continues to fail to match Canon in the video department. In every other way I believe Nikon to produce superior cameras, but that's just my personal taste. I was hoping that the D5100 would be a good intermediate camera for smaller jobs while I wait for more serious cameras to be released (The D800, for example).
 
I have my reasons for choosing the D5100 over the D7000- it's smaller, cheaper, in stock, has 1080p 30fps, has a flip-out LCD and accepts Pre-AI lenses. I don't need the D7000's advanced photo abilities because I have a D700.

Per the manual, D5100 allows in video mode manual control over Aperture but not shutter or ISO. However, I believe I found a work-around (annoying but it seems to work) that works with AF lenses.

But the jury is out on AI (manual focus lenses). Anyone have a work-around for Manual Focus lenses?
 
Please see my post in this other thread:

http://www.dvxuser.com/V6/showthread.php?248496-D5100-manual-settings-with-AF-but-not-MF-lenses

The information that was posted to begin this thread appears to be incorrect. In my experience, you CAN manually set ISO even within Live View in the PSAM modes. I have not tested this function with manual lenses. I have three Nikon AF-S lenses.

When I have a chance, I will test shutter priority mode and manual mode. But as per my usage with an AF-S lens, I am able to manually select an aperture (in aperture priority mode), activate Live View, and manually select an ISO, lock and unlock AE-L, and use exposure compensation (if needed).
 
Bikergeek said:
Please see my post in this other thread:

http://www.dvxuser.com/V6/showthread.php?248496-D5100-manual-settings-with-AF-but-not-MF-lenses

The information that was posted to begin this thread appears to be incorrect. In my experience, you CAN manually set ISO even within Live View in the PSAM modes. I have not tested this function with manual lenses. I have three Nikon AF-S lenses.

When I have a chance, I will test shutter priority mode and manual mode. But as per my usage with an AF-S lens, I am able to manually select an aperture (in aperture priority mode), activate Live View, and manually select an ISO, lock and unlock AE-L, and use exposure compensation (if needed).
Click to expand...

I have tested aperture, shutter, and manual modes on the D5100 and have confirmed that the ISO adjustment can be set manually. Here is a simple methodology for using the camera manually for video:

If you switch the camera to manual mode, you have metering and independent manual control of aperture, ISO, and shutter via the info screen. To adjust shutter, rotate the thumb dial. To adjust aperture, press the exposure compensation button and rotate the thumb dial. To adjust ISO, map the ISO adjustment to the function key and press the function key and spin the thumb dial. When you have the settings you want, press AEL and enter Live View. The settings will hold.

I am still experimenting to see which settings can be adjusted with effect in Live View.

More to follow.
 
Bikergeek said:
I have tested aperture, shutter, and manual modes on the D5100 and have confirmed that the ISO adjustment can be set manually. Here is a simple methodology for using the camera manually for video:

If you switch the camera to manual mode, you have metering and independent manual control of aperture, ISO, and shutter via the info screen. To adjust shutter, rotate the thumb dial. To adjust aperture, press the exposure compensation button and rotate the thumb dial. To adjust ISO, map the ISO adjustment to the function key and press the function key and spin the thumb dial. When you have the settings you want, press AEL and enter Live View. The settings will hold.

I am still experimenting to see which settings can be adjusted with effect in Live View.

More to follow.
Click to expand...

The settings don't really stick if you follow this procedure. To convince yourself, try this: after pressing AEL but before entering LV, either cover up the lens with your hand to make it much darker than the surroundings or point it at a light source to make it much brighter. Then enter LV and wait a second or so before removing your hand or removing the light source. In that second, the camera will have reset the exposure to match the very dark or very bright exposure, and only then will it lock the exposure.

So to make the camera uses the settings you want, you have to point at a scene that matches your settings (i.e. where it shows 0) before entering Live View, and that scene must remain constant for at least half a second or so until the camera adjusts, and then you can press AEL.

Unfortunately, not only is this somewhat cumbersome, but it also does not work with non-CPU lenses.
 
Radikal said:
The settings don't really stick if you follow this procedure. To convince yourself, try this: after pressing AEL but before entering LV, either cover up the lens with your hand to make it much darker than the surroundings or point it at a light source to make it much brighter. Then enter LV and wait a second or so before removing your hand or removing the light source. In that second, the camera will have reset the exposure to match the very dark or very bright exposure, and only then will it lock the exposure.

So to make the camera uses the settings you want, you have to point at a scene that matches your settings (i.e. where it shows 0) before entering Live View, and that scene must remain constant for at least half a second or so until the camera adjusts, and then you can press AEL.

Unfortunately, not only is this somewhat cumbersome, but it also does not work with non-CPU lenses.
Click to expand...

Your method isn't appropriate. If I meter the camera for my hand or a lens cap, it is impossible to zero the meter. Therefore, when I lock exposure, enter Live View and remove my hand or the lens cap, I have an overexposed image. How is this result unexpected?

Yes, you should zero the meter for your subject as it is lighted at that moment, lock the exposure, and enter Live View to begin recording. If your light levels and subject matter are consistent, the results will be good. If your subject and lighting are constantly changing, manual settings on any camera won't work.

The only downsides I can see with the Nikon approach are the loss of a meter in Live View and the need to establish exposure before entering Live View.

Regarding lens compatibility, how is this method not compatible with manual lenses? If you are in manual mode and adjust ISO and shutter speed in camera and then tweak the aperture/iris ring on a manual lens, the Nikon meter will allow you to zero your exposure. Without changing the lens aperture, you then lock exposure, enter Live View and begin recording. If you change the lens setting after or during exposure lock, obviously the meter won't be zero.

I don't have a non-CPU lens to test this workflow. Are you saying that the meter doesn't show or work when a manual lens is connected to the camera?
 
Bikergeek, you are overlooking the greater implication of what Radikal has pointed out - he has shown that when switching into live view, the camera essentially takes an exposure reading and resets its own parameters before re-engaging the AE-lock. Whether or not you exposed correctly for your subject or whether the lighting is constant is irrelevant.

Also I'm not sure if you understood Radikal's method entirely... he is not metering the camera to his hand or the lens cap - He is metering and setting exposure correctly for the scene then placing his hand infront of the lens after engaging AE-lock and before entering live view mode in order to better demonstrate that the camera will not hold its settings. If the camera was holding it's settings, it would not adjust the exposure to his hand, but it seems it does.

Now, what this means for a properly light live scene is that the camera might come up with exposure settings which give an overall equal exposure to that which you chose, but the parameters might be completely different. For example, you might have dialled in settings of f/2.8 & ISO 100 & 1/50th. However the camera might set it to f/4 & ISO1600 & 1/400th for an equal exposure. It will then engage AE lock so these settings won't change again.
 
deltoidjohn said:
Bikergeek, you are overlooking the greater implication of what Radikal has pointed out - he has shown that when switching into live view, the camera essentially takes an exposure reading and resets its own parameters before re-engaging the AE-lock. Whether or not you exposed correctly for your subject or whether the lighting is constant is irrelevant.

Also I'm not sure if you understood Radikal's method entirely... he is not metering the camera to his hand or the lens cap - He is metering and setting exposure correctly for the scene then placing his hand infront of the lens after engaging AE-lock and before entering live view mode in order to better demonstrate that the camera will not hold its settings. If the camera was holding it's settings, it would not adjust the exposure to his hand, but it seems it does.

Now, what this means for a properly light live scene is that the camera might come up with exposure settings which give an overall equal exposure to that which you chose, but the parameters might be completely different. For example, you might have dialled in settings of f/2.8 & ISO 100 & 1/50th. However the camera might set it to f/4 & ISO1600 & 1/400th for an equal exposure. It will then engage AE lock so these settings won't change again.
Click to expand...

Ignoring the lens cap test, I can consistently lock exposure and bounce in an out of Live View without a change to the aperture/shutter/ISO values and with no apparent change in the image. The lens cap test may not be appropriate because the camera could be sensing a drastic metering change.

The bottom line is that with the D5100, I can mount a lens, establish a reasonable exposure, lock that exposure, and shoot a nice looking video clip without highlight issues or mud in the dark bits. The camera shoots a nice clean video clip that gives me plenty of latitude for correction and/or grading in FCP. I normally do portable shooting outdoors with uncooperative subject matter (nature, wild animals, motorsports). I simply need the camera to behave within certain parameters and deliver clean video clips. I don't do studio shooting and obsess over histograms or waveform monitors.
 
deltoidjohn said:
Now, what this means for a properly light live scene is that the camera might come up with exposure settings which give an overall equal exposure to that which you chose, but the parameters might be completely different. For example, you might have dialled in settings of f/2.8 & ISO 100 & 1/50th. However the camera might set it to f/4 & ISO1600 & 1/400th for an equal exposure. It will then engage AE lock so these settings won't change again.
Click to expand...

Deltoidjohn, you hit the nail squarely on the head, but the situation is slightly better than that. We know from the manual that the camera does respect f-stop settings made before entering live view, but not the ISO and the shutter speed (see page 108). Zeroing out the exposure before entering live view seems to guarantee the exposure you want, but not necessarily the combination of ISO and shutter speed. I wish there was more to be said on that issue.

With non-CPU lenses the camera does not give any exposure readings at all before entering LV, so I don't know of any way to address this.
 
Radikal said:
Deltoidjohn, you hit the nail squarely on the head, but the situation is slightly better than that. We know from the manual that the camera does respect f-stop settings made before entering live view, but not the ISO and the shutter speed (see page 108). Zeroing out the exposure before entering live view seems to guarantee the exposure you want, but not necessarily the combination of ISO and shutter speed. I wish there was more to be said on that issue.

With non-CPU lenses the camera does not give any exposure readings at all before entering LV, so I don't know of any way to address this.
Click to expand...

It is not clear in the manual, but one would think that the camera has a consistent behavior between picture mode and video mode in live view. I wish there was metadata captured with the video. Unless the camera is writing the wrong metadata to its still images, it is clear that with picture-taking, I can lock aperture/shutter/ISO and enter live view with no apparent change to the settings (confirmed via test photograph metadata). Once in live view, I am able to take a photo and record a video clip at one ISO setting, change to a different ISO value (without leaving live view) and take a new photo and video clip with the new ISO documented in the still image metadata. The other night, I bounced back and forth between ISO 100 and ISO 1600 indoors under halogen lights and could clearly see the difference in the still image and the video. I am not certain that the information on page 108 of the D5100 manual is correct. Your lens cap test proves that the manual is correct. But I am still questioning the camera's behavior.
 
Bikergeek said:
The lens cap test may not be appropriate because the camera could be sensing a drastic metering change.
Click to expand...

Drastic metering change or not, it proves that the camera is still able to make up it's own mind and over-ride your settings even though exposure should be locked.

Bikergeek said:
I can consistently lock exposure and bounce in an out of Live View without a change to the aperture/shutter/ISO values

Once in live view, I am able to take a photo and record a video clip at one ISO setting, change to a different ISO value (without leaving live view) and take a new photo and video clip with the new ISO documented in the still image metadata.
Click to expand...

The displayed ISO and shutterspeed settings do not reflect what is being recorded in videos - they will not update to display any settings the camera chooses to over-ride for video shooting.

The reason for this is that the parameters you entered do still apply to still photos taken in Live View mode - hence why the metada also reflects those settings. However, those settings do not also apply to videos. In live view mode, the ISO and shutterspeed values displayed on screen do not reflect what is recorded in video, and neither does the metadata from any still photos taken in Live View mode.
 
Bikergeek said:
The other night, I bounced back and forth between ISO 100 and ISO 1600 indoors under halogen lights and could clearly see the difference in the still image and the video.
Click to expand...

I haven't been able to replicate this result with the video. If you can replicate it again and post the exact steps you used to achieve this result, it would be a big break-through.
 
Radikal said:
I haven't been able to replicate this result with the video. If you can replicate it again and post the exact steps you used to achieve this result, it would be a big break-through.
Click to expand...

I will do it again with better subject matter this weekend and post and link the results via Vimeo. I understand what deltoidjohn and the Nikon manual say. But I have video clips that seem to show otherwise. There was no magic to my method. Sitting in my evening office with moderate lighting from halogen track lights, I set the iris wide open in aperture priority mode, entered live view, and shot test shots and video clips at 100 ISO and 1600 ISO (alternating back and forth without leaving live view). After dumping the pictures and video to my computer, I verified as per the picture metadata that the F-stop didn't change and that the ISO did. In viewing the video clips, there is a drastic difference in the look of the video clips that I presume were shot at 100 ISO versus 1600 ISO. The 1600 ISO (presumed) video clips are noisier with a different color rendition than the presumed 100 ISO video clips. If the manual page 108 is correct, one would assume that the video clips should look identical (because the camera supposedly selects shutter and ISO automatically). I pointed the camera at various locations in my office and repeated the test with the same results. The following evening, I repeated the tests in shutter priority mode and manual mode.

To make the test more scientific, I will mount the camera to a tripod and shoot the same subject exactly (this weekend).

Radikal, please try your camera and see what happens. Make sure that your ISO auto sensitivity menu selection is set to OFF which will gray-out the selections for maximum ISO and minimum shutter. Map the ISO choice to the function button.

Perhaps there is some other function in the camera that is fooling me. But with my method, I fixed one variable which is iris and changed only one other variable which was ISO. And as a result, the video quality changed. If the shutter and ISO are always adjusted automatically by the camera in video recording mode, there should have been little or no change to the video under the same lighting with the camera pointed at the same subject.

Am I missing something?
 
I have posted a comparison video that shows the results of my ISO experiment from the other evening:


The unprocessed video clips were edited together with added text via FCP.
 
Last edited:
I can see in your test clips there is a big difference between the clips, but I don't see how this is a conlusive test.

To really see what's going on (without having metadata) you need to film a moving object such as a fan and measure the motion blur, by comparing it to a control sample (ie a video camera that allows you to set shutter speed, or still photos taken in Manual mode). Once you know your shutterspeed (from the compasion to the control sample) and aperture (that you set), you can roughly work out the ISO for any particular clip with the same overall exposure.

By doing this, you'll be able to work out if you are actually able to adjust ISO on a consistant, repeatable basis without the camera also adjusting other parameters.

Of course all of this is academic as, given the required effort to even achieve anything comparable to proper manual control, I would never use this camera when there are perfectly viable alternatives (D7000 if you have to use Nikon or 550D if you have to stick to the same budget)
 
deltoidjohn said:
I can see in your test clips there is a big difference between the clips, but I don't see how this is a conlusive test.

To really see what's going on (without having metadata) you need to film a moving object such as a fan and measure the motion blur, by comparing it to a control sample (ie a video camera that allows you to set shutter speed, or still photos taken in Manual mode). Once you know your shutterspeed (from the compasion to the control sample) and aperture (that you set), you can roughly work out the ISO for any particular clip with the same overall exposure.

By doing this, you'll be able to work out if you are actually able to adjust ISO on a consistant, repeatable basis without the camera also adjusting other parameters.

Of course all of this is academic as, given the required effort to even achieve anything comparable to proper manual control, I would never use this camera when there are perfectly viable alternatives (D7000 if you have to use Nikon or 550D if you have to stick to the same budget)
Click to expand...

Thanks for the comments. I did not post this experiment as a conclusive test. The video samples raise a question as to how the camera is performing. Despite the controlled environment, the camera clearly adjusted something between the two video clips that caused a clear exposure and noise difference. The only settings change was ISO.

Yes, as I noted in my Vimeo posts, I will be performing a motion test to further determine the camera's behavior. My sense it that the camera allows manual control over two factors but not all three.

Regarding the D7000, although it is a fine camera, it is limited to 24P in 1080HD mode (Like the Panasonic GH cameras) and does not have an articulating LCD (I will never buy another DSLR without one). These issues are a deal-breaker for me. I am not obsessed with shooting at 24P (like many on this forum). I prefer to shoot at 30P (29.97) or higher. I am not trying to achieve a soft and jerky "film" look. I am trying to achieve a sharp, smooth HD video look. Most of my video opportunities involved fast-moving animals in nature, sports activities, and motorsports events. My dream DSLR camera would shoot 1080P/60. 30P is the next closest thing. The D5100 does 30P very well. The last time I looked, 29.97 was the broadcast standard for video in NTSC countries.

People on this forum have the tendency to project their own needs and biases regarding camera operation on others. 24P and manual control are not definitive VDSLR needs for everyone. Memories are short and people have forgotten that the popular Canon 5DMKII was released without manual movie control or 24P capability (NTSC 30P only). For nearly a year until the firmware update, many users of that camera produced awesome results. Without the GH13 hack, the stock GH1 shoots a kludgy 60i that requires pull-down to achieve a jerky 24P (solved in the GH13 and GH2).
 
Last edited:
Has anyone ever tried calling Nikon and telling them what's up? I've thought about it for a while but I don't know how productive that would be. Even a firmware update that would allow you to see what settings the camera is using while in live view/recording would be helpful.
 
You must log in or register to reply here.
Back
Top