View Full Version : having to use macro to see ground glass.
Stuart Campbell June 3rd, 2008, 05:14 AM We've just taken delivery of our 1st Letus Extreme from Zacuto and are having the following issues.
In order to focus on the ground glass you have to use the macro function. This is after zooming in far enough to see grain of course. I don't need to go on about how this renders the unit useless!
Are we overlooking something completely obvious here? Surely we are.
Of the units we've hired until now, we have always noticed how close to the end of the focus limit the ground glass sits. Here's what we are using the Letus on and how we did it.
JVC GYHD200e with Fujinon 16x5.5
26" monitor for focus, 7" LCD, 9" CRT all to check against each other.
Letus Extreme with Nikon attachment.
Siemens star chart
Daylight conditions
Studio conditions
all same result!
lenses, Nikon 24mm, 50mm, 85mm 135mm Zeiss Planars 50mm 85mm 105mm
We followed the instructions as per Letus. We've also tried focusing on GG with lenses off and on.
If anyone can point out the blindingly obvious that would be great.
2nd issue;
Whilst I understand that alignment of image sensors is not always perfect, the horizontal position of the image is WAY off!!!! Ignoring the focus issue, you'd have to zoom in a long way to fill the screen! This presents a problem when our customers ask for vignetting! (which they do). To fill the camera with the image you've zoomed way past the left hand vignette before the right hand side has filled the screen. The vertical alignment of the image is also off, by about 20 degrees, but not too much of a bother.
Is there a way of centering up this image within the Letus as its current position is far from acceptable? and how is it that of all the Letus units we have hire until now they have all had the images in different positions!! but none as bad as the one we now own!
Thanks
Stuart
p.s Also, is it just me or does anyone else find the cheap photocopy instructions page very unprofessional and lacking in detail? My opinion of Letus as a bona fide company is rapidly waning!
Ted Ramasola June 3rd, 2008, 05:42 AM i use an hd100 with an adapter.
To use an adapter, our 16x lens is at FULL macro, focus at Infinity. Tweaking is done with the zoom. In our set up its up at 30mm. different adapters and users could vary. My settings is based on a +10 achromat.
Bob Hart June 4th, 2008, 01:17 AM Stuart.
With any of the adaptors, relaying via the camcorder's own lens, you have to zoom in a fair way, some more than others, in the region of 32mm to 54mm for 1/3" sensors.
With the JVC standard zoom lenses I have played around with when modifying my own AGUS35 to fit, the filter thread is contained in a plastic housing which the lens hood attaches to.
The lens focus movement is free to move inside this housing. The housing can be flexed enough to put your image centre off.
I observe that the Zacuto rails system on some illustrated kits has a step or dogleg in the structure. The system for the JVC may not have this.
If there is a step or dogleg in your kit, this means that there are four points of fixture in the combined structure where off-axis angular or torsional deviations from optical centre can be introduced, six if you include the Letus and camera holdfasts.
Therefore angular loadings on the lens hood filter mount on front of the Fujinon could deviate the optical centre axis of the Letus relative to the Fujinon lens optical centre axis.
It takes very little deviation at the long end of the zoom-in you have to use, to move the Letus screen image frame across the camera image as viewed.
There might be variation between individual Letus units but variations introduced in the combined assembly might more likely account for the centering deviations you are observing across different individual units you have attached.
As for having to use macro focus, attaching my home-build appliance I found the standard Fujinon presented a serious challenge to my efforts. The filter mount was too far forward for best position of the Century achromat I used and I was forced to move it to the lens itself.
My imagining is that you if want to avoid macro use, you may need to dismount the inbuilt Letus achromat from the Letus body and attach the achromat directly to the front of the Fujinon focus barrel as close as you can get it.
You would need to remount the Letus body as close to the front of the Fujinon lens as you could with a custom shorter attachment adaptor ring you would have to get made in a machine shop.
You would need to make a custom ring to mount the now free achromatic inside the moving front of the Fujinon lens itself.
The new Letus attachment adaptor ring might likely need to have the allen screws replaced with short grub screws in radial holes within the 82mm filter thread itself as there would likely be only enough allowable front-to-back thickness to permit a threaded shoulder and not enough extra for a plain shoulder for the normal screws.
Because the optical axial alignment of the Letus to the camera is maintained by the threaded adaptor pieces, you will most likely have even more problems with centering if you follow my hack without making a new machined adaptor ring.
This hack would probably eliminate your need to use macro which I found to be unsatisfactory for both focus and edge distortion on my own adaptor.
Please do not attempt my hack unless you have good engineering skills.
This comment is also speculative theory so please do not take it at all seriously. I have made a few mistakes recently and the engineering for the Zacuto system you are using may be perfectly fine.
Perhaps you could post an image of your adaptor including a close-up of the LEtUS to front of Fujinon attachement to assist or correct any of the number of assumptions I am making which could be wildly incorrect.
Finally if you are achieving good edge-to-edge sharpness and no other optical defects when using the macro function of your Fujinon lens, leave best alone and concentrate on getting the alignments of the combined assembly correct.
Stuart Campbell June 4th, 2008, 01:59 AM Firstly Bob,
Many thanks for your reply. Some interesting suggestions!
I'm loathe to tinker with the unit (I'm more of an operator than engineer) because the unit was purchased to fit a purpose, not to be adapted and bastardised to do so. What's really disappointing is that I've used other Letus units that don't display the same problems! This can only mean that there are massive differences between each unit sold! I wonder though, if this 'supposed' relay lens that Letus are 'bringing out' would solve the problem. I'm really beginning to wonder about this company. This relay was supposed to be released in March, April, then May, then June and so on............
The edge distortion (when somewhere near focus) is terrible, so we cannot even 'make do' with a fudge of correct focus.
The excellent Zacuto unit does allow for doglegging and alignment alteration but I find the filter ring on the Fujinon too stiff to skew the positioning. I'll try again, but I'm always reluctant to start forcing things into a position they were not designed to be in!
I'll attach an image of the unit assembled for you nonetheless.
Many thanks for your thoughts Bob.
Ted,
You use full macro and infinity zoom????? Really? I can't see how that could work. You've got my mind ticking now. Can't think of a situation where I've ever had any lens on any camera like that but, hey, I'll give it a go.
Many thanks
On the whole I'm massively disappointed. We've just ordered a totally useless lump of glass in metal case. I'm nearing the point where it goes back. Not surprisingly I've had no reply from Letus, and I really don't expect one either.
Ted Ramasola June 4th, 2008, 05:10 AM quote :
"Ted,
You use full macro and infinity zoom????? Really? ..."
Stuart,
Just to clarify its full macro and zoom between the 20 and 40 mark. Theres no number 30 on the barrel but counting the notches its approx in between.
Its the Focus ring thats at infinity. with this method. You lock down two variables and just tweak one-the zoom.
The setting up of a lens adapter would be less cumbersome than if you have to tweak 3 variables every time.
Again, let me add that settings vary between diff adapters.
Heres an old article in showreel mag of Taylor Wigton on how he set up his hd100 witha n adapter for a shoot.
http://447productions.com/M2.pdf
Ted
Stuart Campbell June 4th, 2008, 05:36 AM Hi Ted,
Interesting article. Thanks.
Well, I've spent the morning trying to get this lump to look sharp. I tried your infinity method and came up with the following for 'acceptable' sharpness.
Focus Infinity
Zoom 22mm
Macro somewhere in the middle!
Now, it amazes me how different each adapter can be and smacks of bad manufacturing processes.
I tried the 20mm, infinity and full macro option, but it doesn't work.
The whole thing is so fragile once set that I'm scared to look at the bloody thing in case it shifts! I really don't think I'm prepared to use this in a job. Too risky.
I tried Bob's suggestion for distorting the alignment, and it certainly improved things but I now have misgivings for stresses on both the camera and Letus, and also the effects on focal distortion by not having the Letus sitting flush with the Fujinon.
Right now I wish I'd spent the extra and bought the P+S.
Stuart
Ted Ramasola June 4th, 2008, 06:10 AM Stuart,
I consider myself an extensive HD100 + adapter user, in fact my adapter stays on my cam even inside the pelican case. I have done TV spots and cinema projected material using that setup.
I must say that be patient with it and it will pay off. once i locked focus I tape down the fujinon and never touch it.
The creative options that the adapter presents to you as a cameraman is in my opinon a great motivation to get it locked down.
heres a guage for you to try.
Put macro in full and focus into infinity. Iris at at least f2.8. any wider and you get CA and the adapter image tends to loose sharpness at corners.
place a 50mm lens on your letus and put it in full open or at 2.8
focus on a chart and counter check the distance on the 50mm with tape. (here you locked down the 50mm)
Lets go back to the fujinon adapter setup.
zoom in to the GG till you get sharp image. In my case its good to have the fujinon iris at f5.6 -8 for best calibration results.
if you think its good, take note of the image corners.
Take off your jvc and using a caliper or a piece of paper, note the image size you are getting on your adapter GG.
I dont know what the letus makers claim on the image size but in my adapter i can capture a 30mm x 16.8mm image.
cinema 35mm is at 24 mm x 18mm and full frame still 35 is at 36mm x 24mm.
if you are getting a very small image then adjust the distance between your adapter to the front element of the fujinon.
I had a hard time with this since the fujinon's front element is recessed deep inside the barrel. I can actually remove the ribbed anti flare ring in front of the lens and allowed my achromat to go in deeper and closer to the front glass element of the fujinon.
Doing so gave me much better results.
keep trying .
Ted
Bob Hart June 4th, 2008, 09:28 AM Stuart.
Did you source the Letus Extreme and the Zacuto kit together from Zacuto as a turnkey solution or were they bought in separately? You may have already said this so forgive me if I am repeating myself.
Sometimes there is different interpretation between US English and Strine on the same sets of words, so all I am doing to trying to eliminate any misunderstanding on my part. I hope I have not created mischief here.
Distorting alignment? - That sounds cruel bad. I was meaning your rig may have had some latent stresses already in existence deviating the two axes off true and moving the fittings might bring the alginment back in.
If you are having to do some serious bending then maybe the Letus has had a serious clout in shipping and moved something inside but the co-incident focus issue suggests something else.
It would be interesting to find out which dioptre is in back of your Extreme. I understand there are more than one version, some intended for a particular camera type.
Hopefully this may be the source of your problem with this particular adaptor and represent the potential cure.
I don't know whether there is a particular optimisation for the JVC GY-HD--- family.
Where your camera mounts to the rails, are there lateral adjustments and vertical adjustments?
When setting up, it is possible to introduce deviations with camera weight alone whilst trying to match it all up by hand.
I find it helpful to point the whole assembly nose upward when I tighten the assembly it to avoid camera weight bending on the Letus Extreme / Camcorder junction. This is not necessarily factory approved practice, just my own possibly bad habit.
One more experiment you might try is to take out any spacer or ring which is between the back of your Letus Extreme and the front of your Fujinon filter mount to enable you to move the Letus Extreme physically as close to the front of the Fujinon lens as you can get it.
This spacer will be the mounting ring which has three allen screws on radial centres locking it onto the back of the Letus Extreme. Taking this out and supporting the adaptor by hand, loosening the Letus support on the rails and moving it closer to the front of the Fujinon will give you an extra 9mm or so closer position to front of the Fujinon lens.
If this makes the focus worse by requiring more Fujinon macro to be turned in to focus or is even moving out of its range, then try adding some distance between the Letus and the Fujinon lens.
- The Fujinon macro may be comprised of a simular retrofocussing element as the larger ENG lenses. For best results without edge softness I think your Fujinon backfocus must be spot-on and no macro turned in or out.
This may seem counter to Ted's solution however he has optimised his arrangments around custom positionings of his achromat and his dual duty 35mm and medium format adaptor. I think his achromat is also from Wayne Kinney for a non-flip adaptor.
There are many effective ways to achieve the same cat skinning.
There is some interaction between achromat position in front of the Fujinon lens and this rear focussing element in the Fujinon.
If you can post the images of your setup and a frame grab of objectionable images, this might steer us in the direction of a solution for you.
However - before doing any more experiments, wait for a few more responses here unless you are up against a tight deadline to get this sorted.
Bob Hart June 4th, 2008, 11:48 AM Furthur to my post above.
It seems that Zacuto make their own version of a rails kit and buy in Letus Extremes to resell with their kit. It might be worth a phone call or email to them.
Letus make their own rails kit for the Extreme.
Stuart Campbell June 5th, 2008, 04:57 AM Hi Ted and Bob,
Many thanks for your time spent considering this problem it's really appreciated!
I've conducted a few more tests this morning in the office as you both suggested. Before I did anything I re-double checked back focus on the Fujinon. Its absolutely spot on.
I found that to achieve sharp focus on the grain (without macro or infinity focus) I had to slide the Letus 26mm away from the front of the Fujinon. This, however, then gave conflicting focal distances as marked on the 35mm lenses. If I'm right then all focal distances should include the difference of 26mm from what's marked on the lens, as the focal plane position has effectively been shifted 26mm. So a true distance of 474mm from the Letus focal plane shows as 500mm on the lens. I tested this and it was consistent. I don't really want to be sharpening my maths skills on every shoot as well! However, this does achieve the sharpest image of all my testing so far, but not still not great.
So, I went back to Ted's method and found the best focus I could with all focal distances correct and Letus fitted as it should be. See attachment for the best focus.
Settings: Infinity focus, 22mm zoom, macro mid way and marked.
50mm lens f4, Fujinon also at f4. I've tried various other aperture configurations, but you start to get aberrations elsewhere, which appear very quickly! etc.
(Ted I tried your method for measuring the image plane on the ground glass, but I'm not sure I follow where you measured. Did you disassemble the Letus to see the ground glass? It's not possible to take a measurement from the anchromat end or have I plainly misunderstood?)
Unfortunately this is still soft, although its the absolute best I could achieve. I thought I would double check again on the Letus instructed method and had the same results. Unable to focus on ground glass without using macro. I decided to try with the GG soft, but obviously all shots are also soft too.
I've since had an email from the manufacturer who suggested that many people use macro to obtain focus, and that doesn't seem to be unusual. However, I truly don't believe using macro gives you the sharpest image so I am awaiting my next reply.
Bob, I've attached an image of the rails and support. I purchased the rails and Letus from Zacuto together for this camera. The quality of the Zacuto gear is extremely good and the support system is as stable as it gets. I've tried the Letus support and found them fragile, wobbly and not stable enough for a shoot with the JVC. I've also tried Cavision and another one who's name escapes me. Zacuto is by far the best. It offers adjustment on all planes and I have set it up so it lines up with the Letus exactly. The base plate, wedge and tripod plate all stay together as a single unit now so its very quick to set up.
Well this is now 3 mornings or 12 hours I've spent trying to get a sharp image from this unit. From the point of view of my business this is more than unacceptable. I'm wasting time on this unit when I have a thousand other things to do as it is. Am I not entitled to buy something and have it work without all this effort? Needless to say I'm a little frustrated! I'm really hoping this is something really daft on my part that I'm overlooking (which I'm always willing to accept as a possibility) and that one simple thing will get it working, but.....I'm really beginning to doubt that now!
Your efforts are really appreciated guys. Many thanks.
Ted Ramasola June 5th, 2008, 05:56 AM Stuart,
Your basicaly shooting a very small image, and thats the GG image, about the size of a post stamp and so close to your lens, this makes it normal to use macro. Engaging it fully can not only give you a better image, but also it locks down the variable to just your zoom to fix when setting up.
Ive been in your path around two years now so i had to contend with a lot of trial and error.
In your setup photo, is there a step down ring threaded on your fujinon filter thread? if so, try removing this, i assume its an 82-72 step down ring, after removing it push the achromat deeper into your front lens. You could get better results.
Thats how i set up mine. The achromat is pushed closer to the front glass element of the fujinon. -in fact i went further and removed the anti glare ring, its that round plastic inside your focus barrel at seem to look like a series of terraced rings, its just threaded there, removing it widens the front opening and allowing the achromat to get real close to the front glass element. This improved my edge to edge sharpness in my set up.
Ted
Stuart Campbell June 5th, 2008, 07:11 AM Hi Ted,
No, I don't have a step down ring. Just the normal 82mm fixing to the Letus. I'll try to remove the inside of the focus ring like you said and try the full macro test tomorrow. Does it just thread anti clockwise to come out?
Also, straying away from my problem, you say your Letus stays on the camera and goes straight into a Peli case. What size Peli are you using.....is must be huge! Have you got a pic by any chance? If I can ever get this unit to work satisfactorily I'd consider using it on all of our corporate work and as the JVC is used exclusively for any corporate stuff we get then it could remain as a JVC/Letus rig. But can you really fit it in a case with Matte Box attached etc?
Stuart
Ted Ramasola June 5th, 2008, 08:59 AM Stuart,
yes, it unthreads CC. Heres a shot of what goes in the peli. Thats my own custom made adapter and custom made rails.
It uses the SGpro Achromat. I had a Brevis achromat but it didnt pair well with the fujinon and caused excessive CA.
I must also add that i too was already frustrated with my results in the start but because of selfless people in this forum it kept me going, Bob Hart was one of those who was patient enough in helping me fine tune my own rig.
Ted
Bob Hart June 5th, 2008, 10:33 AM Ted.
At the back of the Letus Extreme flip enclosure, there is an achromat, fastened in conventional optical practice inside an approx 8mm shoulder with a threaded ring. Its rim sits just inside the shoulder.
The various adaptor rings with their filter threads slip onto this shoulder and secure with three radius screws.
The depth of these rings and thus the setback of the camera from the Extreme varies for different filter thread diameters. The achromat would have to be removed from withn the Letus Extreme housing to move it to the front of the Fujinon lens.
I'm not sure about recommending this as any warranty open to him relating to the Letus would be voided.
When you removed the plastic ring out of the front of the Fujiono lens, did you find a 72mm diameter filter thread inside the rim?
I have observed two styles of lens attached to the JVC HD100 camera, both Fujinon. I did not study them closely.
One of Steve Rice's Fujinons looks like it either has no plastic ring or that it has been removed or fallen out. I did not study it to see if there was an internal thread there and have no opportunity to study it before I see it next.
If there is a 72mm thread, this raises a few options for experiment over here.
I have the 82mm ring so this would fit up to a JVC GY-HD100 standard Fujinon as I know it. I also have a 4+ and 7+ achromat which will fit up to a 72mm internal thread. I could do a limited bit of extra testing.
This won't likely happen until this Saturday and will depend on the crew numbers at the shoot as to whether I can take time out to play with the second camera.
Stuart.
The Siemens chart suggests there might be room for improvement by reference to what I have observed with another Letus Extreme on a different camera type. Unfortunately I don't have the same chart to make a valid comparison.
A MINI35 will resolve the "B" block (862 lines ) on a Lemac chart on a JVC and on the Sony Z1 and gives signs on both that optically, the cameras see the "A" block as there is a distinct moire pattern visible.
The Extreme on the Sony resolves the "B" and a moire pattern on the "A" block at a comparable framing as the Mini35 and when reframed for the wider view and the Lemac chart repositoned to frame in the wider view, the apparent resolution improves.
From my own opinion and independent of the Letus designers intent, my concern relating to performance does not concern the centricity of the groundglass frame but the centricity of the projected image which falls upon it.
In easyspeak the important bit is centre of Nikon lens image falling in centre of image seen by the camcorder. The generous groundglass area on the Letus Extreme and I think the Brevis and SGPro is a bonus where sometimes some added resolution can be sought with more zoom-back, but is not an absolute given. For consistent and predictable results across a range of SLR camera lenses I think 28mm image width and no more is about as wide on the groundglass as you should want to go.
If the client wants a vignette, then I would introduce it as an effect in post. If it absolutely has to be done in-camera I would make a cardboard do-nut to shove in the back of the Letus against the achromat and trim the inside diameter of the cardboard do-but until I got it centred. My personal preference is to shoot clean footage as you can put salt in the stew but you can't take it out again.
I observe with the Z1 and PD150, the centre of image shifts with zoom-in.
The Panasonic has been anecdotally reported here at dvinfo to do this as well, which direction I do not know. How the JVC behaves, I no not know.
It is a pretty tough call to expect adaptor builders to customise to the variety of offsets of different camera types with thread filter ring attachments.
It would require a lockable adjustable offset within the ring, a dog to build, a dog for the operator to set up and a potential mongrel for off-axis errors.
The alternative is designing to permit user access to the guts of the adaptor to move prisms or mirrors about.
P+S recommend people do not climb inside their MINI35 to get at the internal adjustment for reasons of introducing contaminants but they do advise that it does exist.
Letus have been more generous with user access information in a new manual about to be published but they also recommend against operators going beyond a certain stage in roaming around inside their adaptor.
The MINI35 for the Z1 relay focus point hits sharp uncomfortably close to the infinity point. My personal preference and it seems Letus also, has been to hit sharp focus much closer to the close-end of the relay focus range.
I can't speak for Dennis's Brevis, Wayne Kinney's SGPro or Brian Valente's RedrockM2.
The P+S Technik approach has obvious advantage in that operator relay focus errors at the faraway end might be less critical than at the close end. This requires some premium design and build quality management which is what you pay for with the MINI35.
Their "connecting kits" (relay lenses included) are also camera specific and are for a fairly narrow range.
The Letus has to please a wider range with generic connecting kits based on filter mount diameter, so relay focus biased to the closer end makes good sense as there is more scope for coarse adjustment.
This also means that the practical placement of their achromat remains within the rear body of their adaptor.
If the 82mm diameter adaptor ring is intended to be specific for the JVC, then I am surprised if it has been set up to require additional macro focus on the JVC standard Fujinon lens. If is generic, then this is understandable.
I wonder if the intended direct relay option for the JVC GY-HD--- camera family, also applicable to the Sony Z7 is the main priority of their intentions for this camera type. The 82mm adaptor ring may therefore be generic for other camera types.
As to the Zacuto rails system. Please conduct one more experiment for me.
Place the camera in a safe place like over a mattress or suitable padding and no height for anything to fall.
Release the Fujinon lens/JVC lens mount locking ring. Observe any movement of the entire combination, then without assisting by pushing or shoving on the lens or adaptor or the rails, attempt to re-secure that mount ring to the closed position.
If the lockring baulks and requires pushing or shoving to enable it to close to its locked position, the Zacuto rods need to be re-aligned so that in a pre-loaded position, the lockring will release and re-secure without other assistance needed.
The MINI35 can be set up to do this. The Extreme on its Letus rails can be made to do this, even with the small pillar under the front tube. The Zacuto rails kit should be no exception.
If it cannot be set up this way, then the structure is too compliant and will introduce some bending loads on the camera/adaptor junction and camcorder case structure in normal use.
The apparent forward weight bias on the Zacuto kit is also a bit of a worry to me. There may be bounce with long lenses. If you observe Ted's setup and the Letus rail system, there are no junctions but a continous unbroken structure.
The preceding is a big and complicated message. I hope I have not thrown you into total confusion and despair. Hopefully the next few days will find a solution for you.
Bob Hart June 5th, 2008, 11:35 AM Stuart.
I'm covering this point in a separate post.
You have referenced a discrepency between the witness mark info on the side of the SLR camera lens on front of the Extreme and the real distance between the subject and the focal plane.
Just where you place the Letus Extreme relative to the Fujinon lens is totally divorced from the SLR lens focal plane position and its relationship to the SLR lens on front.
You should regard the SLR lens on front of your Letus Extreme to the groundglass as a separate camera.
If there is a discrepancy between lens witness marks and real distance between the focal plane which is the groundglass and the subject, then this is a separate issue to the Fujinon focus issue and a separate problem to resolve.
There will be four causes :
The lens witness marks are incorrect. (Sometimes the witness mark barrel can move on the internal focus barrel and can be re-adjusted by slacking off some small grub screws with the lens mounted on a reference instrument and resetting the position of the outer ring.)
The lens internal collimation is off.
The mount is incorrectly positioned in the Extreme.
The groundglass is out of position.
If you move the Letus body furthur forward of the Fujinon lens, you may find a little variation in the groundglass area you see for a given zoom setting on the Fujinon but that and an altered relay focal distance are the only other variations.
On the Extreme, the genuine Letus pillar which goes under the front tube of the Extreme body is normally attached so that the vertical upright which connects to the bridge is between the red thumbscrew and the flip enclosure, not forward of the red thumbscrew.
If it is in this correct rearward position, then the rear face of the pillar represents the focal plane and is what you hang your tape measure off.
So far as I know, the collimation or backfocus relating to the SLR lens mounted in front of the Extreme should be pretty close to correct with the Nikon mount and the others slipped fully home into the housing.
There may be a few millimetres discrepancy between real distance and lens witness marks. My concern is to have infinity focus available on the lenses. I want the backfocus to be correct as this is a critical issue for wide and especially ultrawide lenses which will simply go softer at all focal positions.
Bob Hart June 5th, 2008, 10:11 PM Stuart.
Furthur to above posts, the JVC camera I want to play with is in hospital with the usual firewire malady and also being firmware upgraded for the DR100, so I cannot do the homework for you until a week or more down the track.
A hint on setting backfocus which I am sure you already know but repeating anyway because I have pedantic tendencies and am simply getting old.
The Fujinon lens needs to be set at its widest aperture in order for the best focus point to be found. This means closing the aperture on the SLR lens to provoke the groundglass texture and also keep light into the camera at an acceptable level.
If the forward offset of the whole appliance forward of the camera gives best sharpness and no edge softness or distortions, then my inclination would be to make a new longer spacer ring or another threaded spacer to add length.
Don't go to any of this trouble just yet. Another solution might be found, maybe even something as simple as an achromat swap in back of the Extreme.
Stuart Campbell June 6th, 2008, 01:37 AM Bob,
Many thanks for all those thoughts. We're taking them all on board and your gesture of a test at your end is most generous. I'm extremely grateful. If I was anywhere near Perth I'd buy you large beer! (mmm, the majority of my family in law live in Perth...)
In reply to some of your points;
I don't think the 82mm thread on the Letus is specific to the JVC. I might be wrong but I'm sure the hvx200 is 82mm also. This may mean it's more of a generic adaptor!?
We will test for misalignment due to load on the rails in a minute. That's good thinking. However, to see how good the Zacuto rails were we loaded the Letus end up with everything before we put on the JVC. I have to say, hardly any movement under load. Really quite impressive. My initial thought of a stepped bridge was one of scepticism, but I can say that it's exceeded my expectations easily.
As for the focal plane point on the Letus being the rear of the Letus support; well, that's how I've always understood it too. However, when we were testing we noticed that all the distances on all the lenses were consistently out by about 10mm! We then assumed that we'd just misinterpreted the instructions so now measure from the front of the support, which now gives accurate distances. I can accept that some internal collimation error on the SLR lens can occur, but not on all 7 that we have!
Now, when my employer sent me on a camera engineering course (many moons ago) I have to admit I never really paid that much attention. (I was always more concerned with making nice pictures than how it all worked etc.. and also, can I go to the pub now please?...) so, that said could you explain what you mean by A and B blocking on the star chart. Fortunately I'm conversant with Moire so I'm Ok on that one at least!
I had a long chat with one of my partners last night and we've agreed to abandon in camera vignetting, so we'll be taking your good advice and doing it in post from now on. Like you say, salt an' all that.
Now, we also came to the conclusion that we are not under any circumstances prepared to alter, adapt or take apart the Letus in any way. If we have to do any of that to obtain a sharp image then its simply not the right product for us and is not truly fit for the purpose it was sold to do. Either it's faulty or it's simply not compatible with 'ALL' JVC GY cameras.
So with that in mind, as a business, we can not afford to risk it on jobs right now. We are due to use it next week (although I'll be hiring again now, which really makes me cross as I have one sat in my bloody office that I can't use and I'm pissing money down the drain on hire again) and again at the end of June. I think that if we get nowhere with it and are left with no option but to adapt and alter then I'll be sending it back, getting a refund and buying another product. Apart from the fact I'm simply not able on a fine engineering level I'm just not in the business of buying things and having to mess around with them so they work.
I've heard nothing but good things about this unit, which is precisely why I bought it. I just hope I can be another one to say good things about it, but right now I can't!
I'm still hoping!
I've not yet had a reply from the manufacturer other than 'what's wrong with using macro? lots of people use it'! so I'm still hoping for some more constructive advice. In the mean time I'll try to remain patient (for a few more days at least).
Stuart
Bob Hart June 6th, 2008, 04:52 AM Stuart.
The Lemac chart is a A4 sized cardboard chart which has 4 Siemens stars one in each corner, other markers, a contrast row and two rows of boxes made up of vertical or horizontal lines and aspect ratio guides.
The upper row of blocks as I call them are "A", "B" etc starting from sharpest which is film, "B" which is 862 lines then on to lower resolutions. The lines are upright for assessing horizontal resolution
The lower row of blocks starts at "F" which is film, the "G" etc onward to lower resolutions. The lines in these are horizontal for assessing vertical resolution.
The 10mm or so consistent error suggests lens flange to focal plane being very slightly off. To get a 10mm increment focal plane to actual subject is a mere whisker on each of the threaded adjustments I have built-in to my own home made adaptor, a movement of less than about 1/10th of 0.5mm, in kitchen terms, maybe about the thickness of clingwrap. I am no engineer, just a hack-artist which is why I had to design in the ability to adjust.
P+S Technik do a full collimation with a projection rig and install superfine metal shims where needed as is normal lens collimation practice. They are ARRI camera engineers after all. Again this is where the cost premium comes from with the MINI35.
What process is used in the Extreme I do not know. I imagine that for product consistency, they would use some sort of assembly jig to get identical placement in all examples.
As to turnkey utility out of the box, that has obviously not happened for you, understandably a real disappointment
A local here opted out of groundglass relay of all flavours for his ad work for reason that the extra time taken to manage the system to his standards and extra lighting requirements amounted to productivity lost and moved up to a F950.
Hopefully your issues will be resolved when the direct relay systems for the JVC, Sony Z7 and Canon XL family are sorted. Of course this does not help you right now.
Please let us know if you still have the unit in about a ten days time. That is about when I will be able to get at a JVC camera.
Stuart Campbell June 6th, 2008, 07:19 AM Hi Bob,
I remember the chart now. Think I might even have an old one somewhere! It'll be one of the things I took home when one of the studios I worked in closed down. I'll have a look for it. Nice to know what its true purpose is!
As for the Letus; well I spent another couple of hours trying to get things sharp this morning without success. I am 100% confident there is no problem with the JVC end, which only leaves the Letus or the 35mm lenses at fault. Given that the same soft images occur on all 7 lenses I'd wager it's the Letus.
I'll have a rest from it this weekend and it looks like we now have Tuesday free, so I'll commit the day to yet more testing. If nothing comes to light by then I'll be sending it back. I can't waste any more time or money on it. Sadly I've still not had any further reply from the manufacturer which all adds to me wanting rid of the damn thing. It seems it was not the right thing for us to have bought. What a shame as we had high expectations from it.
Stuart
Bruce S. Yarock June 6th, 2008, 08:00 AM Stuart,
I don't know if this will help you, but here's what I do with my canon h1 and a1. I've only had the Extreme for a couple of weeks, and had to spend some time fooling with it ,along with another camera guy who is more tech than me, and had experience with the brevis. I had a hard time seeing the ground glass, until I figured out a solution.
1-Without the slr lens yet attached, I zoom in on a lighted white wall. Even zoomed in , it's very hard to see the little black particles for getting focus. I realized that using my "magnifying" function on the Canon made it possible to see the ground glass.
2- Once I had the Canon focused in, i use the "focus preset" button to lock down a focal point. Even if I mistakenly turn the camera's focus ring, I can use the prest to go back (it's right at .9mm).
3-I hear what you're sying about the gg image being off horizontal plane. I tried adjusting, but it seems to always go back. Anyways. i zoom back out to just before vignetting, and set that point with my "zoom preset" button. This way, if i zoom in tight, i can always go back to the correct spot.
4- put on the slr lens and focus.
Hope this helps.
Bruce yarock
www.yarock.com
Stuart Campbell June 10th, 2008, 01:16 AM Well I've enjoyed a nice weekend watching England hammer New Zealand at Trent Bridge and having been away from this infuriating problem I started the day fresh with enthusiasm, hoping to get this sorted.
It got quashed quickly!
I'm fed up now. This bloody thing is useless. It's either not designed for the JVC or I've got a dud.
I'm going to try one more time. The sharpest point of focus with it is simply NOT SHARP!
So far it's been an absolute waste of my time and money, and still no reply from the manufacturer.
Anyone got any more ideas?
Bob Hart June 10th, 2008, 08:24 AM Stuart.
I got hold of the JVCHD100 tonight to set it up with the MINI35 for this weekend's shoot.
I put the Extreme on first.
The Fujinon needed just a touch of macro for sharpest focus.
The results off the Extreme and the MINI35 were comparable, the Mini35 slightly better on a movie sized frame off the groundglass.
I rolled some tape however was not able to capture, as the cam's fire wire seems to be dead despite having been fixed by the doctor or claimed to have been fixed. The battery went flat before I could port it across via component to the Z1 at 1080i.
If you send me an email, I'll mail you the tape.
Stuart Campbell June 10th, 2008, 11:02 AM Hi Bob,
So...maybe I'm expecting too much from the Letus then? Maybe the sharpest image I can get is as good as it gets?
You are a real star for doing a test your end. I'll send you a mail. I'd really love to see your results.
Any chance you could tell me the exact method you used to achieve sharpness (well soft sharpness).
Many thanks Bob
Stuart
Bob Hart June 10th, 2008, 05:56 PM Stuart.
I cannot say your particular unit is not defective. Being so far off-axis as you describe points to something if the actual mounts are straight.
No doubt you have already checked the face where the 82mm ring goes into the back of the Extreme. This should be located firm against the body all the way around with no high places.
Half a hair will put you more off-centre than it might otherwise be. This will also push you a little more off sharp focus without macro selected.
That macro thing does discomfort me a little but that is what it is there for. It is the same element within the Fujinon lens that moves as is moved by the backfocus adjustment.
There is really only one way to be assured the image is as sharp as you can get it and that is with a bigger screen to look at.
With the JVC LCD screen you can get close with the test patterns and using moire to get to the sharpest point, but out there in the real world sharp focus is a harder ask especially if you are nearsighted like me and perfectly adapted to soft images.
Sorry it has not worked for you. Your expectations may be beyond what the system can deliver for you but until I can see a comparative image between two different camera/Extreme combinations I can't make that judgement validly either.
I'll see if I can get the tape captured over the weekend.
Stuart Campbell June 11th, 2008, 08:39 AM One thing I did notice is that when you tighten the grub screws to lock the SLR lens attachment on to the front, it pushed slightly off kilter. You have to have one person pushing the adaptor against the Letus while the other slightly tightens each grub screw. Its really easy to end up with one side about a mm or 2 away from the Letus body!
I'm now wondering whether;
a: the best image I'm getting is as sharp as it gets with this unit (that was the Siemens star previously attached) and that my expectations of the Letus were just too high. I appreciate the DOF adapters give softer images to that of a HD lens for example, but, wow. Softer than I thought!
or
b: There is something slightly out of alignment within the unit giving me soft images.
The results from the tests so far look like the image is sharp but has been softened afterwards. Does that make sense? It's like a sharp image that's been gaussian blurred!
We are going to get it as best as we can and shoot with it next week. I don't know if I'm mad or not to risk this. We will aim to shoot everything without the adapter too as backup!
The only way I think I can get better test results is to get hold of a very large HD monitor, which I can't do right now. On that subject you should be able to get sharp images using normal crts. I even know that mr Phil Bloom shoots a lot of his stuff using the viewfinder only!!!!! And there's nothing wrong with his pictures, although I've never seen them on anything other than a PC.
So, this infuriating saga continues....and oh...STILL no reply from the manufacturer!
Chris Barcellos June 11th, 2008, 09:48 AM Stuart:
I've shot a couple of Letus adapters, as well a few self made on 4 different cameras (VX2000, FX1,HV20, HVX200), and every one of reacts to the achromat combo provided by the manufacturer in a different way. With a Letus35a, I had to replace their achromat, with Cinevate's to make things work. However, with the LetusFE (predecessor to the Extreme) I could mount on all 4 and be able to get a decent focus on the ground glass.
I suggest that what is happening is as you zoom in, you are losing focus capability. You may think you are still in the focus range, but you are not likely there, and this is typical. What you are trying to do is get zoomed just inside the vignetting on the image, but no more. In some instance, you have to live with slight vignetting.
You can resolve by adding tubes to extend distance between camera lens and ground glass, or changing achromats. It takes experimentation. I note that BH Photo sells several different versions of the Letus, for different cameras, because of this issue, I suspect, so have you checked with Zacuto regarding that.
Ted Ramasola June 11th, 2008, 10:20 AM Stuart,
i have downloaded the siemens star you posted. We cant objectively verify the situation you are in since this image doesnt include targets for the corners where we can also help you gauge wether what your getting is the "acceptable" image from the letus adapter.
I suggest this, if you dont have a focus chart try to print additional siemens stars (photcopy?) and mount them on the wall in a way that would fit the corners and sides of your 16x9 framing. check perpendicularity of the wall against your cam/setup, Focus as best you can then post the image, zoom and other lens settings on the fujinon and 35mm lens.
this way, most of us who has experienced working with adapters especialy those with the same setup as yours can confirm or comment on the results you are getting.
With an adapter on the hd100 i suggest to set detail to normal or -1 up to -3. min or lower detail will add to the diffusion introduced by a focusing screen.
Ted
Bob Hart June 11th, 2008, 05:48 PM Ted has beaten me to it. I was about to ask the same question.
There are some contentions as to sharpening detail, the groundglass texture being visible to the camera and the risk that extra detail load on the codec will cause softening due to the codec throwing it away to preserve frame rate.
I contend that in good lighting, adequate groundglass motion and correct shutter speeds, there should be actually less detail load on the codec with selective focus going on.
When it bites is where there is low light combining with tighter lenses already provoking artifacts. The added video gain noise makes the sharpness disappear dramatically.
My personal preference is to bring the detail up a little. You may have to drop it back a little with wide lenses. With wides, lots more background textural detail will be in sharper focus and the camera may want to try to sharpen it all.
Stuart Campbell June 20th, 2008, 09:55 AM My personal preference is to bring the detail up a little. You may have to drop it back a little with wide lenses. With wides, lots more background textural detail will be in sharper focus and the camera may want to try to sharpen it all.
What a great suggestion. The detail is set to min. Why didn't I think of that?
I've just got in from a huge unexpected short notice job filming 6 open heart operations in a hospital. So, as soon as i catch breath and sort myself out for the next shoot (which should be with the Letus) I get back to you with my next test results.
How did you get on with your test Bob?
Stuart
Bob Hart June 20th, 2008, 09:17 PM Stuart.
I shot some footage on the EX1 last weekend and it is posted now at www.exposureroom.com under the membership DARANGULAFILM. Any videos with Congoro in the title header was shot on it, only one minute each of slow-mo of medieval extras slugging it out with swords etc..
My relay zoom setting was about 42mm.
As for the JVC HD100, I did give it a short try but the firewire port was still burnt out and I could not capture it and had to hand the camera back for the weekend shoot. I could not copy out analogue to the Sony on 1080i. It just would not do it - blue screen only.
There was I recall a similar offset to that seen on some of the other images posted here. I was able to get 862 or better lines of sharpness on the Lemac chart.
You can gain some apparent resolution compared to the P+S when you zoom back as far as you can off the groundglass without vignetting. On the same frame size as the P+S, you lose some and the contrast off the P+S is better.
This of course is not a truthful comparative test as the P+S is direct relay and the Letus image is going through more pieces of glass via the Fujinon lens.
Stuart Campbell June 23rd, 2008, 04:37 AM Hi Guys,
well I had a couple of hours spare this morning and re built the unit and set detail back to normal. There was a noticeable difference in the right direction unsurprisingly but chromatic aberation became more noticeable.
Lighting conditions were not perfect at the time of the test, but good enough to shoot in. There could have been a couple of other factors to explain the aberation now I think about it but I'm in too much of a hurry to investigate further.
Having improved the focus a little we are going to take it out and shoot some stuff with it on Thursday.
I'll post some results when we get back.
(still no contact from Letus. pretty poor customer service in my book!)
Ted Ramasola June 23rd, 2008, 07:04 AM Stuart,
take note that CA in this setup can be caused by two things.
1. your stock fujinon when its aperture at f2 and wider. you can isolate this by going for at least f2.8 or lower. Also the CA exhibited by the fujinon has the tell tale color of green and magenta.
2. the achromat. in my observation of the brevis and sgpro, this is tinted towards the blue and yellow colors. Not so sure about the CA from a letus achromat.
also , better edge to edge sharpness is achievable when your, at least, at f8 on the fujinon.
|
|