This is a post that I will be updating as I continue to learn and experiment with the M2 from Redrock Microsystems.

First off, they were excellent about getting it here in 4 weeks. Their shipping seems to be top notch and very well packaged (the $120 due to customs was a pain in the rear!).

Anyhoo, I was pleased to see how sturdy all their contraptions were, and the actual product seemed solid. I was a little dismayed at the lack of any PRINTED instructions on use. No handy diagrams or instructions about adjusting the backfocus (more on this later). Simply a URL with a few tutorials. I would have perferred a book, but what can you do?

Anyway, I soon realized how much set up there was to do. First, I had to remove the spinny GG and then use an allan key (inside the tight little box) to remove the rail adapter and add some shims. Took a few tries and half an hour or bashed knuckles, but I got it to the right height.

Now comes the tricky part... focusing. It is my understanding that the GG must be moved forwards or backwards to achieve Infinite focus (critical) and I suppose backfocus too. There are tiny screws called the "collets" under the GG, and we are supposed to remove the GG (easy to scratch) and loosen these screws a tiny bit and then slide the GG holder forwards or backwards (which way for what excapes me at the moment) by a millimeter and then tighten it all back down (evenly so the GG doesn't wobble).

Okay... that is done (theoretically, I still havent gotten the right allan key for the collets). Seems like that is all the set up :)

From my few early tests, the GG seems Top Notch and the construction and design is excellent. Some of the screws are overtightened at the factory and could use some muscle. Also, the rail-clamps seem overly tight and are a bit tough to slide on a 15mm rod.


I know this seems overwhelmingly negative, but I am happy with the product thus far, and I am sure as I learn to use it properly those consternations will turn to adulation...

More to come.

Redrock Micro M2

Second Impressions:

Well, they say you can't unmake a first impression. I beg to differ.

After a good night's sleep and some breakfast, the M2 is looking like the perfect little addition to my Video-making collection.

Now that I have adjusted it to perfection, the image quality is truly stunning. There is slight fall-off towards the edge of the frames, but no real vignietting, and the loss in light is almost nullified if using a 1.4 lens on the other side.

I'll post some footage soon... but I am impressed.

Matt

I hear ya Matthew. My first adapter footage was a bit of a disaster. Once I had things figured out and screened my first footage, I was struck by the images.

Hey Dennis, where in Ontario are you?

get any footage uploaded yet matthew? (one thing i just noticed... 3 users from eastern (i dont know if you guys from ontario consider yourself east or not) canada... with 3 different 35mm adapters.. and around the same amount of posts... cool..you guys want to meet up and have a shallow DOF party?)

Redrock Micro M2

I'll post some footage soon... but I am impressed.

Matt


Hey Matthew,

It'll be great to see some of your footage. There aren't a lot of samples from the XL2 M2 combo on the RedRock site.

I just read the M2 fared very well in a comparison test on the set of the tv show "24." The testers were using the new crop of HD 1/3 cams, but the DP's were impressed the M2. The freeview article is at this link:

http://www.showreel.org/memberarea/

I did some fairly elaborate tests using the M2 combined with all the cameras mentioned in the Showreel article, and though it was not touched on specifically, the M2/Canon XL H1 combo (using the Canon factory lens) generated excellent results.

Taylor Wigton
DP, Los Angeles

You need to be a member (payment) to view the article.

You need to be a member (payment) to view the article.


Hello Leo,

Actually, it's free to view those articles. Yesterday, I saw a post from the Showreel publisher indicating there was a lot of interest in the articles, so they were made to be viewed as "freeview." I read them yesterday, and I'm not a member, yet. I may subscribe because there were a lot of interesting articles on the site. Just go to the "Cinematography" section and click on either of the two articles about "HDV on the set of 24." Pretty cool website!


I did some fairly elaborate tests using the M2 combined with all the cameras mentioned in the Showreel article, and though it was not touched on specifically, the M2/Canon XL H1 combo (using the Canon factory lens) generated excellent results.

Hello Taylor,

Those were great articles. I enjoyed those very much.

Sorry Matthew. I just read your post from a week back. I'm about 16 hours northwest of Toronto :-)

Cool Dennis, we should get together some time...

Sorry I haven't posted any vids yet. Just got back from Goderich Ontario shooting some super 16mm so that has been consuming my life (and chequing account!). I haven't shot a tonne cause I am waiting on a 7" VGA LCD screen so I can Run&Gun, have an image flip, and be able to critical focus!

Keep your eyes on this post... I'll have something up soon...

Also, what is the best compression for net. I don't want to lose too much detail, and want to keep it anamorphic... anyone? Does WMP support anamorphic?

Windows Media Encoder supports any pixel resolution you'd like. I'd use 872x480 sqare pixels if you're just talking 16:9 anamorphic. For the sample footage I've been posting of late, I'm using 3000Kbps. WME's MPEG4 at 3000Kbps is pretty close (maybe better) than MPEG2 at 9500Kbps. So far no one's complained they couldn't see the clips.

If you're looking for more compact files, try 436x240 at 500Kbps.

And yes, if you're ever up on the North Shore, look me up!

Hi Matthew,

Any second impressions?

Second Impressions coming....

Need webspace to post footage!

Hey Matthew,

So how goes the XL2/M2 testing?

Since there appears to be a lot of interest in a lower cost relay lens by many XL series owners, I was curious if you had had any contact with the RedRock guys about developing one for the XL series. I know there is discussion on the Cinemek forum regarding a future XL relay lens, and I know the Letus35Xl has a few sample images posted. Most people are pleased with the footage from the RedRock M2, so it would be interesting to see what those guys could do with a relay lens.

I appreciate all the info you post.

I'm pretty sure Jamey has this on his mental plate. There has been much talk of it on the Redrock forums.

Yes, I would love a redrock XL relay lens provided it reamined relativly cheap. I think the usability of the rig (Expecially the XL2 comapred to the JVC or any other cam cause the XL lens is humungous) would be vastly improved with a much shorter relay lens. Provided it has an iris and ND filter, it should be good!

Matt

Has anyone tried using a Nikon marco lens as a relay lens? I'm new to this forum, but it seems like many folks would like a shorter lens to use as a relay lens with the xl1/2. For example, a 55mm nikon macro is designed for close focus (such as the distance of the ground glass on the Mini35) and it has an adjustable aperture, but I don't know if it is otherwise suitable. These lenses are high in quality and readily available.

a macro lens would have no focusing abilities or iris control. It couldnt do the job of a relay

Hi Matt - I just found this thread and was wondering if you had any sample footage taken with your XL2 using the M2. How are you liking the results. I am looking for a wide angle alternative to the Canon 3X lens. How do your wide angle shots with the M2 look?

What we commonly call "macro" lenses are often close-focus diopters which don't, infact, have focusing abilities or iris control. However, what Craig was suggesting, I believe, was that we might use Nikon (or Canon, Pentax, etc.) lenses designed specifically for macro work (there are several extremely good ones out there, and all but one have both iris-control and focus, one has a fixed 0.95 iris, I believe).

The thing that I can't answer for him is how much space it takes to get from a Nikon mount to an XL Mount (1/3") and how that would affect backfocus-- whether this is a usable solution or not. As I recall there was a company that strictly made XL to film-style adapters (before the rage of the DOF adapters). They may have a second life, as a relay lens adapter source.

-Jack

I was considering this, but I am having some trouble getting things together.

My biggest problem is focal length. Can anyone guess what focal length in MM, multiplied by 7.6, would give a frame size of 36x24 mm? Once that can be established, I will rent the appropriate Canon EF lens and EF adapter to check it out and post my findings here. If it works out, I will conduct the same experiments with a Nikon SLR lens and Nikon to XL adapter.

Update:

I am finding it difficult to find an EF adapter for sale. If anyone has one and can do the tests, that would be awsome.

Thanks.

I'm absolutely wild-wild guessing here so ignore please.

A 28mm focal length lens, set about 5mm - 8mm forward of its normal mounting flange. Groundglass about 180mm - 200mm from lens. Might get you close to a 35mm x 24mm frame, might be a bit tighter.

Repeating above. Ignore. Don't go out buying in lenses on my say so.

I think the way to do it is figure out what percentage of full zoom must the 20X be on to facilitate a 36x24 frame? As chris hurd points out in his artilce on the watch dog, at full zoom, in 16:9 mode, the XL2 20X IS lens has a 35mm equivelent of 846mm (it's a 108mm lens in 35mm terms, multiplied by 7.6 to compensate for the 1/3" CCD). So... When I set the Zoom on my lens to match the M2, it is usually at about "12". (If you dont know what I am tlkaing about, zoom your 20X in, click on "position preset" and then select zoom and set it. You should see a number near the Zoom indicator...)


Now, whatever, this arbitrary number means, it seems that "12" is almost fully zoomed in. I would say fully zoomed is probably "15"?

Okay... so, lets divide 846mm by 15... then multiply it by 12, so we can see what 12/15th's of 846 is.
I get 676.8mm... now, let's divide that by 7.6 (For the chip size)... I get 89.05. Therefore, the closest prime lens capable of getting around 36" by 24" would be an 85mm lens. Your image might be zoomed out a tad.

Someone check my math...

Thanks

Matthew Nayman

To get full 24x36 frame, you would need to do some calculation. I believe it's 7.2x not 7.6x.
whatever your SLR lens is, multiply the focal length by 7.2x and you will find the distance between the center of the lens and GG. For example, the 50mm lens will need 360mm distance. 28mm will need 201mm distance. If you use 85mm lens, it's 612mm.If you don't mind the distortions of the 24mm or 28mm, they are the best for your need. In fact
24mm lens need 195mm
28mm lens need 229mm
50mm lens need 400mm
85mm lens need 697mm
Hope this info helps.

Quyen

Queyen,

Seening as how you have built relay lenses... can this not be shortened using an Achromat lens? Also... what is, technically, the focal length of your lenses?

So, a 28mm wold work, but the GG would have to be exactly 22.9 CM away?

Doesnt seem like much of a savings on space! How does the Relay in your set up manage to be so short?

I can't speak for Quyen, but with the adapter I'm finishing I found that the space is eatan up after the light makes it through the flipper mirrors. For this reason, I think making an adapter without image flipping could wind up being more complicated if a relay is to be used.

If you operate in defiance of the original designed geometry of the lens, move it forward of the focal plane, the position of the object can be brought closer to the front of the lens to restore sharp focus. This could mess some other things up.

To experiment, I drew a target the size of the pixel area of a 1/3" CCD upon a damaged groundglass. (However, there is such a thing as "effective" pixel area so all my assumptions are now wrong.)

I then cut a 24mm x 18mm hole out of a piece of Weetie packet. I positioned this against a smooth background light source then adjusted distances between the Weetie panel, my selected lenses and the groundglass until I was able to frame the hole sharply in the 1/3" rectangle.

There are other complications relating to video camera like "on-chip" lenses which make my assumptions invalid. Beyond casual playing around with things by hand I did not set up a proper experiment.

Sorry, bob. Not following you... I am not well versed in the world of lenses and optics, so I need a hand understanding this.

1: A 36mm X 24mm image is projected onto the GG.
2: The GG now becomes the focal plane.
3: An XL2 lens, zoomed through an Achromat, reads the 36mm by 24mm image, and relays it to the 1/3" (8.3333_mm) chip (albeit upside down)
4: This is a function of how far the lens is Zoomed in.
5: Could and SLR lens, attached to the camcorder body, with a proper focal length (when multiplied by 7.6) not give you exactly the same image? (A macro lens could focus close enough, no?)

Please explain!

The image projected onto the GG is not 36mm x 24mm; it's a circle of course, and tends to be around 43mm.

If you attached an SLR lens directly to the camcorder body, the lens would be projecting its 43mm circle onto a tiny CCD rated at 1/3" (it's not really 1/3"; dimensions vary between manufacturers but tend to be around 4.8mm x 3.6mm, so, 6mm diagonal). Therefore, the CCD would only be "seeing" a small section of that image. This would give the illusion of a very narrow field of view. It will NOT turn the SLR lens into a telephoto however.

An analogy may help if anybody reading this is lost. Imagine printing a 43mm (~1.7") round image onto a piece of paper, then using scissors to cut a tiny rectangle out of the circle's center that's only 4.8mm x 3.6mm (.19" x .14"). That rectangle is what the CCD would be picking up.

You could use a lens that has a 6.0mm image circle (like the camcorder's stock lens) to solve this problem, but then you wouldn't have shallow depth of field.

I just re-read the first line in your post, Matthew. What we were just talking about wasn't what Bob meant.

Bob was talking about how you can increase the distance between, for example, a lens and its focal plane and shift its close-focus range closer.

For example, say you have a normal SLR camera. Its lens can be focused as close as, say, one foot, or as far as infinity of course. Now place a "macro tube" or "spacer tube" (search eBay for these) between the camera lens and camera body, and you'll shift that lens' range down to, for example, 2" to 40 feet. You'll lose infinity focus. This is how macro photography is done and that's how people get those super-close-up shots of insects on flowers and such.

Oh wait... a tube between the SLR lens and XL2 body? I was confused.


Basically, what I am gunning to know is, is it possible to find an SLR lens, that could fit fairly snugly (macro tube or not), to the XL2 body (With proper adapter), and could work as a relay lens.

If so... what focal length would that lens have to be? I understand that the 7.6 is not REALLY a magnification, and that we are basically seeing only a small chunk of the real 35mm image... but for our discussion, lets use the 7.6 magnification.

So what focal length, plus 7.6 mag, with proper spacer tube to narrow focal distance, would relay the image to a 1/3 CCD chip in the XL2?




I think we're making progress here guys!

Oh wait... a tube between the SLR lens and XL2 body?

No. I said, "a normal SLR camera." Anyway that topic has to do with focal planes and what happens when you move a lens out of its correct registration distance.



Basically, what I am gunning to know is, is it possible to find an SLR lens, that could fit fairly snugly (macro tube or not), to the XL2 body (With proper adapter), and could work as a relay lens.

Yes. Adapters exist and they work. Quyen Le for example, makes such an adapter.

If so... what focal length would that lens have to be?

Have to be to do what? Work without a 35mm adapter and retain shallow DOF? No such SLR lens exists. Or did you mean, work in conjunction with an adapter? People are already doing this and you should look at the Letus subforum.


I understand that the 7.6 is not REALLY a magnification, and that we are basically seeing only a small chunk of the real 35mm image... but for our discussion, lets use the 7.6 magnification.

You can't use magnification just for the sake of discussion. It's cropping, not magnification, and the two are totally different.

So what focal length, plus 7.6 mag, with proper spacer tube to narrow focal distance, would relay the image to a 1/3 CCD chip in the XL2?

I think we're making progress here guys!

The spacer tube was in an example about an SLR camera, so strike that one. You do not want to give up infinity focus.

Also, this train of thought has actually already been thought out here on dvinfo (two years ago! check the archives). Bottom line is, SLR lenses put out way too big of an image circle for our CCD's. That's why we focus on a piece of GG, focusing screen, or what have you.


Sucks, don't it!

:-)

I'd still like to see that footage with the M2 and XL2 ;)...

...but I don't see what the confusion is with the SLR lens-as-relay thing.

I think Quyen spelled it out quite clearly (quoting):



24mm lens need 195mm
28mm lens need 229mm
50mm lens need 400mm
85mm lens need 697mm
Hope this info helps.

Quyen

Getting proper cropping for the 35mm frame is a matter of changing the flange focal distance between the relay lens and CCD array.

Those are the proper numbers to project 36x24 frame on to a 1/3" CCD. For example, a 28mm lens would have to be positioned 202mm (that's a typo above) behind the GG for the proper scaling, and the lens would have to be as far in front of the CCDs as necessary to bring the image into focus (GG/CCD flange distance). The flange distance will NOT be the same as the published distance for that particular lens... we're talking about changing the effective magnification of the lens by modifying flange focal distance here.

BIll, I understand about that, as I own an M2 and am not an idiot :P

I have no interest in putting an SLR lens directly on the body to use as a normal lens, only as a Relay lens.

For example, a 28mm lens would have to be positioned 202mm (that's a typo above) behind the GG for the proper scaling, and the lens would have to be as far in front of the CCDs as necessary to bring the image into focus (GG/CCD flange distance).

Okay, I have drawn an illustraition to show how I interpret what you just said, and how I see it working (Keep in mind, I am a dolt!)

Image 1: http://www.notomatofilms.com/Explination.jpg

The green lines represent how far Justin and Queyen say the Nikon lens would have to be to refocus the image onto the CCD block. 19.5 CM (or 7.5 inches) on either side doesnt seem like a space savings at all!

Image 2: http://www.notomatofilms.com/Explination2.jpg

This image shows how I would like to set up the rig, but the Green Question mark is an unkown focal length. Is it not possible to find the right focal length lens, place it directly on a Nikon to XL adapter (or EF adapter) and then that part of the image circle which shows through to the XL CCD's would be exactly the 36x24 image? I assume this is how Queyen's and P+S's relay works.

Are spacer tubes really nessicary? If it's a matter of not being able to focus close enough on the GG, then cant I use an Acrhomat?

Interesting proposition. I'd be interested to see where this goes.

Okay. Here's some "in air" tests of some Nikon mount SLR lenses as pseudo-relay lenses.



LENS. - FLANGE TO F/P - FLANGE TO OBJECT


28mm---------51mm--------------224mm

50mm---------51mm--------------354mm

20mm---------51mm--------------233mm

14mm---------51mm--------------151mm


THE METHOD.

36mm x 24mm rectangular target cut out of a cardboard sheet. Bright TV screen used as backlight. This target is referred to in the following as the object or object frame and for a relay device, represents the position of the groundglass relative to the flange face of the lens to be used as a relay lens.

The flange face is a constant. Measuring from the front element of the lens is not valid as no two lenses of the same focal length but different brands or styles will be of the same dimensions. So all measurements have been taken from the flange face on rear of the lenses.

1/3" rectangular target drawn on a groundglass to represent the camcorder CCD area. 1/3" is the corner to corner or diametrical dimension. Groundglass in this experiment is NOT the groundglass used in the 35mm relay device, merely a tool in this instance to enable seeing the image as if it were on the camera CCD.

51mm approximately seems to be the optimum forward offset of the flange face relative to the focal plane to keep the image of the object frame "scaled" to the size of the camcorder CCD. So for Nikon, this means 46.5mm + 4.5mm. The 4.5mm forward offset would apply to all lens mount types.

The 14mm focal length lens would appear to be the most practical in terms of the length of any relay lens 35mm adaptor combination. Shorter focal length lenses are not a practical option in 35mm SLR styles.

A 14mm SLR lens is an expensive way to go about setting up a relay path. C-Mount lenses of the same focal length might be a better option.

C-mount lenses in the 14mm ballpark are likely to have a rear optical element which protrudes up to 10mm behind the C-Mount flange and may well interfere with the front internal optics in the XL1, either prism face or an optical filter.

The focus of the SLR lenses tested was set about midway to permit adjustment either way for backfocus adjustment. Given this variation, my results are roughly consistent with those published by Quyen above. His results will be more valid.

The dimensions in the table above should not be regarded as precise, but more of a general guide only as the testing apparatus was very primitive, carboard boxes, books and lots of stickytape to make things stay put.

The general results were consistent with earlier tests I have done with C-Mount crimcam and Bolex lenses.

A 20mm - 25mm focal length C-Mount lens is probably the most practical option as a relay lens. In a non-erecting design, the structure is going to have to be fairly long.

However, less than 14mm focal length lenses will create a problem if a prismatic path is used as the flange to object distance will be too short for the lens + prisms to fit in between the camera and groundglass. I think the 14mm lens might be a bit too short anyway.

Anyway. That's my contribution. Consider the baton now handed for somebody to take it to the next level.

Alright, did some more tests. Here is what I got using a Nikon 50mm shooting through Redrock Achromat and M2 (mounted is a nikon 28mm 2.8).

www.notomatofilms.com/pic.jpg

I used a card to reflect the projected light. It seems mostly infocus, and roughly 1/3 inch digaonal.

This is a crude test at best, but may warrant further testing.