View Full Version : Quality of ball bearing for HDR
GatorNic 02-22-2005, 11:28 PM Hey all, just a couple questions on quality of steel ball bearings for HDR imaging
1. Found two places for ball bearings but different grades (and price). Looking at the 2.5 inch.
Boca Bearings 2.5inch (http://www.bocabearings.com/main1.aspx?p=product&id=7577&n=2-1/2IN_CHROME_STEEL_BALL_GR.1000) Grade/Sphericity = .001000" Diameter Tolerance = +/- .005" - $49.95
McMaster's 2.5inch (http://www.mcmaster.com/catalog/109/html/3366.html#.)diameter tolerance of ±.0002" and a sphericity of .00005" - $83.91
How much of a difference does it make in these grades? And would it *really* have any effect on the quality of the hdr, if shot properly?
2. How much of a differnce is there between shooting one of these and a 10inch glass globe?
Thanks!
edit: woops, wrong link. Now its fixed!
|
|
-Vormav-
02-22-2005, 11:45 PM
Most of the HDR guides that I've seen direct you to blur your HDR map slightly. And if you do blur the maps, I really don't think the difference will be noticable at all. I don't think that the variation of the diameter (within the range of the diameter tolerance) would make any difference at all for your purposes. Doubt you could even tell them apart without very very precise measuring tools.
Can't really comment much on the difference between a 2.5inch and a 10inch globe though. Unless you're really really serious about getting extremely accurate with this, you may just want to try picking up a shiny, silver ornament for HDR stuff. I have a set of ornaments, along with some ball bearings from McMaster, and the difference is almost indistinguishable.
CapnPanic
02-23-2005, 04:23 PM
1. i don't think you will notice any difference in the two bearing grades, assuming they both have a nice mirror finish. The imperfections that would cause problems should these bearings be supporting a large rotating object (as they were designed to so) shouldn't be noticable at all for your uses.
2. the main differences will be that you need a longer lens in order to fill the frame with your 2.5" bearing, you will have more trouble finding a support for the bearing that doesn't cover a large portion of the lower surface, the glass sphere will be a bit lumpy (assuming you are using the lawn-ornament type) and since it is glass mirrored from the inside, there will be a slight double reflection, noticable mostly on the edges. The glass sphere is also alot easier to break.
GatorNic
02-23-2005, 05:11 PM
Thanks for the replys! :thumbsup:
I have shot before with a christmas ball (not so great), and a 12 inch glass globe before (pretty good results).
I think you guys are right on the distortion of the two balls being not that apparent. I think I am going to go with the $50 ball. As for the camera, a friend will actually be shooting it with his Nikon D100, with a pretty good zoom lense. So I don't think it will be a problem filling the frame. The holder, I am pretty good at jerry-rigging stuff up....I am thinkin about using a wooden dowel with a rubber stopper screwed to the top. Just have to go thin enough not to obsure it too much. Time for a trip to Home Depot!
cpnichols
02-23-2005, 06:59 PM
If you are using your HDR for reflections (as important as diffuse lighting when using HDRs), then you are going to want an HDR that is sharp. On the other hand, I agree that you will not see much difference between the two bearings.
One thing to keep in mind when doing HDRs from probes is to find out exactly how reflective your probe is. Take a picture of an object that is reflected in the probe (preferably directly behind you), then mesure its value. Best done with a 50% gray card. Then turn your camera around 180 degrees and take the picture of the object directly. Mesure its value. You will most likely find that your probe is not 100% relfective, and probably close to 50% relfective (or one stop). A good thing to keep in mind. If you are really anal, do the same thing the the reflections on the edge of the probe. This will let you figure out if there is a fresnel falloff to your probe. Once you have done all that, you may be able to make yourself a photoshop script that balances your raw pictures for you.
GatorNic
02-24-2005, 10:17 PM
One thing to keep in mind when doing HDRs from probes is to find out exactly how reflective your probe is. Take a picture of an object that is reflected in the probe (preferably directly behind you), then mesure its value. Best done with a 50% gray card. Then turn your camera around 180 degrees and take the picture of the object directly. Mesure its value. You will most likely find that your probe is not 100% relfective, and probably close to 50% relfective (or one stop). A good thing to keep in mind. If you are really anal, do the same thing the the reflections on the edge of the probe. This will let you figure out if there is a fresnel falloff to your probe. Once you have done all that, you may be able to make yourself a photoshop script that balances your raw pictures for you.
Yeah wow that's great, I had never really thought about that. Normally I am pretty anal, for this upcoming project its more about the lighting (skin), except for I suppose the eye reflection but those are fairly blurry. Anyway its a good thing to keep in mind on future projects. Thanks!
CGTalk Moderation
02-24-2006, 11:00 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.
vBulletin v3.0.5, Copyright ©2000-2013, Jelsoft Enterprises Ltd.