About LV. A few questions.

[uaiomex]

It's being years now since I wanted first to ask these questions:
 
Why LV is possible while tethering to a monitor and not on the back's LCD?

I've read many times CCD's can't be used for LV, that you need a CMOS sensor. Don't P&S cameras use CCD's and they always had LV since the beginning?

And, If tethering (whatever the reasons) allow for LV from a digital MF back, how come nobody comes with a solution to use an iPad or iPhone?

I'm aware that P1 is now marketing some limited form of LV.

Thx in advance
Eduardo

[yaya]

Why LV is possible while tethering to a monitor and not on the back's LCD?

It requires quite a lot of processing power, which until recently was not available (now it is, on the Phase One IQ backs)

I've read many times CCD's can't be used for LV, that you need a CMOS sensor. Don't P&S cameras use CCD's and they always had LV since the beginning?

P&S use Interline CCD technology (as opposed to full-frame technology in MFDB) and this allows for "streaming" of data

And, If tethering (whatever the reasons) allow for LV from a digital MF back, how come nobody comes with a solution to use an iPad or iPhone?

Again this has all to do with processing power and fast I/O.

[MrSmith]

back manufacturers probably didn't want to invest in R&D (or lacked the money) or pay for Licensed IP to implement decent quality live view.
it takes a Canikon to develop stuff like that, problems like processing power and fast in/out are all solvable problems if there's the desire and expertise.

[yaya]

back manufacturers probably didn't want to invest in R&D (or lacked the money) or pay for Licensed IP to implement decent quality live view.
it takes a Canikon to develop stuff like that, problems like processing power and fast in/out are all solvable problems if there's the desire and expertise.

Probably worth a reminder that Leaf had (tethered) Live View since 1996. Not sure what Canon/ Nikon had back then so perhaps we can ignore the comment about expertise? Between Phase One and Leaf we hold quite a lot of imaging related IP so perhaps the IP comment can also be removed...

Of course everything is solvable but there are costs involved and it is not only about money but also about size, power consumption and several quality related issues.

Full frame CCDs are still believed to produce the highest image quality for most typical applications, which was always at the top of the MFDB priority list and therefore features such as on-board live view had to wait until technology allows them to be implemented

[MrSmith]

ther's live view (like on a canon) and then there's live view that looks like footage from a CCTV camera in a snowstorm.
they are both called 'live view' but very very different.
my comments still stand. (unless you are a moderator on this forum?)

[yaya]

ther's live view (like on a canon) and then there's live view that looks like footage from a CCTV camera in a snowstorm.
they are both called 'live view' but very very different.
my comments still stand. (unless you are a moderator on this forum?)

Given a full frame high resolution CCD I'm not sure what kind of Live View Canon would be able to get. And no I'm not a moderator

Yair

[design_freak]

Given a full frame high resolution CCD I'm not sure what kind of Live View Canon would be able to get. And no I'm not a moderator

Yair

+1
Canon would not do anything in that direction. Firstly, as you wrote, the technology has its limitations. Second. Because the MFDB market is not attractive for them. It is too little and too much money to invest.

[Wayne Fox]

back manufacturers probably didn't want to invest in R&D (or lacked the money) or pay for Licensed IP to implement decent quality live view.
it takes a Canikon to develop stuff like that, problems like processing power and fast in/out are all solvable problems if there's the desire and expertise.

ther's live view (like on a canon) and then there's live view that looks like footage from a CCTV camera in a snowstorm.
they are both called 'live view' but very very different.
my comments still stand. (unless you are a moderator on this forum?)

I think his points are valid and I think you are making some assumptions.  In fact, mfdb makers have been serious at this a very long time, and to me a more reasonable assumption would be Canikon could even be paying some licensing fees because others have been at it longer and probably invented it.  As far as paying licensing fees for LiveView, not sure who invented that but since every Point and shoot has live view, i think it predates putting it into canikon dSLR's by over a decade ... maybe isn't even covered by any patents.  It really is about packing enough processing power into the back without draining a battery every 10 minutes.

If I'm paying 10's of thousands of dollars for a MFDB, my 1st, 2nd, 3rd, 4th and 5th concerns are IQ - and you can probably go further as it being my ONLY major concern.  Live view in the field while offering some nice things to tech camera viewers and perhaps some improved focusing, really isn't big on the list.  Nice if they get it, but not at the expense of any IQ.

One advantage that Canon has is they are designing their own sensors (and I think Nikon has started doing some of theirs but not sure about the high end ones) , something which MFDB makers maybe can't afford to do.  So it could be that neither Dalsa nor Kodak are interested in trying to develop a 645 cmos sensor,  perhaps technologically it isn't quite feasible yet, and maybe they've tried and they still can't match the quality of CCD, I can't say.

[uaiomex]

Thank you Yair for your response.
Wayne, I've been enjoying LV since I bought my 5D2 and despite I never needed it, today I would not buy a new camera without it. It's very nice to have it and it's really helpful in (often) certain situations.  It really makes life easier and photography more experimental. A 44X33 sensor is not much bigger than a 24X36 sensor. Why not develop a cmos sensor this size? Many pros using pro caliber FF cameras claim not jumping to the Pentax 645D because LV is missing (among other things).
Eduardo

One advantage that Canon has is they are designing their own sensors (and I think Nikon has started doing some of theirs but not sure about the high end ones) , something which MFDB makers maybe can't afford to do.  So it could be that neither Dalsa nor Kodak are interested in trying to develop a 645 cmos sensor,  perhaps technologically it isn't quite feasible yet, and maybe they've tried and they still can't match the quality of CCD, I can't say.

[TH_Alpa]

Just as an info, and it has been already somehow informed by Yair from Leaf:

Live View was first developped and introduced by Leaf and Sinar on a DCB MFDB. At that time Sinar and Leaf cooperated concerning digital backs. I remember having made the very first tests of Live View, then presenting it the first time to some selected photographers in Zürich, asking them if they were needing this in their daily work. The response was overwhelmingly "yes", and as a result Sinar decided to introduce it to the market.

Thierry

As far as paying licensing fees for LiveView, not sure who invented that ...

[john milich]

just occurred to me that with a reflex body, why not direct the mirror to a second sensor, cmos, with real live view; something of the quality of a...Lumix G1 would do fine

[madmanchan]

Unfortunately that wouldn't help nailing the focus (for that it's best to see what the real sensor sees). 

[ErikKaffehr]

Wayne,

I'm not sure about the CCD vs. CMOS thing regarding IQ. I'd suggest that:

- LV is much about image quality, because it allows you to focus on the actual image.

- There are probably a couple of issues regarding LV. I don't think that the Kodak and DALSA sensor are intended for LV, but also that perhaps the small vendors cannot afford to develop the ASICs needed to mangle the massive amount of data needed by LV.

Best regards
Erik

I think his points are valid and I think you are making some assumptions.  In fact, mfdb makers have been serious at this a very long time, and to me a more reasonable assumption would be Canikon could even be paying some licensing fees because others have been at it longer and probably invented it.  As far as paying licensing fees for LiveView, not sure who invented that but since every Point and shoot has live view, i think it predates putting it into canikon dSLR's by over a decade ... maybe isn't even covered by any patents.  It really is about packing enough processing power into the back without draining a battery every 10 minutes.

If I'm paying 10's of thousands of dollars for a MFDB, my 1st, 2nd, 3rd, 4th and 5th concerns are IQ - and you can probably go further as it being my ONLY major concern.  Live view in the field while offering some nice things to tech camera viewers and perhaps some improved focusing, really isn't big on the list.  Nice if they get it, but not at the expense of any IQ.

One advantage that Canon has is they are designing their own sensors (and I think Nikon has started doing some of theirs but not sure about the high end ones) , something which MFDB makers maybe can't afford to do.  So it could be that neither Dalsa nor Kodak are interested in trying to develop a 645 cmos sensor,  perhaps technologically it isn't quite feasible yet, and maybe they've tried and they still can't match the quality of CCD, I can't say.