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

[Mark David Williams]

QUOTE
In the end, it's up to each person to decide which camera is right for them and what they want to film
However, you'd be better testing your lenses on each camera yourself and then comparing the results, rather than comparing resolution numbers from different tests by different people over the internet.

----------------------------------------------------------
There is just one test that holds REAL weight and that is Alan Roberts and Phil Bloom looks at technology from a camermans point of view.

Others tests favour certain cameras. Or language that does.

I'm not going to buy expensive charts and then seek to borrow cameras so I can test them so I can be sure of the results and certainly if I did and showed these results others would undermine them to sell theirs. This has happened to Alan Roberts being underminedand now testing seems not worth the paper its written on with the exception of a few still trusted.

Instead I joined this thread and confirmed what I suspected. The Sony FS100 and the Panasonic AF101 do not win over an EX1/Letus combo and there are numerous DSLR's that outperform them in many ways. I guess the reason I've joined the debate is because I really wanted the two cameras to be better than I knew they were so I could move up and not held back by one step forward three back that these two offer.

The debate about the big manufacturers crippling cameras and holding back technology has been going on for years Indeed Jim Jannard set up Red because of this.

This is disapointing because it seems we are returning to this. I believe maybe the EX1 was a blip when Sony felt threatened by RED and perhaps no longer do so. If this is the case then the EX1 may be a Camera to keep hold of.

Anyway.... All my own personal opinion. My advice to others is be wary, its confusing but get the tool you need for the job you are doing.

Mark

[Brian Drysdale]

You don't even need charts, you can just push the cameras into those dark corners performance wise and compare them. The final selection would a personal selection depending on your visual style and what you intend to use the camera for.

You're the first person to say that DSLRs outperform them in many ways. The GH2 seems to be the only one that people seem to be making a case for in that regard and that tends to be image quality compared to the AF100.

I'm not saying that large sensor cameras are the answer to everything, they're not. Just as a shallow DOF isn't the only way to direct the audience where to look in a frame.

Yes, the EX series are extremely good bang for buck.

[Mark David Williams]

Brian

Why do you think I have access to any camera I want to test? I don't.

I didn't say DSLR's outperform them. I said numerous DSLR's out perform them in may ways.

Mark

[Brian Drysdale]

You can go to a dealer and test their demo cameras.

In which way do these DSLRs outperform them many ways? Certainly the DSLR's give better stills and if you want an extremely shallow DOF a FF35 stills camera will do that aspect better, but it will still have the other weaknesses for shooting video.

[Mark David Williams]

You can go to a dealer and test them?

None of the dealers here stock them let alone have demo ones. The price of these cameras leave them in the professional domain an area I don't think they deserve.

How do DSLR's outperform them?

FS100 8 bit signal processing and HIGH cost compared to say a 5D that is 14bit £1600 and a much bigger sensor. The 5D may not measure as well in tests but it gives a gorgeous picture that although subjective I think is the best out of all of them.

AF101 Again High cost and an even smaller sensor.

All the DSLR's outperform the AF100 and the FS100 in cost. An idea I have considered is using an EX1 in conjunction with a 5D especially considering the awaited mark 3

Mark

[Brian Drysdale]

If you check here:

http://www.dvinfo.net/forum/sony-avc...s-video-2.html

Alistair has mentioned that the FS100's sensor processing is 12 bit.

However, the 5D also does a lot of line skipping etc and Alan Roberts doesn't even regard it as broadcast quality, It does give nice images, but it has a lot of moire, which restricts it. Plus a very compressed codec. If you want do colour correction, this is isn't the camera do heavy duty correction

Doug Jensen does a good summing up on the FS100, a camera which is very much aimed at the DSLR user.

Sony VideON | Sony Super 35mm Seminar at the 2011 NAB Show - Part 1 (NEXFS100U) | NAB 2011

Go to dealer that does stock them when the FS100 come onto the market or rent one for a day, you're not far from London.

[Mark David Williams]

QUOTE
If you check here:

http://www.dvinfo.net/forum/sony-avc...s-video-2.html

Alistair has mentioned that the FS100's sensor processing is 12 bit.
----------------------------------------------------------
Others have said 8bit signal processing.
-----------------------------------------------------
QUOTE
However, the 5D also does a lot of line skipping etc and Alan Roberts doesn't even regard it as broadcast quality, It does give nice images, but it has a lot of moire, which restricts it. Plus a very compressed codec. If you want do colour correction, this is isn't the camera do heavy duty correction
-------------------------------------------------------------------
And what I said was " An idea I have considered is using an EX1 in conjunction with a 5D especially considering the awaited mark 3"
USED in conjuntion as another tool in my arsenal of tools In other words for specific tasks.
------------------------------------------------------
QUOTE
Go to dealer that does stock them when the FS100 come onto the market or rent one for a day, you're not far from London.
------------------------------------------------------------------
This thread has told me all I need to know and I'd appreciate not being talked down too.

[Alister Chapman]

Originally Posted by Mark David Williams

If I had to choose between the Sony FS100 or Panasonic AF101 or an EX1/Letus combo.

With EX1 only you get 10 bit out 1000 line resolution. You can use this configuration for shots that need no shallow dof or close closeups that do
With EX1 PLUS Letus you get 700 lines 10 bit out which aint far off the resolution of the new cameras.

VERDICT
The EX1/Combo offers the better deal for all round film making.

Mark

10 bits of low contrast, noisy, soft, reduced dynamic range Letus footage, or 8 bits of low noise, high contrast, sharp, high dynamic range footage?

No contest, if shallow DoF is the goal, I'll take quality over quantity thank you.

The 10 bit argument is a red herring with the EX1 anyway. The EX1's noise will limit how far you can grade long before the 10 bit output brings any advantage. You need a noise figure better than 56db for 10 bit to make a worthwhile difference in most cases as any more noise than this and the noise is larger than the 8 bit sample size, so all 10 bit does is record the noise more accurately. You should try grading both 10 bit and 8 bit EX footage, recorded with the same relative bit rate. You will see that there is no real difference to how far you can push either. A higher 8 bit, bit-rate helps reduce quantisation ver the built in codec, but 10 bit over 8 bit makes little difference for acquisition with the EX1/EX3.

IMHO the FS100 walks all over the 5D as a video camera. Better dynamic range, ergonomics, HDMI output, no line skipping, lower noise, less skew. But I'd still rather a 5D over a Letus. The Mk3 will still be a stills camera at the end of the day so it will likely still have many of the same issues as the Mk2.

[Mark David Williams]

Alister
So are you now saying the FS100 is 8bit signal processing? How did you come to believe it was 12bit if that is the case?

[Brian Drysdale]

Originally Posted by Mark David Williams

This thread has told me all I need to know and I'd appreciate not being talked down too.

Sorry, but your own real life tests count more than a thread on the internet or as saying on a well known cinematography site goes;;

"Test,.. Test... Test"

The most that this discussion thread points out are the issues that you may come across, need to test and see if they're significant for what you're trying to do.

[R Geoff Baker]

Originally Posted by Mark David Williams

Alister
So are you now saying the FS100 is 8bit signal processing? How did you come to believe it was 12bit if that is the case?

I don't really mean to inject myself into this debate, but I will point out that signal processing and output are necessarily different things; an 8 bit signal processing would output at best 6 bits, if my high school maths are remembered, it would take at least 11 bits to output 8.

Years ago there was a DV camcorder that had signal processing of 10 bits, and the resulting output had noticeable banding as it was something below 8 bits of output.

Just to make the point that output and processing are different things, and can't possibly share the same value ...

Cheers,
GB

[Mark David Williams]

Cheers GB

Well that's what is mentioned the camera has 8 Bit signal processing so therefore if what you're saying is right it only outputs 6 bits?

[Alister Chapman]

Thanks GB.

Mark, I was talking about recording the cameras output, not about the DSP.

The FS100's HDMI output is 8 bit and the F3's and EX1's is 10 bit.

I have been led to believe by those that should know that the internal processing of the FS100 is 12 bit, not as sophisticated as the F3's and using a simpler DSP.

There seems to be an assumption being made that the DSP and the output bit depths are the same, which certainly in the case of the F3 they are not and in the case of the FS100, I can't see how it's possible to use an 8 bit DSP to get a decent 8 bit output.
To get the kind of dynamic range being seen from the FS100 or F3 hints at the sensor output being at least 12 bit, if not 14 bit. You cannot simply ignore 4 bits of data without having a significant impact on the systems dynamic range. As there is little difference in DR between the FS100 and F3 it is my opinion that at the very least the FS100 DSP must do some processing on all 12 bits of sensor data.

[Mark David Williams]

Test Test Test

Would be nice to walk into a store take the camera outside and test maybe spend a few hours hook the camera up to a monitor film in low light test the highlights do some colour correction try some gamma curves Hook up to a recorder etc The likliehood though is it will be plonked on the counter shown where the buttons are and sweet talked by a salesman who shows you the glossy brochure and tell explains why it is better than the competition.

Objective Information can be gleaned from the internet where real users can upload clips give opinions and where you can judge those opinions by their honesty and integrity from past reviews like Alan Roberts for example. Please don't tell me Alan Roberts hasn't tested the FS100.. It was just an example.

Hiring a camera I agree would be a much better idea.

I have downloaded footage for these cameras looked at pics of what they can do seen the specs tried to uncover how they do what they do which is often hidden in some cases to deflect bad publicity and criticism. The next best thing to hiring one.

Just as an aside. With the EX1 I knew almost straight away from reviews this camera was the real deal and fairly rapidly had one on order.

[R Geoff Baker]

It has been a long time since my maths been tested -- but with certainty I can say that if the processing takes place at 8 bits the output must be lower. Conversely, if the output is 8 bits ... the processing must have been higher. Someone that paid closer attention in school can doubtless give the precise relationship between process and output -- best I can recall is that a couple of bits of headroom is required.

(Hope all is well in Herts; been a few years since I lived in Hitchin but fond memories of springtime)

Cheers,
GB