Welcome to our community

Be a part of something great, join today!

Register Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

3K chickenfeed & mini Primes

Once again, please don't mention 'worry' or 'lose sleep'. I will not be one to do this,
nor will it affect my purchase decisions.
The theme is speculation
I am all curious about what red may or may not be planning for their line of mini primes
since 3K raw / 1080p RGB acquisition took the back seat when at NAB the 4k red ray was demo'd.
 
so.. 2/3 4k would be better.. any doubts?? why?
 
It would not be better if it was at the expense of sensitivity, noise, and dynamic range.
 
reasons why more pixels are not always better

reasons why more pixels are not always better

filmmaker's gang said:
so.. 2/3 4k would be better.. any doubts?? why?
Click to expand...

I forget what the exact sensor size is for their 2/3 format, but based on http://en.wikipedia.org/wiki/Image_sensor_format#Common_image_sensor_formats I will assume it is 8.8 mm wide. To put 4k pixels across that sensor size means your pixel pitch is 2.1 microns per pixel.

For reference, the Canon 50D DSLR has a pixel pitch of 4.7 microns, which I think is currently the densest sensor of any Canon DSLR at any price. Many people think the 50D pixels are too small for its sensor size, causing noise level and per-pixel sharpness to suffer over a similar but less densely packed sensor.

If you did have a 2.1 micron pitch sensor, each pixel would have less than 1/4 the area of a 4.7 um sensor. Other things being equal, both sensitivity and dynamic range scale directly with area, so if your pixel shrinks to 1/4 the area, it looses two stops of sensitivity.

note... Scarlet is already 3k, so perhaps Red's sensors are better than Canon's. But regardless of how good they are (again, other things being equal) for any given sensor technology, smaller pixels means less sensitivity and less dynamic range.
 
Sure, it might be nice to have 3.5K or something like that one day, so you have some headroom for stabilizing in post, or a little reframing or whatever. But you don't need 4K for that.

Maybe one day Red will upgrade the 2/3 to 3.5K or 4K (isn't that the whole idea of upgradeable modules?). But either way, you're going to be finishing at 1080p/2K, not 4K.
 
I hope they don't try to squeeze any more pixels onto the 2/3 sensor. They're small enough already!

Sure I'd love to have more resolution but I'd be even happier if they kept the same resolution and moved up to a larger sensor!
 
ehehehehe larger sensor all on the way!
 
jbeale said:
Many people think the 50D pixels are too small for its sensor size, causing noise level and per-pixel sharpness to suffer over a similar but less densely packed sensor.
Click to expand...

When tested correctly, the 50D noise is the same as the 40D. Several tests were flawed by measuring different spatial frequencies or using different raw processing for each camera (e.g. ACR).

jbeale said:
Other things being equal, both sensitivity and dynamic range scale directly with area, so if your pixel shrinks to 1/4 the area, it looses two stops of sensitivity...for any given sensor technology, smaller pixels means less sensitivity and less dynamic range.
Click to expand...

I kindly disagree. My position is that sensitivity and dynamic range stays the same.

I've been meaning to write a post about this topic for a while. Maybe now I'll finally get a round tuit.
 
Daniel Browning said:
I kindly disagree. My position is that sensitivity and dynamic range stays the same.
I've been meaning to write a post about this topic for a while. Maybe now I'll finally get a round tuit.
Click to expand...

I look forward to reading it should you find the time. I would also be interested to know as to what limits the pitch on a photosite can go down to before we see a shift in dynamic range or sensitivity.
 
I don't think Red are wasting there time or money on the mini primes. There is still going to be a lot of market for the scarlet 2/3 (interchangeable). Even if you start off with a scarlet and a set of 6 primes, but decide you want more resolution after a couple of years, you can still keep the scarlet ad your primes for your slow motion shots, green screening, or as camera b or c.

If your still not finding a reason to use it, sell it. I'm sure there will be plenty of people wanting a 2/3 scarlet kit with 6 mini primes. Don't limit your thoughts to being based in Americas film industry. Places like India, Philippines, Malaysia and even my country New Zealand, are going to have many good reasons to use 3K video, HD video or even 720p video.

I mean hell, we're only just starting to get decent internet here and even with that I cant stream HD on youtube yet.. We only just started to get viewable HD television too.

3K is going to do amazing on the market. Trust me.


Jono
 
most feature films are transferring 35mm negs at 2k. these films look fine so my feeling is the 2/3 3k sensor will even hold up to being screened at the theatre. it won't be spiderman quality but it will probably hold up well enough. remember most older lenses don't even resolve much beyond 3k. also, 2/3 chips require physically smaller lenses. this is a big advantage for a lot of productions. red probably feels 3k is enough for most applications for quite awhile now that 1080p will be the standard for tvs and 2-3k DI will be good enough for theatre projection.
 
Daniel I have no reason to doubt you but your post was way over my head and I didn't understand much of it.

Meanwhile, I've seen with my own eyes that images from the Nikon D3 (12mp on 36x24mm) has less noise at ISO6400 than images from my Nikon D2x (12mp on 24x16mm) at ISO640, smart people have said it's because the Nikon D2x has such tiny photosites, and the 2/3" Scarlet has triple the photosite denisty of the D2x.

So while I'm sure Red is going to prove me wrong (otherwise they wouldn't be releasing a 2/3" cam!), on this one I won't believe it until I see it.
 
Lee Saxon said:
Meanwhile, I've seen with my own eyes that images from the Nikon D3 (12mp on 36x24mm) has less noise at ISO6400 than images from my Nikon D2x (12mp on 24x16mm) at ISO640,
Click to expand...

That sounds about right. :)

Lee Saxon said:
smart people have said it's because the Nikon D2x has such tiny photosites,
Click to expand...

That is incorrect. There are three reasons:

Sensor size

The D3 has a sensor that is 2.5 times larger. Even if it has very tiny pixels (e.g. 48 MP), it would *still* be superior to the D2x because of its size.

Technology

The D2x had very poor performance even compared to contemporary cameras of its time. The disparity has only increased with time.

Processing

The third factor is that Nikon started a new style of image processing with the introduction of the D3, most notably desaturating on a curve so that shadows have most of the chroma information removed, but luma is left in tact. If the same processing were applied to the D2x, it would also get an improvement.
 
uuups I lost half of my text ....

so as you follow the arguments of David about the quality of Oscar winning movies, I think the discussion about how fast 3K could be outdated or not is more or less obsolete .... ;-)
 
Gerald Menclik said:
uuups I lost half of my text ....

so as you follow the arguments of David about the quality of Oscar winning movies, I think the discussion about how fast 3K could be outdated or not is more or less obsolete .... ;-)
Click to expand...

I think you already know the answer to that one. Just look back through history then realize that the speed of technological advances is accelerating.
Best example = Red Ray carries 4K at 10mbps.

Besides, my original post was to speculate about the possibility of another brain for mini primes. Reason: 6 lenses supporting the lowest in the sensor line.

Now look what you've gone and done ... you've made me repeat myself.
 
There is a point of diminishing returns where more resolution does not yield sgnificant visual improvement in the delivered product in proportion to the costs of dealing with the increased acquisition, storage, processing and workflow requirements. One does not need more than 3k acquisition to produce pristine fully resolved 2k or HD deliverables, much less web video, HD or not. In any situation where viewing distance is 2 x screen height or more, the difference in 2k and 4k presentation compared to the limits of human visual acuity is pretty insignificant.
 
You must log in or register to reply here.
Back
Top