Which bubble level to believe?

I've got an RRS lever QR with a bubble spirit level built in on the flat face. With my 5D2 attached via an RRS L bracket, I put a standard plastic 2-D bubble level in the hot shoe. They read differently.

As the L bracket is closer to the plate than the hotshoe level, I suspect that's the accurate one. But is it?

It really only matters what the one on the camera says as it is what your trying to get level. I would go with the level on the hot shoe.

Take Care,
Cheers, Patrick

Try leveling with the hot shoe level and then reversing that (without moving the camera) and see if it reads the same.

I have found the hot shoe levels have too much play in them to be accurate. Check the head level with a quality level to see if it is accurate.

Quad wrote in post #8744023
I have found the hot shoe levels have too much play in them to be accurate. Check the head level with a quality level to see if it is accurate.

I think that's correct. The hotshoe was never made to be a level spot on the camera, just to hold a light.

I leveled the RRS QR without the camera in it, then checked it with a carpenter's level...dead on. Then I put the plastic bubble level onto the QR...dead center in both dimensions.

It's clear that the hotshoe is off level.

tvphotog wrote in post #8744355
I think that's correct. The hotshoe was never made to be a level spot on the camera, just to hold a light.

I leveled the RRS QR without the camera in it, then checked it with a carpenter's level...dead on. Then I put the plastic bubble level onto the QR...dead center in both dimensions.

It's clear that the hotshoe is off level.

Thanks Jay. Good for everyone to know not to trust the hotshoe as place to level your camera.

I've never understood the idea of a hotshoe level.

You look through the viewfinder and you level the horizon across the frame. If you can't tell it's off kilter then, you won't be able to tell in the picture either.

Curtis N wrote in post #8744797
I've never understood the idea of a hotshoe level.

You look through the viewfinder and you level the horizon across the frame. If you can't tell it's off kilter then, you won't be able to tell in the picture either.

'Horizons' are not always level in the viewfinder. If you're one of the people who like receding coastlines (for e.g.) to be level then *shrug*

Is it really that important to get your camera perfectly level then take a picture or is it more important to take the picture then fix horizon in PP? Its easier to just eyeball the horizon and snap away.

Take Care,
Cheers, Patrick

Saint728 wrote in post #8744908
Is it really that important to get your camera perfectly level then take a picture or is it more important to take the picture then fix horizon in PP? Its easier to just eyeball the horizon and snap away.

Take Care,
Cheers, Patrick

i'm still off by a degree or so when eyeballing. my hotshoe level gets me close enough when i remember to use it .

ed rader

Level does make a difference in architecture and panoramas. Often my entire rig can be level, but depending what lens I shoot a panorama with, the camera can be off level somewhat due to the weight in portrait orientation. (360 pano setup so the camera is "hung" over the center and not fixed to the level base) I will use a hotshoe level in both orientations, and then take additional readings from the mount. At least with my rig, I find the tripod level, clamp levels, and the hotshoe level, will be pretty close to eachother once everything is set.

Sometimes the 'horizon' is a slope of a mountain, and that is not level.

You can easily test any spirit level that you're trying to use with your camera.

Suspend a plumb bob (or a home-made equivalent) so that the line (string) is in your test scene. Level the camera as accurately as possible with the spirit level. Position the camera so that the plumb bob line is near one end or the other of the image. Make a test exposure. Now, examine the image (on a computer, not the camera's tiny display) and see if the plumb bob line is parallel to the side of the image.

Do the above for each spirit level setup that you have and see what works best for you.

That trick will also teach you a lot about the viewfinder coverage.

SkipD wrote in post #8746525
You can easily test any spirit level that you're trying to use with your camera. Suspend a plumb bob (or a home-made equivalent) so that the line (string) is in your test scene.

Works great for one dimension but the camera needs to be level in 2 for architecture.

Good to see the 7D has both catered for electronically - getting anything really accurate is a right royal pain. I have 3 bubble levels and all 3 are different ; hotshoe on the 5D is a bit out but the 5DII is reasonable. My current method is to use a Digi-pas electronic spirit level and work around the slight differences in hot-shoe setup from memory. This gets me accurate to 1 degree or so - good enough for most shots.

I guess the only viable solution is indeed the in-camera level - replacement bodies better have them