Despite that list, it actually works against much older devices. I've ran it with a T2i, for example, which is like three generations older than they officially support.
I'm only getting ~4 fps, with substantial latency. Is that your experience?
Image quality on the first lens I had at hand (40mm f/2.8) blows my Logitech 920 away. Field of view on the 920 is equivalent to a ~26mm on full-frame.
Super excited to see if my slightly-broken EOS M will do the trick, too.
Camera Live works with non-canon models as well, there is at least partial support for some Fuji and nikon in the alpha releases and it looks like the maintainer intends to increase support even further. The heavy lifting appears to be done in libgphoto2 which supports a ton of different models.
In terms of “why don’t you just use ____”, I have scoured the web for any way to use a Nikon as a webcam on macOS (over usb, so without dropping $120 on a camlink which is sold out everywhere now anyway), and the only other thing that exists is an app called ecamm live which costs $20/month forever. For windows and even Linux there are other more affordable or open source options.
> I have scoured the web for any way to use a Nikon as a webcam on macOS (over usb, so without dropping $120 on a camlink which is sold out everywhere now anyway)
The Elgato Cam Link is worth checking out if you're serious about this - it's an HDMI video capture device which presents itself on the computer end as a UVC compatible USB webcam and will accept the HDMI streams DSLRs usually send. They're expensive though.
There are other similar devices on the market too but not sure how good they are for this situation.
Elgato sells other capture cards (eg. HD60), which still seem available.
I recently bought a raspberry pi hi quality camera (and lens) to tinker with, and was able to connect it to my mac via pi->hdmi->capture-card to make a nice webcam view with the bokeh effect.
the total cost of this example does add up, but still cheaper than a DSLR setup.
capture cards are devices primarily intended for game streamers to give your computer an hdmi input to stream/record games from a game console, but are also useful for other purposes such as turning your dslr into an expensive webcam.
If you use the usb gadget mode of the raspberry pi to present the PI as a USB WebCam, you can forego the HDMI-connection, and thus the need for the HDMI capture card.
I've been trying to get a FPV drone camera feed into my Windows 10 computer so OpenCV can muck around with it.
OpenCV easily recognizes Webcams, but sucks at picking up 'video adapter' kinda stuff. (That is, I'm going to be using an RCA to USB dongle, such as EasyCap.)
If you can 'trick' OpenCV into thinking a webcam is attached at the USB, everything is golden.
Can you describe this raspberry pi method of 'presenting' the device as a USB WebCam? That sounds promising...
edit: Forgot to mention OpenCV is part of a c++ 64bit program I'm writing.
I used it for a few calls but the latency was too high for me personally. I work remotely so I'm interested in the DSLR DoF for video calls, but it seems the best is still a Capture card.
I use it with a T6s, along with CamTwist. It looks like it's bad latency and resolution initially but but you have to open the Preferences in CamTwist and increase FPS to 30 and size to 1920x1080.
Am I the only one with strong objections to using Zoom as a verb?
Putting aside the fact that it's already a verb, which courses plenty of confusion, I can go along with using Photoshop as a verb, because it is objectively the most powerful and widely used tool with a very long legacy.
But Zoom is a clunky insecure mess with impressive, but not nearly leading market share. Is this really who we want to be gifting the recognition (and money) that comes with generic verb for all videoconferencing? At least Adobe did something to earn their verb...
P.S.: I see no mention of Zoom in the page. Does this fall under the "don't mess with titles too much" rule of HN?
Actually, I totally agree with you. To Photoshop can't be that much misunderstood, but I think a lot of us read the heading as it used zoom in it's original meaning.
If it's already a verb, confusion should come from people using "Zoom" as a noun, the use of it as a verb is the expected situation not the confusing situation.
Either way, citation needed - who has been confused?
"zoom" is a verb (also sometimes a noun), "Zoom" (the company/product) is a noun.
I agree that "Zoom" as a noun is already confusing as capitalization isn't a very reliable way to differentiate words, but "Zoom" (the noun) being used as a verb that is completely different from "zoom" (the verb) is a whole other level of confusion on top of that.
For citations, see your sibling comments. And besides, confusion clearly wasn't my main point.
I was not yelling "hey everyone, stop using this word!", I was asking whether anyone else sees the same problems in its use as I do. And even if I were, telling people why you object to something in hopes they agree and maybe do something about it is literally how discourse works.
I webex with people. I have teams meetings. I zoom with people. I have hangouts. I'm meeting on discord. I have sob conferences (that's what we call old lync/skype for business calls).
I don't care what verb you use as long as the link is correct.
If we couldn't stop people from literally using "literally" figuratively, what makes you so optimistic about stopping them from zooming?
> telling people why you object to something in hopes they agree and maybe do something about it
It's fine to object, but the "do something about it" part is not as simple – it's really hard to police discourse. Previous attempts for the most part have either completely failed or catastrophically backfired
Yeah, I couldn't find a better way to say "do something about it". I meant in the sense of "don't use it themselves and maybe occasionally point out the problem". Kind of like Stallman's infamous GNU/Linux copypasta. I have no intention to police anyone, but I see no reason to not object. At least then people know that not literally everyone uses it - although they will point out that "literally" everyone does.
I was just waiting for someone to go for the "there's more important things going on" bullshit.
Did I ever say this was the most important thing in the world? Did I even say it's important? Even slightly? You're acting like I'm asking healthcare workers to drop what they're doing and become the language police.
I shared an opinion. How dare I?!?
Give me a break...
>Did I ever say this was the most important thing in the world? Did I even say it's important? Even slightly?
You don't have to say it explicitly, there's this thing called context. If you ask people to worry about it and maybe "do something", then you give it importance.
I'm not saying something has to be the most important thing in the world to care about it. Plenty of things that are not "the most important things in the world" are still worthy caring about. Heck, most things we have to care about (like our rent), are not "the most important things in the world" either.
What I'm saying is that this particular thing is so unimportant that we shouldn't care about it.
>I shared an opinion. How dare I?!? Give me a break...
>And even if I were, telling people why you object to something in hopes they agree and maybe do something about it is literally how discourse works.
Yes, but not how language works (same way it's not how the laws of physics work).
Language doesn't develop through discourse, but through actual discussions people have (not meta discussions, about what words are right, just plain discussions about whatever).
> Language doesn't develop through discourse, but through actual discussions people have
True, language is mostly shaped by people speaking it, but the way those people speak is also influenced by their opinions and the way people around them speak. The fewer people use the term, the fewer new people will start. It also provides some pressure to not use it because someone they know doesn't.
For an example of the latter:
If a term is offensive to a certain group, they will tell people about it. Some of those people might stop using it as a result. The people around them might never even be exposed to the original term, and if they are, would be told that some consider it offensive and they might stop using it too.
(for practical examples, see racial slurs, curse words, etc.)
I'm genuinely surprised how many people had a problem with my comment, as I've seen plenty of similar discussions around "GNU/Linux", "Googling", "Windows key", "sending a Gmail"...
I think you have to accept it. Zoom was in the right place at the right time. Now it's a verb. The good news is that once a brand becomes a word, it doesn't really mean much about the company itself. I blow my nose into CVS brand Kleenex. I have a Xerox machine made by Canon. I've captured many a "Kodak moment" with my iPhone's camera. Windows users routinely "Google it" by asking Cortana, which actually uses Bing. And I Zoom with my friends on Discord.
All in all, the COVID-19 pandemic has created a lot of strange victories. Would Tiger King, Animal Crossing, and Zoom have been popular if they rose to prominence at a different time? Probably not. I think the pandemic forced niche things onto people that don't normally encounter them. Adults have video conferences at work, but the pandemic brought them to kids. That is a recipe for a cultural moment and genericizing their trademark. (Animal Crossing I can't explain. I assume that the people that were going to play it were already going to play it, and I assume that the people that were going to reverse engineer the code to figure out how turnips were priced were going to do that anyway. But maybe not.)
You are right, Zoom was a right place right time thing. Doesn’t bother me other people saying it, but I’ve never used Zoom once, nor do I have any plans to, so I don’t see it making it into my vocabulary.
I don’t get it either, but Animal crossing was already one of the most beloved and best selling franchises in Nintendo history, so the pandemic didn’t really do much there, except give people more time at home to play and crack the code faster!
title is 200% wrong: its not zoom inc. Also, only works with Canon.
DSLR are a shitty closed market that prey on the few professionals that still need it. It's like the printer ink cartridge all over again.
Canon had the fortune of being "hacked" (to use proper legal terms after DMCA) and their cameras sales soared because of the free publicity and improved product. Now they are the only ones that maintain a modicum of interoperability that make projects like these a little easier (but still far from decent or stable). From other "DSLR but for now only cannon" projects: this will never support anything else.
Until recently "to Skype" was the generic term for video calling among most of my friends and family. Given that nobody has actually used Skype for that purpose for 5 years, this crisis has accelerated it's decline and Zoom/WhatsApp/facetime is creeping into our vernacular.
So one crappy clunky video conferencing verb being replaced by another.
"Teams-meeting" has become the norm for video-conference. One of my annoying pet peeves is pointing this out and calling it "meeting on Teams" - if that's where it is.
Since I use up to 5 different platforms each day...
“Let’s zoom” means just that; let’s get on Zoom and have a meeting. I’ve never had somebody say, “Let’s Zoom,” and contact me on FaceTime, Google Meet, WebEx, or another service.
It works(ish) but worked better before most of the major players locked virtual cameras out with updates. Issues with unsigned code, so then some folks started unsigning apps smh. Still works with Chrome based version of several tools, and it’s cool when it works, but definitely a bit on the fragile side. The Canon software has the same limitations and just lets you use software from a perhaps more trusted source depending on your perspective…
I just discovered this some days ago through a blog (https://www.crackedthecode.co/how-to-use-your-dslr-as-a-webc...) and it worked out quite well. However I did not yet find a way to control the depth of field through aperture in live view because it's always fully open.
Please take care when using your DSLRs for video. Often the sensors are not adequately cooled for long term usage as a video camera (>30m). You have the potential to damage the sensor if it is not designed for continuous use and doesn't have proper thermal protection. Newer cameras will shut down if they get too hot, or have limits to how long you can run in video mode.
Even the stackoverflow answers say as much. Most of the thermal ramping occurs in the first five minutes and then stabilizes after cache is exhausted. As all these dslr’s allow you to record 29m50s videos back to back all day long it’s also indicative that it’s not a big enough issue in returns to warrant addressing in their firmware.
As you’re not writing to the SD card, hdmi passthrough should yield an overall lower operating temperature after an hour than writing to the SD card when recording video after even 10 minutes.
I tried getting to the source of this rumor. Didn’t find one particular origin, just speculation amongst Redditors and Twitch users about the 30 minute limit and bypassing it with the Nikon or Canon firmware mods will void your warranty.
It will however introduce more noise. Particularly at higher ISO’s. Astronomy photographers (a major driver of the firmware hacks in the first place - see “Nikon stareater bypass”) have cooling solutions available there too.
> At present [2012], digital cameras' video cuts off after 30 minutes to avoid them being classified as video cameras (which attract 5.4% duty because they are considered to be video recorders
Economic Partnership Agreement between Japan and EU had signed that abolishes the video camera tax.
Most camera brands are in Japan but I don't know how many cameras made in Japan.
This Japanese article describing the agreement especially for cameras. I can't find English articles like this.
At the end of the day, it's just a calculated risk and this is one opinion on one side. My DSLR was quite expensive and I love using it for photography, so if there is any risk of damaging the sensor I will not be taking it. YMMV.
>..if there is any risk of damaging the sensor I will not be taking it. YMMV.
curious.
what would it take for you to be convinced it does no harm? Just a psychological curiosity -- not trying to annoy you.
The opinions seem fairly stacked, and the corporations seem to be fairly non-chalantly enabling video/high-volume sampling on these sensors.
The only person in your linked SO thread said "it might wear out the sensor because it wasn't designed for that amount of sampling", so one google's 'Can you wear out a CMOS image sensor from over-use?', and get linked to another SO thread where it's more-or-less concluded that the CMOS sensor doesn't wear practically.[0]
I can understand the 'calculated risk' mentality, but honestly it seems like physical jostling from unpacking the equipment from its' case or exposure to a mid-high level humidity environment, stuff that's done every day by photographers around the world, would be many times more damaging to the investment that just running the solid-state components more often.
Not many DSLR appliance failures are from solid-state device failures.
Not annoyed at all! I enjoy chatting about these things.
In my dream scenario? I think I'd love to see a scientific analysis on the topic and some real data. A blog post with someone taking apart a DSLR, instrumenting the sensor and the surrounding electronics with a thermocouple to measure how hot it really gets over time, and proving it doesn't get to a point of damage would go a long way for me, and I would generally be interested in it. How hot does it actually get??
The wishy-washy "wear out" arguments I don't find particularly convincing (I did FPGA design for an endoscopic camera company for a couple years, so I am familiar with CMOS sensors in general). These DSLR sensors are quite large and I imagine can produce a significant amount of heat, and for all intents and purposes are built into systems that are intended for periodic use with photography. The manufacturing tolerances needed to build a system safe for continuous sampling/use don't particularly make sense when designing a photography camera ($$$). I'm dubious the manufacturers actually did the design optimizing for video use, and I presume this is why there are warnings in the manuals about continuous video use and limitations in the firmware. Maybe there's enough tolerance to handle some video, but not all the time.
I think I was a little careless in specifically calling out "sensor damage" — I'd actually be more concerned about the surrounding components than the sensor itself. Capacitors age and change tolerance in response to temperature cycling, for example.
I mean, look, if we're going to quantify these things, you stand a greater chance of damaging the sensor by using the camera than not. It's a tool. It's a toy. Just like action figures, they're more fun out of the box.
You can see a sharp rise in the first 15 minutes and the camera asymptotically approach 43ºC after an hour.
Nikons are rated for 0-40ºC so there is some truth to them not wanting to warranty cameras used for more than half an hour.
But 43º??? Come on. These components are easily rated for 60º.
Anyone that's done any weather testing on consumer electronics can tell that it's the process of thermally cycling the DUT that will kill you (i.e. going from 20º to 30º hundreds of times, not holding it at 40º).
If you are worried about the heat, I may recommend you mill off the CCD and install one of these bad boys: http://instein.eu/d3000.htm
But of a old thread, but all Nikon's that can take video have thermal sensors and cut off when reaching "too hot". There have been people note in hot locations like Australia that they cannot record for 10 minutes before the camera shut off. So Nikon's at least are "safe to use".
I have often left my D750 running for 8+ by mistake.. it's happy as.
It is the actual recording mode that generates the heat. Transcoding, putting stuff in the buffer, saving to storage, etc. With the HDMI-out, you are just sharing the viewfinder and offloading all those tasks to your computer, so heat is not really an issue unless your room is extremely hot to begin with.
The sony a5100 is really popular with twitch streamers right now because it will run all day in this mode and you can turn off all of the UI elements to get a 'clean' output. You don't get all the cool cinema features but you do get a camera that is better than just about any webcam.
For folks with Nikon DSLRs and Windows, there's an app called SparkoCam which uses the your camera as a video feed source and appears to the host operating system as though it were a webcam. Works for Canon as well though Canon and at resolutions higher than are currently supported by Canon's "use your camera as a webcam" software.
The best results require the use of an HDMI capture device... and all of the good ones sold out almost immediately when the quarantine started.
Switching to a "real" camera has been such a huge, huge improvement for me on video calls all day, I machined a custom vesa mounting arm to hold my camera in just the right spot.
In a word: bokeh. I'm running a 50mm f/1.8 lens on a full-frame sensor. My job requires me to interact with external customers, sometimes in a pre-sales role. Appearances shouldn't matter, but they do. Anything I can use to present a more professional and "expert" image is a win for me.
105 comments
[ 4.6 ms ] story [ 176 ms ] threadhttps://www.usa.canon.com/internet/portal/us/home/about/news...
Does anyone know of a similar linux app that works over USB rather than HDMI?
here is the command
I did wrap in a shell script to test if the module is already loaded.I'm only getting ~4 fps, with substantial latency. Is that your experience?
Image quality on the first lens I had at hand (40mm f/2.8) blows my Logitech 920 away. Field of view on the 920 is equivalent to a ~26mm on full-frame.
Super excited to see if my slightly-broken EOS M will do the trick, too.
In terms of “why don’t you just use ____”, I have scoured the web for any way to use a Nikon as a webcam on macOS (over usb, so without dropping $120 on a camlink which is sold out everywhere now anyway), and the only other thing that exists is an app called ecamm live which costs $20/month forever. For windows and even Linux there are other more affordable or open source options.
check out https://news.ycombinator.com/item?id=23409540
I misplaced the usb cable that came with my nikon d5000 but I was able to find a replacement on amazon yesterday that should arrive later today.
https://fujifilm-x.com/en-us/webcam-support/
First downer: My X-T20 is not supported :(
Second downer: Windows only?
So sad right now :(
I bought a elgato camlink 4K just recently because I wasn’t so happy with Camera Live but now I wish I hadn’t.
This sounds like a bummer. I want to try this anyways to give my old Powershot a new purpose.
Did anybody check this out yet?
There are other similar devices on the market too but not sure how good they are for this situation.
I’ve entered that phase in my cycle where I’d rather buy separate pieces that are good quality and put them together.
I recently bought a raspberry pi hi quality camera (and lens) to tinker with, and was able to connect it to my mac via pi->hdmi->capture-card to make a nice webcam view with the bokeh effect.
the total cost of this example does add up, but still cheaper than a DSLR setup.
I’m interested!
i have this one: https://www.elgato.com/en/gaming/game-capture-hd60-s-plus
you can find others from magewell
i don't have measurements but the latency was indistinguishable from a directly connected webcam.
OpenCV easily recognizes Webcams, but sucks at picking up 'video adapter' kinda stuff. (That is, I'm going to be using an RCA to USB dongle, such as EasyCap.)
If you can 'trick' OpenCV into thinking a webcam is attached at the USB, everything is golden.
Can you describe this raspberry pi method of 'presenting' the device as a USB WebCam? That sounds promising...
edit: Forgot to mention OpenCV is part of a c++ 64bit program I'm writing.
http://www.davidhunt.ie/raspberry-pi-zero-with-pi-camera-as-...
https://www.raspberrypi.org/forums/viewtopic.php?t=148361
Putting aside the fact that it's already a verb, which courses plenty of confusion, I can go along with using Photoshop as a verb, because it is objectively the most powerful and widely used tool with a very long legacy.
But Zoom is a clunky insecure mess with impressive, but not nearly leading market share. Is this really who we want to be gifting the recognition (and money) that comes with generic verb for all videoconferencing? At least Adobe did something to earn their verb...
P.S.: I see no mention of Zoom in the page. Does this fall under the "don't mess with titles too much" rule of HN?
/s
> Putting aside the fact that it's already a verb, which courses plenty of confusion
Actually, I totally agree with you. To Photoshop can't be that much misunderstood, but I think a lot of us read the heading as it used zoom in it's original meaning.
Either way, citation needed - who has been confused?
I agree that "Zoom" as a noun is already confusing as capitalization isn't a very reliable way to differentiate words, but "Zoom" (the noun) being used as a verb that is completely different from "zoom" (the verb) is a whole other level of confusion on top of that.
For citations, see your sibling comments. And besides, confusion clearly wasn't my main point.
People will use it or not use it regardless of objections...
I was not yelling "hey everyone, stop using this word!", I was asking whether anyone else sees the same problems in its use as I do. And even if I were, telling people why you object to something in hopes they agree and maybe do something about it is literally how discourse works.
I don't care what verb you use as long as the link is correct.
> telling people why you object to something in hopes they agree and maybe do something about it
It's fine to object, but the "do something about it" part is not as simple – it's really hard to police discourse. Previous attempts for the most part have either completely failed or catastrophically backfired
The "problem" of using Zoom as a verb for teleconference?
In 2020, you really want people to concern with not "using it themselves" and occasionally "pointing out" this "problem"?
Did I ever say this was the most important thing in the world? Did I even say it's important? Even slightly? You're acting like I'm asking healthcare workers to drop what they're doing and become the language police.
I shared an opinion. How dare I?!? Give me a break...
You don't have to say it explicitly, there's this thing called context. If you ask people to worry about it and maybe "do something", then you give it importance.
I'm not saying something has to be the most important thing in the world to care about it. Plenty of things that are not "the most important things in the world" are still worthy caring about. Heck, most things we have to care about (like our rent), are not "the most important things in the world" either.
What I'm saying is that this particular thing is so unimportant that we shouldn't care about it.
>I shared an opinion. How dare I?!? Give me a break...
And I shared mine on that first opinion.
Yes, but not how language works (same way it's not how the laws of physics work).
Language doesn't develop through discourse, but through actual discussions people have (not meta discussions, about what words are right, just plain discussions about whatever).
True, language is mostly shaped by people speaking it, but the way those people speak is also influenced by their opinions and the way people around them speak. The fewer people use the term, the fewer new people will start. It also provides some pressure to not use it because someone they know doesn't.
For an example of the latter: If a term is offensive to a certain group, they will tell people about it. Some of those people might stop using it as a result. The people around them might never even be exposed to the original term, and if they are, would be told that some consider it offensive and they might stop using it too. (for practical examples, see racial slurs, curse words, etc.)
I'm genuinely surprised how many people had a problem with my comment, as I've seen plenty of similar discussions around "GNU/Linux", "Googling", "Windows key", "sending a Gmail"...
All in all, the COVID-19 pandemic has created a lot of strange victories. Would Tiger King, Animal Crossing, and Zoom have been popular if they rose to prominence at a different time? Probably not. I think the pandemic forced niche things onto people that don't normally encounter them. Adults have video conferences at work, but the pandemic brought them to kids. That is a recipe for a cultural moment and genericizing their trademark. (Animal Crossing I can't explain. I assume that the people that were going to play it were already going to play it, and I assume that the people that were going to reverse engineer the code to figure out how turnips were priced were going to do that anyway. But maybe not.)
I don’t get it either, but Animal crossing was already one of the most beloved and best selling franchises in Nintendo history, so the pandemic didn’t really do much there, except give people more time at home to play and crack the code faster!
DSLR are a shitty closed market that prey on the few professionals that still need it. It's like the printer ink cartridge all over again.
Canon had the fortune of being "hacked" (to use proper legal terms after DMCA) and their cameras sales soared because of the free publicity and improved product. Now they are the only ones that maintain a modicum of interoperability that make projects like these a little easier (but still far from decent or stable). From other "DSLR but for now only cannon" projects: this will never support anything else.
So one crappy clunky video conferencing verb being replaced by another.
Since I use up to 5 different platforms each day...
I use my Sony a7r II as a webcam on Linux using gphoto2 with v4l2loopback:
https://daniel.lawrence.lu/blog/y2020m03d31/
According to following works also on certain GoPro models: https://www.knowhero.com/guides/goprowebcam
Stackoverflow discussion: https://video.stackexchange.com/questions/21932/can-continuo....
Even the stackoverflow answers say as much. Most of the thermal ramping occurs in the first five minutes and then stabilizes after cache is exhausted. As all these dslr’s allow you to record 29m50s videos back to back all day long it’s also indicative that it’s not a big enough issue in returns to warrant addressing in their firmware.
As you’re not writing to the SD card, hdmi passthrough should yield an overall lower operating temperature after an hour than writing to the SD card when recording video after even 10 minutes.
I tried getting to the source of this rumor. Didn’t find one particular origin, just speculation amongst Redditors and Twitch users about the 30 minute limit and bypassing it with the Nikon or Canon firmware mods will void your warranty.
It will however introduce more noise. Particularly at higher ISO’s. Astronomy photographers (a major driver of the firmware hacks in the first place - see “Nikon stareater bypass”) have cooling solutions available there too.
Edit: Yup. https://petapixel.com/2012/05/21/30-minute-video-limit-in-di...
> At present [2012], digital cameras' video cuts off after 30 minutes to avoid them being classified as video cameras (which attract 5.4% duty because they are considered to be video recorders
EDIT: I had it backwards. The tax is on trucks and it gets circumvented by importing them as passenger cars: https://en.wikipedia.org/wiki/Chicken_tax
This Japanese article describing the agreement especially for cameras. I can't find English articles like this.
https://av.watch.impress.co.jp/docs/series/zooma/1171774.htm...
This is the original agreement text that's too long didn't read.
https://www.mofa.go.jp/ecm/ie/page4e_000875.html
curious.
what would it take for you to be convinced it does no harm? Just a psychological curiosity -- not trying to annoy you.
The opinions seem fairly stacked, and the corporations seem to be fairly non-chalantly enabling video/high-volume sampling on these sensors.
The only person in your linked SO thread said "it might wear out the sensor because it wasn't designed for that amount of sampling", so one google's 'Can you wear out a CMOS image sensor from over-use?', and get linked to another SO thread where it's more-or-less concluded that the CMOS sensor doesn't wear practically.[0]
I can understand the 'calculated risk' mentality, but honestly it seems like physical jostling from unpacking the equipment from its' case or exposure to a mid-high level humidity environment, stuff that's done every day by photographers around the world, would be many times more damaging to the investment that just running the solid-state components more often.
Not many DSLR appliance failures are from solid-state device failures.
[0] : https://photo.stackexchange.com/questions/15127/do-sensors-w...
In my dream scenario? I think I'd love to see a scientific analysis on the topic and some real data. A blog post with someone taking apart a DSLR, instrumenting the sensor and the surrounding electronics with a thermocouple to measure how hot it really gets over time, and proving it doesn't get to a point of damage would go a long way for me, and I would generally be interested in it. How hot does it actually get??
The wishy-washy "wear out" arguments I don't find particularly convincing (I did FPGA design for an endoscopic camera company for a couple years, so I am familiar with CMOS sensors in general). These DSLR sensors are quite large and I imagine can produce a significant amount of heat, and for all intents and purposes are built into systems that are intended for periodic use with photography. The manufacturing tolerances needed to build a system safe for continuous sampling/use don't particularly make sense when designing a photography camera ($$$). I'm dubious the manufacturers actually did the design optimizing for video use, and I presume this is why there are warnings in the manuals about continuous video use and limitations in the firmware. Maybe there's enough tolerance to handle some video, but not all the time.
I think I was a little careless in specifically calling out "sensor damage" — I'd actually be more concerned about the surrounding components than the sensor itself. Capacitors age and change tolerance in response to temperature cycling, for example.
Here's a study from 2014: https://www.lmscope.com/en/warming_camera_sensor_en.html
You can see a sharp rise in the first 15 minutes and the camera asymptotically approach 43ºC after an hour.
Nikons are rated for 0-40ºC so there is some truth to them not wanting to warranty cameras used for more than half an hour.
But 43º??? Come on. These components are easily rated for 60º.
Anyone that's done any weather testing on consumer electronics can tell that it's the process of thermally cycling the DUT that will kill you (i.e. going from 20º to 30º hundreds of times, not holding it at 40º).
If you are worried about the heat, I may recommend you mill off the CCD and install one of these bad boys: http://instein.eu/d3000.htm
I have often left my D750 running for 8+ by mistake.. it's happy as.
The sony a5100 is really popular with twitch streamers right now because it will run all day in this mode and you can turn off all of the UI elements to get a 'clean' output. You don't get all the cool cinema features but you do get a camera that is better than just about any webcam.
The best results require the use of an HDMI capture device... and all of the good ones sold out almost immediately when the quarantine started.
https://cascable.se/pro-webcam/
Another option is Ecamm Live: https://www.ecamm.com/
Camera Live does work with other cameras. Nikon is coming in the next release. See github issues.
https://lightroom.adobe.com/shares/088254b7439745b982784c0a4...