Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
02-08-2010, 07:22 AM   #1
Veteran Member
GoremanX's Avatar

Join Date: Dec 2009
Location: Georgia, VT
Photos: Albums
Posts: 1,657
Focus, Focus, Focus (or the usefulness of focus charts)

I'm starting a new thread for this because I thought it would make a good discussion. The original thread is here and the post I'll be referencing is this one.

The relevant quote is:
QuoteOriginally posted by Ben_Edict Quote
And lastly I want to add, that these focus charts are more or less meaningless. As you can see, it expects the camera to focus onto that black line in the centre of the chart. As nobody actually knows, where the central focus point is pointing to (the small illuminated rectangle is only a very rough approximation) and the AF point is bigger anyway, these charts are misleading: the AF point often simply doesw not lock onto the focus line, but a bit obave that (mostly) or bewlow. In these cases it seems, the lens is misfocusing.

The only valid test targets ar completely plano-parallel to the camera sensor, i.e. the famous brick wall.
This is something I've read from other forum members before, and it got me thinking; does it really matter where the camera's focus point really is? When framing a picture and focusing on a subject, isn't it more important that the camera's auto-focus be calibrated relative to the center of the viewfinder rather than where its focus points reside?

I realize that the auto-focus system might be targeting slightly above or below the area shown by the focus indicator. But the way I see it, a proper focus chart sitting at a 45 degree angle allows me to calibrate my camera so that the auto-focus system is in line with what I actually see. The physical location of the focus sensor seems irrelevant.

Meanwhile, a brick wall would only calibrate the auto-focus system for distance from the lens, not for the center of the viewfinder.

By "proper" focus chart, I mean a real chart where the camera only has one specific line to focus on. A ruler is not a proper focus chart because the camera could use any of the ruler's lines near the focus point to make a focusing decision.

This is a proper focus chart:
View Picture EXIF
Name:  GORE4702.JPG
Views: 3200
Size:  15.5 KB

This is not:
View Picture EXIF
Name:  GORE4703.JPG
Views: 2928
Size:  9.2 KB

02-08-2010, 07:53 AM   #2
Veteran Member
jeffkrol's Avatar

Join Date: Sep 2006
Location: Wisconsin USA
Photos: Gallery
Posts: 8,434
Actually........ this is a proper focus chart. Yes I know it's expensive but it's a model too..
LensAlign® Focus Calibration System
02-08-2010, 08:00 AM   #3
Veteran Member
GoremanX's Avatar

Join Date: Dec 2009
Location: Georgia, VT
Photos: Albums
Posts: 1,657
Original Poster
QuoteOriginally posted by jeffkrol Quote
Actually........ this is a proper focus chart. Yes I know it's expensive but it's a model too..
LensAlign® Focus Calibration System
See that looks too much like a ruler to me. The auto-focus system might focus on a different line each time you re-focus, introducing randomness into the results.
02-08-2010, 08:08 AM   #4
Site Supporter
Site Supporter
Lowell Goudge's Avatar

Join Date: Jan 2007
Location: Toronto
Photos: Gallery | Albums
Posts: 17,868
I think that when people look at the lens align focusing chart they confuse the ruler at the side with the target you are focusing on.

The target is large, and vertical (perpendicular to the camera) and forces the camera to focus on a surface at a single distance from the camera.

the ruler is at an angle to allow the user to determine how far back or front of the purpendicular target the camera actually focused but is not used by the focusing system as a target, only a measurement tool

What is important to note is that for this focusing chart and all others you must use the central spot, not auto selection. and they must be perfectly alligned to the camera.

Additionally, All too often I see focusing charts that were not alligned properly, so the reading on the side is done at a different distance than the central line.

for any focusing system, the only way to evaluate it is with a perpendicular target, not a sloping one since there is no way of insuring the focus spot. Assuming either a single line or the half moon is trying to force the camera to focusi on that line in the total range of the sensor by assuming that presenting a high contrast point somewhere in teh sensors range will force the focus point, I am not so sure.

02-08-2010, 08:16 AM   #5
Veteran Member
jeffkrol's Avatar

Join Date: Sep 2006
Location: Wisconsin USA
Photos: Gallery
Posts: 8,434
Very good read:
REVIEW: LensAlign Focus Calibration System - Canon Photography Group
02-08-2010, 08:17 AM   #6
Veteran Member
Ben_Edict's Avatar

Join Date: Jul 2007
Location: SouthWest "Regio"
Photos: Albums
Posts: 3,309
The Lens Align sure is one of the more useful test target and prooves my point: the focus target needs to be plano-parallel to the sensor, to be of use. In the LensAlign arrangement, the ruler is only there to estimate the amount of any back- or front focus, if there is one at all.

To GoremanX:

Ofcourse your question is absolutely relevant. I think there are actually two answers to it:

1 for these test rulers the exact (very ecact indeed) AF point position is extremely relevant, as these targets work by ssuming, that the camera can focus on a tiny hairline.

2. a slightly misaligned focus point (in comparisson to the AF-point marker in the viewfinder) will lead to a test ruler shot showing front- or back focusing, but in real world images, where usually the AF target is larger than that hairline, the focus would be fine.

Also, generally these ruler targets are small and are imaged from too short a distance, unless one tests macro lenses. At these near distances (as discussed in the old reference thread) lenses do not perform well and the DOF is tiny. Both problems may lead to an increased perception of mis-focus.

Ben
02-09-2010, 08:56 PM   #7
Senior Member




Join Date: Aug 2008
Location: Waxhaw, NC, USA
Photos: Gallery | Albums
Posts: 211
I'm a LensAlign user myself and can attest to it's usefulness compared to homemade angled ruler gizmos. As has already been stated, the angled ruler of the LensAlign is only to check front/back focus but you're actually focusing on a vertical target.

My only advice to folks would be to spring for the LRK (Long Ruler Kit) right from the get-go. I didn't and wound up paying more than I needed too after I realized I needed it for testing longer lenses. Great tool and very easy to use.

Regards,
Terry Wyse

02-09-2010, 10:59 PM   #8
Senior Member




Join Date: Sep 2009
Posts: 253
There is also the method of checking focusing accuracy using Moire Interference Patterns. Here are examples:

http://www.northlight-images.co.uk/...ges/cameras/1ds3_af_micoadjustment.html

Using Moire' Interference Patterns to Test DSLR Auto Focus
02-11-2010, 07:26 PM   #9
Otis Memorial Pentaxian
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 42,007
QuoteOriginally posted by photogerald Quote
There is also the method of checking focusing accuracy using Moire Interference Patterns. Here are examples:

http://www.northlight-images.co.uk/...ges/cameras/1ds3_af_micoadjustment.html

Using Moire' Interference Patterns to Test DSLR Auto Focus
Here is the correct link to the Northern Lights page:
AF microadjustment for the 1Ds mark III, 1D Mk3, 5D Mk2, 7D
And another incredibly useful page on the same site:
test images for printing and lens resolution
Steve

BTW...The moire test is VERY interesting. Manual focus using the Katzeye is spot on. AF is...well, I may be looking into having the AF calibrated.

Last edited by stevebrot; 02-11-2010 at 08:15 PM.
02-12-2010, 06:10 AM   #10
Veteran Member
GeneV's Avatar

Join Date: Oct 2008
Location: Albuquerque NM
Photos: Albums
Posts: 9,830
Very interesting thread, despite the number of times it has been discussed.

The instructions for the Moire test bring to mind that the ony thing that the inclined ruler or expensive target saves you is a bit of chimping by trying the focus slightly out and in to determine back or front focus after locking on a flat target with high contrast. Without the inclined scale, calibrating the focus is something like the process at the optometrist, where he asks "better or worse" with corrections in both directions. Am I understanding this correctly?

Also, wouldn't it be useful to calibrate so that focus point shown in the viewfinder is actually the point of focus, or is it too variable to be done? It seems to me that having the most exacting AF in the world is not terribly useful if you can't place the point of focus in a three dimensional scene with any accuracy.

Last edited by GeneV; 02-12-2010 at 06:22 AM.
02-12-2010, 07:04 AM   #11
Veteran Member
GoremanX's Avatar

Join Date: Dec 2009
Location: Georgia, VT
Photos: Albums
Posts: 1,657
Original Poster
QuoteOriginally posted by GeneV Quote
The instructions for the Moire test bring to mind that the ony thing that the inclined ruler or expensive target saves you is a bit of chimping by trying the focus slightly out and in to determine back or front focus after locking on a flat target with high contrast. Without the inclined scale, calibrating the focus is something like the process at the optometrist, where he asks "better or worse" with corrections in both directions. Am I understanding this correctly?
My wife the optometrist would take exception to that. It's not actually a zero-ing in process of "better or worse", but most people (like me) who don't have years and years of University training don't get that.

But in essence, your impression is correct. When using a flat surface for calibrating focus, you're just doing trial-and-error.

QuoteOriginally posted by GeneV Quote
Also, wouldn't it be useful to calibrate so that focus point shown in the viewfinder is actually the point of focus, or is it too variable to be done? It seems to me that having the most exacting AF in the world is not terribly useful if you can't place the point of focus in a three dimensional scene with any accuracy.
This is exactly my point, and is precisely where I feel the inclined scales have the advantage. When calibrating using a flat surface, the camera's actual focus sensor could be anywhere in the field of view and you'd still get "accurate" focus of the flat surface. But when using an inclined scale, you're calibrating your camera to focus on the actual focus point display. By that I mean the red square that flashes when you focus, which I referred to in my OP as the "center of the viewfinder" since the K-x doesn't have that flashing red square.

What prompted me to bring this up is that someone mentioned inclined scales were misleading because the actual focus point might not be aligned with that flashing red square. But I maintain that it's better to calibrate for that flashing red mark, not where the focus sensor really is. My pictures are almost never a flat surface.
02-12-2010, 07:30 AM   #12
Site Supporter
Site Supporter
Lowell Goudge's Avatar

Join Date: Jan 2007
Location: Toronto
Photos: Gallery | Albums
Posts: 17,868
QuoteOriginally posted by GoremanX Quote
My wife the optometrist would take exception to that. It's not actually a zero-ing in process of "better or worse", but most people (like me) who don't have years and years of University training don't get that.

But in essence, your impression is correct. When using a flat surface for calibrating focus, you're just doing trial-and-error.
Ask your wife how she selects the lens corrections for each "trial and error" test. there is a method to insure convergence quickly
QuoteQuote:



This is exactly my point, and is precisely where I feel the inclined scales have the advantage. When calibrating using a flat surface, the camera's actual focus sensor could be anywhere in the field of view and you'd still get "accurate" focus of the flat surface. But when using an inclined scale, you're calibrating your camera to focus on the actual focus point display. By that I mean the red square that flashes when you focus, which I referred to in my OP as the "center of the viewfinder" since the K-x doesn't have that flashing red square.

What prompted me to bring this up is that someone mentioned inclined scales were misleading because the actual focus point might not be aligned with that flashing red square. But I maintain that it's better to calibrate for that flashing red mark, not where the focus sensor really is. My pictures are almost never a flat surface.
the lens align chart if yo ulook closely has a combination of the inclines surface, and the flat target to focus.

the target insures you are focused on one precise plane, and the inclined ruler tells you how far front or back of the plane the actual focus point is.

it combines both issues if used fully, a vertical chart on its own, does leave you with trial and error.

Also, if you are using the flashing red mark as focus information that is just wrong. The focus indication is the green hexagon. the red mark shows which sensor is being used but is NOT the indication of being in focus
02-12-2010, 07:54 AM   #13
Veteran Member
GoremanX's Avatar

Join Date: Dec 2009
Location: Georgia, VT
Photos: Albums
Posts: 1,657
Original Poster
QuoteOriginally posted by Lowell Goudge Quote
Also, if you are using the flashing red mark as focus information that is just wrong. The focus indication is the green hexagon. the red mark shows which sensor is being used but is NOT the indication of being in focus
erm... no, I don't do that, that would be silly. I use the focus indicator (red flashing rectangle) to confirm WHERE I'm focusing, not whether I'm focused or not.
02-12-2010, 08:23 AM   #14
Site Supporter
Site Supporter
Lowell Goudge's Avatar

Join Date: Jan 2007
Location: Toronto
Photos: Gallery | Albums
Posts: 17,868
QuoteOriginally posted by GoremanX Quote
erm... no, I don't do that, that would be silly. I use the focus indicator (red flashing rectangle) to confirm WHERE I'm focusing, not whether I'm focused or not.
Maybe I mis red what you said,

I went back and re-red it and while I understand why you referenced the red square, it still seems to read like you use it as focus indication. but perhapos I am reading too much into it.

Any way, in re-reading the post agian, when you say the focusing point may be anywhere, actually no, it is is one of 11 specific points and those points are not necessairly the size of the red dot.

the use of the flat focusing surface is to insure which ever spot you have selected for focusing is presented with a single plane at that spot, to focus on.. since the sensors are not infinitely small, presenting with a slope induces error since you don;t necessairly present the part of the sensor which is influences the decision with the correct data.

I fact the best target for the center sensor which is a cross sensor, would be a zigzag line on a flat surface.
02-12-2010, 11:15 AM   #15
Veteran Member
GeneV's Avatar

Join Date: Oct 2008
Location: Albuquerque NM
Photos: Albums
Posts: 9,830
QuoteOriginally posted by GoremanX Quote
My wife the optometrist would take exception to that. It's not actually a zero-ing in process of "better or worse", but most people (like me) who don't have years and years of University training don't get that.

But in essence, your impression is correct. When using a flat surface for calibrating focus, you're just doing trial-and-error.



This is exactly my point, and is precisely where I feel the inclined scales have the advantage. When calibrating using a flat surface, the camera's actual focus sensor could be anywhere in the field of view and you'd still get "accurate" focus of the flat surface. But when using an inclined scale, you're calibrating your camera to focus on the actual focus point display. By that I mean the red square that flashes when you focus, which I referred to in my OP as the "center of the viewfinder" since the K-x doesn't have that flashing red square.

What prompted me to bring this up is that someone mentioned inclined scales were misleading because the actual focus point might not be aligned with that flashing red square. But I maintain that it's better to calibrate for that flashing red mark, not where the focus sensor really is. My pictures are almost never a flat surface.
I do understand the the optometrist has narrowed down the possibilities before we get to the "better or worse" part, but I would think that we have done the same narrowing when the camera focuses on the flat surface. The front or back focus part is (hopefully) fine tuning.

What I honestly don't know about the red dot is whether the dot will be off from the point of focus by the same amount and in the same direction always, or whether we are talking about a margin for error. Also, can't the dot just as easily be off from the true point of focus laterally or diagonally as well as just in distance?

For example, if I put the red dot on the tip of a dog's nose to focus on that point, if the camera is actually focusing on a point left or right or diagonally of the nose, I will miss the focus and the photo will appear to have a back focus problem. I'm probably just being dense, but I am not seeing how the inclined scale would help me tell if I am focusing on the wrong point in the scene because the dot is off.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
af, auto-focus, calibrate, camera, canon, center, chart, charts, day, dslr, focus, k-5, lens, line, nikon, pentax, photography, pm, post, price, steve, system, thread, viewfinder
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Charts for Auto-Focus Check acknowledged by Pentax. ebooks4pentax Pentax DSLR Discussion 34 07-27-2015 06:05 AM
10 Reasons why manual focus primes are better than those new auto focus zoom lenses JoepLX3 Pentax SLR Lens Discussion 188 07-25-2012 03:28 PM
K100D Super - Focus point locked to center in Manual Focus ? JGabr Pentax DSLR Discussion 2 01-25-2010 09:41 PM
Front focus/ back focus on Manual lens. Possible? WangJianWei Troubleshooting and Beginner Help 6 04-20-2009 07:50 PM
My experiences with checking FF/BF using Yvon Bourque's focus charts Christine Tham Pentax DSLR Discussion 0 11-17-2008 04:47 PM



All times are GMT -7. The time now is 02:07 PM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top