Thread: Sony PMW-F3

Page 21 of 26 FirstFirst ... 11171819202122232425 ... LastLast
Results 201 to 210 of 253
  1. Collapse Details
    Senior Member
    Join Date
    Feb 2008
    Location
    Philippines
    Posts
    2,790
    Default
    all possible


    Reply With Quote
     

  2. Collapse Details
    Senior Member
    Join Date
    Oct 2010
    Posts
    187
    Default
    Quote Originally Posted by Yak View Post
    A few questions on lenses and the new PMW-F3 1) besides PL mount lenses what other lenses will work on the camera, and 2) when the Zoom comes out what are the chances it will be adapted to work on a AF-100. I could see some one making a grip with a Battery and zoom rocker which could interface with ether Sony's new zoom or any powered 2/3 zoom with a built in 2x (allowing it to cove the 4/3 sensor).
    1) The camera is using a PL to F3 adapter. This new F3 mount is Sony's own and has no other adapter available yet.
    2) No. It's a servo zoom and afaik thats not possible (well it is but would cost more than just buying the camera)


    Reply With Quote
     

  3. Collapse Details
    Senior Member
    Join Date
    Apr 2009
    Posts
    1,176
    Default
    Knowing Sony, the servo zoom they are building for the F3 will be very proprietary. So unless someone manages to break that code and sell a (potentially illegal?) conversion/adapter then I don't think it will be very likely.


    Reply With Quote
     

  4. Collapse Details
    Senior Member phd's Avatar
    Join Date
    Jul 2003
    Location
    Kristianstad, Sweden
    Posts
    477
    Default
    What have Sony PMW-F3 to do with AF100


    Reply With Quote
     

  5. Collapse Details
    Senior Member
    Join Date
    Apr 2009
    Posts
    1,176
    Default
    Quote Originally Posted by phd View Post
    What have Sony PMW-F3 to do with AF100
    Pretty much nothing, aside from the fact that it is an S35mm sensor camera with the ability to use interchangeable lenses (as are the Alexa, D-21, RED ONE, Epic, the F35, the Genesis, the 9000PL, etc.). It's in a very different market segment.

    Oh, and there doesn't seem to be a proper home for it on the forum, so this thread ended up here.


    Reply With Quote
     

  6. Collapse Details
    Senior Member
    Join Date
    Oct 2010
    Posts
    187
    Default
    Quote Originally Posted by phd View Post
    What have Sony PMW-F3 to do with AF100
    What Michael Said. We were formally in love with the AF100 and now have a new girlfriend. Although we might keep the old one "on the side"


    Reply With Quote
     

  7. Collapse Details
    Senior Member
    Join Date
    Mar 2005
    Location
    toronto, Canada
    Posts
    839
    Default
    This is a very cogent post


    Quote Originally Posted by Steve Castle View Post
    First, let's say that this PMW-F3 being a Q67 is still conjecture based on a previous post within this thread. It may be a bayer, it may be stripped, it may be Q67. But given Sony's recent presentations there is a large possibility it is Q67.

    All we know for sure so far is that it the photosite is 4x the size of a dSLR sensor, and that its s35 sensor not related to dSLR sensor. This is all that is confirmed. So let's first establish that we are still talking in the realm of predictive discussion.



    You're missing the point of the graph you are criticizing. Within the example presented, for a Bayer pattern 4K , 4096 x 2160, would be 8.8 million photosites (4096 x 2160 = 8,847,360). In a three-chip camera, we had a 3:1 ratio of photosites to recorded pixels. In a RGB stripe camera, we had 6:1. But in the bayer sensor, we have only a 1:1 ratio.

    So, for given a fixed sensor real estate (s35), it becomes a balance between the size of the photosite relative to the pixel density. Doubling your pixel density obviously has its own set of drawbacks, the choice between bayer, stripped, and Q67 is a balance between those parameters. Each has its benefits and drawbacks, Q67 is quincunx pattern which is a compromise between a full RGB and a bayer.

    Now, if you're saying that your doubled 5.6k sensor is a 'Q71' and is superior to this camera when you put the effort to downsample it to your desired format. It very well may be. However you are again dealing with more pixels crammed in the same sensor size. As of now, we don't have any point of public reference in which to judge the quality of this PMW-F3 by to your fantasy sensor much less any real product, so in either case there is no real way of comparison as of yet. We'll see soon...





    Also, this 'Q67' is obviously an evolution of Sony's ClearVid...except this is a S35 sized sensor.
    I'm not sure about this.

    Quote Originally Posted by Graeme_Nattress View Post
    This is not the place for an argument, but the F35 with it's stripe sensor shows horizontal chroma moire issues from such a pattern and strong vertical aliasing. At high details, the red, green and blue channels don't align, being effectively 1/3 of a pixel out of alignment from each other.
    Graeme, perhaps your methodology, or the set up of the camera was a factor. Maybe you didn't use industry standard methodology...

    The F23 doesn't exhibit any of this and the F35 is a next gen sensor (although differs in being a single sensor). Nevertheless, the macro and unit cells should be aligned perfectly, and the dsp should not be incident in this case. Where's the causality ? The F23 has an optimized Olpf for R,G,B and I expect no less from the F35, so high frequency detail shouldn't be a problem unless you are measuring it 4K. As the standard bearer of the Sony line it should be alias free at 1920x1080.


    You may think that your sharp 1080p is not "that" much lower than R1, but it's actually a plainly visible difference between an R1 and an F35 under cinema projection conditions, not just pixel-peeping. It's not just visible are greater fine detail, but superior MTF at lower levels of detail too.
    Is anyone else bored with the resolution argument ? Whilst one can understand why you chose resolution as a means to an end, with all the attendant design decisions formulated on it, that doesn't obviate the reasons Arri, Sony, Panavision, choose a different route. What is the final viewing mtf of film ? And what does film do that digital doesn't do right now, both in viewing and in real world shooting environments - not charts ? And why ? If one acknowledges those answers, then resolution becomes a question only on the comparison of implementation of a technology, not the specs thereof. Win, place or show. Many image specialists will debate it with you. I could tell you the list of award winning Asc cinematographers who prefer Alexa because they find resolution is not the definitive when choosing a camera, as I'm sure you could list Asc cinematographers who prefer Red.

    At RED we don't count resolution - we measure it. And we don't count aliased detail as resolution either. You can go back through this forum and my posts and find the original ones where people asked what the measured resolution of a R1 would be (long before the camera was produced) and the answer I gave was conservative and has been improved upon.

    I'm very much looking forwards to pointing an F3 at some test charts to see what it's made of.

    Graeme
    The F3 should be about 12 microns. The true test, as always, will be a filmout. If the camera performs as expected, and with the outboard recording, it could be nearer to an F23 light. At a $ 25K price tag (with three quality cine primes) ? Sony marketing should highlight this ...

    Red has benefited the industry by its innovation, and I look forward to the results of all this competition.


    Reply With Quote
     

  8. Collapse Details
    Senior Member
    Join Date
    Apr 2009
    Posts
    1,176
    Default
    Quote Originally Posted by vagarob View Post
    What Michael Said. We were formally in love with the AF100 and now have a new girlfriend. Although we might keep the old one "on the side"
    Well, I wouldn't put it quite like that. They are two different cameras at two different price points that I would really use in two different ways. No need for me to love one over the other


    Reply With Quote
     

  9. Collapse Details
    Red Team Graeme_Nattress's Avatar
    Join Date
    Apr 2005
    Location
    Ottawa, Canada
    Posts
    1,135
    Default
    Or perhaps it's just the F35 that shows strong vertical luma aliasing and chroma rainbows horizontally? You would think it would be alias free, but that's impossible unless the resolution ludicrously low from a vastly over-sampled sensor. Practically speaking all you can ever hope for is negligible aliasing. The F35 shows what I'd term as medium/strong aliasing - it's not negligible like on the RED, or strong like on a 5D2. Seeing this is simple - point camera at zone plate - capture 444 uncompressed over HDSDI. I've not measured the F23 so don't know there, but as you know it's prism+3 chip, so rather different. The HDC1500 though does show a lack of total alignment in the three channels though, for instance.

    If the resolution argument was solely about resolution then indeed it might possibly be boring. However, when done well, high resolution can lead to two other important factors - high MTF at low details, and negligible aliasing.

    Resolution is a necessary, but not sufficient quality in making an image. If you have no resolution, you have no image. If your resolution is too small, the final image is blurry and indistinct. If the low resolution is due to low capture resolution, there could also be low MTF at even the details you do manage to capture. A high resolution doesn't always make for a good looking image though - I'm still haunted by the NHK 8k imagery of a Sumo wrestler.

    If resolution is not an issue, then people are free to choose low resolution cameras. It's their choice. I wish them luck.

    As you say, the true test is the image. A good examination of a camera with a diagnostic chart will tell you how it's designed, what things to avoid shooting and how to get the best out of the camera.

    Graeme
    www.nattress.com - Film Effects and Standards Conversion for FCP
    www.red.com - RED - 4k Digital Cinema Camera


    Reply With Quote
     

  10. Collapse Details
    Senior Member
    Join Date
    Mar 2005
    Location
    toronto, Canada
    Posts
    839
    Default
    Quote Originally Posted by Graeme_Nattress View Post
    Or perhaps it's just the F35 that shows strong vertical luma aliasing and chroma rainbows horizontally? You would think it would be alias free, but that's impossible unless the resolution ludicrously low from a vastly over-sampled sensor. Practically speaking all you can ever hope for is negligible aliasing. The F35 shows what I'd term as medium/strong aliasing - it's not negligible like on the RED, or strong like on a 5D2.
    Well, alias free is as subjective as dr (in a way). The F35 does have large photosites, with oversampling and a decent bit depth - throughput of a/d converters and dsp, which is why I asked about your methodology. It's just hard to believe knowing Sony's high end cams. Their striped engineering at this level is superb.

    Seeing this is simple - point camera at zone plate - capture 444 uncompressed over HDSDI. I've not measured the F23 so don't know there, but as you know it's prism+3 chip, so rather different. The HDC1500 though does show a lack of total alignment in the three channels though, for instance.
    I wish I had the funds to do that Did you do a filmout with a real world scene ? High frequency, dr etc... ?

    If the resolution argument was solely about resolution then indeed it might possibly be boring. However, when done well, high resolution can lead to two other important factors - high MTF at low details, and negligible aliasing.
    That wasn't directed at you. I'm sorry if it seemed that way. What I was opining is that many are dismissing the importance of other image variables, as if Arri, Panavision and Sony chose 1920x1080 or 2K by necessity. There is obviously good reasons, other than technological insufficiency for them to do so. Anyone looking at their Cypress chips would realize they could have a 4K chip if they wanted. Therefore, other considerations weighed more heavily in their decisions.

    If resolution is not an issue, then people are free to choose low resolution cameras. It's their choice. I wish them luck.

    Graeme
    I wouldn't go so far as to say it's not an issue. However, it's not necessarily the defining issue depending on how the implementation of the resolution is achieved. Just as you would say that bayer and redcode are not necessarily the issue but the implementation of the two is.

    As you say, the true test is the image. A good examination of a camera with a diagnostic chart will tell you how it's designed, what things to avoid shooting and how to get the best out of the camera.
    This is where we respectfully differ. For instance, since your cameras resolving power was (or is ?) dependent on the scene and the motion thereof, only real world testing told us how red held high frequency detail in motion, or color fidelity and aliasing in motion. the tradeoffs of a camera and its design are far more evident in the field than in the lab, which is why the F23 seems to best the Nyquist limit in the field.

    All I'd like is more objectivity on resolutions place in the optical field. For you and your designers it is a means to an end, but others have taken up the mantle and started defining success or failure by it just by the specs on the page. That"s what I meant by boring. But, I don't expect to change anyones preset opinions on the matter. It's enough to see the potentiality beyond the specs, and cheer on your team, and the competition.



    Reply With Quote
     

Page 21 of 26 FirstFirst ... 11171819202122232425 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •