utsira

If the acquisition and projection format is 1080p then I don't see any reason to use a higher resolution. Difficult to say whether 500mb is enough without knowing the length of the film. If a blu-ray movie is say, for the sake of argument, 90-120 minutes at 15-30 gb total, then that works out as 160-250mb per minute.

If Sony brought out a similar update for the A6000, I'd buy one tomorrow.

Nice! I liked the V overlay, not sure what to call it. What's the music?

Wow. I guess it was too good to be true. Props to 5D for being so transparent with their corrections though.

I think 5D may have mislabelled their diagram, comparing it to Samuel's figures over at DVX user. He measures GH4 4k at 22.5 ms and 1080p at 13.7, suggesting that Cinema 5D's GH4 measurement of 14ms is actually for 1080p. http://www.dvxuser.com/V6/showthread.php?303559-Measuring-rolling-shutter-put-a-number-on-this-issue! But I agree that the detail looks the same as other APS-C cameras at the moment, i.e. a bit soft. Regarding the full sensor readout, that comes from an interpretation of the Sony press release: It's possible that what is actually being described here is a form of pixel binning, but that would be a somewhat disingenuous description even by PR standards. I don't know a lot about moire, but I thought that even a full sensor readout doesn't eliminate it completely.

Hi Ebrahim, I've been using the E-M5 alongside the Panasonic G6 for a while now. This was all shot on an E-M5 with an Ai-S 85mm lens on a lens turbo: http://marginaliafilm.wordpress.com/2014/08/19/floating-lanterns/ Nb there is also some stabilisation in post added here. Only a very tiny bit added though, a few pixels in some cases, just to ease the vibration, as a heavy 85mm lens, even with a focal reducer, is sometimes just a bit too long for the stabilisation. (EDIT see post below this one) Some of the shots are with the 2x electronic telephoto, ie 244mm, these needed more stabilisation in post. Overall, I find a well stabilised lens, such as the Panasonic Mega OIS lenses, to be more pleasing for video. In my experience these are just as steady, and with less "drift" when a camera move ends. (edit, hmmm, on 2nd thoughts maybe OMD 5-axis edges this one......) Plus the G6 is incredible for video, and I really like the Panny lenses too. Of course, the E-M5 works with absolutely any lens, so it's definitely a great camera to break out if you want to grab some handheld shots with manual glass.

Cinema 5D have tested the A5100, found it to have 13 stops of DR and less rolling shutter than the GH4. Very impressive at this price. It's good that cinema 5D now seem to be testing this kind of thing routinely. No mention though of what difference the 50 Mbps XAVCS makes vs AVCHD. I guess it's harder to devise a repeatable test for the kind of things you'd expect a higher bitrate to be better at, such as the amount of visible mud, blocking when filming detailed motion and so on. It would be interesting to see this up against a hacked D5200, to see who the king of apsc is. More testing to come apparently. http://***URL not allowed***/lab-review-sony-a5100-video-dynamic-range-power/?utm_source=rss&utm_medium=rss&utm_campaign=lab-review-sony-a5100-video-dynamic-range-power

Love it! Great name for a company too I think. Where can I get a seaweed hat?

Yes, this is my experience too. It is interesting that you do occasionally see artefacts that look like post stabilisation. Does this mean that the OM-D stabilisation in video mode is partly electronic? The shooting while walking thing is interesting too. It's the rotation of your shoulders from side to side as you walk that it doesn't seem to be able to correct. So it isn't a steadicam substitute in that sense. But as long as both feet are planted on the ground (or you're in a wheelchair?), it's pretty impressive.

I used the Lens Turbo on my E-M5 at the weekend, with an Ai-S 85mm f2. The shots did need varying amounts of stabilisation in post, but often just a tiny amount to smooth out vibrations. Some of them are with the x2 electronic telefocus as well (244mm equivalent, if my maths is correct? 85mm x 0.72 focal reduction x 2 m43 crop x 2 telefocus). It is incredible that you can get usable shots handheld at that length. You can also see the E-M5's codec falling to pieces at various points unfortunately.

I reckon go for it (though it's not my house either). People have posted films in the main forum before. There was a horror short not long ago. And you're right that posts in the screening room don't seem to get the number of hits.

IMO they messed up the timing of the XAVC-S roll out, in that their mid-range cameras like the A6000 (and even the A7) now look underspecced next to the entry level (here in Japan there's no A3000, so A5100 is now the entry level system camera, apart from all the 5R, 5T, and 5000s still floating around). I might have picked up an A6000 at the end of the summer, but now I'm going to wait until there's something with a decent-sized grip and top dial, an EVF and XAVC-S. I don't know whether that will be a firmware upgrade for the A6000 (I hope so), an A6100, A7000, or what.

Most of the video above is a 1/50 shutter, 30 fps, in a 24p timeline, (with a bit of super slow optical flow in the middle). The reason why is that while the whole of Japan is NTSC, eastern Japan (including Tokyo) has 50hz electricity, so electric lights would strobe at 1/60. By mistake the shot of the flower shop is 1/60 (you can just make out the light source flickering, there's still quite a bit of daylight at this stage so it's not noticeable). The rest of it, certainly the night scenes under electric light, is 1/50. I can't tell the difference in terms of motion cadence between 1/50 and 1/60. I think the 180 degree shutter "rule" can be flexed a little bit!

I don't know whether an E-M5 firmware update is also in the offing, but it could certainly do with an updated codec. It's only 20 mbps and it breaks and turns to mud very easily. Kind of disgraceful for a 2012 camera that was $1000 or something when it first came out. It seems the E-M1 and 10 do a much better job in that regard. Regarding frame rates, I often just throw everything on a 24 fps timeline and retime it for 80% slow motion.

Thanks guys! All shot on the Panny 20 1.7 v1. I like setting myself "just one lens" challenges...

I've not seen any mention of s-log in the coverage or press release so far. The list of picture profiles looks the same as it did in NEX days. Still, given how poor the reporting of video-related features is on some of the more photography-centric sites, I suppose it's possible it's there.

And.... the embargo is lifted: http://***URL removed***/previews/sony-alpha-a5100 Has the same focussing system as the a6000, but with a touch screen, which could make for some interesting possibilities. Clean HDMI out too. This could be a very cost effective brain for a larger rig. From the press release: =full-sensor readout. I haven't found any mention of high fps slow motion modes though. edit: press release again says "1280 x 720 (120p)" not bad. 50 mbps XAVCS 24/60p, full sensor readout, 720p 120fps, clean HDMI out, zebras, focus peaking, 179 focus points, first (?) touchscreen implementation of the new alpha menus. September will be an interesting month for cameras. We're starting to see how far down the range the new codecs and other features are going.

Sony don't exactly have a history of exciting or meaningful firmware updates, unfortunately... they barely bother with the lenses, it's just body after body. but yes, I'd love to be proven wrong here. I've been sitting on the fence about the A6000. 50 mbps would push me over

According to Sony Alpha Rumours. If true, this means that XAVCS isn't just for cameras at the top of the lineup. I wonder whether it's a full sensor readout as well. I wish they'd hurry up and do an A6100..... or better yet a meaty firmware update for the A6000 http://www.sonyalpharumors.com/sr5-this-is-the-first-image-of-the-new-sony-a5100/

Yeah, Tetrachrome 400 is quite a warm LUT, but the greens suffer. I see the LUT as the starting point of a good grade.

OK nevermind. Looked at it again, and it isn't quite flush in the mount, there was a hair's breadth gap on one side. Pushed it in, and then I was able to twist it. Very stiff and grindy though, and worried that I wouldn't be able to take it off. It seems to be getting a little easier after taking it on and off a couple of times. Am I in danger of damaging the mount on the camera do you think?

Hi, I have a Zhongyi Lens Turbo that fits fine on my Panasonic G6, but I recently noticed that it doesn't attach to my neglected Olympus OM-D E-M5. It will sit flush in the mount, but then you can't turn it at all to lock it in place, not even a fraction of a mm. I think i've probably had the Lens Turbo a little too long to return now. It didn't occur to me to test it on the E-M5, as I've been shooting so much with the G6, and it worked fine on that. Has anyone ever experienced anything like this either with a Lens Turbo, or with other adaptors or lenses? Should I try to force it? File something down?? Any pointers would be greatly appreciated.

I had a go at grading Roman Legion's original files using the ImpulZ LUTs: I wanted to compare the generic log LUT that Impulz ships with, to the one that's calibrated for the FS700's s-log 2. I didn't touch the colour board, just a very quick and easy test.

+1 for the 20mm 1.7, one of my favourites. I have version 1, should be quite cheap second hand since v2 came out. I completely understand the appeal of old manual lenses: lovely direct focus throw, all metal construction (sometimes), massive front element etc; ergonomically they're a joy to use for video, and can be incredibly cheap second hand. Sometimes I look at a photo taken with my Ai-S 24mm 2.8 and it looks like a photo from the early 80s, because of the rendering. But to go wide on m43 you do need a speedbooster, and IMO the cheaper imitations (lens turbo etc) don't cut it. This is an unpopular opinion, but I'm not even 100% sold on the speedbooster So don't overlook native glass, some good deals can be had second hand, and the lenses are fantastic, lovely detail and rendering. The assumption that modern glass="sterile" is complete nonsense IMO. 3 panny primes might work out cheaper than, eg 3 Ai-S + speedbooster