MattH

Samsung does state that the total number of pixels is 30.7 million, but that only 28.2 million are used for imaging.  Which is roughly corroborated by the maximum shooting resolution of 6480*4320 = 27993600 pixels.

What we don’t know is whether the effective sensor area (28mp) matches precisely with the standard apsc size of 23.6*15.7mm.   My calculation assumes that it does.

So if we assume the width of the active sensor is 23.6 and the number of pixels in that width is 6480, then the width of each pixel is 0.00364197530864197530864197530864mm.

If we assume that the nx500 uses a 1:1 crop of the sensor when recording quad HD, which as far as I am aware it does.

Given that the horizontal resolution of quad HD is 3840, the active width of the sensor in quad HD mode must necessarily be 3840 times 0.00364197530864197530864197530864, which is 13.985185185185185185185185185185mm.

The width of a full frame sensor is 36mm.
The crop factor of the NX500 in quad HD mode is therefore 36 divided by 13.985185185185185185185185185185, which is 2.5741525423728813559322033898305. 

2.57 is close enough for me.

A 600mm lens would give you the exact field of view as a 1544.5mm lens would give you on a full frame camera in video mode.

If the active sensor area of the NX1 is smaller than APSC then god knows, maybe it is 2.74.  Either way its still fucking long if you put a 600mm on it.

The pixel density is exactly what I worked out.  

The camera with the smallest 4k sensor is the first question I asked.  Everything after that was answering that question.

The Samsung NX500 does not have a crop factor of 2.75. It has a crop factor 2.58 as I worked out above using the sensor width and horizontal resolution to find the actual active width at 3840.  (2.75 sounds like a nice tidy marketing figure).

The studio 4k and micro studio look like they have the same sensor and do have a crop of 2.75, but as you say you need a recorder and there doesn't seem to be much footage from it out there. 

Overall there isn't that much between them when you have a 600mm lens at hand, so I agree the nx500 and gh4 would probably be better choices.

As for the smaller sensors, I would certainly like to see someone rig one up with a 600mm lens.  I suppose the problem is that the smaller part of the lens you use the less sharp it gets.  1500mm equiv' in 4k sounds pretty damn good.

Im not trying to be contrary at all here, but everything I have said is objective fact.

On APS-C a 50mm will give you a 2 times zoom over 23.7mm so yeah you are getting very close with 24mm, but the fact is that it doesn't scale linearly.

As for the *4 vs *2 thing, I don't know how I can put it better than the four thirds example.

This has made me wonder which cameras will give you the greatest reach.
I’ve worked out the active sensor width at native QHD on various cameras.

GH4 is 14.41mm which is 2.497* crop.
NX500 is 13.92mm which is 2.58* crop.
Micro Studio Camera is 13.056mm which is 2.75* crop.

So 600mm with these cameras will give:

GH4: 1498mm
NX500: 1548mm
Micro Studio: 1650mm

If you crop to 1080, the result will obviously be double which is:

GH4: 2996mm
NX500: 3096mm
Micro Studio: 3300mm

By the way the active sensor width when you crop to 1080 native on a 1DC is 13.33mm which gives 2.7 crop.

So your 600mm is giving you 1620mm equivalent.

Sorry sir, you are mistaken.  Using your own example.   In fact your own example isn't even true.  So you are doubly mistaken unfortunately.

Firstly, Angle of view doesn't scale that perfectly with respect to focal length.  A 25mm lens on full frame will give you a horizontal angle of view of 71.5 degrees.  If you were correct a 50mm would give half that which is 35.75 degrees. But it doesn't. A 50mm gives you 39.6 degrees.  The higher you get in focal length the less difference it makes.

However if you had said a 25mm lens and a 56mm lens you would have been correct.  The picture area of the 25mm would be four times greater than the picture area of the 56mm.   But lets take a simpler example.  A full frame sensor is 36mm wide and a four thirds sensor is 18mm wide. Half as wide.  Because we are talking about video we can assume a 16:9 ratio in each case, so we know that the full frame sensor is 4 times greater in area than the four thirds sensor.   But as we know, crop factor for four thirds is 2.  A 25mm on four thirds will give you the equivalent of 50mm on full frame, not the equivalent of 100mm on full frame.  As far as crop factor and equivalents are concerned it's the lengths that matter, not the area.

Pahahahahaa! what a joyous personality!

How the transcript should have read:

Jessica: The Vlog update only costs $99.99, before taxes! We provide FREE Shipping if we place the order today, too!

You: Ok cool, it that next day delivery or is that extra, will fed-ex be delivering my FIRMWARE UPGRADE?  Will it come on CD-rom or floppy disk?

Jessica: Let's go ahead and add it to your cart! Do you have any other questions before we proceed with the order process?

You: Yes, I have a question. what makes you so arrogant as to assume I will need and accept your assistance?

Jessica: To help speed things along, please let me stay with you throughout your purchase, that way we can ensure you complete the order successfully.

You: No! good day sir!

Processing power OR write speeds.  Its already debayered, converted to YCbCr and chroma subsampled.  It could merely record that data if it had the write speeds. Or, true, processing power would be needed to compress.  Maybe the GH4 can't do it, but it isn't beyond the realm of posibility.  I was just going to say that the Canon XC10 manages it, but apparantly its only 8bit, so maybe its more difficult than I thought after all.

I think you answered your own qestion. : )  Its 2.1, nowhere near 4.3.  As you identified, crop factor is a ratio of length not of area, which is why I asumed Inazuma was joking.

In fact, if the sensor ratios are the same, the ratio for the width or height is exaclty the same as the ratio of the diagonal.

I meant internally. The price of a 4k recorder is a significant further investment to match\beat the competition.  If they can output 10 bit, i dont see why they cant record it.

I will assume irony.

Is it 10 bit?  I guess not. I think it should be 10 bit for a paid upgrade if the camera can handle it. If this was the case there would be no question whether it was worth it. People would be going out and buying new gh4s in anticipation.

The lowest physically possible shutter speed is determined by the lowest frame rate at 360 degrees. Any longer and the next frame can obviously not expose. So if you have a 24p mode its 1/24th. My guess is that it will be 1/24th or 1/25th on nearly all cameras.  It reminds me of scenes in chunking express, (which can be seen in the trailer), where ambient light at night was obviously too dark for film, so they had to lower the frame rate to get a long enough exposure. Not really needed nowadays because of high iso but it makes a stylistic effect.

Thanks for the extremely detailed write up.  The photos are very usful at illustrating everything.  It does deem to have quite an impact on detail.  Im actually wondering if that would help with moire on the bmpcc without the need to get a mosaic filter.  The thing that seems most limiting from the newsshoter review is that he says it will vignette on wide lenses.

How come you are no longer participating on the forum?

The bmpc and bmpcc are native iso 800. Unlike a stills camera, when you change the iso the sensor does not change its analogue gain. Instead, they apply a pre compression digital boost or pull.  Indeed, when you record raw the iso changes nothing but metadata.  So by shooting at 200 iso you are loosing 2 stops of highlight clipping lattitude over middle grey. If you are clipped at 800 you will still be clipped at 200. You need to lower the physical exposure to solve the clipping.

That sounds right.

In the latest video, perhaps the tones are less flat, but the colour of some peoples skin is really orange.  The blond woman with the baby at  1 min 14, and the shadows on the face of the boy/girl at 15 seconds.

False Dilemma.  https://en.wikipedia.org/wiki/False_dilemma

In the second example the alexa lut again looks better.  In the second one, the baby looks like a ghoul from 'I am legend'.   Its a shame that the bottle top is blown out. Was it like that before any adjustment?

In this frame I prefer the Alexa lut colors. Easily. That was my instant impression, and still is after close inspection.  Which comes closer to your recollection of how things actually looked?  Does it seem to produce better colors consistently?

They use a skateboard for this, but you will be more amazed with what they are filming.  If you pulled off a shot on these I'd buy you a pint.  Go to 57 seconds to skip the chit chat.

I would love to see someone fall on their face on this thing. Or rather, I would love to see them topple over with their expensive camera, smashing the camera into the floor and smashing their face into the camera, which is what will happen. How can you possibly concentrate on operating the camrea. The example has terrible framing and non existent focusing, It's like trying to stand on top of a ball. A really stupid idea. But dont let me stop you doing it, just make sure you have somebody else to film it at the same time. I could do with a laugh.

As far as the minimum focus of 1m I think its just a case of "no free lunch".  Its would take a seriously entitled person to expect life to be perfect for them all the time.  If the markings on the rangefinder really are accurate at every setting on every lens I frankly find it amazing that is even possible at all.

I'm no mathematician but I was wondering if the focus scales logarithmically:  So, for example, if you calibrated your lens to so it had a 0.5m minimum focus distance at the 1m setting, would the 50m marking then give accurate focus for 25m?   Perhaps this is wishful thinking, but if this or some other useful correlation could be established then maybe the focus markings could still be useful with a closer minimum focus.

Why are they waiting till september to release it?  They should have examples ready right away.

Thanks.  Yeah I gues it will depend on whether the lens has the electronics to change focus, therfore just needing power from the camera.