cantsin

Once again, Blackmagic officially specifies that Resolve 12.5 and - even more significantly - Blackmagic Camera Utility require Windows 8.1 64bit or higher. Windows 7 is not supported. If people use it nevertheless and encounter problems in the workflow, they shouldn't complain. It's the same tiresome story as with people constantly complaining about unreliable recording and framedrops with BM cameras because again and again, they use unsupported SSDs/SD/Cfast cards.

Well, if you are comparing the 64bit edition of Windows 7 to the 32bit edition of Windows 10, you're comparing apples to oranges. Of course I was referring to the 64bit version of Windows 10. And Resolve won't run on Windows 7. With a Blackmagic Camera, you buy into a software/hardware ecosystem consisting of the camera and Resolve. It simply makes no sense using this when you're on Windows 7 - just as it makes no sense to use an end-to-end ProRes workflow without a Mac.

Indeed. Windows 7 doesn't even support USB 3.0, a sheer necessity when transferring camera raw files. Conservative folks who prefer to stick with older yet proven OS technology shouldn't expect that it will work with bleeding edge video technology that requires every bit of hardware and OS performance. Upgrading to a newer Windows version is a necessity if one wants to work with Resolve and a highly GPU-accelerated workflow. For a Windows 7 system, better work with software like Avid and Premiere, and cameras like the C300.

No, because reaching s35 or Full Frame/Vistavision DoF equivalence on a 1/3" sensor would require impossibly large apertures. But emulating Full Frame aesthetics on MFT or s35 aesthetics on 1"/s16 is indeed just a matter of two more f-stops. Noise/grain doesn't differ that dramatically between those sensor formats when the target resolution is 2K or lower. (But I really wonder why I am even replying to such flamebait.)

Aesthetics of different film/sensor formats is a myth left from analog film and older digital days. When shooting 16mm, emulating the look of 35mm film was nearly impossible - even with equivalent focal lengths and DoF - because 16mm had four times the grain and a quarter of the optical resolution. With modern camera sensors, this difference no longer is relevant, at least not when shooting/mastering at 2K or lower resolutions.

Simple advice: for 4K, only buy lenses (a) that are optically corrected and don't rely on software/firmware distortion compensation, (b) whose resolution is rated at 12 MP on dxomark.com or comparable review sites. These two criteria rule out most Panasonic and Olympus zooms, and most of their wideangle primes.

The Minolta/Sony A-mount is, like Canon EF, a 100% electronic mount for lenses with no manual aperture control. AFAIK, it uses a non-open, proprietary protocol and hasn't been reverse-engineered by adapter makers. In theory, it should be possible to make a focal reducer from A-mount to MFT. In practice, A-mount lenses probably didn't have enough market share to make such a development economical. The upside of this: If you have a Sony E-mount camera, and Sony's official A-mount to E-mount adapter, you can use old Minolta AF lenses as you please - and find them dirt-cheap on flea markets. (Bought a mint condition Minolta 100-300mm zoom for 10 bucks on a flea market last weekend.)

Check out this - found this a really good, informative, to-the-point critical overview:

Samyang 12mm/2.0 MFT, an excellent affordable lens.

In most countries, broadcast standards require at least 4:2:2 color sampling in broadcast-quality codecs. This excludes practically sll DSLR and mirrorless cameras proposed here in this thread unless you combine them with an external recorder. The cheapest camera to record broadcast quality video internally is the Canon XC10 (or the BM Pocket, yet with some strings attached regarding IR filtration, sound recording and extra cost for lenses).

Likely not doable because the housing of the pocket (with the rear TFT display) doesn't provide enough heat dissipation for 60p processing.

It doesn't make sense to store your project in a higher bit depth or color sampling rate than in the source material, unless the edit includes a lot of SFX or stills in a higher color depth. The pragmatic advice is to use h264 high profile, intraframe, 10bit if necessary, at the bitrates normally used for ProRes, DnXHD/HR , Cineform. The resulting quality will be better than ProRes/DnX/Cineform because of the more advanced codec. h264 also the advantage of being an open standard similar to JPEG, mp3 etc.; it will likely remain playable for a few decades if not longer. If you want absolutely highest archival quality, export DPX or TIFF sequences + uncompressed audio files.

The two cameras have the same sensor.

The Yi4k has a ProTune-equivalent "flat profile". And GoPro never put Cineform into its cameras, and seems to have abandoned the codec. - Besides, there is little sense in exporting 4:4:4 video created from applying a LUT to 4:2:0 video.... The whole GoPro studio/Cineform workflow is outmoded today (at least for action cam footage) where computers have fast enough for editing the native camera footage without transcoding quality loss.

It is fascinating how much insight members in this forum have into engineering processes, even if official documentation says otherwise.

Believe it or not, Blackmagic cameras - including the Pocket - have active sensor cooling (which BM officially states on its website, "internal active cooling for the cinematic sensor") for a reason. And the 2.5K cinema cameras have bigger and heavier housings for the same reason. Active sensor cooling, finally, is the reason why the BM Pocket eats up batteries so fast.

Here's a prediction: raw video in consumer cameras won't happen any time soon because of the steep requirements for fast storage media. Companies like Canon cannot afford, like Blackmagic, to list only a few SD cards as being fully compatible to their cameras, and deal with resulting user confusion and complaints. (Even Blackmagic's own forum is full of people complaining about frame drops and corrupted recordings, and it turns out every single time that they ignored the list of recommended storage devices.) Conversely, Blackmagic can't upgrade the Pocket and Micro Cinema with a higher resolution sensor as long as the existing 2.5K sensor (of the orginal Blackmagic Cinema Camera) has too high cooling requirements to be put into a small camera housing. Unless there's any 1" or MFT sensor on the market, or announced, whose power consumption/heat dissipation is on the same level as the existing Fairchild 1" 1080p sensor, a small camera with higher resolution won't happen. (EDIT: Blackmagic could use the 1" 4K sensor of the Micro Studio Camera for a Pocket-like small camera, but the gain in resolution would come at the price of a significant loss of dynamic range.)

Wow. If these forum reactions are any indicator, then we know why Trump will get elected this year.

IMHO: There's no really good reason to choose a GH2 over a G6 except maybe the price (if you find a really cheap used GH2). And regarding the question Nikon vs. Panasonic/MFT vs. Nikon DX, I'd compare videos and decide which I look to prefer: https://vimeo.com/search?q=d5500 https://vimeo.com/search?q=panasonic+g6 Panasonics have much better handling while Nikon is superior for stills (and might produce more pleasing video colors, depending on taste). Don't underestimate the costs for switching systems either. MFT lenses are generally more expensive than their Nikon DSLR counterparts; buying a Nikon-to-MFT Speed Booster isn't cheap either.

Don't underestimate other issues connected to the 50D: It can only record 30p (other frame rates not natively, but only by dropping frames of the 30p stream), the maximum raw video resolution is under 1080p, there is moiree and aliasing (worse than BM Pocket), and you'll need a potent computer (i7 + at least 16GB RAM + GPU power on the level of the Nvidia GTX960 or better) if you want to natively edit and grade the CinemaDNG footage. In addition, the 50D records uncompressed CinemaDNG on expensive, fast CF cards; the cheapest option, an off-brand Komputerbay 64 GB card for about $60, will only give you about 15 minutes recording time. Frankly, I wouldn't recommend the 50D anymore because of these issues unless one has most of the required accessories (fast CF cards, fast CF reader, fast computer, backup batteries, Canon-compatible lenses etc.) already lying around.

Here's an attempt at friendly advice: You might do yourself a favor by removing that video from youtube. Honestly, it doesn't flatter you. (EDIT: Unless it was your intention to establish yourself as a YouTube personality similar to "The Angry Photographer".)

Since you bring up academic standards and your scholarly background, you will also be familiar with academic standards for selection and qualification of peer reviewers. So here's a simple question: Have you ever actually had a URSA Mini in your hands and tested it yourself? Or are the points you are bringing up here from third-party sources? And are those third-party sources (a) sourced by yourself and (b) reputable? If the latter shouldn't be the case, you're not only damaging your reputation on this forum, but you're also damaging your reputation as a scholar.

Well, only zooms can be parfocal...

There are AFAIK no true parfocal photo zooms - true parfocality requires complex optical constructions which are too expensive to make sense in a photo camera. On top of that, even parfocal cine zoom lenses need manual shimming adjustment because of mechanical tolerances of camera mounts. This isn't possible with photo zoom lenses either.

Dug into some of my old 50D raw footage. It didn't show fixed pattern noise at high ISOs, but very strong noise nevertheless. Maybe this is dependent on the camera, perhaps also the running time (and resulting sensor heat).