Panning shot blues

cletus

I'm really disappointed with my panning shots from a race a couple weeks back. I've got tons of shots where part of the car is nice and sharp, but other parts of the car are a blurry mess. Here are some examples:





Notice in both shots the front of the car looks pretty good but the rear of the car is a mess.

It's got to have something to do with angles and motion relative to the camera, but I'm not fully understanding what is going on. I guess the more important question is, how do I keep this from happening?

adbsgicom

Far from an expert here, but what DOF should you have expected? (FF/Crop, focal length, aperture, distance to subject?)

cletus

adbsgicom wrote: »

Far from an expert here, but what DOF should you have expected? (FF/Crop, focal length, aperture, distance to subject?)

DOF should be pretty big. #1 was 200mm @ 1/125 & f/22, #2 was 85mm @ 1/125 & f/18. I wasn't very close to either car.

jhefti

I'm no expert either, but it looks like a DOF issue. The blur on the back of the cars does not look like motion blur, and it's hard to think of a reason you'd have motion blur on one part of the car but not the other if all of the car is going the same velocity.

Just my $0.02's worth...

jhefti

cletus wrote: »

DOF should be pretty big. #1 was 200mm @ f/22, #2 was 85mm @ f/18. I wasn't very close to either car.

Just saw this reply after I posted my previous comment. I'm stumped on this--very weird! Were the cars sliding or drifting? Maybe it's something with the AF servo??

cletus

Here is another shot. Pretty close to the same location as shot #2, taken about the same time. For this shot, I dialed the ISO down to 100. 200mm @ 1/125 & f/14.



Here the whole car looks pretty good.

Molotov Everything

jhefti wrote: »

it's hard to think of a reason you'd have motion blur on one part of the car but not the other if all of the car is going the same velocity.

Just my $0.02's worth...

What I'm thinking maybe in regards to that, is that the car in shot 2 is moving uphill and coming towards him slightly. That's adding 2 more axes of movement, so during the time the shutter was open, the front and back of the car could have different changes in distance relative to the camera.
I don't know if that's the real reason, but it seems like it could be.

cletus

Molotov Everything wrote: »

What I'm thinking maybe in regards to that, is that the car in shot 2 is moving uphill and coming towards him slightly. That's adding 2 more axes of movement, so during the time the shutter was open, the front and back of the car could have different changes in distance relative to the camera.
I don't know if that's the real reason, but it seems like it could be.

I think this is the answer but I'm just having a hard time believing that there could be that much difference in only a fraction of a second.

Also I should note, this is the first time I've ever shot with a VR lens. All of these shots were with a Nikon 70-200 f/2.8G AF-S VR-II. The VR was on and in "Normal" mode.

jhefti

cletus wrote: »

I think this is the answer but I'm just having a hard time believing that there could be that much difference in only a fraction of a second.

Also I should note, this is the first time I've ever shot with a VR lens. All of these shots were with a Nikon 70-200 f/2.8G AF-S VR-II. The VR was on and in "Normal" mode.

Ah, you're holding out on us! Was this on a tripod too? Not sure about Nikon, but Canon VR (IS) does some very odd things when panning and/or on a tripod. If this is the first time you've seen this phenomenon, try again with VR off and see what you get.

adbsgicom

Could it be rear vibration as they hit the front breaks and rear is lifting slightly?

cletus

jhefti wrote: »

Ah, you're holding out on us! Was this on a tripod too? Not sure about Nikon, but Canon VR (IS) does some very odd things when panning and/or on a tripod. If this is the first time you've seen this phenomenon, try again with VR off and see what you get.

Nope. Handheld.

Here's another example. The car looks awful, but if you look close the driver's helmet looks pretty good and you can see details on the drink and radio cables!





70mm @ 1/125 & f/13

cletus

adbsgicom wrote: »

Could it be rear vibration as they hit the front breaks and rear is lifting slightly?

Possibly in the first shot. The others the cars are accelerating.

jhefti

That last shot and crop are especially puzzling. Motion blur looks to dominate the lettering on the side of the car, and seems to persist across the top of the car too. However, the driver looks to be in good focus. With those numbers I doubt it's a DOF issue--and besides, the lettering on the windshield is OOF too.

I am truly stumped...

cletus

I've got a theory. Maybe the rental 70-200 was insulted when it realized that it had been mounted on my poor, lowly little D90. The lens turned on a hidden "funky picture mode" to encourage me to go out and purchase a more befitting, and more expense, Nikon body. Good thing I didn't rent a 300 f/2.8... it might have refused to even let me take pictures if I mounted it to my camera!

adbsgicom

One question from a completely naive point of view. Is 1/125 fast enough to deal with little vibrations and subtle movement in the cars, even if your panning is dead on? My DOF question was initially spurred by an assumption that you'd have the aperture wide open to get the shutter speed as fast as possible, or is the goal in this sort of shooting to get the blur on the wheels, and thus the slower speed?

jhefti

cletus wrote: »

I've got a theory. Maybe the rental 70-200 was insulted when it realized that it had been mounted on my poor, lowly little D90. The lens turned on a hidden "funky picture mode" to encourage me to go out and purchase a more befitting, and more expense, Nikon body. Good thing I didn't rent a 300 f/2.8... it might have refused to even let me take pictures if I mounted it to my camera!

This is better than anything I have come up with...

zoomer

Shutter speed.
Go back and shoot at F4, as far away as you are that will give you plenty of depth of field to cover the car.
Make sure you follow through with the shutter.

cletus

adbsgicom wrote: »

One question from a completely naive point of view. Is 1/125 fast enough to deal with little vibrations and subtle movement in the cars, even if your panning is dead on? My DOF question was initially spurred by an assumption that you'd have the aperture wide open to get the shutter speed as fast as possible, or is the goal in this sort of shooting to get the blur on the wheels, and thus the slower speed?

I'm sure there would be some vibrations that 1/125 would not freeze, but I don't think they would account for the amount of blur in these shots.

Shooting wide open at a high shutter speed can sometimes result in shots that look like a car parked on the track. The shutter speed is fast enough that there is little to no motion blur in the background as you pan with the car. However, there are times you can shoot wide open and still see that the car is in motion. Here the slight lean in the car, the tilt of the driver's helmet and the distortion in the tires are all visual clues that the car is going through a high speed left hand corner:



200mm @ 1/1600 & f/2.8

zoomer

Work with your shutter speed and find a happy medium.
There will be a speed and f-stop that will freeze the car and let the background blur both from panning motion (assuming good techique) and from depth of field and your photos will be much sharper than they are when using those tiny f stops. On a bright sunny day an ND filter might be a big help to allow you to use a larger f-stop.
If you want motion in the car you will have to live with some blurriness of the car.
I think you will find that professionals who shoot motor sports are wide open or close to it....around f4...certainly not at F22....maybe for an occasional super blurry panning shot but not on a regular basis.
Shots at F22 will be a little blurry no matter what because of lens limitations.

JBHotShots

From my experience 1/320th in bright daylight will give you wheel blur to show motion but that is dirt track racing.

aj986s

I'm not a pro photog, but taken enough pics at road racing events to learn that panning can be very location specific. I finally learned that to get decent panning shots, I had to spend some time experimenting from a given vantage point. Speed of the cars, angle relative to the camera, available light, stability of the camera/photographer, shutter speed and f-stop are all contributing factors. I try to pick a particular view spot, and then try a couple of different shutter speeds with passing cars to see what's working at that time. I'll zoom in on LCD to examine results, and when I've found a set of settings I like, then I'll shoot a bunch of cars at that specific spot. If I change viewpoint, I'll do some more sampling again. I generally try to use the smallest f-stop possible, to get the most DOF. I've shot pans from 1/30 to 1/400, depending upon the circumstances. I usually end up hand holding most shots. Some of my Canon lenses have a IS for panning, that allows left - right movement of the camera.

mattdan12

Molotov Everything wrote: »

What I'm thinking maybe in regards to that, is that the car in shot 2 is moving uphill and coming towards him slightly. That's adding 2 more axes of movement, so during the time the shutter was open, the front and back of the car could have different changes in distance relative to the camera.
I don't know if that's the real reason, but it seems like it could be.

I believe this is the most accurate description of what is happening. Unless you are absolutely parallel to the car as it is moving past you, you're not going to get the entire car in focus/sharp as you pan with it. When it's coming at you at an angle - the front and back parts of the car are moving at two different speeds in relation to your camera - you can only pan on one "section" of the car - the other part you may get some motion blur like you are seeing.

I used to shoot a lot of local track stuff and races a few years ago. Here are a few examples from the 2006 ALMS race at Miller Motorsports Park in Utah. In these examples, there are a few of just part of the car in focus like your images, then a few where I was able to be parallel to it's direction of travel and got the entire car in focus.

Panning with the car coming from an angle:







Panning with the car parallel to the camera:








I'm no expert on this, so take my opinion with a grain of salt....

JSPhotography

I think it is vertical motion of the vehicle. Slightly faster shutter should fix it and still get the blurr you are after.

JSPhotography

One of the tell tales I am seeing from the original poster as well the shots that just got posted is the partial silver ring on the tire on the corner that is not sharp. That tells me there was up down movement at that instant. To me it looks pretty consistant in all the examples.

cletus

aj986s wrote: »

I try to pick a particular view spot, and then try a couple of different shutter speeds with passing cars to see what's working at that time. I'll zoom in on LCD to examine results, and when I've found a set of settings I like, then I'll shoot a bunch of cars at that specific spot. If I change viewpoint, I'll do some more sampling again. I generally try to use the smallest f-stop possible, to get the most DOF. I've shot pans from 1/30 to 1/400, depending upon the circumstances.

A very, very good idea. I'll be putting that to use next time I'm shooting motor sports!

cletus

mattdan12 wrote: »

...Unless you are absolutely parallel to the car as it is moving past you, you're not going to get the entire car in focus/sharp as you pan with it. When it's coming at you at an angle - the front and back parts of the car are moving at two different speeds in relation to your camera - you can only pan on one "section" of the car - the other part you may get some motion blur like you are seeing.

I think you're right. It's surprising that there would be enough difference in the relative motion over that brief a time, but I guess there is.

Gorgeous shots by the way!

cletus

JSPhotography wrote: »

One of the tell tales I am seeing from the original poster as well the shots that just got posted is the partial silver ring on the tire on the corner that is not sharp. That tells me there was up down movement at that instant. To me it looks pretty consistant in all the examples.

I kept thinking vertical motion as well, but checkout the blur in the LEDs on the side of the car. Each one of the three "lights" is a small triangle shaped cluster of plain, round LEDs. That looks like good old faction horizontal motion blur... at least to my eyes

adbsgicom

cletus wrote: »

I think you're right. It's surprising that there would be enough difference in the relative motion over that brief a time, but I guess there is.

Gorgeous shots by the way!

I had a little doodle on a notepad here, was was trying to see if it could be the relative motion difference and how to put it in to some equation. Empirical data looks convincing, though....

JSPhotography

You sure about those lights? That is a lot of blurr that is not consistant with the rest of the car. Whats the yellow by the door handle?

JSPhotography

adbsgicom wrote: »

I had a little doodle on a notepad here, was was trying to see if it could be the relative motion difference and how to put it in to some equation. Empirical data looks convincing, though....

OHH, like the way you thinkl.

cletus

JSPhotography wrote: »

You sure about those lights? That is a lot of blurr that is not consistant with the rest of the car. Whats the yellow by the door handle?

Wish I had a good static shot of the side of that car. I'll dig and see what I can find.

Edit: I'm such an idiot. I did have another shot of that car:



95mm @ 1/250 & f/10