Bolex Viewfinder Area

[Benjamin Sarabacha]

I've got a Bolex REX-5 (with Super 16 + Bayonet mount modification by Les Bosher, see photo)

I shot a test roll of Double X negative film with a live preview camera (uncropped) looking through the viewfinder.

Upon comparison of the live preview screenshot and the processed/scanned film (by Negativeland in New York), you can see that maybe 15% or more of the exposed image is not actually visible through the viewfinder (see photos attached). In fact, the image is not even centered the same, and the left and right width is much wider on the exposed image than the viewfinder would have you expect. This would seem to make framing shots a bit of a guessing game...or else you crop in quite a bit to try to match the viewfinder? 

Is this normal for a Bolex super 16 modification? Or can something be corrected to show the entire exposed area? 

[Dom Jaeger]

It’s always been common practice on movie cameras that the viewfinder frameline for composition is cropped in a bit from the full aperture. In the past it was because a projection frame always would crop the full image, giving a cleaner boundary than the gate aperture, and allowing for choice of aspect ratios and room for slight reframing. Even now, for a professional final edit it’s much cleaner to crop in from the often dirty gate edges, or past where the perfs might poke into the image. Depending on the accuracy of the mount recentering and the lenses you use, you might also find that certain wide angle lenses vignette slightly on the full S16 aperture, so a little crop in can protect for that. 

On some cameras the viewfinder groundglass has room outside the frameline showing the full aperture, but Bolexes use a hard mask on top of the reflex prism to show the frame, which is always a little smaller than the gate aperture. 
It‘s possible to remove that mask, file it larger and realign it to match the gate, but it is a bit of a job, requiring complete front disassembly and care with alignment.

One further possibility is that your viewfinder is not properly aligned, as there are actually two masks, one on the prism and one in the eyepiece. If not aligned properly, the eyepiece mask could be offset and further cropping the image created by the prism mask. You can usually see the second eyepiece mask edges through the viewfinder, as a darker frame just outside the image frame. 

[Mark Dunn]

Before the days of scanning there were huge areas on the film that you'd just never see on the screen, so this never mattered except for critical uses such as titling and to a lesser extent TV transmission, for which there was the concept of 'safe action' and 'safe title' areas within which you would keep titles or action. Sometimes these framelines were added to the viewfinder.

Now that you always see the entire negative there isn't really an answer for the older cameras, except Dom's modifications, unless you crop to the viewfinder image in post-production.

In the example it doesn't spoil your composition IMO.

[Benjamin Sarabacha]

Thank you for the replies.

I understand that some cropping occurs to hide the messy edges of the film frame/vignetting etc.

However, in this case it seems like an extreme amount of cropping occurs into the usable exposed image if I was to try to only use the viewfinder area as the final image. 

Furthermore, you can see the center of the viewfinder is not the center of the exposed frame. See photos 1 through 3. The viewfinder image has to be shifted to the left to match the center of the exposed image quite a bit. 

If I was to try to match the viewfinder and film scan framing exactly (to ensure I was only getting only what I wanted in the frame) by cropping the viewfinder image to 16:9, I am loosing a great deal of the usable exposed image. See the photos 4 through 6. Photo 4 is cropping the viewfinder image into 16:9. Photo 5 is the film scan cropped in to match that view. Photo number 6 shows the overlayed minimum crop on the film scan over the cropped in viewfinder image. That looks like a 15mm or 20mm lens perspective difference to me—not negligible in my opinion. I would say it looks almost like the same amount of cropping that would occur if this wasn't a super-16 camera and you wanted a 16:9 image from a standard 16mm frame. 

My theory is that on my camera at least, it seems the super-16 conversion did not center the frame viewfinder correctly, and that the viewfinder is not showing the super-16 area. 

I don't know that, but, if this is normal for super-16 conversion Bolex cameras, my only options to be sure of my framing is to crop into the viewfinder image and even more into the exposed frame (reducing quality and making use of even wider lenses) or else to guess at the edges of the frame based on this testing...

A note on framing: This example is a test shot and should not be evaluated on artistic grounds. In a real project with artistic merit, I consider it vital to know 1) what is in frame and what is not and 2) what is in the center of the frame and what is not. While there are many valid uses for this camera where frame lines definitely do not matter that much like handheld documentary shoots that may not care about their frame lines or even notice if it doesn’t match their view exactly, in a carefully crafted film project, I do not think it is asking too much to know where the frame ends and where it begins. Say you’re shooting a period piece and wanted to avoid framing the satellite dish that is sticking out on the edge of the victorian house in the background. How could you be sure it wouldn’t be in the shot? In this case it appears major cropping is the only sure-fire way...

 

Link to view images: https://drive.google.com/drive/folders/1wsYP6YykogEXmRimwAFr5uCCyaLb0aBW?usp=sharing 

Edited September 27 by Benjamin Sarabacha

[Dom Jaeger]

It's hard to tell from your photos but the viewfinder image does not look like a normal, clean 10x viewfinder frame. It has what looks like vignetting on the left side, which is not present on the film, so it's not the lens. How are you taking the pictures of the viewfinder image, from a DIY video split or a phone up against the eyepiece? Either way, it looks like there is cropping or vignetting happening there. So I can't tell what the actual viewfinder frame looks like.

So first step is get a better image of the actual viewfinder frame. Does it look like that viewed by eye, with a vignette on the left?

Second step is to determine if the eyepiece mask is aligned with the ground glass mask. As I mentioned before, you should see a second darker frame around the image border. The image border is created by the hard mask on the top of the prism, which is the ground glass surface. You can flip out the prism and see the mask mounted on top. Now there should be a second mask in the eyepiece, which needs to be aligned so that it does not crop into the ground glass framed image. I don't know exactly how Les Bosher used to do his conversions, so maybe he removed that mask. Anyway, just see if you can tell whether there is a second mask perhaps cropping into the first mask. That calibration is done by adjusting the right angle prism at the end of the viewfinder tube. It's fairly easy to remove the tube cover to look inside and see if there is a mask inside the eyepiece. There's a hole to access the one cover screw under the mag saddle, just need a long driver. The prism adjustment needs a two prong driver. 

If that is OK, then you either need to have the ground glass mask enlarged and/or re-positioned, or live with the framing you have. You can shoot a framing test (film a frame on a chart precisely lined up with the viewfinder frame) to know exactly what to extract from the scan to match what you framed.

Les wasn't doing great work in his later years (or someone else was doing the work for him), so it's also possible there are other issues at play. The front does need machining to open up the viewfinder window to see the expanded area, masks need expanding and careful positioning, certain optics need removal or replacement, the gate needs to be properly machined and centred. 

Easiest thing to work out and potentially fix the issue is to send it to a good Bolex tech if you have someone nearby. Maybe Duall or Pigeon Creek Camera or Austin Beatty.

[Benjamin Sarabacha]

Thank you for that information! 

 

This viewfinder video tap camera was a DIY setup using a microscope camera. I also have the bolfix video tap and took a photo comparing both cameras in front of a lens chart. I noticed the microscope camera (Amscope) was causing the vignetting seen, but there wasn't really noticeable cropping. The bolfix tap however does have cleaner border lines, but I don't think it is actually revealing more of the image. You can compare both images here: https://drive.google.com/drive/folders/1rglsX1RvTeWuhoYQwcouoSrD-Pn0B1Mf.

 

I also took some photos of the prism and masks (same google drive link) so that maybe someone can tell something from them?

 

If nothing looks obviously wrong, I don't think I'll send it out for checkup/repair, but we'll see..

[Benjamin Sarabacha]

Perhaps this is a clearer demonstration of the situation. Here is a preview image of the viewfinder without vignetting. Beside it is a film scan from my Epson flatbed scanner (home developed, scanned for this test). The camera was pointed on a lens chart and the viewfinder used by eye to line up the chart within the boundaries of the finder. 

 

Approximate minimum 16:9 ratio framing is indicated in red on the film scan. Comparing the two, it seems like the viewfinder is not showing the right side most of all--the exposed image goes completely passed the chart.

 

It does seems like the issue was exaggerated by the DIY preview camera setup...however it seems a bit wasteful to crop in to match finder image. Generally, it is not enough of an issue to worry about that framing, I'll just have to keep that edge in mind if I want to keep it...