[audiobomber]

Can anyone link me to a decent chart for the job. I know I could search myself but I want one someone has experience of using to point me in the right direction.

I find this chart is simple and effective. http://focustestchart.com/focus21.pdf

[bossa]

I find this chart is simple and effective. http://focustestchart.com/focus21.pdf

Yeah but it really works when you add this one to it:

http://focustestchart.com/focus10.pdf

[bossa]

Instead of putting the camera on a tripod and aiming at some sheet of paper on a desk do the following:

1: Use that second PDF link I posted: print and make the focus target and chart contained in the PDF.
2: Tape the chart to a door (Tape all around the edge of the chart to try and keep it flat against the door)
3: Make sure all the bits are lined up in the VF by opening/closing the door the required amount
4: Make sure that the camera's mirror is in lock-up mode as well as using the self timer for shooting from the tripod.
5: Make sure EVERY test shot is made from an OUT OF FOCUS starting point. (Force the camera to focus on your hand waving in front of the lens prior to making the next shot)
6: Start taking shots from the maximum MINUS through to the maximum POSITIVE adjustment settings on the camera. This will allow you to count through pictures in LR if you get a bit confused so you can work out which is which.

PS: The camera needs to be around 50x the focal length from the chart if possible for accurate results.

Regarding the focus target: You might want to change the text on that to a bunch of 'X" or dots to make sure it's the strongest contrast area visible to the camera. The focus target is definitely worth making as it makes sure the camera is focusing on a plane parallel to the sensor and not some vague text sloping away on the chart.

* Make sure your lighting is as bright as possible.

[vanyagor]

Instead of putting the camera on a tripod and aiming at some sheet of paper on a desk do the following:
The focus target is definitely worth making as it makes sure the camera is focusing on a plane parallel to the sensor and not some vague text sloping away on the chart.

The author himself now thinks that the parallel-to-the-sensor focusing target is redundant and he explains it in detail in the link posted by audiobomber. Speaking about focusing on some tilted page of text, of course you do not know 100% what part of the letter has your camera chosen to AF on, however it doesn't matter much when you try to focus on a certain line and get another line far from the one you aimed at in focus.

My feeling is you do not need test charts etc. The little extra precision you gain is not worth the time you spend on printing this A rough test (even without a tripod) is enough IMHO to see if you have the issue or not, then if you are really interested in the exact amount of back/front focusing you may want to do a more thorough test. Maybe I'm wrong, but this approach worked for me.

[bossa]

The author himself now thinks that the parallel-to-the-sensor focusing target is redundant and he explains it in detail in the link posted by audiobomber. Speaking about focusing on some tilted page of text, of course you do not know 100% what part of the letter has your camera chosen to AF on, however it doesn't matter much when you try to focus on a certain line and get another line far from the one you aimed at in focus.

My feeling is you do not need test charts etc. The little extra precision you gain is not worth the time you spend on printing this A rough test (even without a tripod) is enough IMHO to see if you have the issue or not, then if you are really interested in the exact amount of back/front focusing you may want to do a more thorough test. Maybe I'm wrong, but this approach worked for me.

I found my own primitive way of doing all of this and it was to place a small black pebble on a concrete path with a couple of twigs a few inches either side of it. It's best to get the light earlier or later in the day so that the surface texture of the concrete helps in the identification of the focal zone. Getting the angles right is hard but you get a feel for it after awhile.

Regarding the target: I've tried the other methods and found that the target certainly helps get a consistent focus.

[audiobomber]

I have had great success with the Nikon chart. I know it's reliable because the results are consistent. I've stopped using a tripod to test focus, I just put the page on the sill of a daylit window with diffused sunlight so the shutter speed stays up and take a few shots. I'm sure other methods work too, but i know this one does, so don't knock it until you try it.

[veato]

Thanks for the advice. I'm going to try the guide audiobomber linked.

[veato]

When testing a zoom do you test many focal point? And if you get different readings at 17mm and 50mm?

[vanyagor]

When testing a zoom do you test many focal point? And if you get different readings at 17mm and 50mm?

Yes to both questions.

[Marc Sabatella]

Can you explain what a correctly exposed image means in this case.

Not in just a sentence or two. But any number of books on the subject will. The short version is just what I already said: any simple subject should be rendered with 12-13% reflectance what this actually means in practice is the tricky part. at least, that's correct from the camera's perspective. Whether or not it happens to meet your persoanl expectation for that particular scene is another matter.

I thought that a correctly exposed image is when the whole histogram is within the range

that's not even always possible. Depending on the metering mode you use and the scene, it's entirely oossible a correct exposure will clip at one end or the other.

In fact one has to look at color histogram too, not just the luminance histogram. Is that correct?

one desn't "have" to do any suchthing. Regarldes of what is techically correct from the standpoint of metering standards, what matters is how like the image itself.

Can you explain why wrong exposure wide open implies that it is camera's fault?

Because the camera meters with the lens wide open. It measures the actual amount of lens reaching the sensor, not the amount that would reach the sensor if the lens were stoppsed down.

Could it be that the aperture just doesn't open fully for whatever reason

if that were happening, then the camera would be metering that way, too. So the exposure should still match the shot.

[Marc Sabatella]

For the purpose of just finding out if your lens is back focusing or not you don't really need a chart. You just need to have you camera (and the test subject) fixed (tripod and table top serve perfectly well). A good measure of front/back focusing is how much AF correction you need to apply to fix it, so you don't really need a test chart imho. Please correct me if I'm wrong.

you're wrong. With a plain sheet of text, there is no way to know on which line the camera is actually trying to focus. The whole point of a focus chart is to provide an unambiguous target. It's not the *only* way to do this, but a sheet of text definitely won't.

[Penumbra]

Well met!

I am interested in this lens (17-50mm Tamron). I have no way of testing it prior to purchase (wont get into details as to why). I do know about the.. inconsistencies of this particular lens but i was hoping that by buying it new, in 2012 i will be getting a new version that probably is free of the known QC problems.

Doesn't seem like this judgement of mine is right? Can one be sure that by buying now, in 2012, a new versions of this lens he'll get a good copy or not?

Thank you!

[vanyagor]

Well met!

I am interested in this lens (17-50mm Tamron). I have no way of testing it prior to purchase (wont get into details as to why). I do know about the.. inconsistencies of this particular lens but i was hoping that by buying it new, in 2012 i will be getting a new version that probably is free of the known QC problems.

Doesn't seem like this judgement of mine is right? Can one be sure that by buying now, in 2012, a new versions of this lens he'll get a good copy or not?

Thank you!

I don't think it depends on the year. I purchased mine 2 month ago. It was a "Made in Japan" copy and brand new from authorized dealer.

3 more things I'd like to say:

1) despite all the troubles, I still love the lens
2) back/front focus may be a natural/unavoidable effect (see the article on back/front focusing on this forum for instance). Most problems might be due to user error
3) Tamron seems to have great customer service so if you live in the US or Canada, you should be able to get your lens fixed sooner or later (if it is wrong at all). In my case it is complicated. My lens is a US model, I'm in Canada now and then I'm moving to Brazil (and nevertheless I will keep the lens)

[loco]

Penumbra, I don't think you can be sure of anything. I would suggest you buy new from a reputable seller who will accept returns because then you can at least return it and try another copy. Plus, you get the six year warranty from Tamron if you buy new.

I have had this lens on my list as a kit lens replacement for some time but these issues have delayed my purchase. I have considered buying it used but am a bit afraid to go for it without the security of the Tamron warranty.