I have shot my 3rd wedding now with the D700, and I gotta say that I am pretty disappointed with the auto focus in low light. I shoot mainly with fast primes (non-afs). I think I might try and rent a D3 to compare, but I find that the autofocus on my old D2h was way better.

I find my 85/1.8 is giving me the most trouble. Has anyone found that they had to play with the AF fine tune much?

BTW I don't use any of the AF assist lights, even though I know that by using them it would help, but I never needed them on my D2h/x and I find they draw to much attention to the photographer.

Basically I am just looking to hear other D700 photographers insights on low light autofocus, preferably how it compares to the D3.

>I have shot my 3rd wedding now with the D700, and I gotta say
>that I am pretty disappointed with the auto focus in low
>light. I shoot mainly with fast primes (non-afs). I think I
>might try and rent a D3 to compare, but I find that the
>autofocus on my old D2h was way better.
>
>I find my 85/1.8 is giving me the most trouble. Has anyone
>found that they had to play with the AF fine tune much?
>
>BTW I don't use any of the AF assist lights, even though I
>know that by using them it would help, but I never needed them
>on my D2h/x and I find they draw to much attention to the
>photographer.
>
>Basically I am just looking to hear other D700 photographers
>insights on low light autofocus, preferably how it compares to
>the D3.

The D700 has the same AF module as the D3. Have you considered setting your D700 to auto-ISO and set max ISO to 3200? I can't speak for the 85/1.8 but my 85/1.4 and 60/2.8 primes have no issues with focusing in low light with the D700

>The D700 has the same AF module as the D3. Have you considered
>setting your D700 to auto-ISO and set max ISO to 3200? I can't
>speak for the 85/1.8 but my 85/1.4 and 60/2.8 primes have no
>issues with focusing in low light with the D700

1. Auto-ISO (or ISO setting in general) will affect IQ, but has nothing to do with AF performance. The only possible relationship to the problem at hand *might* be if the softness is due to camera shake (i.e. shutter speed) rather than mis-focus, and a higher ISO would allow faster shutter (all else being equal). The original poster sounds "sharp" enough (pun intended) to know the difference between camera shake and mis-focus.

2. Faster lens (wider aperture) will benefit AF performance in light-challenged situations. Having said that, you'd have to compare min lens aperture, metered shutter speed, ISO, EV comp, and AF mode in order to directly compare the "low light AF performance" from one body to another... otherwise one shooter's idea of "low light" might be quite different from another's.

My two cents.

- Bob (eteam)

Not having a D700, I can't answer your question directly, but if it helps...

Since changing from a D2Xs to a D300 as my main camera, I find the D300 to be noticeably better at AF in low light conditions (typically, local village hall stage productions). I'd be surprised if the D700 AF is generally worse than the D2 series.

Brian
Welsh Nikonian
Check out the Nikonians Team pages

I don't know of a suggestion as to why you are getting focusing errors. If it's only with one lens, is it consistently out of focus in the same way--before the subject or behind the subject?

As a workaround, for really important shots, unless your eyesight is poor, have you tried the old-fashioned way--manual focus?

I have heard from a few professional photographers that all they use are manual (exposure, focus, ISO, WB) settings on their latest and greatest digital gear. They know what they set and what to expect. No automation "surprises". The quality is consistent.

Seriously. That's how it was done for many decades, successfully.

>I don't know of a suggestion as to why you are getting
>focusing errors. If it's only with one lens, is it
>consistently out of focus in the same way--before the subject
>or behind the subject?
>
>As a workaround, for really important shots, unless your
>eyesight is poor, have you tried the old-fashioned way--manual
>focus?
>
>I have heard from a few professional photographers that all
>they use are manual (exposure, focus, ISO, WB) settings on
>their latest and greatest digital gear. They know what they
>set and what to expect. No automation "surprises".
>The quality is consistent.
>
>Seriously. That's how it was done for many decades,
>successfully.
>
>

This is an excellent point. I've been A/B-ing the D300 and the D700 in low light conditions. My main interest is a noise evaluation, but I've come to recognize that the two systems indulge in a bit of negotiation wrt focusing in low light (as expected actually). I've found it less frustrating to take a bit of executive control in such situations.

Are you using single-area focus, dynamic area (and if so, 9,21 or 51pts?), or auto area?

my experience is that the percent of shots that are in clear focus using dynamic area with 21 or 51pts, or auto area, in low light is problematic. Using either 9pts, or single area, I find to produce a higher percent of in focus shots. I am shooting a wedding this coming weekend and will be in a church that does not permit flash and I intend to use only single area focus in order to have control and not rely on the brain of the camera. In moderate or good light I have had not problem with dynamic area 21pt.

Larry Miller, Annapolis, MD
D700/D800
http://www.pbase.com/lmmiller9
http://lmmillerphotography.smugmug.com/

Alright maybe I should clarify a bit. I shoot with the 11 af points, on the single af mode selecting the points manually but usually using the middle point. I use the 28 1.4, 50 1.8 and the 85 1.8 mainly, and I just find that in low light it searches a lot, like I can acquire focus eventually, and when I do it is sharp, it is just slow. I am talking 3200iso, 1.8 @ 1/125 dark. I should mention I shoot mainly on Continuous servo, so I might try single servo at my wedding tomorrow and see if it helps any.

i tried these settings the other night see my post

http://www.kenrockwell.com/nikon/d700/af-settings.htm

if your pictures arent good enough
your not close enough..

Reading Ken Rockwell's advice to just put the autofocus on Continuous servo and auto area focus, he does not discuss the effectiveness of this setting in low light.

I am curious if others agree with using this setting, for example in a low light church wedding. My concern is that the camera may be focusing on someone other than the bride or groom (like the Minister) and leave the primary subject out of focus. What is your experience?

Larry Miller, Annapolis, MD
D700/D800
http://www.pbase.com/lmmiller9
http://lmmillerphotography.smugmug.com/

My low-light shooting is of dim village hall stage productions rather than churches, and I'm using a D300 rather than D700, but the principle is the same

My preference is for using Single-area AF with the appropriate focus point for the desired competition, to make sure the camera knows where I want the plane of sharpness to be.

Brian
Welsh Nikonian
Check out the Nikonians Team pages

>Alright maybe I should clarify a bit. I shoot with the 11 af
>points, on the single af mode selecting the points manually
>but usually using the middle point. I use the 28 1.4, 50 1.8
>and the 85 1.8 mainly, and I just find that in low light it
>searches a lot, like I can acquire focus eventually, and when
>I do it is sharp, it is just slow. I am talking 3200iso, 1.8
>@ 1/125 dark. I should mention I shoot mainly on Continuous
>servo, so I might try single servo at my wedding tomorrow and
>see if it helps any.

Unless your subjects are moving and you require continuous tracking there is no sense in using CS. I would agree, try the Single Servo setting and let us know how it turns out.

>>Alright maybe I should clarify a bit. I shoot with the
>11 af
>>points, on the single af mode selecting the points
>manually
>>but usually using the middle point. I use the 28 1.4, 50
>1.8
>>and the 85 1.8 mainly, and I just find that in low light
>it
>>searches a lot, like I can acquire focus eventually, and
>when
>>I do it is sharp, it is just slow. I am talking 3200iso,
>1.8
>>@ 1/125 dark. I should mention I shoot mainly on
>Continuous
>>servo, so I might try single servo at my wedding tomorrow
>and
>>see if it helps any.
>
>Unless your subjects are moving and you require continuous
>tracking there is no sense in using CS. I would agree, try the
>Single Servo setting and let us know how it turns out.
>

If you are shooting at apertures like f1.4 and f1.8 CS does make sense since if you get focus lock with a select point and recompose or the subject moves the focus will adjust. There is not enough dof to get away with it.

>If you are shooting at apertures like f1.4 and f1.8 CS does
>make sense since if you get focus lock with a select point and
>recompose or the subject moves the focus will adjust. There is
>not enough dof to get away with it.

Just to clarify, so that everyone is on the same page:

Autofocus and metering is performed with the lens at minimum aperture (that's "wide open"). When the shutter release is pressed, the body dials in the specified (or metered) aperture. So AF is *not affected* by selection of f/1.4 or f/8, only by the lens' minimum aperture.

On the other hand, as narc correctly points out, selected aperture *will* affect depth of field, where tiny DOF provides little margin for AF error.

- Bob (eteam)

Just to clarify MAXIMUM APERTURE: Wide open aperture for the lens. i.e. You would need this for low light conditions.

AF NIKKOR 85mm f/1.4D IF, Tech Specs: Maximum Aperture 1.4, Minimum Aperture 16.

>Just to clarify
>MAXIMUM APERTURE: Wide open aperture for the lens. i.e. You would
>need this for low light conditions.

Yup, I got this backwards. If you change "minimum" to "maximum" in my previous post, it should (still) make sense: Auto-Focus is accomplished with the lens wide open, always, regardless of the dialed aperture setting.

- Bob

I'm sorry to hear that. I'm not having any problems with AF and don't use the head light either. It is better than the D2X, which I am happy with as well.

Regards,

Don Kesler

http://www.donaldkesler.com

Through the judicious use of adjustment layers, no pixels were actually harmed in the processing of my shots..

There is a podcast on how to use the Af on the camera...not sure which one it is but you can look it up on the website here at nikonians

www.jsa-photo.com

- because you are using the outer AF points in portrait mode.
When you do this in portrait mode the sensors try to read detail horizontal to the ground and there is often insufficient readable detail across many subjects faces.
With static subjects taking focus with the camera in landscape position, or thinking carefully about where on the face you place the sensor and using focus lock can be a solution.
When the AF cannot accurately read the subject using AF fine tune is not a reliable solution.

Photography is a bit like archery. A technically better camera, lens or arrow may not hit the target as often as it could if the photographer or archer does not practice enough.

Len Shepherd

Personally, I specifically tested the D300, D700, and D3 AF. While we all know the D300 is way slower than the D700 and D3, I found the D700 to be observably slower than the D3. This was actually later validated on pop-photo's D700 review. While most people seem to not notice any difference, I did and I am not the only one.

I would recommend that you try a D3 and see if it gives you the results you are looking for.

Greg

"D700 to be observably slower (than D3)... validated on pop-photo's D700 review" - the problem is that nowhere PopPhoto describes their camera testing procedures except in this hard to find How to Read a Camera Test written in 2005.

Why did the test @ ISO 100 when the native base ISO is 200? Did they test AF-S or AF-C, 3-D tracking, how many AF points? How do they measure these hundredths of a second differences? Do they start off focus @ infinite, with which lens, on what target? What is their margin of error?

In this side by side of popphoto's D700, D3 & D300 tests results you can see that D700 has less noise than the D3 @ ISO 400, 800, 3200 but more @ 200, 1600 & 25600. Margin of error?



Note: save the image & view @ 125 or 150% to see the details better.

"I found the D700 to be observably slower than the D3" - you need to be pretty good to observe a .12 second difference @ 6 EV for example.

"the D300 is way slower" - well according to Popphoto's tests D300 above it's faster than the D700 from EV 12 to 0 & at -2! Look at the numbers, not the curves which are a bit skewed, for example @ 6 EV; D700 .61 s vs. D300 .51 s.

Their "Certified Tests" are certified by whom? Apparently by their own staff.

Attachment #1, (gif file)

But the Pop Photo tests do not support your findings that the D300 is "way slower" than the D3, so whose results should we trust? Without knowing more about your testing methodology (and theirs), it's not possible to draw any sensible conclusions.

Brian
Welsh Nikonian
Check out the Nikonians Team pages

I will admit my D300 findings should be considered skewed as far as absolute values because my testing was not in an identical environment as the D3 and D700 tests. But, I do feel they were represntative enough to conclude the D300 was slower than the others. However (and the more critical point for this thread) is I tested the D3 and D700 side-by-side with the same subjects, lighting, lenses, settings, and vantage point. This was done in fixed lighting. Straght-up fixed point focus may have been almost identical between the D3 and D700...what I concentrated on was continuous autofocus 3-D tracking. I tried both subject and camera movement. With numerous tests using a selection of lenses (first on one camera, then the other, and sometimes back again) I personally found the D3 to always track better. This isn't so much initial focus as it is focus tracking, but it's still focus. Personally, I can't figure out how others are not seeing the difference. It was fairly obvious for me. And same results regardless of lens, so I have to rule that out. All I can figure, I don't think focus tracking latency is what others are testing.

I have been using my new D700 with both a 300 f4 and a 70-200 VR lens to shoot dogs running agility courses. In good light, the focus is wonderful. In low light conditions (which is what I bought the camera for) I am having a very hard time getting a really sharp shot of the moving targets. I have tried most combinations of autofocus tracking choices. I seem to have plenty of light, but no real sharpness. It is better with subjects close up. So then I tried single shot with still subject in the same light conditions - and I find them to also be less focused in low light - unless my subject is close to me.

I took my D300 and compared them side by side with the same lenses and aside from the difference in noise under low light Hi ISO conditions, the D300 is sharper using the same lenses.

I have been shooting dogs doing agility in less than optimal light for several years with a D2H, D300 and a Canon 1DMarkII - so I know when it's not all operator error. Otherwise I am very happy with this camera, but something's not right with the low light focus.

Debating on sending it back for exchange or to Nikon for adjustment.

Sara

The last two posts remind me of a theory I proposed a few weeks ago. It seems to me that "focus tracking latency" would have a lot to do with cpu speed. Does anyone know yet if there is a difference between the D700 and D3? Seems to me there would have to be, just based on the fact that the operating voltages are different.

The fact that one post finds the D300 BETTER than the D700 would support my other contention that the D300 might have better lock and/or tracking ability due to the fact that the AF points cover more of the sensor area. And given the application - dogs running an agility course (when you are probably struggling just to keep the dog somewhere in the frame) - it would be expected that this weakness would show up under those conditions.

As for the Pop Photo charts - we can't have our cake and eat it too, i.e. diss them on the one hand, then cite them to support our argument. I would stick with what I've said before. The charts show that ON PAPER and in the same (unknown) test conditions chosen by PopPhoto all are excellent and in the same ballpark.

As for me, I pay very close attention to people who I know are shooting in conditions similar to the conditions in which I sometimes shoot. The last two posts were very helpful to me in terms of evaluating what to expect when I (probably) eventually get a D700 and how it will fit into my shooting repertoire. I feel no need to get defensive about it.

PEACE,
Steve

Steve,

Good point. I wonder what the difference would be with different settings for a4: Focus Tracking with lock on?
Default is Normal, options are Long, Short, or Off.

I remember having trouble with my D200 tracking a dog running at me (friendly). Then I read Thom Hogan's D200 book, and changed the a4 option to off. This increased the speeed at which the autofocus would track a moving subject.

Also what about a2 AF-S priority Selection. Focus is default, Release is optional?

Mark

MarkF

D800
I still own an F100; do you think film will make a comeback?

Visit my Nikonians gallery.

Visit my Nikonians gallery.

Visit my Nikonians gallery.

I agree w/Steve and Mark's thought processes.

It would seem to me from an engineering standpoint that you could obviously expect to be able to draw more current from the D3's battery. As such, I would think small design modifications done to keep the peak load on the D700 a little lower than it would be on the D3 might be just enough to create these situations that only some of us doing specific shooting styles are seeing. Certainly focus tracking would be one of these areas since you are moving mechanical parts (more current draw). If you make just enough design adjustments (limitations) to keep the current draw within a safe range for the 3 series battery then it would seem to me this is where the most observable difference would be between these two camera bodies. Certainly if I was on a design team trying to cram a D3 into the D700's body and use a 3e battery, I would suspect I'd have to put a cap on the focus motor current draw and assume worst case (no battery grip). This would make sure the camera didn't shutdown from lack of current/voltage to the electronics.

Greg

Greg,

Interesting theory.
We know the D700 shoots at a maximum of 5FPS with the standard battery. With the grip and the D2/D3 battery or AA batteries it will fire at 8FPS. I wonder if it will also autofocus faster?

Mark

MarkF

D800
I still own an F100; do you think film will make a comeback?

Visit my Nikonians gallery.

Visit my Nikonians gallery.

Visit my Nikonians gallery.

Mark,

It would be an interesting experiment. I could see it going either way. If someone runs this test, please post the results!

Greg

> With the grip and the D2/D3 battery or AA batteries it will fire at 8FPS. I wonder if it will also auto focus faster?
Going back in time the F100 with grip had a faster fps, film rewind speed and IMO (not confirmed in the F100 manual) just slightly faster AF acquisition from a standing start.
Things like start up time can affect AF speed.
As a D3 and D300 user I go along with the Naturescapes.net view that compared to the D3 the D300 can take slightly longer to acquire focus lock but once it does it tracks very fast.
Whilst I do not own a D70 I am confident overall the AF on the 3 new cameras under discussion is the best yet from Nikon - and at least the equal of anything from Canon.

Photography is a bit like archery. A technically better camera, lens or arrow may not hit the target as often as it could if the photographer or archer does not practice enough.

Len Shepherd

"D300... AF points cover more of the sensor area..." - D300 AF area, D700 AF area images.