Upcoming hands-on comparison of F3, FS100 and AF100 from Philip Bloom

[Glen Vandermolen]

^^ this.

If you can't afford an F3 and need a full-size sensor video camera (not DSLR), the FS100 and the AF100 are perfect. They both can make very pretty pictures. The rest is up to you.

[John Godwin]

Originally Posted by Glen Vandermolen

^^ this.

If you can't afford an F3 and need a full-size sensor video camera (not DSLR), the FS100 and the AF100 are perfect. They both can make very pretty pictures. The rest is up to you.

I agree and I'm glad to hear that. There are *some* people posting here who apparently want to dictate what my choice of camera must be.

My local dealer has me on The List to receive an FS-100 from the first batch they receive, hopefully within the next few weeks. Unlike B&H I don't have to make a deposit or pay an potential restocking fee. I just have to go look at it, handle it, evaluate the image with my own eyes, and then decide if it works for me and my clients in a way that'll pay for itself and make money.

Truthfully, I won't know for sure until I actually hold the camera and get a sense of the ergonomics and if the couple of disadvantages (no ND, mostly) are something I'm comfortable working around.

In the meantime I'd appreciate practical recommendations on variable ND filters for the kit lens. Seems like a better option than messing with my matte box for some of the shooting I do. And I'll probably get the kit lens for starters because I don't have a lot of other lenses sitting around here, as I come from betacams and full-sized dvcams and Z1 and EX 1 & 3. So I'm willing to start with the kit lens and add others as things develop.

[Galen Rath]

Reconsider the kit lens, at least do some investigating, I've not read much good about it on the forums. All you need is ANY lens (plenty of used Canon FD lenses on ebay for as little as $50, get a F 2.8 or better ) , and a cheap adapter if you just want to experiment with the camera. The kit lens is an $800 retail lens that you get for $600 on the package, but maybe you can find a way to spend $600 better.

[Asif Khan]

Originally Posted by David C. Williams

And, the F3 and FS100 have the same chip. And, neither the AF100 nor the GH2 are S35 or APS-C, both are Micro 4/3".

Exactly. The chip can't be the "same" if one is S35 and the other is APSC. All the rest long rambling by Steve about "design" are pure speculations. VG10 and Nex-5 had an identical chip. Both were APSC.

[Steve Mullen]

Originally Posted by John Godwin

In the meantime I'd appreciate practical recommendations on variable ND filters for the kit lens. Seems like a better option than messing with my matte box for some of the shooting I do. And I'll probably get the kit lens for starters because I don't have a lot of other lenses sitting around here, as I come from betacams and full-sized dvcams and Z1 and EX 1 & 3. So I'm willing to start with the kit lens and add others as things develop.

A 200ISO sensitivity camera in bright sunlight set to the correct video shutter speed of 1/60th second and set at f/2.8, for a shallow DOF, requires an 8-stop neutral density filter.

An 800IS sensitivity camera requires 10-stops of light reduction.

Unless you buy a full set of ND filters, you’ll need to buy a vario-ND filter. (Remember, no stacking of ND filters.) Unfortunately, the maximum useable filtration from a vario-ND filter is 5-stops.

[Brian Drysdale]

A variable ND is effectively two polarising filters stacked.

[Alister Chapman]

Who says you can't stack ND filters?

Sure I would not want to stack low quality filters of any type and perhaps it's not the best way to go, but with good quality filters, especially those that have IR cut it should not be that terrible, certainly no worse than a vari ND.

The F3 stacks ND filters internally.

There will be lots of ND options for the FS100. There will be Nikon adapters with built in ND's, you can get thin film ND's that can be placed behind the lens, so you could consider a .9 (ND8) behind the lens and then a vari ND in front.

The thousands of DSLR shooters out there have been dealing with this issue for a couple of years and it has not stopped them.

[Steve Mullen]

Originally Posted by Brian Drysdale

A variable ND is effectively two polarising filters stacked.

Open an Vario ND filter and look at the non-gap between the glass elements. You'll see immediately this is not the same construction as two filters in two mounts screwed together.

From Philip on the AF100: "Great to have a 4 setting ND filter wheel."

[Brian Drysdale]

Yes, but it's still 2 filters being stacked. Where's the difference?

In this case, with the densities involved here, I'd use IR ND filters. It's standard practise with cameras like the RED.

[David Heath]

Originally Posted by Noah Yuan-Vogel

Sorry, which observed facts are you saying this theory fits? Only that it is cheaper and draws less power?

No, not at all. Mainly observations of zone plates and circular resolution charts that have been put on line. The beauty of such is that the patterns they cause cameras to produce tend to be independent of a host of factors - lens sharpness, even focus, for example. (Defocussing a camera causes aliases to be less distinct - it doesn't change their pattern.) The pattern shapes and colour are a direct function of the sensor characteristics and the way the data is read out.

For the F3, I've used Alan Roberts report, and as said in the last post, figure 2, the b-w zone plate, top centre. You can clearly see the coloured aliases in the out of band, and I measure the circle centres to correspond to roughly 2490H and 1400V - I think it's more likely to be 2560 and 1440 if I could measure it accurately. That implies some fundamental resolution of 1280x720 (assuming the second figures), and it seems reasonable to assume that is the photosite resolution for red and blue - so a total sensor resolution of 2560x1440 if Bayer. A total of 3,686,400 - which I believe is close to what Sony have previously stated? I do note Alan concludes his report by surmising that for the F3 "....the sensor has approximately 12.9 Megapixels, typical of a digital stills camera." I can't see what in his results leads him to think that.

For the FS100, I've used the chart in Nigel Coopers test. It's possible to extrapolate the aliases visible in that to get an idea of the resolution at which real detail turns into aliasing, and that seems to also be around the 1250H and 700V mark. To be sure, you'd need to put the camera on an out of band zone plate, and use real circle centres to measure more accurately.

I don't believe that's a co-incidence, given what we know of the sensor - and the idea of much simpler readout on a 2x2 basis then would answer a lot of questions.

The storage media alone, for example, could easily account for 2w, and then the use of older XDCAM chipsets instead of newer and more available chips for AVCHD encoding could account for a big difference, .........

The FS100 is quoted by Sony at 5.6 watts, the F3 at 24 watts!! And with the EX1 at 12.5 watts (same codec, same media as the F3) I can only conclude it's the F3 chip and the way the F3 reads it that is responsible for the high power consumption. So why is the FS100 so low? If it uses a far simpler readout method, as suggested, that question is answered.

Again just a guess, but is your theory about the FS100 debayering in this unusual way also a guess?

As David Williams says, I'd also say "hypothesis" is a better word than "guess"! But it's a hypothesis that explains all the observed facts - measured features on charts, power requirement differences, cost differences between the cameras, and others which I don't have the time to go into. I'll freely admit that I'm not sure of the exact numbers, but am more confident of the basic principle.

And if my hypothesis (or "guess" if you must...) isn't right, then how do you explain some of the fundamental differences between the F3 and the FS100 on charts? Why the nature of the aliasing is so different between them if (as I think we all agree) they share the same physical sensor? As said at the start - the nature of alias patterns isn't affected by optics, focus etc - only the intensity.

[Steve Mullen]

Originally Posted by Brian Drysdale

Yes, but it's still 2 filters being stacked. Where's the difference?

Photo 101: It's the air gap creating different densities for light to pass through.

[Brian Drysdale]

Yes, but there'll be an air gap using two rotating polarising filters.

By the nature of things, there'll be an air gap putting together any filter pack and some DP's use quite a few filters in their packs. The usual problem that worries 1st assistant is the reflections between the filters and some matte boxes allow you to angle filters to reduce the problem.

[Doug Jensen]

Originally Posted by Galen Rath

Reconsider the kit lens, at least do some investigating,

That is good advice. Optically the stock 18-200 looks good, but it's too slow for shallow DoF, plus the inevitable problems with ramping. What was Sony thinking with a f/3.5-6.3 lens? It defeats the whole purpose of having a Super35 camera for shallow DoF. They would have been better off bundling it with a nice fast f/1.5 50mm prime. And if you want to use the zoom "live", well, there's no servo plus the ramping issues.

Save your $$ and apply it to buying a 3rd party adapter and start building a selection of Nikon or Canon lenses if you don't own some of those already.

[Piotr Wozniacki]

...and the ramping is considerable, as I could witness first-hand while playing with the VG10 featuring the very same lens.

That said, I do not quite share Doug's opinion a 50mm prime should be bundled with the FS100. Those who are going to actually use prime lenses, probably already have some legacy stock or would like to carefully pick them themselves (along with appropriate adapter for the E mount). Those who are going to learn this camera from the basics would be better off to start with a zoom - but certainly a faster one, even if at the expense of a shorter range, and/or bigger size/weight/price...

Yeah, I know such a zoom doesn't exist yet for the E-mount, just wishful thinking here.

[Brian Drysdale]

From a size point of view a limited zoom range with a f2.8 stop makes sense, otherwise you'll need to go for a slower stop for the longer zoom ranges.

To a certain extent these lenses already exist, you could use RED zooms on these Super 35 sensor cameras for example. Or, use stills zoom as varifocal length lenses. Although, that's not the same as a stock zoom lenses and these tend to be made for a price, rather than quality.