I really should know this by now, but I can't quite find concise info in my notes.
Several of my alt lenses don't seem to be dead-on for infinity, on my 5D, and I need to know just exactly how you test this as I calibrate.
Also: My Zeiss ZF50 macro, with Kindai adapter, reaches the end of its focus-travel with subjects as close as about 30 feet. Does that seem normal? It also kinda-sorta- seems dead-focused on distant landscapes, but I want to be sure.
Live-view would help, but we're talking 5D here
------------------------------------------------------------------------------------------------
I do understand the basic ideas behind the adjustment itself: Thinning the shim or adapter, or changing the stop-position on those lenses that have them. But how the heck to you test?
Just shoot a distant landscape with the lens full-out, then zoom-in on the image? That seems incredibly time-consuming.
If you do that, and the image is slightly soft, how do you determine if the problem is actually IF adjustment, vs the lens simply being at it's sharpness limit?
Also- I've read hear that some folks consider "near infinity" (or some such term) to be more useful. Can you explain this? For instance, with a 50mm lens at f/8, what is the closest distance that should be dead-crisp with the lens adjusted full-out?
Does this distance change for a 24mm lens?
Can adjusting to this number cuse the lens to focus PAST infinity?
Right, heres the best way beacuse I feel your pain as no one came to my rescue on this forum either when i started out in alternative lenses 6 months ago -
Take the lens and fit the adapter exactly as it is, no shims just stock as you bought it....
Fit to camera.
Set the aperture to wide open
Set the lens focusing to ~ (infinity)
Go outside your house, find a scene with a view to infinity (like the gate in the field opposite my house for instance) and set the centre of the frame to pick out a feature (like a wooden gate in the field)
Starting at the lenses ~ mark take a shot.
Move the focus to just off infinity (lets say 1mm off infinity for instance) Take a shot
Keep creeping the lens and taking shots, around five should do it.
Open the files in RAW software (Capture One is excellent for this as it allows you to see many files open at the same time)
Analysis -
If the adapter is the correct thickness the first shot taken at infinity will be in focus
If the adapter is thinner than required the lens will focus past infinity. Shot number 2 or three will be in focus.
If the adapter is thicker than required the lens will never focus on infinity.
Dont let your eyes deveive you. Its possible even for a f1.4 lens shot wide open to show good rendition of infinity, even if lacking contrast. If its not in focus, its not in focus.
Solutions -
If the adapter is too thin, get some parcel tape (thin brown stuff) and some electrical tape. Stick a small length of electrical tape to the table and using a craft knife cut some shims, three to be precise. Unclip the adapter from the lens and stick the shims to the adapter on the lens side, not the camera body side at 2, 6 and 10 o'clock. Put the adpater back on, it will be much tighter and more tricky to fit. Go outside and repeat the test. If its still not thick enough, undo the adapter and place shims of parcel tape on top of the electrical tape to increase the thickness. Retest and keep going until ~ is where it should be, at the ~ mark on the lens. Now leave this adapter on the lens for the rest of its life! Never take it off again.
If the adapter is too thick, I suggest looking for a thinner adapter. Alternatively you can get the adapter, stick it flat on some wet and dry (fine sand paper for rubbing down paintwork on cars) and carefully thin the adapter by rubbing in circular motions.
Some thoughts -
You will find more than often the adapters are not thick enough. I have an adapter on my Contax 28mm f2 lens that had to be ludicrously thicker than the normal 1.42mm (its never exact from my findings) I had to be 1.54mm, no way could I shim a 1.42mm adapter to that thickness and get the thing on the lens! Its a crapshoot as someone said. No one lens will require exactly the same thickness lens as the rest. All these super expensive adpaters may have better quality chips, aluminium etc but the end result is more than likely you will need to do modification ot the lens if you are to achieve ~ where it is supposed to be. No $250 adapter is the answer. I have adapted 6 lenses and all have el-cheapo £10 adapters bought off Ebay. The only expensive adpater I bought, a Happypage one, is sat on my shelf doing nothing because it was exactly the same as the others.
Just take your time testing out the lenses, the results can be unbelievabe when set up right.
From a practical point of view, with normal and wider lenses, I'm happy with infinity focus when I see a crisp image of a power transmission tower that's located on a hilltop about 5 km (3 mi) away from my house. For longer lenses, "infinity" is a lot further away and you're often limited by atmospheric clarity, and so it's not easy to do. OTOH, if you happen to have a collimator on hand (as my local Camera Repair Centre does), then you can test for infinity focus with a bench rig.
When you start to discuss "dead-crisp" focus (I'd call it "critical focus"), then that's what you get at the exact point of focus. Zones of "acceptable focus" in front of and behind the subject (aka Depth of Field), are related to focal length, distance, aperture and the "circle of confusion" (CC), and CC is related to the size and distance from which you want to view the image. google "the ins and outs of focus" to find a downloadable version of Harold Merklinger's excellent book on the topic - warning, it's not a trivial read.
Answers to your questions -
"I do understand the basic ideas behind the adjustment itself: Thinning the shim or adapter, or changing the stop-position on those lenses that have them. But how the heck to you test?
DO not rely on stopping down to bring infinity in to focus, thats a bad idea. Some lenses have to have elements in alignment to bring the edges into correct focus like ones with floating elements (the Contax 28mm f2 for example) Set infinity to the right place by shimming / choosing the right adapter only. By chaniging the focusing ring position you will lose depth of field. My Oly21mmf3.5 was never adapted properly before i sold it. It never had as much DOF as my 17-40 f4L. I could bring IF in to line by stopping right down but i am sure this is not the right thing to do.
"Just shoot a distant landscape with the lens full-out, then zoom-in on the image?
That seems incredibly time-consuming."
Some lenses are not parfocal, in other words when the lens is zoomed in, infinity wavers a little from when it is zoomed out. I have been lucky in that the lenses I have adjusted have never exhibited this problem as they are all primes (bar one). Yes lens testing is very time consuming but once the har dwork is done, its done.
For a given focal length and aperture, there's a hyperfocal distance that will render everything from that distance to infinity in focus, via depth of field. These charts should be online somewhere. That's a good start towards understanding some of the above questions.
David Clapp wrote:
Answers to your questions -
"I do understand the basic ideas behind the adjustment itself: Thinning the shim or adapter, or changing the stop-position on those lenses that have them. But how the heck to you test?
DO not rely on stopping down to bring infinity in to focus, thats a bad idea.
That's not what I wrote. I'm talking about certain lenses that have built-in, adjustable stops for their infinity setting. It would be great if all lenses had this, but most don't.
I would not rely on hyperfocal distance at all, this is no substitute for a correctly set up lens. Take the time to sort out the ~ point and then get into hyperfocal distance later.
Cableaddict wrote:
Great stuff, so far. Thanks gents.
That's not what I wrote. I'm talking about certain lenses that have built-in, adjustable stops for their infinity setting. It would be great if all lenses had this, but most don't.
I am sorry, I dont understand what you mean, do you mean to be able to reposition and adjust the focus ring by losening a screw or something like the Olys for example?
David Clapp wrote:
Set the aperture to wide open
Set the lens focusing to ~ (infinity)
Go outside your house, find a scene with a view to infinity (like the gate in the field opposite my house for instance) and set the centre of the frame to pick out a feature (like a wooden gate in the field)
Starting at the lenses ~ mark take a shot.
Move the focus to just off infinity (lets say 1mm off infinity for instance) Take a shot
Keep creeping the lens and taking shots, around five should do it.
Open the files in RAW software (Capture One is excellent for this as it allows you to see many files open at the same time)
Analysis -
If the adapter is the correct thickness the first shot taken at infinity will be in focus ...Show more →
Here's two obvious problems with this technique (and the reason I started this thread)
1: Some lenses exhibit focus-shift when stopped down. Since you would most likely be taking distance shots with a larger DOF, wouldn't it be safer to make your calibration at f/8?
-Or perhaps test at f/16 AND wide open, and find the happy medium if they don't match?
2: You method only works if the lens is currently focussing PAST infinity. This is the only way you'd know for sure that one of the 5 shots was actually in critical-focus. (I'll adopt Jim's term from now on)
As I wrote originally, let's say shot #1 was the crispest of the 5. How do you know that shot # "1/2" would not have been even better? (i.e., you still have a ways to go) The only way I can see this working is to first PURPOSELY thin the adapter or shim, so that maximum crispness is achieved within shots 2-4, then add tape until shot #1 matches the original best shot.
jcolwell wrote:
... if you happen to have a collimator on hand (as my local Camera Repair Centre does), then you can test for infinity focus with a bench rig. .
Now just where DID I put that collimator? I knew I should have kept it handy!
-But seriously, are they any wall-charts, or home-made gizmos that can be used?
jcolwell wrote:
When you start to discuss "dead-crisp" focus (I'd call it "critical focus"), then that's what you get at the exact point of focus. Zones of "acceptable focus" in front of and behind the subject (aka Depth of Field), are related to focal length, distance, aperture and the "circle of confusion" (CC)....
And there's the rub, exactly. Confusion is right!'
It makes me start to wonder if the infinity-focus on even my Canon lenses is really "exactly right."
Are they relying on hyperfocal focusing, to keep QC costs down?
If a lens or camera conistently front- or backfocuses, then you should consider sending it in for calibration. Otherwise, you're in charge; if you don't like where it's focused, then change it. If you're shooting fast action and you want reliable AF, then you should go to the 1D series - it's not perfect by any means, but it's not bad either.
I've read many times (but not in any Canon publications) that the EOS AF system promises to put the point of focus within the DOF, but really doesn't matter, because AF doesn't always focus where you want it to. That's one reason that AF lenses still have manual focus rings, and the better AF lenses usually have better manual focus capabilities.
As was suggested to me previously, try taking your test shots from an elevated position (I used a freeway overpass near my home). This will take the immediate foreground out of the picture (no pun intended).
Mike, can you explain how that is helpful? I don't quite understand.
Jim,
I almost never use AF, anyway, so that's a non-issue for me. I have zero problems using manual focus, evn when the subject is moving, unless the light is very low or I need a huge DOF.
Cableaddict wrote:
Mike, can you explain how that is helpful? I don't quite understand.
Jim,
I almost never use AF, anyway, so that's a non-issue for me. I have zero problems using manual focus, evn when the subject is moving, unless the light is very low or I need a huge DOF.
I'm using a manual focus Mamiya 645 300/2.8 APO lens on 1DII to shoot rugby games - I get many more keepers than I thought I would. I guess I forgot that I successfully shot with MF lenses for about 25 years, before getting into AF gear.
David,' no worries! You got this discussion going in a very useful way. (thanks)
Stay tuned, it might get more interesting.
--------------------------------
Jim, it is truly amazing what one can do with MF if you try hard enough!
I like the challenge of MF, even with sports, not to mention the accuracy, as you always know what's being locked onto. There's even an advantage: With MF, you can shoot burst mode while the action moves by you / away from you, and adjust focus at the same time. (granted, your keeper rate may go down a bit!) Unless I'm missing something, you CAN'T do that with AF, even on a 1DsIII.
Cableaddict wrote:
With MF, you can shoot burst mode while the action moves by you / away from you, and adjust focus at the same time.
I do exactly that. The 1DII is on high speed drive, and the lens is MF.
Cableaddict wrote:
Unless I'm missing something, you CAN'T do that with AF, even on a 1DsIII.
I do a similar thing with my 70-200/2.8L IS that I'm also using on my other 1DII body. Normally, I use AI Servo (with AF activated on the * button). When there's foreground clutter (i.e. players) in front of the action, I "pump" the AF using the centre AF point on whatever part of the action that I can see, with the centre AF point as my "registered" AF action on the WB button.
Cableaddict wrote:
Mike, can you explain how that is helpful? I don't quite understand.
This will give you infinity across the entire frame...at least that's how it had been described to me in this forum. I tried a search, but couldn't locate the exact thread where this advice came from. It may have been in one of the "Oly xx is a gem" threads...
EDIT: Here ya go...found the thread. The recommendation to use an elevated vantage point was mentioned early in the discussion.
