Hi all,

Anyone experimented with 35mm optics on HD10?

I understand that Agus35 or Aldu35 adapters should do the job... but has anybody actually tried?

The advantages should be:

- Film-like Depth of Field (because of the optics)
- Excellent Follow Focus on the external lens without any problem to hit the marks repeatedly (compare to the current wandering servocontrols)
- Less need for ND filters as both adapters do consume a couple of F-stops naturally :)

---------
If you don't know what Agus35/Aldu35 is, it's a Mini35 adapter with rotating or static ground glass accordingly. Just make a search for either term on dvinfo.net for more. There's also a Professional Mini35 adapter from P+S Tek (Germany) available, but it's in the $8K price range, as far as I know - and they don't make it for HD10 yet.

We've been talking about this. Do a search under 35 mm lenses and HD10.

heath

I could only find Eric Bilodeau's quick mentioning of the HD10 and Mini35 (P+S made) tests.

But it doesn't seem like anyone here has practically built a aldu35 adapter for HD10?

(Please correct me if I'm wrong - and the link to the appropriate thread(s) will be apreciated.)

I thought someone was experimenting...Sorry...

heath

Alex,

I am with a small museum in Los Angeles making documentaries and intending to shoot our upcoming project on an HD-10. We have been following the Agus35 and Aldu35 threads with interest and have developed a prototype adaptor based around an older Canon Auto Bellows, Canon still lenses and the JVC camera.

After working in fits and starts on the project for a few weeks now, our initial reactions to the adapter are quite positive and seem to agree with your list of advantages. In addition to the far greater control of depth of field, we are most interested in the ability of such a system to give us greater and repeatable focus control. Although still very much in development with this project, we have been quite pleased with the image quality thus far. All of the good people participating in the Agus35 and Aldu35 threads have made a wonderful contribution.

David - what about the GG grain?

I imagine that with HD10's *much higher than DV resolution*, it is also much harder to make GG (ground glass) with such fine grain that won't be recorded by HD10?

I tend to like Aldu35 design better. Questions:

- What do you use as adjustable spacers?

- How did you make a grain-less GG?

- What did you use as a condensor lens?

- Did you even need a macro in front of HD10, and if yes, what +?

...and yes, the people on 35mm>Digital adapter threads are amazing... they made these threads the largest in DVinfo.net history, according to Chris Hurd :) But can we implement their solutions on HD10?

the lens length dont need a macro to focus on the gg.

cause the hd10 can focus real close to the subject. the 37mm lens is just great for the adaptor.

ive built an adaptor for my trv18. im just waiting for my hd10 to arrive so i will adapt it too.

i think i have to grain my gg a little more. for the resolution.

ciao

With short depth of field, how are you going to focus?
To keep HD quality, you don't want to end up with SD looking focus.
Just wondering out loud,
-Les

The general comments have been: "You just have to be good at pulling focus" This doesn't make sense if you can't even see well enough to know if you're in focus or not.

Virtually all of the test samples (SD or otherwise) I've seen have suffered from focus problems with such shallow DOF. I'll be that even the examples which 'look' in focus would soften up considerably at HD resolutions.

Medium and large format cameras use the ground glass (or 'focusing screen') with a loupe with high magnification to look very closely at tiny little areas of the image to check focus.

Possibly some serious optional magnification to the eyepiece would at least give you the optical tools to know that you're in focus. You could place your focus marks in a follow-focus situation using the magnification. You'd need to be able to disable the magnification (which would likely only show you a portion of the image) for real shooting/framing considerations. I'm not sure how you'd move such a magnifier around on the image.

Similarly, Just thinking out loud...

Les, Jonathon: in addition to my "normal" monitors, I use a 30" LCD HD monitor (it's not to heavy) to check the focus on the fly. Because HD10 only outputs SD via its Component Out during recording, I have a workaround to actually get an HD signal during recording. For that, I simply feed HD10's FireWire Out into the monitor. FireWire Out does have HD signal even during recording. Because 30" monitor is big enough, focusing is no longer a problem.

Daniel: what did you use for the Condensor lens??

Do you have any pics of your rig?

Also: HD10 has *much* higher resolution that DV cameras, so I'm worrying that Ground Glass couldn't be made fine enough for HD10 not to see the scratches... or?

Alex and all,

We have tried both the static and rotating (Aldu as well as Agus) version of the adaptor (as well as a couple of vibrating ground glass systems that proved to be a bit less hopeful.) My instincts suggest that either the Aldu or the Agus could be made to work even with the HD-10’s increased resolution.

However, for the moment we have opted for the classic Agus with spinning ground glass primarily for personal aesthetic reasons. Though we seen very hopeful results from both systems, we seem to prefer the liveliness of the image created by the spinning glass. I’m very sympathetic to the appeal of the size reduction and simplicity of the Aldu, but in our case the camera is riding at the end of a little motion control rig anyway and a couple of inches here and there will not really make much of a difference.

We are building our rig around a cannibalized Canon Auto Bellows macrophotography device which has the advantage of providing essentially a small optical bench on which to adjust the components.

Our ground glasses have been made after Alain Dumais’ method following Dick Dokas’s excellent tutorial using 600 grit aluminum oxide.

We are using condensers – actually 2 at the moment which we aren’t ecstatic about (There seems to be far too much glass in this contraption). Still the images are looking pretty hopeful even with all of that glass.

We are, in fact, using diopters on the front of the HD-10. Currently we have +7 and are waiting on a +10. As Daniel pointed out, it is definitely possible to focus close enough with the HD-10 to dispense with the diopters but we have found we’re getting better results at this stage with rather than without. Still, as is the case with most such devices reported in these pages, ours is very much a work in progress and will undoubtedly change over time.

Les and Jonathon’s points about even establishing good focus are well taken. We are using Aspect HD and have high hopes for Phil Wright’s 1394 laptop-as-field-monitor application. We anticipate putting a motor on the Bellows front plate which, with any luck, will provide something like controllable focus - a real glimmer of hope after struggling with the lack of control of DV focus systems like we all have for so long.

We are painfully aware that a rig like this is far from ideal, but in our experience to date, the advantages are seeming to far out weigh the disadvantages. Thanks again to all of the people who have propelled these thoughts ever forward.

David said:

>>Though we seen very hopeful results from both systems, we seem to prefer the liveliness of the image created by the spinning glass. I’m very sympathetic to the appeal of the size reduction and simplicity of the Aldu, but in our case the camera is riding at the end of a little motion control rig anyway and a couple of inches here and there will not really make much of a difference.<<


Actually, with Agus35 it's not only the size - it's also:

- Noise factor (motor vs. no moving parts in Aldu35);

- Need for the extra power for the motor;

- Unwanted Gyro effect of the rotating GG;

- Larger GG must be harder to produce - compare to the use of pre-fabricated clear UV filter with its existing fitting as GG in Aldu35;

- It must be much harder to ensure that the rotating GG is strictly parallel to the condenser lens, than when GG is static.

You said that Agus35 is more "lively" than Aldu35. What exactly do you mean by that, please?

Alex,

I agree with everything you have said about the Agus35, noise, power, gyro effect, larger ground glass.

Still, in our particular case, we use dual system sound; the power is 2AA batteries that seem to last forever and the whole contraption rides on a pan tilt head on dolly and track all run by motors so that the gyro action of the disc has not been noticable to us. And then again, it may all change and we may go back to the static ground glass after all.

But for now, what we have seen in all of our tests is that, when projected, (in the venue that our finished movie will be seen (1280X720 projection in a small theater)) there is a subtle but to us visible difference between the static and the moving ground glass.

With a lock-off shot using our static prototype, the image takes on (what can be a beautiful) appearance of stillness almost like an old photo or painting while the spinning gg somehow provides, even with a lock-off, a more kinetic sense to the shot. For the movie we are preparing the latter is more to our liking. I'm sorry not to be more specific. I think this is a really very subtle distinction and, were our shooting conditions otherwise, I'm sure we would be choosing the static ground glass approach instead. To me the important thing is that these contraptions actually seem to work and provide us with better control and more pleasing images even with the HD-10.

David, what your spinning GG is made of?

Some posters suggested Bosscreen as grain-less GG.

Supposedly this should eliminate the issue with the visible grain on GG - and take away the need to spin GG as well, I guess (because if you can't see GG's grain structure, then what's the difference whether it's moving or not?)

However I could not find any reports of the actual Bosscreen experiment as GG.

Alex,

Many thanks for the suggestion of Bosscreen. I had missed those posts.

Sadly our spinning gg is at the moment plastic. We are having optically flat, 40 thousands thick glass discs made by a scientific glass company here for a quite reasonable price but we will not receive them for a couple of weeks yet. The images created even by the spinning plastic are quite good really so I'm hopeful about the glass. Still, we've invested so little into this that we would change directions in a heartbeat if it were a better approach.

Have you made or are you considering making such a device?

David: >>are you considering making such a device?<<

Yes! But I'm not very proficient in Optics at all, so I'd like to have some proven, step-by-step instructions with the parts numbers etc. before I begin, if at all possible...

Alex,

I'm not exactly an optical engineer either but I have been surprised how relatively easy it has been to create a contrivance that works, at least, acceptably well. We're feeling pretty confident that we will use some version of such an adapter on our upcoming project for at least some if not all of the shots. How the with and without shots would intercut remains to be seen.

If you were only a bit closer I ask you to drop by and show you what we've been able to cobble together thus far.

My name, for the record, is David Wilson (not Wislon) - a long standing miss-type I just recently discovered and am just now trying to correct.

...
i too am constructing a mini35 adapter for my hd10. its almost done, i just have to finish refining my GG after finals this week. i plan on shooting a short film with it after its constructed. i bought a +10 macro lens for it as the close up focus wasnt good enough on its own. so far tests have been promising....definately will need a lot of light. and right now im using a 50mm f2.8-16 lens. i should have some tests to show once i get all the scratches out of my GG and get everything cemented together.

Original HD10 lens system, although controlled by servomotor, does Not accept remote controls a-la LANC. Also, manual rotation of the focus ring has the weirdest lag that makes it impossible to hit the same focus marks twice.

35mm adapter should fix this problem, as the quality lens is now attached externally. So manual follow focus should be no problem with the adapter (what are you guys using for FF?)

But let's go 1 step further. I noticed that practically *all* of the optics is AutoFocus lenses - which means, we can control the focusing servomotor inside electronically, and possibly have a completely electronic follow-focus system - with the advantage of *not* touching the camera/rig while focusing.

Is this feasible? Has it been implemented? Am I crazy and this can't be done?

Auto focus 35mm lens's usually have a drive socket that the camera body turns. It would be possible to disable the electrical connection to the video cameras focus motor, and bring the connection outside the JVC to a connector. A small gear motor could be mounted to the 35mm lens and the JVC could then focus that. Most people would not be keen on hacking the camera like that, I think.
This would solve some of the problem of getting good focus with the short DOF lens.
-Les

Les, that is not what I meant.

I proposed to use AutoFocus motor in 35mm lens as means of Follow Focus.

The question is, whether anyone knows what protocol AF servomotors use in 35mm lenses, so then I could make a totally independent (from the camera) electronic device that would be programmable to do Follow Focus electronically.

Just to clarify: I'm not talking autofocus at all.

35mm AF lens's don't have a drive motor. The camera body has the motor in it. It's driven by a coupling to the lens.
See my above post.
-Les

Thanks for clarification... I guess I got it backwards. So there's a Sensor, and not a servomotor in 35mm SLR lens? In this case, looks like the manual FollowFocus rig would be the best bet.

BTW, I'm working on the aldu35 adapter, but with a twist. I'm using old Nikon F body (same like F2) as SLR lens adapter and GG holder.

For GG, instead of the UV filter, I currently use D-type Nikon focusing screen (FS).

My current version has the FS removed from its position on top of the camera, and rather attached straight behind the lens, where the film is supposed to be. (Of course, the 45deg. mirror must be out of the way as well).

This way, HD10 is positioned straight behind the adapter, and not at right angle vertically to the Nikon - as others did.

Advantage: the whole rig can be mounted normally on tripod etc.

Current problem: the ground glass (brand new focusing screen) does not seem to be fine enough. HD10 can clearly see the grainy structure of it. Waiting for the fine aluminum oxide powder to try and improve the FS so the grain is no longer detectable.

Will report on progress (anyone else?)

No sensor in the AF lens either. The focus is determined by the optics and sensors in the SLR body as well.

Back to what I thought was your intent: I like the idea of driving the focus of an external lens from the JVC's servo signal. It would not know that it is actually focusing an outside lens.

-Les

Off-topic, but I just need to understand... you wrote:

>>No sensor in the AF lens either.

So no sensor and no servomotor in the SLR lenses? Then what the electric contacts on the AF SLR lens are for?

F-stop info, and lens ID, perhaps focal length. Depends on manufacturer.
Do you have an AF SLR lens? Take it off, look on the back, and see the drive coupling, it may look like something a small screwdriver would turn.
Trust me, manufacturers are smart, they don't want to put a motor in each lens, when one in the body will cover any number of lenses. Same for focus electronics.
-Les


<<<-- Originally posted by Alex Raskin : Off-topic, but I just need to understand... you wrote:

>>No sensor in the AF lens either.

So no sensor and no servomotor in the SLR lenses? Then what the electric contacts on the AF SLR lens are for? -->>>

OK, while waiting on the aluminum oxide powder to refine the ground glass, I decided to make yet another experiment.

I thought - why not use the 35mm photo film itself as ground glass?

It's already of the correct size and acceptable granularity.

So I shot uniformly white and grey out-of-focus surfaces of different intensities, and developed the film.

Result: failure. The film does not seem to "catch" and diffuse the light from the lens like Ground Glass does. Instead, I can see right through the film and into the lens, no matter how dark or light the film is. Bummer!

Seems like aluminum oxide and refined Ground Glass is about the only option now. I'll report on progress when the powder arrives.

Seems like with HDV resolution, the GG grain issue becomes catastrophic.

After some experimentation with grounding of the UV filter glass, which proved still unacceptable even with the very fine grit (started with 9, finished with 25), I'm waiting on the Bosscreen material... but I don't hold my breath for it either.

So, is there positively no way to eliminate GG altogether? What if we use not 35mm SLR, but 16mm movie primes, thus making the "projected" image smaller and closer to the native size of the prosumer cams' CCD?

I'm willing to rip the JD10's original lens system out and replace it with the PL-mount and 16mm Zeiss Superspeed lenses. Why won't it work?

If you removed the lens, the camera control electronics would error out and probably lock up the camera.
Why not just move the GG to get rid of the grain?
-L

Les, moving GG means noise, possible vibration, definitely *much* larger size than static GG system, much lheavier weight of the adapter, and power concerns.

Will I go rotating GG route? Yes, if Bosscreen won't work. Supposedly, this should be clarified tomorrow, Friday.

Bad news with Bosscreen, folks: although it has by far the least visible grain, I can still clearly see the paraffin structure - looks like fine "smooth" grain. SD cams probably wouldn't see it, but my HD10 does show it perfectly on HD monitor.

Dang!!

I guess Germans tried all that before making the adapter with the ROTATING glass.

Okay, plan B.

Who knows of an optician to manufacture a thin, 3-5 micron grit ground glass with the exact dimensions of a CD?

Keep in mind you don't have to rotate a huge disk the size of a CD to get rid of the grain. I'm working on a system that orbits a GG barely bigger than the 35mm frame.

-Les

Just tested the vibrating GG - successfully!

I simply used the Nikon's focus screen as GG, and attached a micro vibrating motor from old cell phone to the side of the GG's frame.

While GG stands up vertically, the motor is attached horizontally to its side.

As a result, the motor's vibration does not change the GG's distance from the cam's lens, but rather moves GG across the cam's view.

Result: it definitely eliminated all small imperfections which were otherwise highly visible with static GG.

Larger areas where GG's structure is uneven were less affected - due to the fact that the amplitide of the vibration is rather small.

So if GG is quality ground with say 3-5 micron grit, no scratches/blemishes/eneven areas, then my test shows that the vibration will completely eliminate any visible grain.

I'm now cautiously optimistic that a HD QUALITY, wonderfully small home-made adapter is possible.

Because no-one (unless i missed it) produced such adapter, I guess we should dedicate this thread to it and call the adapter vibro35.

(It seems custom to call adapter variants by the name of their developers, but alra35 just does not sound right to my ear :)

Good news Alex!

It's very important not to have the GG move in the in and out direction, as the lack of depth of field will cause an out of focus final image, making the HD image look like SD.
Post a still image grab if you can. I'd love to see the DOF effect we desire with this HD cam!
-Les

please,
post a full native mt2 hd resolution clip!!!


we need to check it out for ourselfs

thanks

ciao

After checking the 5 micron GG made of the 52mm UV filter, I have to say that the original Nikon focus screen is still better in terms of the fine, consistent matte.

Will continue experiments with it later today.

I sure will post more detailed info as soon as it is available.

Please note that I'm not a qualified optician or craftsman, so my efforts are slow and expensive to me (buying tons of different stuff for experiments). I'm still waiting for the delivery of the set of micro screwdrivers that would allow me to have the right tools for this work. I'm more into electronics, you know.

However I'm very enthused with the fisrt tests of the vibro35, and will continue developing it until the results are definite (be it positive or otherwise.)

Hi Alex.

Your Vibro 35 sounds like the ticket.

Following all the exchanges I think the static GG is a dead end,
As mini DV gets sharper the grain will resolve needing finer grain et,- chasing one's tail!

Spinning the GG takes lots of space and rotating it doesn't sound right.-There will always be a pivot area somewhere on the GG that is not moving and be sharp- like time exposure of star field.

I've tried moving the GG like a wig wam- side to side- but again at the ends of the oscillation the screen will be static.

Vibration - one one plane is the way to go I'm sure.

How is it working out???

Vincent

I'm working on the Vibro35 version 0.7

Should be finished Monday with the test footage etc.

I'll try to post the pics of the rig, and the footage stills ASAP after the tests.

Alex, how did you isolate the motor from the housing so that the vibration won't affect the camera? It could potentially become a problem on telephoto... just thinking out loud :) Keep up the good work.

Hi Alex,

sounds great- looking forward to pics.

How did you find Nikon screen without markings??

I've had to resort to an Exacta reflex camera from the 1950's to get a plain GG- it is a chunk of glass and may have too much mass for your vibrate breakthrough.
In which case I may resort to a small electric motor with a bevel gear.

PS- I think we have to call it something with your name on it- Vibro35 sounds like the stuff people try to sell me on Spam mail.

keep up the brainstorms

Vincent

Vincent, Nikon has focus screen type D that is completely clear of any markings.

That's what I use as GG.

It is sold for about $30 at B&H (http://bhphotovideo.com (http://www.bhphotovideo.com/bnh/controller/home?O=productlist&A=details&Q=&sku=37515&is=REG))

Glass GG is probably a no-go as it is times heavier than the plastic D-screen in its tiny plastic frame... thus the glass will require much more powerful vibrations and they certainly will transmit to the camera's lens as well...

Stay put for the photos of my tests later today, hopefully.

>> PS- I think we have to call it something with your name
>> on it- Vibro35 sounds like the stuff people try to sell me
>> on Spam mail.

Hmmm, we dont want any obscene names for such a beautiful thing as mini35 adapter...

how'bout this: if you guys do like my results, then you can call the adapter alex35.

If you rather don't like the results, then it's vibro35 :)

Deal?

Ok- it's a deal Alex

I'm ordering the screen you recommend

Thnks--Vincent

So how is it working out?

I've been alternately incapacitated (operation)/insanely busy lately, so sorry for no progress yet.

I'm dying to continue working on it, should be able to do so later this week!

Yikes... I like your humor !!!!

Take care , I'll stay tuned for when you have time to play again!
-Les

I hadn't followed this thread for a long time, but things are now sounding very interesting. I look forward to more info on the "Alex35".

P.S. I gather these tools flip the image backwards and upside-down and that this will be corrected in post. We were just thinking if there was interest we could put the image flip as part of the HDLink capture app. This would take no CPU time and save a bunch of post processing steps.

I can't speak for anyone else, but the rotating in post is such a simple operation that I almost view it as a non-issue. However, I'm sure it wouldn't hurt to have it as a bullet-item on your product, which would also allow you to appeal more generically to the seemingly growing number of home-made adapter-builders. Despite the ease of the rotate, I have stumbled across a not-so-small number of people who don't seem to see how to do it easily. Your inline rotation during capture would appeal to those folks for sure. Doesn't seem like it would be a bit hit to your dev team either...

It is true that users who making such devices are probably not all that concerned with workarounds in post. :) Still if someone had to do a lot of work with a flipped image, I guarantee your current workflow would be painful compared with it being handled automatically. ;)

P.S. I just looking for the excuse to do it. This is very cool.

Don't get me wrong - I was trying to make the point (eventually) that you should do it! In fact, I'd probably use it if it meant one less step in post. I'm biased because in my particular case, I run *everything* through post, so its a given in my process that I'll have a certain amount of simple 'cleanup' tasks (de-interlacing, rotating) prior to any post-effects work or grading. I think an inline solution would be great - so there's your excuse!

David, Jonathon: thanks for your interest in this adapter.

I'm on the path to use piezo elements as an oscillator for the ground glass (rough idea). I don't like micromotors too much, because their vibration is uncontrollable.

In my version of the adapter, camera is mounted *upside-down* in front of the adapter, thus capturing the *correct* image on tape/outputting it on the production monitor.

Therefore no problem with flipping in post - it simply is not necessary in my case.

Also, this way you can combine any number of segments on tape shot with and without the mini35 adapter. They all will have "normal" image orientation, because without adapter, the cam is obviously not mounted upside down.

David, you sure can introduce the flip-switch in your capturing software, but then you're assuming that the whole tape is shot one way or another - say, with mini35 adapter - which may not always be the case in practice.