KnightsFan
Members-
Posts
1,190 -
Joined
-
Last visited
About KnightsFan
Contact Methods
-
Website URL
https://gobuildstuff.com/
Recent Profile Visitors
The recent visitors block is disabled and is not being shown to other users.
KnightsFan's Achievements
Long-time member (5/5)
752
Reputation
-
Firmware updates could become subscription model?
KnightsFan replied to Andrew Reid's topic in Cameras
I totally understand paying for different software features, like separate purchases for raw recording, or log recording, or higher frame rates Subscriptions with monthly payments, on the other hand... how does that even work? Are there enough theoretical updates to warrant paying each month? I can't think of more than 1-2 software items I'd want updated on any camera I've owned. What can they do month after month? They can't increase dynamic range. So if the only updates are little things I either never use or work around easily, then I'd just unsubscribe. Not sure why I would pay for a subscription and not get functional updates that I care about. Unless the manufacturer shuts the camera off remotely if I don't pay monthly, which would of course be hideous, and I would never buy into that system. However... if the subscription model was like a constant rental, where you pay $x/month and you not only get software, but also hardware upgrades as they arrive, that might be a neat pricing model. When I buy a camera for X, sell it Y months later for Z, I always calculate my cost to own over that time period. If a manufacturer cuts out the buying and selling part, I see it as a win. For example, I could conceivably pay some fixed price per month to always have the latest Fuji XHx model--with some kind of future reporting on when new models will arrive and what they will have, of course. This is a very unlikely model for any company to move to, I'm just writing it out as something that would be interesting and actually pretty acceptable to me. -
This. I have no interest in mirrorless cinema lenses locking me into a particular mount, particularly manual cinema lenses. Nikon lenses in PL mount, however, that would be extremely welcome.
-
I think the patent says 2k, not 4k and I don't think many people want to shoot <2k even for compressed raw. https://patents.google.com/patent/US8872933B2/en (A rumored reason is that the SI-2K, which existed prior to Red's patent, shot 1920x1080 cineform raw. Red added 80 pixels and stepped into history).
-
KnightsFan reacted to a post in a topic: My crowdsourced rolling shutter sensor readout speed repository
-
My crowdsourced rolling shutter sensor readout speed repository
KnightsFan replied to horshack's topic in Cameras
I fully support your endeavor! I'm not negative on what you're doing. Ideally it is best to standardize, I just worry that the list won't grow very large, because of the purchase requirement. Unless you can get buy in from a big reviewer who gets their hands on a lot of models (or maybe you are a reviewer who gets your hands on lots of models personally) -
My crowdsourced rolling shutter sensor readout speed repository
KnightsFan replied to horshack's topic in Cameras
Sounds like either you or the other person measured it wrong (or possibly both of you did). -
kye reacted to a post in a topic: My crowdsourced rolling shutter sensor readout speed repository
-
My crowdsourced rolling shutter sensor readout speed repository
KnightsFan replied to horshack's topic in Cameras
Here's my list https://outerspaceoatmeal.com/tools/RollingShutterComparison.html You can add or not , up to you. Most of my numbers come from DVXuser, CineD, and a couple from other primary sources where the test method has been shared. Global shutter cameras are self explanatory so I link to the product page. Yeah I mean it's ideal to always do it the same way, I'm just not sure many people will buy a specific arduino to fill out this table. Difference with DR is that it's extremely subjective. Rolling shutter is not. People can measure it incorrectly-- which they can do whatever their intended test method is -- but they can't measure it correctly and then arrive at a different conclusion than someone else. Edit: And to be clear, measuring signal to noise ratio is also objective. -
My crowdsourced rolling shutter sensor readout speed repository
KnightsFan replied to horshack's topic in Cameras
In that case none of my values will go into your table. Seems like a waste, though--it's a raw speed so there's no subjectivity. -
horshack reacted to a post in a topic: My crowdsourced rolling shutter sensor readout speed repository
-
My crowdsourced rolling shutter sensor readout speed repository
KnightsFan replied to horshack's topic in Cameras
Nice! I have my own database of rolling shutter values that I can get to you. The one column I would add to the table is the ratio of the rolling shutter to the frame rate. That value normalizes the skew per frame. -
KnightsFan reacted to a post in a topic: My crowdsourced rolling shutter sensor readout speed repository
-
IronFilm reacted to a post in a topic: Let's bring back the good, old-fashioned camcorder of the 1990-2000s, but with modern specs.
-
IronFilm reacted to a post in a topic: Nikon buys Red?
-
I have no idea who licenses what anywhere in the industry lol. Between Nikon buying Red, Blackmagic joining the L mount alliance, several global shutter cameras announced in quick succession, and AI video generation/editing taking off, I do think this might be the point at which I least recommend buying a brand new camera. 2024 might be a big year for market shifts.
-
Probably Nikon has wanted to acquire a cinema company for a while. When Nikon first announced Z mount, they mentioned cinema a lot, but never backed that up with a full feature set. There were little things, like the tripod locating pin (so critical for solid cinema rigs). And of course supporting Raw outputs. But Nikon didn't make the big jumps, like timecode, internal NDs, XLRs. Maybe they wanted to, but didn't quite have the tech, personnel, etc. and buying out a smaller company was the easiest option. So really there are three possible futures. 1. Red's tech moves to Nikon's mirrorless cameras. Redcode perhaps, accessory compatibility e.g. their new EVF, the global shutter sensors. 2. Nikon's tech moves to Red. Z mount, autofocus, or even simple things like LCD screens, mirrorless-size EVFs. And the other huge category: lenses. Perhaps some of Nikon's excellent optics will find their way into cinema housings, either in PL or Z mount. 3. Nothing changes, but Nikon owns a more diversified product line
-
KnightsFan reacted to a post in a topic: Sony A9III with Global Shutter
-
SRV1981 reacted to a post in a topic: 8bit? 420? If you record with a LUT baked in
-
That was a Z Cam E2-M4. I saw similar results with the XT3 back when I used that primarily. Fwiw, I also saw similar color issues comparing Canon 5D3 8 bit vs magic lantern Raw. The color blocks are most obvious in relatively uniform gradients, such as skies. This tree shot isn't the best viewing since it's so busy with high frequency changes, but you can still see it pretty easily when you zoom to 100% or view on a 4K monitor, especially in motion. Most obvious is the greenish splotch in the bottom area that I highlighted, and the upper highlighted area has red and green splotches. If it's not a big enough deal for what you do, then great! To me, it's a big enough deal, since I have the option of 10 bit. There's no downside: file size is the same, and I've never encountered overheating on any camera ever (shooting narrative I take relatively short clips with time in between). That's not to say the difference is uber important... I mean I am splitting hairs about something that has very little bearing on the final product. I'm posting here to show what the difference is not to tell you that it matters for you. All else being equal, I'll always use 10 bit on the cameras I've tested.
-
I don't know if this is universal, or just the cameras I've tested, but I've found that recording 8 bit produces blocky color artifacts that are visible even without log recording or color grading. See my example in the other thread, and note that the comparison is with roughly equivalent bitrate (In this particular example, the 10 bit file ended up slightly smaller but within a couple %).
-
KnightsFan reacted to a post in a topic: 8bit? 420? If you record with a LUT baked in
-
When people talk about motion cadence, I think it's a conflation of many possible sources, where rolling shutter is just one piece and is often not the prominent one. The various motion problems I have identified are below. I don't see a lot of discussion on 4, 5, and 6 on film forums. I see stuff about 4 every now and then. Obvious Settings 1. Frame rate (24 vs 30 vs etc) 2. Shutter speed Sensor Tech 3. Rolling shutter 4. Weird artifacts almost like double exposures. My camera explicitly has a mode that captures at two shutter speeds simultaneously for higher dynamic range. I've seen artifacts like that on other cameras, but not advertised as a specialty mode. Display Issues 5. Display scan rate, frame rate, ghosting, trails. Some screens scan slowly, like rolling shutter on the display end. Others have ghosting effects, where the previous frame is still slightly visible, or trails. I see a lot more information about displays on video game tech sites, rather than film tech sites. Refresh rate, as mentioned earlier, includes pull downs or judder, to fit nondivisible integers. 6. Decode speed (laggy motion with H.265 on older computers, and stutter from high bitrate files on old mechanical drives).
-
Yup, it was always amusing when people watched 24p on a 60 fps monitor and claimed it looks better than 30p. Maybe some people like the effect of pulldowns or frame blending, but to me it's a strong BS indicator. I am fortunate enough to have a 120 fps monitor, which is a great number because it's divisible by 24, 30, and 60. It's great that high refresh rates on our screens are the norm now! I really dislike rolling shutter. It's one of my least favorite imperfections. I'm not saying "global shutter or bust" but the faster the better, and 10ms is around the cutoff where I'm happy. Quick controlled pans don't bother me so much as the vague wobble when it's on a steadicam, or handheld. My favorite movies to make have plenty of action, running, fighting, etc. so it's way more present to me than for most corporate or wedding shooters. I'd sacrifice a stop of DR and noise from modern full frame sensors to get rs in the 5 ms range instead of the 20ms range. The nice thing is that plenty of budget cinema cameras have fast readouts these days, like the UMP 4.6k G2, FX6, and of course Komodo.
-
KnightsFan reacted to a post in a topic: Motion Cadencemo
-
KnightsFan reacted to a post in a topic: Motion Cadencemo
-
KnightsFan reacted to a post in a topic: Let's bring back the good, old-fashioned camcorder of the 1990-2000s, but with modern specs.
-
KnightsFan reacted to a post in a topic: Let's bring back the good, old-fashioned camcorder of the 1990-2000s, but with modern specs.
-
PannySVHS reacted to a post in a topic: Let's bring back the good, old-fashioned camcorder of the 1990-2000s, but with modern specs.
-
John Matthews reacted to a post in a topic: Let's bring back the good, old-fashioned camcorder of the 1990-2000s, but with modern specs.
-
Theoretically I agree, but in practice no box cameras have built the required ecosystem of accessories to create compact camcorder ergonomics. The two missing pieces are are the side handle, and the monitor. There are plenty of "dumb" side handles, but to match camcorder ergonomics it needs lots of buttons. The FS7 handle with its multiple function buttons, joystick, etc. is a starting point. There are very few good monitor options under 5", and you need bulky batteries or lots of cables. By that time you've got a cinema rig. We could really do with some more open standards for camera controls, video, and audio. Lots of vendor lock these days in terms of accessories.