MattH

I do think it would be good if it had h.264 as an option ast well.   However, this way, at least storing footage on a shoot will never be a problem.  The reletavely low bitrate will mean you can use more regular sd cards, and you wont need as many.   Far cheaper than CF cards or ssds shooting raw.

Looking at the still image comparer, compared to a 7d mk 2, the samsung smokes it at 3200 iso.  But it still isn't quite a match for the 5d mk3.   I think samsung are just paranoid about the high ISO thing.  Plus their kit lens is faster than normal kit lenses so you will use lower isos in the same situations anyway when compared to another camera with a kit  lens.  Review sites purposely underexposing to turn up the ISO when they dont even need to makes the camera look bad.

"We havent even seen anything from the c100 mk2 yet"

http://***URL not allowed***/canon-c100-mark-ii-review-exclusive-footage/

Thanks. Some of that looked really good but other parts not so much.

But i'll let the thread get back to the JVC.

Pretty vague information. The picture of the document doesn't even show the right side of the text.   I doubt they are moving a filter across the sensor, as in the whole fucking sensor.  It will more likely happen on a pixel level.  The filters will probably be electronic.  I can't see having 4 million miniscule moving parts to be very easy to make or being very reliable.  What is interesting is that it talks about effective pixels, not photosites, and each effective pixel area is about twice as wide as it is tall. It will be interesting when more info comes to light.

I almost wish I didn't hear about these cameras untill they were in users hands.  It's going to be months before we see a sample from this.  We havent even seen anything from the c100 mk2 yet.   I'm not holding my breath for a miracle with this.  There will be something wrong with the image, there always is with these semi consumer cams.  semi consumer meaning anything from a mainstream company under 20k.

60p on youtube only works in google chrome.    If you are using firefox or another browser you wont see it.  (I will continue to use firefox and just paste the url into chrome whenever I need it which will be hardly ever).

It is definitely a very noticable difference from 24p movies.

Personally I think saying it looks like video is off the mark.  I think people are searching for a term to describe what they felt and "video" or "soap opera" was the best word they could come up with.  I think any video-ness is probably more to do with the sharpness of the camera than anything.

The impression I get more than anything is that it looks like a computer game.

CGI at 24p with motion blur tends to blend the real elements with the fake quite well.

But at 60p the CGI is less blurred and so your mind can critique it more easily.

60p also has the effect of making live action seem more real.  And this sense of realness makes the clear cgi look even faker.

So this trailer basically looks like a 3d computer game with 2d live action video sprites layered ontop, which is kind of what it actually is really.

As to whether It is good, I think if the whole thing was cgi it could work stylistically.

As for live action narative I'm not sure.

interpolated 120hz smooth motion is very disconcerting but maybe genuine 60p is better. 

Its certainly different, and a definite departure from the traditional cinema look.  I would have to watch a normal live action film without cgi to see if I could get used to it or whether it continued to be distracting.

One thing is for sure, even if people get accustomed to 60p motion, CGI will have to up its game to not look fake.

How many times are you planning on repeating what you don't like about the LX100, and that you plan to get the FZ1000? I'm sure Andrew will make a thread about it soon, or you can make one yourself.  Others like the fact that it does not use a mode dial, and most people understand that this was not meant to be a camera to replace the GH4.

 

I think the point is that your post gives nothing in this discussion. If you only can accept 4k with more bits, then there are cameras there for you but they are x thousands of dollars, and there is certainly no-one in this thread about a relatively cheap compact camera that happens to take great video that think it will replace those. The Lx100 is a camera that will take nice quality 4K, and downscaled to 1080p (which is what most probably will do at this point anyway), it seems to compete favorably to far more expensive 1080p cameras. Also, there is an old discussion about downscaling it could yield 1080p 10-bit. (Math seems to support it, but I don't recall any definitive confirmation in testing the quality differences when doing so)

 

People, generally, if you don't feel this is the camera for you, there is nothing gained by moaning here about missing features. It is what it is. Focus on other cameras instead, or maybe make a wishlist thread for future LX cameras. (but be realistic :))

I didn't know this was the "Be 100% possitive and nothing else about the new camera of the week" forum.  But thanks for setting us all straight.

Personally I find it a little puzzling that the issues the camera labs review exposed about the camera havent been addressed in this discussion.  And like all enthusiast gear lust cameras its over priced.   A new camera is a write off for a gear reveiwer but not everyone can spunk over every camera that comes out.  Is this post too negative?  here's a challenge: write a possitive response to it.

The "beach" shots were really good.  Looked other worldly so I didn't know whether it was a composited background, but when she was walking round it didn't seem like green screen so I guess the location just looked fantastic at that point.  Was that the 5d3 raw sunset shot you were refering to or was that the earlier one looking over the city?  The bird flying off behind her looks great too.  Really lucky to get that in the shot, unless it actually was composited.

Andrew, I am not sure what you mean by: "So I find myself using stills mode to shoot video, but when I do that the mechanical shutter has to be enabled if I am to use S-LOG"

It means exactly that.  The point of the sentence is that it doesn't make sense.

You can shoot video in either video mode or stills mode.

Video mode doesn't allow you to take stills at the same time but stills mode does.

When shooting stills the mechanical shutter is used as with all stills cameras, but their is also an electronic shutter mode for taking silent stills (with the dissadvantage of rolling shutter).

But for some unkown reason when silent shutter is activated the camera will not alow you to shoot video in S-log2.

But when mechanical shutter is activated you can shoot slog2 even though there is seemingly no connection between them.

This means you cannot take silent stills at the same time as shooting slog2.

According to Cameralabs the 4k video involved both a crop and scale to get to 4k.  Has anyone been able to confirm this?  Is moire more prevalent in 4k video vs the GH4?

 

http://www.cameralabs.com/reviews/Panasonic_Lumix_LX100/

 

"But bizarrely Panasonic has opted to employ a mild crop and a scale to generate the desired frame size. This means the LX100's 4k footage not only has a slightly narrower field of view to 1080 video and 16:9 stills (so instead of 24mm at the widest zoom, you're getting 26mm), but it also now suffers from moire. I feel it's an unnecessary compromise too, since a nice clean 1:1 crop would have only involved losing 4mm in the diagonal field of view.

 

I put my disappointment to Panasonic and I got the impression the scaling along with absence of Cinelike D profiles was partly for the LX100's target audience and partly to protect the GH4, but I still think a 1:1 crop would have been better, and rather than detract from GH4 sales, I could see existing GH4 owners picking up an LX100 as a B-roll camera. Maybe flat profiles and 1:1 video crops will be offered in future third-party firmware."

Thats a really useful review.  The moire is a dissapointment.  If they downscalled from the full 16 by 9 still it would at least make sense.  But to actually crop but not crop to 1:1 even though its so close, well that must be intentional crippling.   The image stabilisation doesn't seem quite good enough to compensate for the micro jitters from the rolling shutter either.  No flat profiles.  Weirdly he says early on in the review that he will have a video showing the flatest settings, but it is nowere to be found.  Looks like it's not a GH4 in your pocket after all.

By downloading the video and frame advancing when the tram passes I have measured the total frame skew to be aproximately 80% of the distance a simgle point moves in 1 frame.  Given that the frame rate is 24fps, this puts the readout speed at around 33.3 miliseconds, which is disapointing after what samsung were saying about the processing ability of the camera.

I recently acquired a Titan. Holy smokes. I'd agree that pairing a 5K panel with that would be something!

http://www.geforce.co.uk/hardware/desktop-gpus/geforce-gtx-980/specifications

http://www.geforce.co.uk/hardware/desktop-gpus/geforce-gtx-titan/specifications

http://www.geforce.co.uk/hardware/desktop-gpus/geforce-gtx-titan-z/specifications

Scroll down to look at maximum digital resolution.   (4096x2160 for those who believe me).

According to this you wouldn't be able to use the whole screen natively.  Seems stupid to pay for the pixels when you are going to risk artifacts with upscalling or not use the intire monitor with 1-1 pixel mapping.

A DCI 4k resolution monitor (as opposed to quad HD) would make more sense to me.  Altough they may be rare and more expensive.

People said it doesn't have much moire.  It is very difficult to find a camera that doesn't have any.  As maxotics said, you can sometimes see moire with the naked eye and you don't critisise your eyes.  Im guessing in this instance that wasn't the case so it is the camera that is doing it, but it dosn't look that bad. At least its not rainbow moire like a black magic.  But I guess its good for people to be aware that plain but finely textured ties can be an issue, and maybe keep some moire proof ties just in case.

Just watch any videos on youtube.  Iphone 6 plus is excellent.  Great colours and range (it looks almost semi hdr) Excellent stabilisation, really good detail without having to record 4k then downscale it.   The normal iphone 6 is really good too, but the +'s optical stabilization just makes the stabilisation that little bit better.

The Z2 has 2 levels of digital stabilization. The 2nd level can only be used in HD which isn't that detailed.  4k can be taken with the first level, but If you compare its 4k to that of a samsung it is really soft.  I think its the lens.  Stills look soft as well. So even with 4k I dont think you are getting the detail of the iphone 6/6+ in HD.

The samsungs in 4k have more detail, but to me the image is too contrasty with no range and over sharpened.

The fact that you already use Iphone, Its a no brainder.  Get the +

You probably cant go that far wrong with an a6000.  Fairly small, APSC sensor, viewfinder, decent video if you need it. compact zoom kit lens available. Other lenes to expand to. many with OSS (although nex lenses are fairly expensive).  You can use adapters to use nikon lenses.  

EOS-M has very good quality stills and is cheaper and has great lenses which are also cheap, but it doens't have a viewfinder and video is meh. 

I think if you personally want a camera with a particular feature like 4k you should save up and buy one for yourself.

5210 by 2880 Works out at just over 15 megapixels.  Great for displaying high quality photos.  But many new graphics cards can only support 4096x2160 rez, so I don't know If the amd laptop video card apple have got in this thing is capable of native 5210x2880 or if there is an upscaller in there (which seems pointless to me).

MattH: I don't think the chart you posted shows the comparison I am after, ie fullframe vs APS-C with Metabones at the same focal length. Real world experience from someone who has Metabones Ultra would be great but it is probably bit too early.

 

At this point, I can either spend $399 for a standard Metabones adapter for fullframe shooting with Canon lenses and $649 for Metabones Ultra for APS-C shooting. I am wondering whether I could get away with shooting APS-C all the time with the Ultra adapter.

No it doesn't show a comparison with the speedbooster.  But it shows the detail difference regardless of the speed booster.  The difference between them at 60p is particularly noteworthy if 60p is important to you at all.

You may have seen this but it is a comparison using a speedbooster, though probably not the ultra:

Hopefully someone with the ultra will be able to give you some feedback.

The differernce between full frame and crop detail can be seen here:

http://www.similaar.com/foto/slashcam-de/slashcam-2.html

Resolution is prety much the same.  Except apsc crop is better at 60p than full frame.

There is a slight narrower feild of view using the speed booster.  If the speed booster ultra is as good as it says it is then there is no reason not to go with it really.

They are talking about rolling shutter dudes... I'm almost speechless now...

 

Every single CMOS camera uses a different readout speed than what is recorded. The readout is just how fast the sensor gets read. 24p, 25p, it has nothing to do with the end result. And in that white paper it should say 1/60s readout speed, not fps.

 

You do know that every single CMOS camera (except global shutter cameras) do this?

 

So keep your horses in your stables.

It says 60 frames per second, not 16 millisecond read out spead.

60 frames per second may equate to 16.66 ms readout, but its at 60fps not at 24fps.

Iv'e always suspected that something like this happened.   Think of how difficult it would be to get the sensor to start each frame at precise devisions of time for pure 24,25,30p etc.  Not impossible, but much easier to read at 60fps all the time and only record some of the frames.  The first and third of each 5 frames in the case of 24p.

If this was played back at pure 24p timing, every second frame would be shown 1/120th of a second early.  I can't tell you if that would be perceptible or not.  I suspect I wouldn't be able to tell unless I trained to look for it.

But since most people are viewing web material on 60hz monitors, the conversion back to 60hz may play the frames at the exact same timings they were recorded in.

However I am not aware if a 24p video file gives any indication of which frame is which if it initially came from a 60p source.  Maybe someone can enlighten me.  But I would guess it doesn't.  If not then there is a 50% chance that every second frame would be played 1/60th of a second late instead of 1/120th of a second late if it were a pure 24p.

Again I couldn't tell you if this difference would be perceptible.

I don't have a panasonic but this looks good.  Much more natural,

Does 1/60th of a second vs 1/50th of a second really make any difference?  As far as matching 50hz light sources, the frame rate is the most important thing surely?

The amount of exposure per frame will be slightly different depending on at what phase in the oscillation you press the record button, but at 25fps the next frame begins at exaclty the same point in the oscillation so every frame will have the same exposure.   

Matching the shutter angle to the light source is only relevant if your frame rate isn't matched.  But then each frame starts at a different phase so you will still have an amount of flicker anyway.

So If you're shooting 25fps, the difference is just a 16.66 millisecond exposure time instead of a 20 millisecond exposure time.  So you are getting 17% less motion blur and 17% less exposure.  If there is any more differernce than that then I would like to see an example.

They are very close in terms of colour and contrast.  But when viewed at native resolution, the 5D3 has the edge for detail.  The bottle and lamp stand out from the background and look 3d.  The D750 seems to have a fine layer of smudge over it.  The bottle looks like a flat object.  You cant sense its shape.

However, when the image is moving it may be harder to tell them apart.  And if this is the level of D750 detail at 60p, can the 5d3 raw even match that?

Lets also not forget that 5d3 raw is behind 4k cams downscaled to 1080 in terms of detail.  So if detail is paramount, you arn't getting the best with 5d3 raw anyway.

Very good detail for 60p slow motion.  Not 4k downsampled to 1080 detail, but then what camera gives 4k at 60p?  It looks like youve gone for a punchy grade here to show off the colour which is good. It would be cool to see some ungraded samples.  Just corrected for superwhites if the d750 has them?  So far on youtube the suposedly flat profile samples didn't really seem that flat.  Aparently one way to combat the A7s blue highlight blow outs is too shoot with a warmer colour temperature, and correct in post.   Setting Colour temp for yellow lights probably pushes the blues up to make it less yellow, but then when a bright blue led comes on it is too much.

Just a brainstorm idea, but you could invite them to a viewing if youve got a place good enough to do it, then theres no chance. Unless he runs off with the disk/player/laptop.

I think we all know what the deal is.  You have said it before, so theres no confusion:   Video sucess on dslrs was an accidental success for canon.  A nice suprise at the extra interest and sales.  So they created the spin off cinima eos lin to carry the pro video forward and charge as much as the market will pay for it.   The dslrs then just continue to be 100% stills orientated.  The 7d mk1 and mk2 are mainly for sport protographers.  Where the crop is an advantage (as long as the sensor is higher resolution) and high frame rate (with good autofocus) is of prime importance.  The video is just an added extra.  Its just there if they happen to need it.  But kept at consumer level quality, with all video improvments like better autofucus squarely in the consumer arena.