Fujifilm's image sensor based phase detection explained

An interesting article:

http://www.dpreview.com/news/1008/10...fujifilmpd.asp

This could potentially be used on DSLR's as well for both mirror up
shooting and video modes - however the cited lag (0.158s) is still 3x
longer than the shutter lag on most DSLRs (about 45 - 55ms).

--
gmail originated posts are filtered due to spam.

On Sat, 07 Aug 2010 14:20:34 -0400, Alan Browne
wrote:

An interesting article:

http://www.dpreview.com/news/1008/10...fujifilmpd.asp

This could potentially be used on DSLR's as well for both mirror up
shooting and video modes - however the cited lag (0.158s) is still 3x
longer than the shutter lag on most DSLRs (about 45 - 55ms).

WRONG.

Shutter-lag on the average DSLR runs from 55ms to 150ms.

http://www.lightningtrigger.com/CameraCompatibility6/CameraCompatibility6.htm
Read it and weep you lousy misinformation spewing pretend-photographer ****
of a DSLR-TROLL.

Alan Browne wrote:
An interesting article:

http://www.dpreview.com/news/1008/10...fujifilmpd.asp

This could potentially be used on DSLR's as well for both mirror up
shooting and video modes - however the cited lag (0.158s) is still 3x
longer than the shutter lag on most DSLRs (about 45 - 55ms).

The figures of around 50ms for many current dslrs is shutter lag only.
Focus times "as fast as 0.158 second" doesn't really mean much, as it
could mean that the camera was pre-focused, or perhaps not but they
don't say in that article.

On 10-08-07 18:36 , Me wrote:
Alan Browne wrote:
An interesting article:

http://www.dpreview.com/news/1008/10...fujifilmpd.asp

This could potentially be used on DSLR's as well for both mirror up
shooting and video modes - however the cited lag (0.158s) is still 3x
longer than the shutter lag on most DSLRs (about 45 - 55ms).

The figures of around 50ms for many current dslrs is shutter lag only.

Sure. However one can manually (and quickly) focus a DSLR or shoot with
no AF delay if it is already close or at focus. AF with most P&S is slow.

Focus times "as fast as 0.158 second" doesn't really mean much, as it
could mean that the camera was pre-focused, or perhaps not but they
don't say in that article.

I would assume that it is a 'typical case' and neither a near ready or
extreme case. Far better than many P&S' that take half a second and up
to focus and fire.


--
gmail originated posts are filtered due to spam.

Alan Browne wrote:
On 10-08-07 18:36 , Me wrote:
Alan Browne wrote:
An interesting article:

http://www.dpreview.com/news/1008/10...fujifilmpd.asp

This could potentially be used on DSLR's as well for both mirror up
shooting and video modes - however the cited lag (0.158s) is still 3x
longer than the shutter lag on most DSLRs (about 45 - 55ms).

The figures of around 50ms for many current dslrs is shutter lag only.

Sure. However one can manually (and quickly) focus a DSLR or shoot with
no AF delay if it is already close or at focus. AF with most P&S is slow.

One might be able to - but I can't manual focus any SLR with either the
speed , or precision (with AF slr viewfinders) that AF offers.

Focus times "as fast as 0.158 second" doesn't really mean much, as it
could mean that the camera was pre-focused, or perhaps not but they
don't say in that article.

I would assume that it is a 'typical case' and neither a near ready or
extreme case. Far better than many P&S' that take half a second and up
to focus and fire.

I wouldn't assume anything - the figure of 158ms doesn't sound like it
was pulled out of a hat, and their statement seems to indicate from "as
fast as" that "no faster than" 158ms is possible.
Combined focus/shutter lag in ideal conditions really isn't a good
indicator of overall AF performance.

On 10-08-09 16:47 , Me wrote:
Alan Browne wrote:
On 10-08-07 18:36 , Me wrote:
Alan Browne wrote:
An interesting article:

http://www.dpreview.com/news/1008/10...fujifilmpd.asp

This could potentially be used on DSLR's as well for both mirror up
shooting and video modes - however the cited lag (0.158s) is still 3x
longer than the shutter lag on most DSLRs (about 45 - 55ms).

The figures of around 50ms for many current dslrs is shutter lag only.

Sure. However one can manually (and quickly) focus a DSLR or shoot
with no AF delay if it is already close or at focus. AF with most P&S
is slow.

One might be able to - but I can't manual focus any SLR with either the
speed , or precision (with AF slr viewfinders) that AF offers.

That depends a lot on FL, subject and aperture. And to be sure the
"old" prism based MF was a lot more accurate than any AF which all
depend on a hysteresis band.

That said, with the 135 f/1.8, for some situations it is a lot faster
and accurate to use AF than the Mk I eyeball. For wider angle (say 85mm
and down) I can manually focus and shoot faster and accurately than
worrying about which sensor is on, recomposing, and so on.



Focus times "as fast as 0.158 second" doesn't really mean much, as it
could mean that the camera was pre-focused, or perhaps not but they
don't say in that article.

I would assume that it is a 'typical case' and neither a near ready or
extreme case. Far better than many P&S' that take half a second and up
to focus and fire.

I wouldn't assume anything - the figure of 158ms doesn't sound like it
was pulled out of a hat, and their statement seems to indicate from "as
fast as" that "no faster than" 158ms is possible.
Combined focus/shutter lag in ideal conditions really isn't a good
indicator of overall AF performance.

Nothing wrong with a reasonable assumption. They're not going to paint
it bad, but they should be cautious about over stating performance and
then being lambasted later.



--
gmail originated posts are filtered due to spam.

On Tue, 10 Aug 2010 10:03:36 -0400, Alan Browne
wrote:

On 10-08-09 16:47 , Me wrote:
Alan Browne wrote:
On 10-08-07 18:36 , Me wrote:
Alan Browne wrote:
An interesting article:

http://www.dpreview.com/news/1008/10...fujifilmpd.asp

This could potentially be used on DSLR's as well for both mirror up
shooting and video modes - however the cited lag (0.158s) is still 3x
longer than the shutter lag on most DSLRs (about 45 - 55ms).

The figures of around 50ms for many current dslrs is shutter lag only.

Sure. However one can manually (and quickly) focus a DSLR or shoot
with no AF delay if it is already close or at focus. AF with most P&S
is slow.

One might be able to - but I can't manual focus any SLR with either the
speed , or precision (with AF slr viewfinders) that AF offers.

That depends a lot on FL, subject and aperture. And to be sure the
"old" prism based MF was a lot more accurate than any AF which all
depend on a hysteresis band.

That said, with the 135 f/1.8, for some situations it is a lot faster
and accurate to use AF than the Mk I eyeball. For wider angle (say 85mm
and down) I can manually focus and shoot faster and accurately than
worrying about which sensor is on, recomposing, and so on.



Focus times "as fast as 0.158 second" doesn't really mean much, as it
could mean that the camera was pre-focused, or perhaps not but they
don't say in that article.

I would assume that it is a 'typical case' and neither a near ready or
extreme case. Far better than many P&S' that take half a second and up
to focus and fire.

I wouldn't assume anything - the figure of 158ms doesn't sound like it
was pulled out of a hat, and their statement seems to indicate from "as
fast as" that "no faster than" 158ms is possible.
Combined focus/shutter lag in ideal conditions really isn't a good
indicator of overall AF performance.

Nothing wrong with a reasonable assumption. They're not going to paint
it bad, but they should be cautious about over stating performance and
then being lambasted later.

Gotta love all these DSLR-Trolls that base their equipment on how automated
it is. Point and shoot much?

LOL!

Alan Browne wrote:

That depends a lot on FL, subject and aperture. And to be sure the "old"
prism based MF was a lot more accurate than any AF which all depend on a
hysteresis band.

I know what hysteresis is but I don't get the connection with AF. Please
explain, especially wrt phase detect.