[ThorSanchez]

Yes, some people prefer an EVF. But they also have to be willing to put up with the image quality degradation and short battery life created by running a hot sensor full time.

What is the practical implication of having a sensor that is on more often and might be somewhat warmer? If you were shooting hundreds of frames in a few minutes at high ISO how noticeable is the noise from the heat? Is a DSLR at ISO 3200 the same as a mirrorless at 800? Or is it more of an edge case, where you really have to pixel peep a comparison scene and know what you're looking for?

[reh321]

Currently, the sensors are near the physical limits.

Even if someone made electronics with zero lag in the transfer from sensor to display, the EVF would still have lag imposed by the time the sensor needs to gather light to form an image.

Even if some could make totally noise-free electronics, the low-lag image would still be grainy because scenes on planet Earth simply have too few photons to create a clean image with a short shutter speed.

As for power, the "less and less power" trend in electronics is created by making smaller and smaller semiconductor features. That's excellent for digital electronics and can even be done with some kinds of analog electronics. But the only way to apply that strategy to image sensors is to make smaller format sensors. Thus a 24 MPix smartphone sensor is the "low-power" version of the 24 MPix FF sensor. But if you want to keep the sensor big, then the power requirements stay high. Oh, and the other way to reduce sensor power is to slow down the frame-rate. But that performance dimension is going in the wrong direction in MILCs -- the better the EVF experience, the hotter the sensor!

This is all wonderful theory .... I'm guessing the pixel-peekers are truly grateful. But me, I used my Q-7, with its tiny, over-crowded sensor, in 90 degree F weather last week with no noticeable degradation.

Furthermore, I really believe this is a "rabbit trail" - Pentax might provide a 'K-02' as an entry-level camera, but it is beyond my {very active} imagination that they would replace the K-3ii by an MILC.

[photoptimist]

What is the practical implication of having a sensor that is on more often and might be somewhat warmer? If you were shooting hundreds of frames in a few minutes at high ISO how noticeable is the noise from the heat? Is a DSLR at ISO 3200 the same as a mirrorless at 800? Or is it more of an edge case, where you really have to pixel peep a comparison scene and know what you're looking for?

Sensor noise basically doubles for each 8° C temperature rise which means the camera loses about 1 EV of DR or high-ISO performance for each 8°C rise. Threads about the Sony A9 suggest that sensor runs as hot as 48°C before triggering the over-heat warning. Probably a study of camera EXIF data would give a good idea of . That means the A9 might lose up to 3 EV of DR or high-ISO performance on a 15°C day day. So it is like the "DSLR at ISO 3200 being the same as a mirrorless at 800."

The bigger issue is that on a DSLR, it would take prolonged bursts (or using live view) to heat up the sensor. The sensor on a DSLR stays cold until the shutter is fired. And each shot on a DSLR only uses the sensor for a small fraction of the second.

But with a MILC, all it takes to using the EVF. Almost anything you do with a MILC such as framing, AF, metering, tracking the target, waiting for the decisive moment, etc., will heat up the sensor. On a MILC, the sensor might be on for some time and be quite toasty and noisy before the first picture of the day is taken.

[Trickortreat]

That means the A9 might lose up to 3 EV of DR or high-ISO performance on a 15°C day day. So it is like the "DSLR at ISO 3200 being the same as a mirrorless at 800."

Huh? This must be one of the biggest nonsense ive read in my lifetime.
Dxomark tests of the mirrorless are within 10% performance of their dslr counterparts. They must be running those tests in some kind of a cryogenic chamber :/

[reh321]

Sensor noise basically doubles for each 8° C temperature rise which means the camera loses about 1 EV of DR or high-ISO performance for each 8°C rise. Threads about the Sony A9 suggest that sensor runs as hot as 48°C before triggering the over-heat warning. Probably a study of camera EXIF data would give a good idea of . That means the A9 might lose up to 3 EV of DR or high-ISO performance on a 15°C day day. So it is like the "DSLR at ISO 3200 being the same as a mirrorless at 800."

The bigger issue is that on a DSLR, it would take prolonged bursts (or using live view) to heat up the sensor. The sensor on a DSLR stays cold until the shutter is fired. And each shot on a DSLR only uses the sensor for a small fraction of the second.

But with a MILC, all it takes to using the EVF. Almost anything you do with a MILC such as framing, AF, metering, tracking the target, waiting for the decisive moment, etc., will heat up the sensor. On a MILC, the sensor might be on for some time and be quite toasty and noisy before the first picture of the day is taken.

This is all wonderful theory .... I'm guessing the pixel-peekers are truly grateful. But me, I used my Q-7, with its tiny, over-crowded sensor, in 90 degree F weather last week with no noticeable degradation.

Furthermore, I really believe this is a "rabbit trail" - Pentax might provide a 'K-02' as an entry-level camera, but it is beyond my {very active} imagination that they would replace the K-3ii by an MILC.

[BROO]

We might all have our opinions on the technical merits of OVF v EVF and I am sure the Ricoh engineers are more versed on this than any of us but at the end of the day Ricoh are in the business of making money. It is their duty to ensure that money invested on R&D is returned via product sales. I think the question is, where will they get the best bang for their buck? Will the investment required on the development of a high speed ASP-C DSLR be recouped in sales to K3 owners wanting an upgrade path? Can anyone on this forum answer this question?

[fsge]

Sensor noise basically doubles for each 8° C temperature rise which means the camera loses about 1 EV of DR or high-ISO performance for each 8°C rise. Threads about the Sony A9 suggest that sensor runs as hot as 48°C before triggering the over-heat warning. Probably a study of camera EXIF data would give a good idea of . That means the A9 might lose up to 3 EV of DR or high-ISO performance on a 15°C day day. So it is like the "DSLR at ISO 3200 being the same as a mirrorless at 800."

The bigger issue is that on a DSLR, it would take prolonged bursts (or using live view) to heat up the sensor. The sensor on a DSLR stays cold until the shutter is fired. And each shot on a DSLR only uses the sensor for a small fraction of the second.

But with a MILC, all it takes to using the EVF. Almost anything you do with a MILC such as framing, AF, metering, tracking the target, waiting for the decisive moment, etc., will heat up the sensor. On a MILC, the sensor might be on for some time and be quite toasty and noisy before the first picture of the day is taken.

So a sensor cooler is welcome ?

[reh321]

We might all have our opinions on the technical merits of OVF v EVF and I am sure the Ricoh engineers are more versed on this than any of us but at the end of the day Ricoh are in the business of making money. It is their duty to ensure that money invested on R&D is returned via product sales. I think the question is, where will they get the best bang for their buck? Will the investment required on the development of a high speed ASP-C DSLR be recouped in sales to K3 owners wanting an upgrade path? Can anyone on this forum answer this question?

I was hoping someone else would respond ... this does deserve an answer. What kind of camera should replace the K-3ii is ultimately not a technical decision - it is a marketing decision {one aspect of marketing is deciding what a new product should consist of to meet needs of the market}

I have already made clear my opinion that an MILC would be a failure as the upgrade path for the K-3ii, my reasons in part being

(1) As several have indicated in past pages of this thread, an EVF would never be accepted by many in group K3+ (*)

(2) In several threads, members of group K3+ have indicated dissatisfaction with the size of the KP. All kinds of issues are packed in this reason, including a desire for sufficient mass to 'balance' long lenses, and a desire for a top-side LCD. Presumably an MILC would be even smaller.

(3) In several threads, members of group K3+ have indicated a desire for a battery that would last a long time. MILCs tend to use more power and to have less room for a battery than a K-3 does.

These are the three most pressing reasons that come to mind right now. Now that I've started a list, I'd invite anyone else to add thoughts to it.



(*) for brevity I will use the term "group K3+" to denote those in the market for a K-3ii successor

[BROO]

(*) for brevity I will use the term "group K3+" to denote those in the market for a K-3ii successor

So what kind of specs do you require for a K3 sucessor? How much will the R&D cost and will the camera sell for enough money and in the required volumes to turn a profit?
And what is the opportunity cost of ignoring mirrorless?
I would think that a high end OVF ASP-C DSLR is becoming an unattractive market segment for savvy camera manufacturer.
Its all about having a business case, and not whether mirrorless is the appropriate upgrade path for K3 users.

[surfar]

Its all about having a business case

As long as that fits in the Basket!

[ThorSanchez]

Sensor noise basically doubles for each 8° C temperature rise which means the camera loses about 1 EV of DR or high-ISO performance for each 8°C rise. Threads about the Sony A9 suggest that sensor runs as hot as 48°C before triggering the over-heat warning. Probably a study of camera EXIF data would give a good idea of . That means the A9 might lose up to 3 EV of DR or high-ISO performance on a 15°C day day. So it is like the "DSLR at ISO 3200 being the same as a mirrorless at 800."

The bigger issue is that on a DSLR, it would take prolonged bursts (or using live view) to heat up the sensor. The sensor on a DSLR stays cold until the shutter is fired. And each shot on a DSLR only uses the sensor for a small fraction of the second.

But with a MILC, all it takes to using the EVF. Almost anything you do with a MILC such as framing, AF, metering, tracking the target, waiting for the decisive moment, etc., will heat up the sensor. On a MILC, the sensor might be on for some time and be quite toasty and noisy before the first picture of the day is taken.

I appreciate the response, and I get that sensor temperature has an impact on noise performance. But... I don't think this makes a lot of sense from a practical standpoint. I don't know that a mirrorless sensor is really 20, 30, 40, 50°C hotter than a DSLR sensor in typical use. Fuji, Sony, even Pentax have had mirrorless cameras for years. I don't think that I've heard much criticism of those cameras based on sensor noise. And knowing the internet, there would be a gigantic backlash against mirrorless manufacturers if the typical use case resulted in an ISO 800 performance equivalent to ISO 3200 in a DSLR. Or base ISO being the same as 400 or 800 in a DSLR.

You're basically saying that the best Fuji or Sony should have the high-ISO noise characteristics of a DSLR from 15 years ago. Or worse. That certainly doesn't appear to be the case.

[reh321]

I appreciate the response, and I get that sensor temperature has an impact on noise performance. But... I don't think this makes a lot of sense from a practical standpoint. I don't know that a mirrorless sensor is really 20, 30, 40, 50°C hotter than a DSLR sensor in typical use. Fuji, Sony, even Pentax have had mirrorless cameras for years. I don't think that I've heard much criticism of those cameras based on sensor noise. And knowing the internet, there would be a gigantic backlash against mirrorless manufacturers if the typical use case resulted in an ISO 800 performance equivalent to ISO 3200 in a DSLR. Or base ISO being the same as 400 or 800 in a DSLR.

You're basically saying that the best Fuji or Sony should have the high-ISO noise characteristics of a DSLR from 15 years ago. Or worse. That certainly doesn't appear to be the case.

Looking at the DxOMark "sports" ranking, where noise is a major issue, the top five FF entries are by Sony - then comes the K-1 - then a couple of low-mp Nikon's.

Camera Database - DxOMark

[beholder3]

Sensor noise basically doubles for each 8° C temperature rise which means the camera loses about 1 EV of DR or high-ISO performance for each 8°C rise. Threads about the Sony A9 suggest that sensor runs as hot as 48°C before triggering the over-heat warning. Probably a study of camera EXIF data would give a good idea of . That means the A9 might lose up to 3 EV of DR or high-ISO performance on a 15°C day day. So it is like the "DSLR at ISO 3200 being the same as a mirrorless at 800."

I once did some measurements on a K-3 and 5D3. In both cases video recording for about 30 minutes brought the sensor from 22 to about 50 °C (roughly speaking, after 45 min they would go to about 55 °C). Yes, sensor noise increased substantially when checking with a dark. During normal low ISO daytime shooting this was not something that the average phonescreen image watcher would recognize though.

So I assume your comments are quite correct for DSLMs, but few people recognize it.

[Kunzite]

We can't suspect DxOMark of any degree of useful accuracy, can we?
Their tests are good for Internet "fights" though. "My camera has 1 point over your camera! Your camera is garbage!"

[reh321]

We can't suspect DxOMark of any degree of useful accuracy, can we?
Their tests are good for Internet "fights" though. "My camera has 1 point over your camera! Your camera is garbage!"

I have no reason to question the accuracy of their measurements.