Is the Ground Glass / Focusing Screen necessary for 35mm adapters?

[Peter Chung]

I've recently started getting into the 35mm DoF adapters and trying to learn how they work.

So I ghetto-setup my SLR lens and placed my camera far enough from the lens so that it can zoom in and focus on the image coming from the SLR lens.

However, I didn't have any ground glass or a focusing screen. I did try to use one of those clear CD holder things from a CD spindle but it didn't seem to do anything, I guess because it is too clear. I tried to use a frosted plastic CD case and it didn't seem to do much either.

Anyways, I was able to focus my camera and get an image(!) even without a GG.

So, my question is, do I really need a GG? Wouldn't it decrease light loss by not having a GG anyways? It would also get rid of dust proofing issues and such since there is no optics between the SLR lens and camcorder.

Interestingly, the focal flange of the Canon EOS lenses are supposed to be 44mm. But if you look at my last photo, 44mm from the lens is about where my UV filter is(!) What's up with that? Can anyone help me to understand what is going on? My guess is that the macro lens is focusing on a "virtual image" behind the lens???

With further experimentation, I've put other translucent materials where the focal flange should be and when the material is in focus, so is my image. But it was already in focus without the material. This is why I am questioning the necessity of the GG. I think that without the GG, the backfocus would be much easier to adjust and correct since it doesn't need to focus on a GG.

Anyways, if anyone has any ideas or can explain why I still need a GG, I would greatly appreciate some enlightenment.

Thanks!
Peter

[Bob Hart]

Peter.


This one gets an airing every now and again quite often.

There is nothing to stop you from using the aerial image which is the coincidence of focal planes of the two lens systems. In fact, I do this with long zooms and telephoto lenses to gain images the camcorder is otherwise incapable of producing.

The available field-of-view of the lens placed in front may be reduced by vignetting if the exit pupil of the lens is small. What you observe as achievable shallow depth-of-field is more likely to be a product of having to zoom in closer to get inside the vignette.

In that event, you will have a constricted field of view relative to the usable groundglass relayed image and there is limited advantage.

You will likely get a sharper image and improved contrast over a groundglass relayed image.

However, the depth-of-field characteristics remain those of the viewing camera, not the lens you have placed in front of it, so you will not be able to realise all of the creative options the lens offers when used via a groundglass relay.

The apparent flexibility of backfocusing you mention is a product of the deeper depth-of-field characteristic of the camcorder lens. You may find interaction between the two focussing systems which will retain the relayed aerial image in sharp focus oveer quite a wide combined range. You will discover at the extreme thresholds one end or the other that the image will start to go soft or distort in the corners. You may also find that the camcorder autofocus works quite happily through the lens on front.

[Peter Chung]

Thanks for the explanation, Bob.

Can I explore this topic some more with you? I couldn't find anything with a search...

If I understand you correctly, you are saying that the advantage of the ground glass is that you are projecting the full 35mm frame onto the GG at the focus plane. Without the GG, if your camcorder lens is too big, then you'd have to zoom in more to get rid of the vignetting, which in effect, is cropping the image, which means that the field of view compared to the SLR lens is limited.

I did notice that in the photo taken with my SLR at the same spot vs. the image taken with my camcorder, the SLR photo had a wider field of view. Would I get the same field of view as my D-SLR by using a GG?

One comment you made caught my eye...

"You will likely get a sharper image and improved contrast over a groundglass relayed image."

That seems to be quite a huge benefit of not using a ground glass projection surface. Does this also mean that there would be less light loss since there are no optics for the light to go through?



I didn't quite understand the last few explanations, though:

However, the depth-of-field characteristics remain those of the viewing camera, not the lens you have placed in front of it, so you will not be able to realise all of the creative options the lens offers when used via a groundglass relay.

The apparent flexibility of backfocusing you mention is a product of the deeper depth-of-field characteristic of the camcorder lens. You may find interaction between the two focussing systems which will retain the relayed aerial image in sharp focus oveer quite a wide combined range. You will discover at the extreme thresholds one end or the other that the image will start to go soft or distort in the corners. You may also find that the camcorder autofocus works quite happily through the lens on front.

What creative options do I lose by not using a GG?

With regards to the softness or distortion in the corners, is that a result of spherical and/or chromatic aberrations? Would this be resolved by using an achromatic macro lens?

Thanks for your help!
Peter

[Ben Winter]

Originally Posted by Peter Chung

Thanks for the explanation, Bob.

Can I explore this topic some more with you? I couldn't find anything with a search...

If I understand you correctly, you are saying that the advantage of the ground glass is that you are projecting the full 35mm frame onto the GG at the focus plane. Without the GG, if your camcorder lens is too big, then you'd have to zoom in more to get rid of the vignetting, which in effect, is cropping the image, which means that the field of view compared to the SLR lens is limited.

I did notice that in the photo taken with my SLR at the same spot vs. the image taken with my camcorder, the SLR photo had a wider field of view. Would I get the same field of view as my D-SLR by using a GG?

One comment you made caught my eye... That seems to be quite a huge benefit of not using a ground glass projection surface. Does this also mean that there would be less light loss since there are no optics for the light to go through?



I didn't quite understand the last few explanations, though:


What creative options do I lose by not using a GG?

With regards to the softness or distortion in the corners, is that a result of spherical and/or chromatic aberrations? Would this be resolved by using an achromatic macro lens?

Thanks for your help!
Peter

The main use of adapters is to take advantage of the inherent depth of field characteristics of the lens you use, i.e. focus separation between the foreground and background.

Without a ground glass, the depth of field stays the same as the camera's original characteristics. Thus there is no real advantage to adapting a lens to a camera other than to achieve a longer focal length.

http://en.wikipedia.org/wiki/Depth_of_field_adapter

[Peter Chung]

I wanted to see for myself if there really is no difference by not using a ground glass.

Here's some footage with my ghetto set up...


Low-Res 320x180 8MB (1min 17secs)

High-Res 1920x1080 65.4MB (1min 17secs)

There is an observable difference between using the adapter and not.

It seems that I am able to bypass the need for the ground glass altogether. What seems to be happening is that the macro lens is focusing at the exact plane of the focal flange of the SLR lens. The image still exists and all the light from the SLR lens converges at that point, we just can't see it because there is no screen there, or GG. But the macro lens is still able to focus on this "virtual projection screen." At least that's my theory, anyway ;)

Anyways, let me know what you think about the footage.

Thanks!
Peter

Edit: Attached photos of setup with HV20. I know, it's ghetto...

[Bob Hart]

Peter.

Ben's explanation sums up in concise form what I have been attempting to convey in my own convoluted and confusing style and demonstrates one of the advantages of tertiary education which I lack, - ability to get straight to the point whiich is not one of my communicative strengths

Here's a little more to confuse you perhaps.

There are purely optical ways to make a camcorder see the world through a lens built for a larger image size like film and reproduce exactly the features of that lens.

However it is a very complicated task and very costly, probably to the point it might be more cost-effective to go out and buy the 35mm film camera and shoot film.

Groundglass relay costs less.


(If I understand you correctly, you are saying that the advantage of the ground glass is that you are projecting the full 35mm frame onto the GG at the focus plane. Without the GG, if your camcorder lens is too big, then you'd have to zoom in more to get rid of the vignetting, which in effect, is cropping the image, which means that the field of view compared to the SLR lens is limited.)

In practical terms thats what happens. If you go to YouTube, look for "DARANGULAFILM" and look at two clips (URLs to be posted)

The "35 versus 300" clip was shot with the groundglass in place. The "Tiger is 75" clip was shot with the groundglass removed. With the groundglass in, I was able to pull back with the camcorder zoom and pick up more of the groundglass. There is a trace of lens vignette in the top left corner on the "Tiger is 75" clip.

The dark left edge is a construction issue related to my adaptor and not a lens vignette.

(I did notice that in the photo taken with my SLR at the same spot vs. the image taken with my camcorder, the SLR photo had a wider field of view. Would I get the same field of view as my D-SLR by using a GG?)

Maybe. It would depend on whether you can position your camera close enough to the groundglass and retain sharp focus, yet frame the full DSLR image sized frame.

You might need to add a close-up lens, preferably an achromatic dioptre. It also depends on the available wide aperture on your still-camera lens on front.

Some DLSR lenses work, some don't. There is a tendency for some of the DSLR lenses to have brightness falloff in the corners.

(One comment you made caught my eye...
Quote:
"You will likely get a sharper image and improved contrast over a groundglass relayed image.")

That seems to be quite a huge benefit of not using a ground glass projection surface. Does this also mean that there would be less light loss since there are no optics for the light to go through?)

What the groundglass giveth with one hand it taketh away with the other. It gives you ability for shallower depths-of-field for a given field(angle)-of-view. It softens the image by scattering and thus losing some light. On my own setup, the light loss by going through the adaptor versus aerial image was barely observable. The camcorder aperture, shutter and ND settings were the same.

This suggests to me that the lens on front is the significant limiting factor. Adaptors seem to lose between 0.5 and 1.5 of a stop at SLR lens widest aperture setting, depending on design and whether a flip path is included.

(I didn't quite understand the last few explanations, though:

Quote:
However, the depth-of-field characteristics remain those of the viewing camera, not the lens you have placed in front of it, so you will not be able to realise all of the creative options the lens offers when used via a groundglass relay.)

Take this one on faith. - It has been arrived at by lots of experimentation by others here who have shaken down to the same conclusion.

The shallow depth-of-field effect available from the lens on front can only be achieved by using the groundglass or one of thhose horrensdously expensive optical relay adaptors.

(The apparent flexibility of backfocusing you mention is a product of the deeper depth-of-field characteristic of the camcorder lens. You may find interaction between the two focussing systems which will retain the relayed aerial image in sharp focus oveer quite a wide combined range. You will discover at the extreme thresholds one end or the other that the image will start to go soft or distort in the corners. You may also find that the camcorder autofocus works quite happily through the lens on front.)

The gap between the camcorder and SLR lens has to be the same or near to it, groundglass or not. Without the groundglass in place, the SLR lens focus can be used to move the focal plane and the camcorder focus can be used to track it.

The witness marks on the SLR lens barrel however will not longer tell the truth. There are complicated optical reasons for the distortion which are beyond me to explain.

(What creative options do I lose by not using a GG?)

As Ben suggests, you are unable to use a shallow depth-of-field to focus selectively on your subject and leave the background and foreground soft. The creative purpose of this is to direct audience attention where you want it to be.

The human eye at a fixed point of focus also does not see as consistently sharply from 0 to infinity as the normal 1/3" camcorder. (Okay somebody will now tell me the human eye has better resolution.)

(With regards to the softness or distortion in the corners, is that a result of spherical and/or chromatic aberrations? Would this be resolved by using an achromatic macro lens?)

There are complicated optical reasons for the distortion which are beyond me to explain.

[Svyatoslav Pylypchuk]

Unbelievable footage!
If I did not see, would disbelieve :) But I can not do the same.
I use Mamiya 80mm f2.8 (63.5mm focal plane) and "B+W" macro lens +5d on hv20. Nothing is visible absolutely:(
Which one macro lens do you use?

[Rich Hibner]

This has been discussed before in a previous thread

http://dvinfo.net/conf/showthread.php?t=87243

I did the same thing and thought "wow, I don't need a gg". I did some tests, and seen a few things.

1st is you don't get that shallow dof like you would with a GG vs no GG
2nd is you don't get bokeh.
3rd is it's nice for close ups, but it's the same principle if you take your camera and take it about 10feet away and zoom in on the object. if you have this setup you don't have take and move the camera 10+ft away.

right click save as:
http://www.fileden.com/files/2007/1/...7;20Video1.mov

with gg: right click save as:
http://www.fileden.com/files/2007/1/8/612706/Family.mov

[Peter Chung]

Originally Posted by Bob Hart

The shallow depth-of-field effect available from the lens on front can only be achieved by using the groundglass or one of thhose horrensdously expensive optical relay adaptors.

(The apparent flexibility of backfocusing you mention is a product of the deeper depth-of-field characteristic of the camcorder lens. You may find interaction between the two focussing systems which will retain the relayed aerial image in sharp focus oveer quite a wide combined range. You will discover at the extreme thresholds one end or the other that the image will start to go soft or distort in the corners. You may also find that the camcorder autofocus works quite happily through the lens on front.)

The gap between the camcorder and SLR lens has to be the same or near to it, groundglass or not. Without the groundglass in place, the SLR lens focus can be used to move the focal plane and the camcorder focus can be used to track it.

The witness marks on the SLR lens barrel however will not longer tell the truth. There are complicated optical reasons for the distortion which are beyond me to explain.

(What creative options do I lose by not using a GG?)

As Ben suggests, you are unable to use a shallow depth-of-field to focus selectively on your subject and leave the background and foreground soft. The creative purpose of this is to direct audience attention where you want it to be.

The human eye at a fixed point of focus also does not see as consistently sharply from 0 to infinity as the normal 1/3" camcorder. (Okay somebody will now tell me the human eye has better resolution.)

Ben, can you take a look at the footage I posted and tell me what you think? I believe I was able to manipulate the focus point and selectively focus without the ground glass.

I don't track my camcorder's focus at all. I set my SLR lens to focus at infinity and point it at a far distance. Then I set my camcorder's manual focus until that far object is in focus. I don't touch my camcorder's focus and I am able to adjust the focus of the SLR lens and selectively focus.

I do appreciate all of your help and explanations. I hope you don't feel I am trying to argue. I am just trying to understand the limitations and how it all works.

I can confirm that without the Ground Glass, the field of view is much narrower. Meaning I have to zoom in more with my camcorder to get rid of the vignette than if I had a ground glass, which wouldn't require as much zooming in.

I would assume that using a ground glass would actually give me a wider field of view than my DSLR because DSLRs have a crop factor due to the sensor being smaller than the 35mm film size, right?

[Peter Chung]

Originally Posted by Svyatoslav Pylypchuk

Unbelievable footage!
If I did not see, would disbelieve :) But I can not do the same.
I use Mamiya 80mm f2.8 (63.5mm focal plane) and "B+W" macro lens +5d on hv20. Nothing is visible absolutely:(
Which one macro lens do you use?

I set my SLR lens to focus at infinity and point it at something very far away.

Then, with the macro lens on my camcorder, I zoom in until there is no vignette. I then set my manual focus to it's nearest focus distance (scroll the focus wheel on your HV20 all the way up). Then I move the HV20 closer or further away from the SLR lens until the far away object is in focus in my camcorder's viewfinder.

I have a wide angle lens that has a screw on macro lens as a part of it. It's really cheap and is not achromatic so the picture gets blurry and color fringes on the edges. Right now, I'm just experimenting so it's not much of an issue but I will eventually invest in an achromatic macro lens.

What the macro lens allows you to do is zoom in and focus at a closer distance. It can still work without the macro lens, but your picture will look like you're looking through a telescope like in the movies (black all around with a center circle that contains the image from the SLR lens).

[Ben Winter]

Originally Posted by Peter Chung

Ben, can you take a look at the footage I posted and tell me what you think? I believe I was able to manipulate the focus point and selectively focus without the ground glass.

I don't track my camcorder's focus at all. I set my SLR lens to focus at infinity and point it at a far distance. Then I set my camcorder's manual focus until that far object is in focus. I don't touch my camcorder's focus and I am able to adjust the focus of the SLR lens and selectively focus.

I do appreciate all of your help and explanations. I hope you don't feel I am trying to argue. I am just trying to understand the limitations and how it all works.

I can confirm that without the Ground Glass, the field of view is much narrower. Meaning I have to zoom in more with my camcorder to get rid of the vignette than if I had a ground glass, which wouldn't require as much zooming in.

I would assume that using a ground glass would actually give me a wider field of view than my DSLR because DSLRs have a crop factor due to the sensor being smaller than the 35mm film size, right?

Peter, I think you're misunderstanding the purpose of these adapters. By far, the single most popular reason people use them is to achieve shallower depth of field, similar to that of a "Hollywood" 35mm camera. Without a ground glass, you gain no advantage. Unless the size of the rendered image on a focal plane changes, the depth of field characteristics cannot change.

In the footage you presented with the stock HV20, it is clear that you already have significant depth of field separation between the chair and the background; indeed, you even pull focus between the two. In this situation, I would make the decision as the camera operator to NOT use a 35mm adapter in the first place--there would be no real advantage as you are framing what you want, with the DOF you want, so why change anything? In your footage with the 35mm lens in front of the camera, you'll notice you have the same depth-of-field characteristics as the original HV20!

Now, on to the more technical: There is a direct relationship between the size of the sensor in the camera, the focal length and aperture of the lens, and the amount (shallowness) of depth of field that one can feasibly achieve. In a situation, outside for example, where I want a medium shot of a person in-focus but I want the background to be "blurry" I can do two things: Open up the iris of the camera, back the camera up to use the telephoto end of the lens (aka: ZOOM IN) and focus selectively so the background is out of focus. HOWEVER, because of the limitations of the SIZE OF THE HV20's SMALL SENSOR, depth of field separation is limited. I can only zoom in so much before my medium shot becomes a tight shot, and I can only open the iris on the lens so much before it becomes overexposed. Even with the lens zoomed in as much as possible with the framing I want, and opened up while still exposing correctly, I still may not have enough depth of field to separate the subject from the background. Thus, I want to use a 35mm adapter.

Now I can use the depth of field characteristics of a full 35mm lens. This means selective, shallow depth of field even at medium focal lengths.

Try this out: Zoom out your HV20 all the way so it's as wide angle as it goes. How close do you have to put an object to the HV20's lens before the background must go blurry to focus on it? On a 35mm adapter, you may be able to put an object two, five, even ten feet away while using a wide angle lens with the same field of view as your HV20's wide angle. But on the HV20, the distance is in INCHES! Thus the advantage to shooting with a 35mm sensor is shown...you can have things within a reasonable distance of the camera but still have a solid focus separation between objects.

So by using a 35mm adapter, you are basically increasing the size of the HV20's sensor to 35mm by using a "phantom" sensor (the ground glass) in front of it. And with that, you are gaining all the advantages of 35mm optics, mainly the shallower depth of field!

But do not discount the optics of the standard camera, as you have found: if you can achieve a shot without use of a 35mm sensor, by all means! I'm sure some DP's in Hollywood have drooled over the possibility of shooting wide open indoors and getting everything in focus like you can with your HV20..because in 35mm, without a LOT of light and/or fast film, it's simply not possible!