Are DSLRs doomed? Tony Northrup says no.

[Ed_Ingold]

With a MILC, image quality degradation will start right away and get worse over time as the sensor heats up, before a possible cut-off to protect the sensor and electronics.

You have provided no additional information to support this opinion? In fact, your "conclusions" are contrary to my experience and nobody else on this forum, besides frans_waterlander, have suggested any problem. You, in fact, have no personal experience, and are simply repeating what you have read or surmised from internet "experts."

 

As I have noted, I frequently shoot video continuously for two hours or longer at a stretch, and there is no degradation in image quality from beginning to end. The entire sensor is active, as well as the rear LCD, for the entire duration, processing and transmitting data continuously.

 

There are many reasons one would choose an MILC over a DSLR, and vice versa. Sensor heating is not among them.

 

The only documented issues regarding thermal noise pertains to astrophotography, which can involve exposure times from a few seconds to several hours. Thermal noise is additive (root-sum-squares), which limits the dimmest object which can be recorded. Cooling is often used to extend this limit to a few (e.g., 50) photons/second. That's 1,000,000 times less than the dimmest object you can see with your eye on a dark night.

Edited June 24, 2019 by Ed_Ingold

[frans_waterlander]

You have provided no additional information to support this opinion

With MILCs, the sensor and other electronics are constantly active, going through expose/read/display anywhere from 30 to 100 times (and maybe more) per second. With DSLRs in OVF mode, there is only action when you take a shot, say once or twice a minute. In this scenario the energy generated in the MILC sensor is best case 900 times larger than for the DSLR and worst case 6,000 times larger; as a result the sensor temperature increase in MILCs is 900 to 6,000 times bigger than in DSLRs, all else like cooling being equal.

 

So where is the hard data obtained through careful testing or are we not supposed to ask these kind of questions?

[Nick D.]

With MILCs, the sensor and other electronics are constantly active, going through expose/read/display anywhere from 30 to 100 times (and maybe more) per second. With DSLRs in OVF mode, there is only action when you take a shot, say once or twice a minute. In this scenario the energy generated in the MILC sensor is best case 900 times larger than for the DSLR and worst case 6,000 times larger; as a result the sensor temperature increase in MILCs is 900 to 6,000 times bigger than in DSLRs, all else like cooling being equal.

 

So where is the hard data obtained through careful testing or are we not supposed to ask these kind of questions?

Just buy whatever you comfortable with, as I pointed before, B & H still have large format cameras available brand new.

[andylynn]

With MILCs, the sensor and other electronics are constantly active, going through expose/read/display anywhere from 30 to 100 times (and maybe more) per second. With DSLRs in OVF mode, there is only action when you take a shot, say once or twice a minute. In this scenario the energy generated in the MILC sensor is best case 900 times larger than for the DSLR and worst case 6,000 times larger; as a result the sensor temperature increase in MILCs is 900 to 6,000 times bigger than in DSLRs, all else like cooling being equal.

 

So where is the hard data obtained through careful testing or are we not supposed to ask these kind of questions?

 

You’ve asked your question more than once and you have your answer. Between Ed and I, we have at least 15 years experience with the things and no instances of images ruined (or even degraded in any noticeable way) by sensor overheating. You can’t point to any examples of this actually happening. It’s not an actual problem.

 

If MILC sensors do heat up (which is not in evidence) and if heat does degrade image quality (which is not in evidence) then it’s not an amount of degradation that’s visible, so who on earth cares? It’s a photograph. It’s meant for looking at.

[Uhooru]

With MILCs, the sensor and other electronics are constantly active, going through expose/read/display anywhere from 30 to 100 times (and maybe more) per second. With DSLRs in OVF mode, there is only action when you take a shot, say once or twice a minute. In this scenario the energy generated in the MILC sensor is best case 900 times larger than for the DSLR and worst case 6,000 times larger; as a result the sensor temperature increase in MILCs is 900 to 6,000 times bigger than in DSLRs, all else like cooling being equal.

 

So where is the hard data obtained through careful testing or are we not supposed to ask these kind of questions?

Before you can prove heat issues do not happen, there needs to be concrete evidence that the the problem occurs in the first place. You haven't done that. Where is you evidence, besides your unsupported assertions, that the problem even occurs in the first place. It seems no one here has experienced the problem, including yourself. Why should anyone believe you then?

[frans_waterlander]

No, I personally have not experienced the problem since I don't have a MILC or DSLR with Live View. You want concrete evidence? How about these:

Mirrorless cameras and sensors - the good, the bad and the really ugly - PentaxForums.com

A7R III Overheat Issue: Sony Alpha Full Frame E-mount Talk Forum: Digital Photography Review

SONY a7 III a73 OVERHEATING issue: Sony Alpha Full Frame E-mount Talk Forum: Digital Photography Review

Image noise in DSLRs and mirrorless cameras...: Photographic Science and Technology Forum: Digital Photography Review

And what about warnings from manufacturers I quoted earlier:

Automatic shutdown to prevent damage to the camera; overheating - caused by high ambient temperatures, extended periods of using Live View, movie recording, etc.- resulting in fog, bright spots, deteriorated image quality, irregular colors.

Get your collective heads out of the sand and start applying some critical thinking! Ask DxOmark and others to do measurements under those conditions so we understand exactly what we are dealing with! Don't accept marketing hype like the gospel!

[andylynn]

Okay, so a Pentax DSLR can produce noise when trying to record successive 30s exposures (though whether it produces a visible real world effect of that is unclear - the comparison is black frames with a 4-stop push, and I know plenty of people who use their digital cameras to take multiple long exposures and don’t have complaints) and a Sony can overheat when shooting 4K video (no surprise there, DSLRs do it too) and an X-T1 does... something involving a histogram, what the heck are we looking at there and what does it do to the photos? And Sony over-smooths noise. And then we have some speculation about MILCs overheating. But these links don’t make the case. The closest one to addressing the subject is the one with the X-T1 histograms but it’s not clear what the effect of that is on photos, and it doesn’t compare that to a DSLR under the same workload to see if that’s a MILC effect or a digital effect in general.

[Ed_Ingold]

I reviewed the reference supplied above, discussions (not articles) in DPReview. On two or three occasions, someone started a thread claiming overheating, with negative effects on image quality or leading to automatic thermal shutdown. While the threads were several pages long, there were no "me too" responses, and many to the effect of "not me". There was a claim that the lens itself got hot, which seems highly unlikely under even strenuous conditions, operation unless heated by the sun.

 

The subject of Sony overheating comes up regularly in DPR forums, mostly from Canikon trolls, and always from third-hand information.

 

The Sony manual is vague, advising against operation in high ambient temperatures, or storing in a hot vehicle. There is a temperature limit in the menu, which can be set to "Standard" or "High." Even my RX0 (I said MX0 before, in error) does not shut down if set to "High", but it does get quite warm on the exterior.

 

CMOS devices have extremely high impedance, hence any current flow is minimal, consequently little heat is generated. I strongly suspect most of the heating comes from the battery, memory storage, and the rear display. In my experience, only the grip, which contains the battery and memory cards, gets warm, as well as the rear display. Displays, which consume energy to produce light, are know to get hot, as witnessed by LCD computer displays. The 1000 NIT display in my Atomos Ninja V is fan cooled, and the air discharge is uncomfortably hot.

 

With batteries, the internal resistance increases with temperature, which causes more heating at the same load. I suspect that thermal limits are in part to prevent runaway battery heating.

 

While the instruction manuals do not say sensor heating is at fault, they do cite time limits for recording in juxtaposition to temperature limits. The 30 minute limit (actually 29:59) was falsely attributed to overheating in the popular media (e.g., forums). In fact, it is imposed to avoid a 15% tariff imposed on video cameras, defined as continuous shooting 30 minutes or more.

 

My conclusions are baed on actual experience. I shot 4K30p video (100 MB/s), with an A7iii, continuously for over 4 hours last Saturday, outside in 80 degree heat. I used an external recorder and camera battery. There was no overheating, in fact the camera was barely warm to the touch, at the end of the day. At the same time, I took 459 still images with an A7Riii camera, which didn't get hot either, on a single battery which had 60% of its charge remaining.

[frans_waterlander]

It looks like we are repeating ourselves without making much if any progress. I have no dog in this fight other than to find out if mirrorless is a candidate to replace my aging DSLR. Reported issues and manufacturers' warnings make me very uncomfortable, if not down-right skeptical. I'd love to see test results at elevated ambient temperatures and extended operations with Live View, but so far I haven't found those and apparently the people that have read this thread have neither. The people that promote mirrorless should be able to allay my concerns but so far no takers. DxOmark and others I have contacted agree that image quality degradation occurs under those conditions, but say they haven't tested for it and are not planning to do so. Up the proverbial creek.

[Ed_Ingold]

You can't prove a negative. No matter how many times a process works, there's always the chance it won't. Negative evidence is an oxymoron. (I guess that applies to politicians too.)

 

There are many studies regarding the effect of temperature on CMOS (and other) imaging sensors. Short of structural damage, the main effect is on dark noise, which roughly doubles for every 8 degrees C increased temperature. These tests are performed at a controlled ambient temperature. There is nothing to suggest the sensor is subject to self-heating.

 

You should take precautions to avoid concensation when moving the camera between cold and warm areas. That's true for MILCs and DSLRs alike, and their lenses.

 

You may prefer an optical finder to an electronic one, and a MILC may be too small for some hands. To get the best performance, you may wish to buy native lenses to replace others in your possession. If you continue to obsess about sensor overheating, you should stick with DSLRs. If fact, buy two, in case you need one for parts later on. As far as overheating goes, nobody else seems to care.

Edited June 26, 2019 by Ed_Ingold

[andylynn]

So this whole silly line of thought comes down to a health and safety label (if we believe those, let me tell you about a Tamron lens that can give you cancer if you inhale it), some internet trolls and some employee at DXO who wrote Frans an email in which he said he has not done any tests. (So did he divine the knowledge? Do DXO employees have supernatural powers?)

 

Even by the exceptionally low standards of fanboy posts on web sites the information here is of very low value. I’m calling this myth busted. MILCs have been around for 10 years and have millions of users, and the fact that between us Frans and I can’t uncover one single image that has been ruined by the supposed MILC sensor overheat problem is clear and convincing evidence that no such problem exists.

[Ed_Ingold]

I see that Sony issued a firmware update for the A7Rii, which adds a "standard/high" option for temperature cutoff. Nothing about the camera is changed except for the menu item. The update notes repeat the warning that the camera can cause "low temperature burns" from prolonged contact with the skin. This is an ultra-conservative, statement from company lawyers. Low temperature burns are real, and be caused by something barely warm to the touch, but above skin temperature (typically 86 deg F). A cell phone in a hip pocket might suffice.

 

You are also cautioned to avoid leaving the camera, exposed to the sun or in a hot car. A hot car can easily reach 160 deg F, which could burn someone breaking in and stealing your camera. Since you have been duly warned (in a lawyerly fashion), you could be sued for damages.

 

I shall be careful to keep my fingers away from the exposed sensor when I change lenses.

[digitaldog]

You have provided no additional information to support this opinion? In fact, your "conclusions" are contrary to my experience and nobody else on this forum, besides frans_waterlander, have suggested any problem. You, in fact, have no personal experience, and are simply repeating what you have read or surmised from internet "experts."

And worse, he had to ASK those experts back in March of this year a question specifically about this subject, then after asking dismissed, as we've seen here, what the experts and others told him. This is nothing new in his posting agenda: ask then argue. Here's that March question from him, to experts and others, about with his question and premeditated answer before asking:

How much of an issue is Live View sensor heating, noise increase, dynamic range degradation?

 

[frans_waterlander]

You can't prove a negative.

You can prove whether or not the sensor heats up and whether or not it degrades image quality. All it takes is for people like DxOmark, photonstophotos, etc. to run some tests. Here's a simple scenario:

1) Keep a MILC switched off at room temp for enough time to get thermal equilibrium; switch on and measure dynamic range and sensor temperature right away.

2) Keep camera on in photo mode (Live View is constantly doing it's thing) and measure dynamic range and sensor temperature at e.g. 30, 60, 90, 120 minutes

3) Repeat the process with the camera in video mode

4) Repeat at elevated ambient temperatures

That will tell you if the sensor heats up and by how much and if dynamic range deteriorates and by how much.

I find it highly ironic that camera manufacturers warn about sensor heating and image quality degradation, but they and test houses seem to take the fifth (refuse to provide incriminate information).

[Allen Herbert]

For myself I've never had a problem with over heating on my various mirroless cameras. I'm about to purchase a Sony A7r11 ;is over heating going to be a problem for me? I take a lot of photos.

 

Just as a footnote.

 

Obviously it will be inferior to the latest A7 incarnations but how inferior? should I dig deep and buy the latest? landscapes/street is what I mostly photograph.

 

Thank you for any replies.

[digitaldog]

For myself I've never had a problem with over heating on my various mirroless cameras. I'm about to purchase a Sony A7r11, is over heating going to be a problem for me?

As yet, there's hardly any actual data to suggest let alone prove it's a problem. But the solution for those who wish to be diligent consumers is to test this out. Which is why a VAR is so useful and why they may charge a bit more for the same product but it's going to work out so much better in the end. Assuming you have a good VAR in your area, they should absolutely allow you to examine and test that Sony in conditions you need to do a good test. After which, if you decided to buy the product, it would be useful to support that VAR so they are around in the future to support you and other photographers. I say this having some experience AS a VAR and consider myself as a consumer willing to do due diligence when making such purchases. Not everyone of course can take the time to actually test equipment they hope to buy, easier to assume how it will work (or fail for them), then complain about the product.;)

[Ed_Ingold]

Obviously it will be inferior to the latest A7 incarnations but how inferior? should I dig deep and buy the latest? landscapes/street is what I mostly photograph.

I never had a temperature problem with my A7Rii in four years, tens of thousands of images, and many hours of video. I recommend at least two spare batteries, however. They're good for 2.5 hours or 300 images, which ever comes first. It locked up once or twice for unknown reasons. I had to remove the battery to even shut down, but it was okay once power was restored.

[Ed_Ingold]

I passed my A7Rii to my middle son after buying the latest version, A7Riii. There were many improvements in the model 3, but these were of particular significanct to me...

• Double the battery capacity, and 40% of the power requirements - a net gain of 3x the life per charge
  
• High resolution EVF, making it possible to focus directly without peaking or focus magnification
  
• Buffer capacity of at least 40 uncompressed RAW images (instead of abut 8)
  
• Dual Memory Cards. With latest firmware, select - dual storage, sequential storage, or split between RAW and JPEG or RAW and Video (my preference)
  

The M3 AF performance was very good, but improved even more with firmware v3.01, nearly to A9 speed and accuracy. The sensor, resolution and image quality are unchanged from the M2.

[frans_waterlander]

Yet you appear unable to provide outside data

 

Here, again, are some warnings from the camera manufacturers themselves:

Nikon Z7: August 2018; MILC/mirrorless; FF sensor; no optical viewfinder; electronic viewfinder; monitor

Shooting may end automatically to prevent damage to the camera's internal circuits if the ambient temperature is high or the camera has been used extensively for burst shooting, movie recording, or the like.

 

Nikon D7200: March 2015; dSLR; APS-C sensor; optical viewfinder; no electronic viewfinder; Live View on monitor

Parts of the camera become hot while on or in use for extended periods and may cause fog and bright spots.

Live View will end automatically to protect internal circuits from overheating.

Overheating may occur because of high ambient temperatures or extended periods of using Live View, movie recording, or Continuous Release mode.

 

Canon EOS R: September 2018; MILC/mirrorless; FF sensor; no optical viewfinder; electronic viewfinder; monitor

Shooting in high temperatures may cause noise and irregular colors.

Continuous shooting over an extended period may cause the camera's internal temperature to rise and effect image quality.

 

Sony A7III: February 2018; MILC/mirrorless; FF sensor; no optical viewfinder; electronic viewfinder; monitor

When the temperature of the camera rises, the image quality may deteriorate

 

Is that sufficiently 'outside data'?

[frans_waterlander]

Nope. Not a single example of image degradation followed by an explanation of specific user wrongdoing.

Several image degradation warnings directly from Nikon, Canon and Sony for randomly selected recent cameras. And this is not an issue of user wrongdoing, but camera behavior.

[Ed_Ingold]

I would like to give you the benefit of the doubt that this is a serious question, which will help you decide between a DSLR and an MILC.

 

Using the search term, "Sony Overheating" resulted in several useful hits. The results seem to concur with my personal observations. Two are cited below.

 

How to Prevent Your Sony Camera From Overheating - Videomaker

 

Overheating seems to occur when recording 4K video internally, using battery power. According to one source, the A7iii overheats faster than the A7Riii because it records in 6K then downsamples, whereas the A7Riii records 1:1 in Super-35 mode or by binning in FF mode. 4K takes twice the processing and recording power as HD, which doesn't seem to cause a temperature shutdown. Thermal shutdown is a protective measure, and no mention is made of image quality degradation.

 

The main source of heat is the Li-Ion battery, which can get too hot to hold comfortably after 30 minutes or so of operation. It may help to keep the battery door open, but a better solution is to use an external power source. A USB battery only supplies about 50% of the power needed for an A7/A9 camera. The internal battery takes about twice as long to run down in conjunction with USB power, but does not heat up appreciably in that time frame. For an extended shoot, it's better to use a battery substitute, connected to a lines adapter, or a 14 V video battery with a D-Tap output. I've used all three methods for shoots of 2 hours or so, in 1080p60 mode (50 MHz) without overheating.

 

The processing board is another source of heat. It is located directly behind the sensor, where it is covered by the rear LCD screen, also a significant source of heat. It may help to tilt the LCD away from the body for better heat dissipation.

 

Power to store data on the SD cards is probably more significant in terms of heat than processing and display, if only the amount of current it takes from the battery. It may help to keep the memory door open during extended use.

 

For the type of video projects I shoot, I use 1080p60 rather than 4K, to reduce storage requirements and processing time. The typical duration of a session is 2 hours, but can be as long as 10 hours for a competition. The camera runs continuously, as well as the sound recorder, which makes synchronization easier in post. I use an external video recorder, like an Atomos Ninja V, which has a 7 to 14 hour capacity on a 1 TB hard drive and no 30 minute timeout. I can comfortably shoot for 2.5 hours on a single battery, or all day with an external source.

 

I can safely rule out the image sensor as a significant source of heat. Whether viewing only, shooting stills or shooting video, the sensor (and processor) run continuously without overheating, whether in 1080 or 4K. The HDMI output is not compressed, so I record 4,2,2 video at 220 MHz. Consequently the camera does less processing to do (i.e., compression), and nothing is written to the memory cards.

 

Last weekend, I recorded the entire event, a hot air balloon rally, using an A7iii, in 4K, continuously for over 4 hours, in 80 degree air and (mostly) sunlight. Again, there was no overheating. I used a 14 v video battery and a battery substitute. I also recorded on a Ninja V monitor/recorder, which got hot, but harmlessly via fan cooling.

[Uhooru]

Also, if you Google the question "Sensor overheating issues in modern digital cameras", the articles seem to be from a few years ago and are talking about long videos and live view on DSLRs.

[Nick D.]

I love Photonet. Most opinionated people are those, who don't and probably never will buy/have camera/lens in question, but have strong expert opinion about it:)

[digitaldog]

Several image degradation warnings directly from Nikon, Canon and Sony for randomly selected recent cameras. And this is not an issue of user wrongdoing, but camera behavior.

If you never experienced it, you've only imagined it. As expected, the only answer you'll accept is that you have predicted, not experienced

See since my last comment anyone now taking you seriously? Nope. CWOBaT (colossal waste of bandwidth and time) apparently the posting goal seen once again here

[digitaldog]

I love Photonet. Most opinionated people are those, who don't and probably never will buy/have camera/lens in question, but have strong expert opinion about it:)

Or have any images they've made (or just shot, there's a difference) to show us.

BTW, love your dog shot avatar. Not sure why it doesn't show up in the forums but only when clicking on it, but super cute!