|
Sony NEX-FS100 Camera Test
Just a simple test.
Yesterday in the late afternoon, I got an overnight loan of the yet to release Sony NEX-FS100 video camera. With about 1-2 hours of daylight left, I quickly drove to a nearby place and put the camera to test. It was at the end of a long day of work fixing some Avid issues and I initially thought I was just going for a demonstration, so I didn't bring along any support equipment. But I remember watching Den & James's video blog of the camera and how they had shot one of the scenes of "Vertigo" music video without permission. So I tried to go "stealth" and walked around a local old tourist street with the barebones FS100 (removing mic, handle and eyepiece) to see how unobtrusive I can get. I am using the E18-200mm F3.5-6.3 OSS Zoom kit lens. Settings -------- There is no user manual and the camera went through several hands, so I am not 100% sure what are the defaults. But I believe I am testing the camera under default settings. There are 6 picture profiles and I chose the 5th - Cinematone 1 which is visibly darker on the camera's LCD screen but brighter than Cinematone 2. But this being a cinema camera, I felt most people would want to use the cinematone profile and so I tested with it. The gain settings are set to 0db (low), 15db (mid) and 30db (high). It is configurable at an interval of 3db from 0 to 30. I don't know if this is the default but I didn't change it since it is also good to test the entire range. It is a PAL unit and should be the final production unit (but I am not able to confirm at this point). - 1080p25 24Mbps Cinematone 1 - Edited in Premiere Pro CS4 with no grading - Encoded in WMV Average 8Mbps, Max 10Mbps. I did a Mainconcepts H.264 encoding at similar bitrates as well but I thought the WMV version is better in some ways. Note: Some artifacts are due to the WMV encoding. Built quality ------------- The built feels similar to EX1 and like all cameras of this range, better handle it with more care. All plastics but like most Sony products, they don't feel cheaply constructed. Buttons are of course not as big and well-spaed as shoulder mounts, but I don't feel they are that different from other cameras of this price range. There are mounting holes in many places and you can actually mount the camera at its side on a tripod to do a vertical shoot (for vertical mount displays). Unobtrusive? --------------- I can wrap my two palms and fingers around the main unit, leaving the lens exposed. The size feels like placing two DSLR bodies back to back, if you get what I mean. In fact, the internal battery compartment takes up a lot of space (maybe 1/3 to 1/4 of the main unit). You can insert a NP-F970 inside without protruding out. Due to the position of the LCD screen at the top, I do find myself often shooting at below the eye level. So am I really unobtrusive? Not exactly, but I feel I am not much different from a DSLR shooter. Apologies for the shaky footage, I am quite tired at the end of a workday and the camera is not that light. I am also holding the camera in a unfamiliar manner (try visualize holding a barebones FS100 with your hands). Lens ------ Equipped with the same F3 super 35mm CMOS sensor (I was told), I feel the kit lens is really not making use of the sensor's superb capability. - Looks unprofessional with the silver plastics (personal opinion) - Focus ring is at the back and zoom ring on the front, which is contrary to all professional video lens. I kept making the mistake of zooming when I want to adjust focusing. - Focus ring is smaller than the zoom ring and I have to turn a very big round to go from full blur to sharp. - Not sharp or fast - Not quite easy to get shallow depth of field with slow F3.5 - F6.3 (and no ND filter) - Forget about smooth zoom. Difficult, if not impossible? Too much friction. - Infinite focus ring. I always prefer a finite ring.. - Silent and fast focusing/autofocus. Good. - No noticeable barrel distortion at wide/telephoto ends. - No noticeable chromatic aberration. - Okay performance. I heard it's the same lens as the consumer interchangeable lens camera. Note: There is a small iris ring on the main camera unit. LCD Display -------------- I can view the LCD reasonably well in daylight and it's rotatable except facing the talent. Sharp LCD screen like the EX1. The nice thing is that you can press and navigate on the LCD screen directly. But it leaves too much fingerprint for me and I prefer not touch that screen. The small histogram can be permanently left on which is very useful. There is only one level of zebra. The peaking can be set to white, yellow or red with different strengths. I have an issue with the expanded focus just like Canon XF305's. I prefer Panny's solution which only shows an expanded focus in the middle. With a full screen expanded focus, I often forget to turn it off as it looks not much different from the regular screen. Image quality --------------- Low light and low noise is definitely its selling point. Definitely better than AF100 in this regard. I also tested AF100 briefly before with some noticeable noise. 15db gain is surprisingly very usable although I feel 30db is reserved for documentary use with its noticeable noise. A better lens will definitely do wonders for this camera. The absence of a ND filter wheel is also an issue but it can be overcome with a mattebox. However, the aliasing is quite noticeable and somewhat disappointing. I wonder why? Maybe an inferior filter? There is apparently a need to lower the sharpness in the menu settings for most situations. If I remember correctly, it was 0 at default and I shot at 0. To be fair, it may not be any worse than other cameras unless we do side-by-side comparisons. Although not shown in the footage, I did some quick panning of the camera and and I can hardly see any rolling shutter effect. If you look at 02:10, the fan in the shop does not suffer from any rolling shutter effect. Miscellaneous --------------- Battery life is excellent. I shot with the LCD display on at normal brightness. I don't know if there is a power saving feature but it's almost always on throughout my 1-2 hours of shoot and the NP-F770 battery only dropped about 30%. By the way, the loan package box comes with a NP-F770 battery and a Sony Shotgun Microphone. I am not sure if the actual package contents would be similar. Conclusion ------------ Pros: - Excellent sensor with good low light capability and low noise - Excellent battery life - 1080p50 at 28Mbps - Both HD and SD - Modular (very configurable) - Barebones is reasonably unobtrusive - 15db gain clean and very usable Cons: - Consumerish kit lens - No ND filter wheel (others have suggested it's impossible to build it in due to form factor) - Noticeable aliasing? - Not NTSC/PAL switchable |
Re: Sony NEX-FS100 Camera Test
Thanks for the test; it confirms my initial impressions on this camera.
Particularly attractive (and expected) is the low noise level. Unfortunately, lack of ND filters will make many potential buyers look towards the Panny AF-100, with the Sony's own F3 in quite another price league... Piotr Dear Sony, if you read this - it's still not too late to make for this omission in the actual production units. Please, don't make this mistake - avoid it in the interest of us loyal users, and yourself! |
Re: Sony NEX-FS100 Camera Test
I've been browsing the site for years but not posting much... anyway.
I'd like to back up Piotr's statement. The ONE thing that's making me want the AF100 over the Sony is the lack of ND filters built in. This is a major problem for me. Also thanks for the review. Very nice. :D |
Re: Sony NEX-FS100 Camera Test
Let me put it this way. Sony, if you were to have ND filters on this camera, I WILL buy it as soon as
it is available. If it does NOT have ND filters, I will not buy it. |
Re: Sony NEX-FS100 Camera Test
Quote:
Or, as Dr. Seuss would say: Sony, Can it, will it, have ND? No, It cannot, will not, have ND! |
Re: Sony NEX-FS100 Camera Test
I understand that it isn't coming with ND filters. I was just putting Piotr's message to Sony
a little more strongly. In my case, the lack of ND filters is THE reason that I am not going to buy it. If it had them, I'd have already preordered one......even if I had to pay the money up front. And yes, I understand the AF100 is out, but for other reasons, I don't want that camera either. |
Re: Sony NEX-FS100 Camera Test
30db gain? Is that a true 30db gain or are they renaming the conventional 18db to 30 for some reason?
|
Re: Sony NEX-FS100 Camera Test
Low light performance seems very good .. aliasing is disappointing.
|
Re: Sony NEX-FS100 Camera Test
ND filters are something I forgot how much I missed until I went back to the AF100 from years of DSLR. Didnt realize the FS100 didnt have them until this thread!
|
Re: Sony NEX-FS100 Camera Test
Quote:
|
Re: Sony NEX-FS100 Camera Test
He probably does, Peter.
Interestingly, this clip does indeed show bad aliasing when played back in WMP, but none whatsoever when played back full screen from Vegas Pro 10 timeline... Piotr |
Re: Sony NEX-FS100 Camera Test
I suppose that leaves the aliasing open to question, is it the original sensor image or does it depend on how you replay the on board recording?
|
Re: Sony NEX-FS100 Camera Test
UPDATE: I have allowed some people to download the raw file and they have tested no aliasing on other editing software. The aliasing could be a problem to Premiere Pro CS4 and WMP. I will investigate this on Avid Media Composer 5 and re-edit it there if this is indeed true.
|
Re: Sony NEX-FS100 Camera Test
Sorry to say, but built-in ND is not possible on this camera or any Sony camera with the E-mount because there is very little distance between the sensor and flange.
To me, no ND is not a problem because I couldn't imagine using these SLR lenses outdoors without a mattebox. Under controlled lighting...well, it is CONTROLLED lighting. However, my main issue is no HD-SDI and no 10bit. Could you imagine using this camera and the Gemini 444 for the same price as the F3? That would be awesome. Expect aliasing from any single sensor camera as the OLPF must allow more resolution through than the sensor can output. The most expensive single S35 sensor has a good amount of aliasing - the F35. |
Re: Sony NEX-FS100 Camera Test
Aliasing was investigated to be a non-issue. Please ignore my comments on aliasing.
|
Re: Sony NEX-FS100 Camera Test
Quote:
Its exciting!!! |
Re: Sony NEX-FS100 Camera Test
Apologies. Aliasing is a non-issue. It's an editing problem with PP CS4.
|
Re: Sony NEX-FS100 Camera Test
Quote:
|
Re: Sony NEX-FS100 Camera Test
My observations...YMMV
1) Body to body, the AF100 is $1000 cheaper and has both ND filters and SDI out. The savings goes a long way to paying for the Gemini. 2) Arguably the MFT lens mount, as a non-proprietary one, with more lens availability/compatibility/adoption is another positive of the AF100 over the proprietary e-mount on Sony cameras. |
Re: Sony NEX-FS100 Camera Test
With both cameras, the Gemini is total overkill. The nanoFlash fits perfectly with both because neither camera can output anything other than 8bit 422.
The E-mount is not so proprietary because of the Alpha adapter which provides for full electronic control of Alpha lenses. Compare Alpha lenses from Sony and Sony/Zeiss to 4/3 lenses and the Alpha system is the winner. |
Re: Sony NEX-FS100 Camera Test
Quote:
However, once you read tests of these lenses, you'll want to not close down more than f/11 to f/16. So you generally need to stay in a range of f/5.6 to f/11 with f/8 being the "sweet spot." Get out your lightmeter and set minimum ISO with a shutter-speed of 1/55. Measure light on a really bright day. Now see how many stops of light need to be removed to get to f/4 to f/5.6 for a shallow DOF. If it is more than 5 stops, a Vario-ND will work because of image quality loss that sets in. Now calculate how many stops of light need to be removed to get to f/11 to f/16 for a deep DOF. A Vario-ND will work, BUT as light falls you may have to remove it because it's minimum reduction is 2-stops. Now, you may need to use a 1-stop ND filter. Thus, for every lens filter size you use, you need: a 1-stop ND, a Vario ND (2 to 5-stops), and a 6-stop ND. With snow you may need 8- and 9-stop ND filters. Moreover, with a prime lens such as the 16mm F2.8 or an F2 or F1.4 -- to get a shallow DOF you need a monster ND filter. In low light with a big chip and a slow lens (F5.6) you may have only a few inches of DOF! To get the aperture to f/8 you may need 4- or 5-stops of gain (+30dB) AND/OR additional light. PS1: A big chip gives you the option to CONTROL DOF. Everyone seems to forget that with the shutter-speed at 1/50th or 1/60th and a 15 to 17 stop range of real world lighting to cope to with, getting a SPECIFIC DOF (by setting the aperture) requires the use of a huge range of ND filters (bright light) or a 3 to 5-stops of gain (dim light). Once you choose a DOF -- the only thing left under your CONTROL is the amount of light entering the camera or the gain setting. Bottom-line what one thinks they know from shooting with tiny or small chip camcorders is moot. You'll need to call on your experience shooting 35mm slides -- not negative -- film with a still camera. PS2: Once you use non E-mount or A-mount lenses and use "adaptors" -- you encounter HOW you'll control aperture for focusing and shooting. So things become more complicated than shooting with a still camera. |
Re: Sony NEX-FS100 Camera Test
Quote:
Also, for $1000 less, the af100 body has a hires vf. |
Re: Sony NEX-FS100 Camera Test
The argument for internal ND filters is strong only for ENG (or run-n-gun) and in any controlled situation changing filters and lenses is normal operation, and there the lack of internal NDs is of no consequence. Only under controlled situations is DoF a prime concern and there the ability to use wide aperture primes coupled with a large sensor gives all the control needed. Those arguing that the FS100 is not suited to ENG may well be right (they said the same of the EX1, buttons too small, etc). The point about the benefits of bigger sensors is that resolution is determined by the sensor characteristics together with lens characteristics, but in small sensors (0.5") the sensor is so powerfully limiting that maximum resolution can be achieved only with large lens apertures (f/4). Given a large sensor of fine pixels then the lens might become the limiting factor and then indeed you can shoot at f/8 - f/11 with full resolution. This means you have 2 to 3 stops more to play with and hence less need to be changing NDs; but of course only run-n-gunners would be concerned by the time needed to change a filter.
EDIT: just to help out with ball-park figures: the sensor is nominally 13.3mm "vertical", so for 1080 lines, red light of 630nm, any aperture wider than roughly f/16 resolution will not be degraded by diffraction. For a 1/3" sensor, the corresponding aperture is f/3.5. So using 1/3" camera with a built in zoom lens you have a working range of aperture of, say, f/2.8 to f/4 (or even f/2 to f/4), so you really need to be rolling those NDs. The FS100 will allow you another 4 stops. Now, Steve seems to think I'm arguing that this means that ND filters are unnecessary, which is an odd misreading of my post. |
Re: Sony NEX-FS100 Camera Test
Having shot 16mm on fast moving documentaries, you can pretty quickly change filters. Perhaps the main problem is having somewhere to put the filter and can then quickly pull it out again. Although, it's not as easy compared to having the built in filters, which can be used for the course ND settings.
However, this process does rule out screw in filters, they're too slow. I'm not sure that in practise a lot of people worry about the diffraction limitation. I know quite a few people who aren't even aware of it becoming a factor at about f5.6 on the 1/3" cameras. As mentioned, especially with these sensitive sensors, you need a wide range of NDs to control the light. You could use a variable ND, but you'll still need the finer grades of ND for precise control when the variable is too dense. |
Re: Sony NEX-FS100 Camera Test
Quote:
2) E-mount is proprietary, but it is much easier to find appropriate lenses for... there are very few native m43 lenses that actually function natively with AF etc. |
Re: Sony NEX-FS100 Camera Test
On a 1/3" sensor diffraction softening comes in at apertures smaller than f/4 (green light) so at f/5.6 it is noticeable. But certainly people who ask "why are my HD images soft?" don't know about diffraction but my observation is that serious cinematographers are well aware. On 16mm and 2/3" sensors the limiting aperture is closer to f/8 (less of an issue), which might be the basis for some people talking about f/8 as being a "sweet spot" for lenses (although actually that is generally about 1 stop closed from full aperture). The convenience of built-in NDs is undeniable; I just don't consider this issue to be of the significance claimed in discussions.
|
Re: Sony NEX-FS100 Camera Test
Quote:
I read an article from Photokina 2010 that 7 new lenses from other manufacurers for emounts would come out "over the next few years" here's a nice chart detailing 20 of the mft lenses an af compatibility on the gh2. Perhaps there's one for the af100. And maybe Sony has one for the fs100. That would be informative. Compatibilities of DMC-GH2 | Compatibility | Digital Camera | Product Support | Support | Panasonic Global There's an interview with Sonys Bill Drummond explaining the positioning of the fs100. It is clear they were targeting the budget filmmaker hence no ND and the cheaper hdmi. It strikes me that those on a budget will look pretty hard at saving $1000 and buying in to a lens ecosystem that easily let's them use all that old glass. In contrast, the fs100 is $1000 more and feature to feature is missing the vf, nd, and Sdi. So as you point out, it's chip has a different lineage and so it's more of an upgrade and maybe not so much for the budget filmakers. Hopefully the fs100 image is observably $1000 better than the af100. |
Re: Sony NEX-FS100 Camera Test
Quote:
|
Re: Sony NEX-FS100 Camera Test
Quote:
I'll admit that I'm not sure of exactly what points you are trying to make in your lengthy posts, but I'll just comment on a couple of things. 1) I've been shooting quite happily with several of my f/2.8 Nikon lenses on a prototype FS100 outdoors in bight sun. A "monster ND filter" has not been necesarry. I've just been using a $150 variable polarizer to quite easily control the light. No big deal. 2) I don't need to buy filters for every lens. Ever hear of step-down rings? I always buy 77mm filters and use $10 step-down rings to use those filters on any size lens -- including Sony's stock 18-200mm lens. No problem at all. One filter fits all. Also, the stock lens is a joke with it's f/3.5-f/6.3 aperture range. Junk. Totally defeats the whole reason of buying a cool camera like the FS100. My advice to people is to buy the body only and outfit the camera with your own Nikon, Canon, or PL glass. I don't know ANYONE who is advocating using any E-mount lenses, so debating their performance is moot in my opinion. The great thing about the FS100 is that you have the entire world of Nikon and PL lenses at your disposal. E-mount?? What the hell is that? :-) The more I use the FS100 the more I appreciate what it can do -- especially at this price point -- and I own an F3. You're in Vegas, right? Let's meet at the Sony booth for some one-on-one time with the camera. |
Re: Sony NEX-FS100 Camera Test
Doug,
Only a polarizer in the sun? That is odd. I was just outside today with an EX3 and had to use the 2nd ND and go to f4 just to get some faces in correct exposure. Heck, last summer, there were days where I had to use a circular polarizer in addition to the 2nd ND just to keep my aperture at f8. Now, imagine shooting a f2 or f1.4 lens wide open and you do need MONSTER NDs unless you have a large grip truck to bring down the sun (and necessary space to do so). |
Re: Sony NEX-FS100 Camera Test
Steve,
I'm not talking about a regular polarizer. I'm using a variable ND polarizer. I'm surprised you're not familiar with it, especially since someone else already mentioned it on this very thread and I've been using mine on DSLRs for a couple of years. It is a screw-in filter with two polarizers sandwiched together and they rotate independently. Depending on the relative angle of the two polarizers, you can vary the light transmission from about 90% to 0%. Yes, you can completely block out all light. And since it is totally variable, you can get the perfect amount of ND you need. In some ways it is better than normal built-in NDs that only give you two options (1/64 or 1/8). I don't care how bright the lighting is, I can get the exposure I need with the lens wide open and the shutter speed at 1/60th. The variable polarizer will knock it down. And as I said, I only need one large filter, and then I can use step-down rings to fit the various lenses I own. A cheap, easy, and effective solution. I've done it with the FS100, and have plenty of footage that I'm showing at NAB. |
Re: Sony NEX-FS100 Camera Test
Not sure which Steve you are talking to, but a Vario ND only enables a 2- to 5-stop reduction.
You need much more and less. In snow the much more is a lot more. The idea that simply buying a Vario-ND is a solution is, frankly speaking, not supported by those that use these cameras. |
Re: Sony NEX-FS100 Camera Test
Singh-Ray claim up to 8 stops.
Singh-Ray Filters: Vari-ND Variable Neutral Density Filter However, you'll need more than that to get f1.4 using an ISO 800 sensor on a sunny day. Are these cameras IR sensitive with all this ND stacked up? |
Re: Sony NEX-FS100 Camera Test
Claims is the operative word. Yes, you can UNEVENLY cut light by 8-stops. Past 5-stops you get color shift and uneven density.
And, yes you can stack ND filters. But should you, NO. An ASA800 camera and a fast lens at 1/48th -- I'll check my lightmeter but you are going to need welders glass. |
Re: Sony NEX-FS100 Camera Test
I suspect you'd be using a hot mirror to get that level of ND ( I don't think it's available yet at that factor). No square/rectangular standard ND filters currently goes that far and ND 3.0 (ten stops) is about the limit for screw in, plus the IR could become an issue.
|
Re: Sony NEX-FS100 Camera Test
Doug is speaking about his experience with the camera, and his observations must carry a great deal more weight than speculation. If the camera were to have internal NDs they would be much the same as in other cameras (1/8 & 1/64), and since it doesn't have them the user will have to mount an equivalence externally. Inconvenience? Yes, for some. Problem? No. Matte-boxes, like tripods, serve many different cameras.
And not having internal NDs means that the camera accepts a wide range of lenses without having to make focus allowances for internal glass. |
Re: Sony NEX-FS100 Camera Test
The internal NDs would be (or should be) factored into the camera's own flange distance set up, so you would unaware of it unless there was a mismatch of the filter thickness.
This camera's ND arrangement is no worse than that found on the RED Epic, it seems to have a pretty similar ISO, so will need the same levels of ND filtration. |
Re: Sony NEX-FS100 Camera Test
Quote:
|
Re: Sony NEX-FS100 Camera Test
Quote:
I saw that there was an adpater to put A-mount glass on the native e-mount but somehow I missed that there is an e-mount adapter to put Nikon glass on the FS-100. That changes everything for me. What adapter are you using. Also, what, variable polarizer are you using. |
Re: Sony NEX-FS100 Camera Test
Hi Les,
I'm in the Sony booth in the afternoon today, but teaching an FS100 / F3 workshop in room N102 from 10:00 - 12:00. Please come id you can make it. By the time I wrap up the Q&A and grab something to eat, it's 2:00 before I get to the booth. On Thursday I should be in the booth all day, with a few quick excursions to other areas of the show. Would love to meet up. Not only is there a Nikon adapter for the FS100, there are about a dozen adapters already available or just about to come to market from 3rd party manufacturers . . . Nikon, Canon, PL, Alpha, Olympus, etc. I can confidently say that if there is a lens that covers the Super-35 image circle, somebody is going to make an adapter so you can put it on the FS100. Count on it. It's exactly this massive lens compatility feature that led to the camera not having ND filters. The flange depth is so short that there simply isn't room for ND filters. Sony traded ND filters for lens compatiblity. Did they make the right choice? That's open for debate, but that's the way it is. You either deal with it, or move to another camera that suits your needs better. I try to explain the relative differences between cameras to customers at NAB, and then let them determine which camera fits their needs. For some people the FS100 is perfect. For other people, I suggest that they consider stepping up to an F3, or even sideways to an EX1R or EX3. Different cameras for different folks. The biggest surprise to me has been the number of customers who say they are keeping their EX1 / EX1R/ EX3 and adding an FS100 instead of replacing there other camera. In my opinion, that is a very smart move. The lack of ND filters has already become a non-issue for me. Its easy to work around. |
| All times are GMT -6. The time now is 06:50 PM. |
DV Info Net -- Real Names, Real People, Real Info!
1998-2025 The Digital Video Information Network