I�ve got to comment here. After reading Bob�s post, I went outside (Wednesday, February 12, 2003) with my camera and Gossen Luna Pro light meter, # 4D51102. I shot a few scenes to check both instruments, then shot them in tandem against a solid tan/light brown/stucco exterior for my own references. I will insert my observations after the relevant text. I mounted a Hoya HMC 1b Skylight filter on the lens (my old FA* 28-70mm F 2.8 zoom) and set the lens as near to a 30 degree angle as I could. **My model of Gossen Luna Pro meter covers approximately 30 degrees. ________________________________ �--- Bob Blakely <[EMAIL PROTECTED]> wrote: All air/glass interfaces are an abrupt change in impedance for the electromagnetic energy impinging on/passing through the interface.�
Light goes through the glass. A (any) lens acts just like the overflow/outlet pipe in a toilet tank. As long as water (light) stays below the top of the overflow pipe, the water (light) can�t enter. Once the water (light) reaches the top of the pipe, it flows downward into the pipe. (Omni-directionally in case of light). In a camera, the flow (light) strikes the sensor/film plane. In both instances, the �flow� is in one direction. Sure some light �bounces� around inside the lens, but interior baffling in the lens usually neutralizes any stray light long before it can reach the film plane. �A portion of the light must be reflected at each interface. In fact, the same amount of light will be reflected at the inner interface as from the outer interface.� The light goes straight through. I don�t understand why Bob stressed �must be�, though I had often wondered about the way light behaves as it entered the lens. Radio waves (light) generally travel in straight lines unless acted upon by an outside force. A filter is not such a force (unless it�s super filthy). �Lost light is not really the problem for a single glass.� �Lost� me here: �lost light�? Usually �Lost light� is an issue only if there is a �filter factor�. �What would you loose, a third of a stop at most?� None, unless my PZ1p (set for eight segment, multi-pattern readings) and Gossen light meter are broken in the same manner. No loss of light with the filter on! Had there been, the Luna Pro would have registered the loss, or at least differed in its reading, since there is was no filter covering its sensor. No measurable differences in the light reading by either instrument. In any case, a �third of a stop�? �Light reflected from the inner interface, reaches the outer interface and a portion of that is reflected back toward the film. Some is transmitted.� I think I understand here that the light trapped in-between the rear (uncoated) surface of the filter and the surface of the front element, is nonetheless forced down onto the film plane. It seems to me that any light �bouncing� around, trapped in the space between the two elements, would simply dissipate� or have no effect at all. Surely it would not, of its own volition, travel down onto the film/sensor. �If the light is bright, such as a sun reflection off chrome, the reflection will be seen because it's now bright compared to the image. This is flair.� ?? �You may try single coated optics, but by their very nature, they can perform a perfect impedance transformation at only one wavelength. ?? �For this reason, single coated optics must have some affect on color, that is, demonstrate some color cast, however mild.� Almost without exception, �optical� glass must meet certain standards, one of which is its ability to transmit light without imparting color to it. If the filter is �optical� glass, coated or not, it will not transmit color. FYI: The light passing through Pentax prime glass is generally more Gamma neutral than either Canon or Nikon. That is: some glass imparts a �warm� Gamma reading while others may impart a �cool� (Nikon) reading. �The reason for multi-coated optics is broad band impedance transformation.� ?? �Most of us don't think this way, but it is what's happening.� > Regards, > Bob... _____________________________ > From: "Gregory L. Hansen" <[EMAIL PROTECTED]> > > > How important is it to have a coatings on all your > filters? A plain UV > > filter might cost $10 while an SMC filter would > cost $35. I can't see how > > it would matter much on the outer surface, a > little bit of light would > > just be lost. But on the inner surface?� Part of what I know is this: (in layman�s terms): A. A properly fitted lens hood stops more �flare� producing light than any filter. B. Shooting with wide to ultra-wide angle lenses severely increases your chance for flare. C. Multi-coating reduces flare. D. Our world-famed Pentax SMC lenses usually don�t require multi-coated filters, just a filter to protect the front element from�the elements. E. Mounting a coated or multi-coated filter on a multi-coated lens always seemed redundant to me. F. A clear, impeccably clean, uncoated filter stops nothing (except dirt and grease from touching the front element of the lens). G. A filter reduces the incidences of cleaning the front element, which might harm the multi-coating on the front element. H. The filter also helps protect the filter�s screw-in threads and the front element itself from impact damage. I. Any light that strikes the filter/lens elements at such an angle that lets light into the interior surfaces of the lens might cause flare. J. Light striking the film plane will cause flare. *Some people say �flare� as if it�s always a bad thing. Many photographers shoot in a manner that deliberately introduces flare. In my unscientific measurements, taking a set of readings, filter on and filter off, there was no discernable differences in the two sets of readings: none. If there were, tunnoticeablell, unnoticeable even to my not so agile eyes. As I changed angles of attack on the wall, the light and camera meters kept in quantitative (light value) sync. I shoot a lot of 100mm and up telephoto. At 200mm and more, I always have an impeccably clean multi-coated filter (usually a Hoya 1b) screwed onto the lens, then a steel lens hood (usually a HAMA). Why my lens hood fetish? To increase saturation and block stray light. --- Bob Blakely <[EMAIL PROTECTED]> wrote: > All air/glass interfaces are an abrupt change in > impedance for the > electromagnetic energy impinging on/passing through > the interface. A portion > of the light must be reflected at each interface. In > fact, the same amount > of light will be reflected at the inner interface as > from the outer > interface. Lost light is not really the problem for > a single glass. What > would you loose, a third of a stop at most? Light > reflected from the inner > interface, reaches the outer interface and a portion > of that is reflected > back toward the film. Some is transmitted. If the > light is bright, such as a > sun reflection off chrome, the reflection will be > seen because it's now > bright compared to the image. This is flair. You may > try single coated > optics, but by their very nature, they can perform a > perfect impedance > transformation at only one wavelength. For this > reason, single coated optics > must have some affect on color, that is, demonstrate > some color cast, > however mild. > > The reason for multi-coated optics is broad band > impedance transformation. Whatever, since this simple mind cannot comprehend that. I know some filters have "filter factors" (alters or slows down light transmission) bby one to four and half stops. Some filters alter color rendition as in blue, 80 series filters letting in more blorangeblocking red-orLikege (or something lke that). B&W filters almost always have filter factors. But UV, Skylight and a few others do what they do without altering anything other than maybe color. Matt I get it done with YAHOO! DSL! _______________________________ > From: "Gregory L. Hansen" <[EMAIL PROTECTED]> > > > > How important is it to have a coatings on all your > filters? A plain UV > > filter might cost $10 while an SMC filter would > cost $35. I can't see how > > it would matter much on the outer surface, a > little bit of light would > > just be lost. But on the inner surface? ===== Matt Greene I get it done with YAHOO! DSL!
