Culmination of a month of DOF research

guskersthecat

The bokeh doesn't bother me from your footage because I know its dependent, at times, on your 35mm lens.
As for the 'soft glow' I have experienced that in "Version1" of my adapter which used a cheap magnifying lens instead of the present stack of 3pcs +3 close up filters. My suggestion to yours is, try changing your macro or close up to a diff brand to remove the softening or subtle halo.
Also, I noticed that the close up filters, when not placed close to the videocam lens, softens the sides of the image, or gives an uneven focus.
 
Ok, thanks for your feedback Ted. I've got a few more experiments to try out. We used fairly expensive lenses for this setup...the DCX condenser and 120mm achromat. I'm going to try a version without both of these, as I may not need them with the GS400. The condenser reduces frame size, but the 400 has a 1.6 inch focus distance...which would cover a 24x36 frame. Hotspots will the issue.

Did you check out the racked focus in the clip? I'm seeing a definite DOF issue extending across the figurine. So am I missing something?

Just a caveat. I went through a day of experimentation to address some colour blooming, only to find you that it was the GS400, and not the adapter! Folks critiquing the footage suggested ghosting due to insufficient diffusion. I later tested the camera alone to find the same ghosting (blooming in this case). I guess it's just a suggestion to others working on these things to make sure you're troubleshooting an adapter issue, and not a DV compression/camera artifact issue.

Thanks again, I appreciate your feedback!
 
guskersthecat said:
Just a caveat. I went through a day of experimentation to address some colour blooming, only to find you that it was the GS400, and not the adapter! Folks critiquing the footage suggested ghosting due to insufficient diffusion. I later tested the camera alone to find the same ghosting (blooming in this case). I guess it's just a suggestion to others working on these things to make sure you're troubleshooting an adapter issue, and not a DV compression/camera artifact issue
Click to expand...
That's right! But how can one discuss the adapter's performance when the camcorder itself is in question?
Cameras by themselves take lousier pictures (still or video) here than they do using the image converter. Here is a repost of two stills 2.3Mp from GS200:
This is the camcorder:
http://dandiaconu.com/gallery/album05/IMGA0486?full=1
and here is the same camcorder photographing the GG (Nikkor 4/200 on adapter)
http://dandiaconu.com/gallery/album05/IMGA0485?full=1
 
not a clue. (one year ago... one test of who knows how many..)
wild guess: ? wide open (camcorder) Nikkor @4 (wide open as well) shutter could have been @15 or 30. (I can see the gain noise and that's why I guess I must have had the aperture wide open) Both had the same settings though (whatever they were)
 
I'll have to try this out with my GS400. At close range it takes very respectable stills (but nowhere close to the 4Mp Panasonic claims). I've taken some stills with my adapter attached, and of course they looked just like pics from my SLR, but not as good of course. I have not tried the same scene with and without the adapter.
 
Tedd, the previous pics are 2.3M stills with/wo MPIC and moving. I have some static to demo the difference (grain on soft areas) like this: (static)
http://dandiaconu.com/gallery/album13/A_static?full=1
vs. this: (motion)
http://dandiaconu.com/gallery/album13/A_motion?full=1
The above two are 720/480 frames from the footage.
I noticed ALL stills/footage being SHARPER using the adapter. Reason?
The gg itself reduces the light's contrast ratio bringing it within the camcorder's limited response range. End result? The "sharpening" circuit does not "trigger" unless there is a high contrast between reflecting surfaces. Going down to pixel level, a white line will "bloom" in the adjacent line of pixels thereby reducing the resolution of the CCD in HALF!(if not worse) Talk about resolution charts. See countless examples when the lens of Z1 is blamed for the CCD's inability to respond to highlights and so on.
 
Thanks Teddy, the quality is neither better nor worse than a Beattie can display and a camcorder can record. They make the best screen (I am aware of) and I move it. That's all.
 
It seems the beattie has the same optical quality as the focusing screens on the hasselblad cameras, like when you look through them it seems brighter than the actual scene!
Im trying to finish an occilating frame Dan,than can hold the focusing screen of the mamiya rb67. I hope it turns out ok. Im closely supervising our craftsmen thats making a magnet based ocillating device. It would make the adapter smaller.
 
Beattie is actually one step forward from original Hasselblad screens.

I have two Kowa medium format lenses (80 and 150) and I contemplated making one but did not have enough interest; not enough lens options for the format.

Why not order a medium format MPIC from me? (I do custom work as well).That would be the first and last one I would make.
 
Hi Dan. A prime or even a cheap low zoom ratio lens will be superior to any video large zoom ratio lens regardless of cost. It's just physics. All good cine lenses are primes for a reason. The ENG style 20x (or any type of camcorder 12x or 20x lenses) are a huge compromise and have got tons of abberations in most parts of the zoom even when you pay $30,000 for them. Nobody makes large zoom ratio professional photography lenses for the same reason. Better than 3x zoom ratio like 24-70 or 70-200 would be a very bad design. And primes are always preferred of course. Some have 0.2 pixel maximum abberations at 11megapixels compared to 5pixels for cheap zooms.

In your adapter setup you are using the camera fixed lens in a very sweet spot and that makes all the difference in abberations. The camera has more resolution without the adapter but nobody would notice that since the output resolution is very small for the capabilities of the optics in both cases and the abberations will be very bad in low quality smaller than 1/3 inch lenses zoomed in. The loss of contrast is also turned into a plus since the digital cameras cannot handle large contrast ratios. So that leaves loss of light as the only real optical minus of an adapter.

Any professional should take care of highlights using ND or gradient filters and proper exposure settings to make it work within the limits of the equipment. Most samples with blooming and blown highlights usually come from kids using automode or having limited knowledge on the subject. Or tiny little $10 sensors on $5000 to $10,000 "prosumer" cameras. Film is better in dynamic range than modern 2/3inch sensors in real cases but is used as a reference in marketing. Every new product is more filmlike:) We don't want it to be the same with film in exposure on motion because then we wouldn't be able to make the next product more film like:) CMOS cameras are very good in the highlight artifact department. Still in the hands of a kid film would produce similar results with digital (probably worse!).

It is interesting that even if I scan and compress a film at 1mbit/sec it still looks film like with all its blocks, low res, terribly low S/N, 8bit best case dynamic range etc etc. Of course. It has productions values and professional photography. The user is not shooting natural light on a bike in auto mode! Some things do not come from B&H in a box.
 
Guskers, do you feel that the adapter will perform well enough to work with an Andromeda (uncompressed, HD output) equipped DVX?
 
Yes, I believe you would have no problems...here's why:

1. The adapter image is very sharp, free of distortion, and moving grain.

2. Stills taken at 2288x1728 from the camera with adapter attached look very good.

3. I've set up this adapter to work with the GS400 16:9 mode, which uses a wider field of view than the DVX. This means you would have to zoom in 15% or so less than I do, which means more flexibility with shutter speed, and a slightly better image than what you see in the clips. All of the GG's work fine at 1/120s shutter speed, which gives you an idea of how well GG grain is being blended.
 
guskersthecat said:
Yes, I believe you would have no problems...here's why:

1. The adapter image is very sharp, free of distortion, and moving grain.

2. Stills taken at 2288x1728 from the camera with adapter attached look very good.

3. I've set up this adapter to work with the GS400 16:9 mode, which uses a wider field of view than the DVX. This means you would have to zoom in 15% or so less than I do, which means more flexibility with shutter speed, and a slightly better image than what you see in the clips. All of the GG's work fine at 1/120s shutter speed, which gives you an idea of how well GG grain is being blended.
Click to expand...

2) Could I see some of these stills? Or are they in the thread and I missed them?

3) How does zoom affect shutter speed? :embarasse

BTW, your kid is adorable. :)
 
Here you go. They are stills captured from the GS400 with the adapter attached at 2278x1728 (resized for web)

GS400 still shot 1, adapter attached.
GS400 still shot 2, adapter attached.

Note that with the 50mm f1.4 at this range, the DOF is extremely shallow...only about an inch. Keep in mind too that the GS400's so called 4MB stills are more accurately compared to a 2.5ish MB digital camera with respect to actual resolution.

Thankfully our toddler seems to have inherited better looks than her genetic contributor!

The zoom/shutter speed issue is related to grain. The less you zoom in, the higher shutter speed you can use without moving grain showing up. If you are forced to zoom in a great deal and or increase shutter speed, eventually moving grain will become apparent in any moving GG design. The only way to avoid this is to move the GG much faster than 2700 rpm, in which case noise and vibration become an issue.
 
Last edited:
guskersthecat said:
Here you go. They are stills captured from the GS400 with the adapter attached at 2278x1728 (resized for web)

GS400 still shot 1, adapter attached.
GS400 still shot 2, adapter attached.

Note that with the 50mm f1.4 at this range, the DOF is extremely shallow...only about an inch. Keep in mind too that the GS400's so called 4MB stills are more accurately compared to a 2.5ish MB digital camera with respect to actual resolution.

Thankfully our toddler seems to have inherited better looks than her genetic contributor!

The zoom/shutter speed issue is related to grain. The less you zoom in, the higher shutter speed you can use without moving grain showing up. If you are forced to zoom in a great deal and or increase shutter speed, eventually moving grain will become apparent in any moving GG design. The only way to avoid this is to move the GG much faster than 2700 rpm, in which case noise and vibration become an issue.
Click to expand...
Those stills are beautiful.

I'm pretty much sold...I had been thinking I was gonna get a Guerilla35, but it seems yours competes very well with it for a much lower price. However, I'm also keeping an eye on bklyndv's.

Still waiting for the shootout and for you to finish your final flip version before I make a decision.

Oh, and my next paycheck. :D
 
Last edited:
Thanks V. Well, worst case scenario I'll have an adapter collection on my hands! Given the collection of engineered diffusers and carbon fibre on my desk, I suspect the next adapter will be quite different. Heavier, costlier, but likely about .5 to 1 f/stop better with better diffusion, and 0 grain.
 
You must log in or register to reply here.
Back
Top