very unscientific test

[Nawty]

Phase-detect AF works by comparing the image received through either side of the lens. As the f/number rises, the aperture becomes physically smaller and there comes a point when the two sides are too similar for the AF sensor to differentiate accurately. Canon calls a halt to proceedings over f/5.6 (with all but a few high end DSLRs that will run to f/8 ) and above that the AF simply switches off. Any lens or lens/telecon-combo that's above f/5.6 needs to fool the camera to keep the AF working, but that doesn't change the fact that the aperture is beyond the limit and performance is compromised. Some cameras continue to work reasonably well, while others are less good. As Bob says, no Canon lens has a maximum aperture higher than f/5.6.

Normal phase-detect AF can work in surprisingly low light, but needs good contrast, ie light/dark transitions and edges. It's worth noting though that the newer on-sensor phase-detect AF doesn't work so well in low light (at least, at the current stage of development). There's another step with Canon AF systems, when additional sensitivity on the AF sensor is activated at f/2.8 and lower f/numbers. Focusing is more accurate, and faster.

Contrast-detect AF in live view works at any f/number, and can be extremely accurate, but likes both good light and contrast.

Every day's a school day I hadn't realised that was why AF performance dropped off - it's just since using DoF preview on film cameras causes the scene to go so dark I assumed it was that.

 

[dean messenger]

Well...
He started it. He had the option of backing down, but he chose to argue

no phil you just like to try and be your usual bully self and get personal rather than stick to the subject.
calling someone clueless. sorry but where in the thread is there earlier mention of issues with the 5d and certain tc's?
no my earlier TC wasnt chipped and the one i have now is far more sophisticated.
where have i questioned the "bloke" as you put it who has wrote the tests exactly?
and then you end with your childish he started it comment.. just about sums you up really phil..
tbh i couldnt give two stuffs about what other tests have been done becasue as i stated right at the start of the thread and the opening title. Non scientific silliness, but instead the usual forum gobs decide to start getting there knickers in a twist and having there pathetic immature personal little digs rather than take the post as it was meant to be.
but of course i bow to your know it all knowledge because of course you can sit there and say for sure my kenko TC doesnt correctly report the aperture because your the expert arent you.
its no wonder so many people have left this forum and so many beginners feel unwelcome when you have such clever no alls around who are only to happy to throw there verbal weight around rather than actually do something helpful.
and chris you talk total rubbish. never did i say or assume anything concerning AF or AE, and as for experimental.. no taking 4 shots at the same subject with 4 different focal lengths is not an experiment.
i suggest you stop trying to read so much into things and actually make an effort to try and understand that it was simply to see the difference between the lens when its pointed at a static subject.. that in no way constitutes an experiment ,

and yes goody its second nature on this forum there are quite a few whos sole purpose seems to be to insult and belittle others as another thread on here recently regarding bird shots clearly proves

 

[Phil V]

no phil you just like to try and be your usual bully self and get personal rather than stick to the subject.
calling someone clueless. sorry but where in the thread is there earlier mention of issues with the 5d and certain tc's?
no my earlier TC wasnt chipped and the one i have now is far more sophisticated.
where have i questioned the "bloke" as you put it who has wrote the tests exactly?
...

Calm down Dean.
I corrected your initial obvious error with humour and good nature.

Rather than accepting that, you decided to argue the toss. Which then led to everyone else telling you that you were wrong . Including the bloke what does the testing (I'm sorry if you have no idea who that is, I'm not naming him - this isn't his argument).

Rather than writing this rant, you could have looked up what happens to an aperture when you add a TC (I tried to be helpful by giving you links ), then you'd have learnt something and maintained a little respect. Instead you've ranted at me and looked even more clueless . You'd then realise why your new TC might be optically superior, but is electronically inferior to your previous one, I'm not making this stuff up, I didn't enter this to argue, I did it to help you out.

But sure - if you want to blame me because you went off on one rather than doing some research (to reassure you I was right) - feel free. It matters not a jot to me

Maybe next time someone questions or corrects you, rather than digging in for a fight, you could do the leg-work, then you'd learn a little and have less stress in your life.

 

[Deleted member 66768]

Calm down Dean.
I corrected your initial obvious error with humour and good nature.

Rather than accepting that, you decided to argue the toss. Which then led to everyone else telling you that you were wrong . Including the bloke what does the testing (I'm sorry if you have no idea who that is, I'm not naming him - this isn't his argument).

Rather than writing this rant, you could have looked up what happens to an aperture when you add a TC (I tried to be helpful by giving you links ), then you'd have learnt something and maintained a little respect. Instead you've ranted at me and looked even more clueless . You'd then realise why your new TC might be optically superior, but is electronically inferior to your previous one, I'm not making this stuff up, I didn't enter this to argue, I did it to help you out.

But sure - if you want to blame me because you went off on one rather than doing some research (to reassure you I was right) - feel free. It matters not a jot to me

Maybe next time someone questions or corrects you, rather than digging in for a fight, you could do the leg-work, then you'd learn a little and have less stress in your life.

Why do you always have to be an absolute bullroot mate? Anytime I see your name come up its always some smart arse remark to someone else with the obligatory oh I am only being humorous after the fact.

 

[Phil V]

Why do you always have to be an absolute bullroot mate? Anytime I see your name come up its always some smart arse remark to someone else with the obligatory oh I am only being humorous after the fact.

Definitely not for the same reason you had to point it out

Did you read the hundreds of informative and fun posts, or just the ones where I came up against someone stubborn and or I'll-informed?

I entered this thread to help Dean out, he took offence, it escalated. I'm not the only one who pointed out Dean's errors, I'm the only one he felt the need to have a go at.

 

[Byker28i]

Phase-detect AF works by comparing the image received through either side of the lens. As the f/number rises, the aperture becomes physically smaller and there comes a point when the two sides are too similar for the AF sensor to differentiate accurately. Canon calls a halt to proceedings over f/5.6 (with all but a few high end DSLRs that will run to f/8 ) and above that the AF simply switches off. Any lens or lens/telecon-combo that's above f/5.6 needs to fool the camera to keep the AF working, but that doesn't change the fact that the aperture is beyond the limit and performance is compromised. Some cameras continue to work reasonably well, while others are less good. As Bob says, no Canon lens has a maximum aperture higher than f/5.6.

Normal phase-detect AF can work in surprisingly low light, but needs good contrast, ie light/dark transitions and edges. It's worth noting though that the newer on-sensor phase-detect AF doesn't work so well in low light (at least, at the current stage of development). There's another step with Canon AF systems, when additional sensitivity on the AF sensor is activated at f/2.8 and lower f/numbers. Focusing is more accurate, and faster.

Contrast-detect AF in live view works at any f/number, and can be extremely accurate, but likes both good light and contrast.

This perfectly explains why in low light it's easier to get it to work if you use the edge of a light to dark area.

 

[Nawty]

Why do you always have to be an absolute bullroot mate? Anytime I see your name come up its always some smart arse remark to someone else with the obligatory oh I am only being humorous after the fact.

Definitely not for the same reason you had to point it out

Did you read the hundreds of informative and fun posts, or just the ones where I came up against someone stubborn and or I'll-informed?

I entered this thread to help Dean out, he took offence, it escalated. I'm not the only one who pointed out Dean's errors, I'm the only one he felt the need to have a go at.

The OP either understands the basic theory in his head but can't communicate it or doesn't (which is what his posts suggest) and is wilfully ignoring advice and taking offence from it.

Either way you can't teach people who don't want to learn so I stopped bothering but I think Phil was perfectly within reasonable conduct here.

This is a forum, if you are wrong it is the duty of others to correct it irrespective of the OP so as misunderstanding isn't propagated as truth (as I was by Hoppy, and am glad he did because I learned something).

 

[boyfalldown]

removed because as usual the forum bully boys decide to get pathetically personal
think some people need to get off there arse and spend some time with there cameras rather than sit on here all day having digs at people and making personal comments

Nice edit - not. Toys out of your pram much? On the personal comments maybe you ought to take a long look at yourself.

 

[chris malcolm]

The OP either understands the basic theory in his head but can't communicate it or doesn't (which is what his posts suggest) and is wilfully ignoring advice and taking offence from it.

It's always difficult to tell the difference between a confused expression of a clear idea or a clear expression of a confused idea. There's not much chance of becoming less confused if you find correction insulting.