A7r4 announcement today

[Tony Jay]

Sony writes for a7rIV: 15-stop dynamic range under Sony's test condition.

Sony's a7rIII page:

A new front-end LSI and BIONZ X™ boost processing speed and are teamed with the Exmor R™ sensor to deliver up to 15-stop dynamic range at low ISO sensitivity for stills, enabling richer tonal expression.

Sony's a7rIV page:

The latest BIONZ X image processing engine works in tandem with front-end LSI to maximize high-resolution sensor performance while achieving up to 15-stop dynamic range at low ISO sensitivity.

No use directing this post to me - I have read the same thing...

Jim Kasson is making it clear that he wants data, not conclusions...

[Martin Kristiansen]

No use directing this post to me - I have read the same thing...

Jim Kasson is making it clear that he wants data, not conclusions...

And the data coming from the manufacturer won’t be trusted and since the camera was launched so recently that the first batteries have barely had time to be charged no one has had the time to do proper tests and make graphs. Give it a few days.

Last night I watched a short YouTube video comparison between the iii and the iv. I didn’t bother keeping the link so can’t post it here but it shouldn’t be hard to find. I was surprised at how big an improvement the new camera shows. I have no idea as to the testing protocol so would like to see more information.

[JaapD]

“The latest BIONZ X image processing engine works in tandem with front-end LSI to maximize high-resolution sensor performance while achieving up to 15-stop dynamic range at low ISO sensitivity” tells me primarily that the bit depth of the image processing pipeline is able to handle a 15-stop dynamic range. It doesn’t specifically say that the 15-stop dynamic range directly comes of the sensor itself, it could for instance come from averaging a sequence of frames by the BIONZ.

In my opinion a 15-stop dynamic range directly out of the sensor is for now a step too far, technology wise.

I know, still speculating, no real facts yet 

Regards,
Jaap.

[Jim Kasson]

Sony writes for a7rIV: 15-stop dynamic range under Sony's test condition.

Sony's a7rIII page:

A new front-end LSI and BIONZ X™ boost processing speed and are teamed with the Exmor R™ sensor to deliver up to 15-stop dynamic range at low ISO sensitivity for stills, enabling richer tonal expression.

Sony's a7rIV page:

The latest BIONZ X image processing engine works in tandem with front-end LSI to maximize high-resolution sensor performance while achieving up to 15-stop dynamic range at low ISO sensitivity.

The question of what the test conditions are is not even addressed in those quotes, but less answered.

To be precise:


How is the denominator calculated? (I assume the numerator is simply full scale)
Is the pixel-level DR scaled to some arbitrary print size?


By the way, I can understand the utility of normalizing PDR to a fixed viewing angle. I cannot understand what the utility is of normalizing EDR that way -- except to marketeers.

By the way, I see some people saying that Sony made the same claim for the a7RIII. Here's the EDR that I measured:

https://blog.kasson.com/a7riii/sony-a7riii-edr-vs-iso-setting/

Here's what I got for PDR with Claff normalization:

https://blog.kasson.com/a7riii/sony-a7riii-and-nikon-z7-shadow-noise-numbers/

Jim

[SrMi]

The question of what the test conditions are is not even addressed in those quotes, but less answered.

To be precise:


How is the denominator calculated? (I assume the numerator is simply full scale)
Is the pixel-level DR scaled to some arbitrary print size?


By the way, I can understand the utility of normalizing PDR to a fixed viewing angle. I cannot understand what the utility is of normalizing EDR that way -- except to marketeers.

By the way, I see some people saying that Sony made the same claim for the a7RIII. Here's the EDR that I measured:

https://blog.kasson.com/a7riii/sony-a7riii-edr-vs-iso-setting/

Here's what I got for PDR with Claff normalization:

https://blog.kasson.com/a7riii/sony-a7riii-and-nikon-z7-shadow-noise-numbers/

Jim

Agree with everything you wrote.

My understanding: Sony's 15-stop DR is mainly a marketing point. The only conclusion I deduce from Sony's brochures is that a7rIII and a7rIV have similar DR (assuming they used the same Sony's test conditions). I also assume that Sony's DR is not written in the octal number system ;-).

To compare a7rIV's DR with other cameras, we need to wait for Bill Claff's measurements.

[faberryman]

It doesn't seem that Sony's actual DR is as important as its DR as compared to other cameras. Is it better, the same as, or worse than the competition.

[Jim Kasson]

It doesn't seem that Sony's actual DR is as important as its DR as compared to other cameras. Is it better, the same as, or worse than the competition.

At this point, who can tell?

[DP]

The 36MP sensor announced by Sony alongside the 60MP sensor earlier this year looks very much like an A9II sensor, or closely related to it.

A7S line might need 8K ... A9 line does not

[DP]

I would much prefer 24MP on a A9ii and totally uncompromising performance in everything else. If I need more MP there is the R series.

exactly the point - before moving above 24mp on А9 line Sony needs to match or best everything that top line (D* & 1D*) dSLRs do ... 36mp will be only an obstacle to AF and FPS - there is no point to have 36mp on a sports camera if 20-24mp dSLR still does better in some aspects

[DP]

And that 36MP sensor announced earlier sounds suspiciously A9II-like in its capabilities

A7SIII capabilities... not A9II's

[DP]

Sony could have included the a9 AF in the a7rIII or a7rIV but didn't.

no, A9 has stacked sensor assembly ( https://www.sony-semicon.co.jp/products_en/IS/sensor2/technology/exmor-rs.html ) ... neither A7RIII nor A7R4 do.

[DP]

This allows to read the sensor data at 60 fps

just a note 60 fps can be done w/o stacked design too ... see new non stacked 36mp sensor --- 60 fps @ 10bit (to feed EVF/LCD/CDAF/PDAF)

[DP]

I believe there are way more wedding photographers buying the A9 than sports photographers.

I believe in aliens ! wedding is not a repro/product/fashion

[DP]

The question is not whether they said 15 stops of DR, but rather is it achieved in single shot mode. Of course , there's the larger question of the protocol and whether they scales the measurements to some arbitrary, non-native resolution.

Jim

https://www.dpreview.com/articles/3979321923/sony-a7r-iv-what-you-need-to-know?slide=7

"...that '15 stops of DR' figure Sony is advertising is with downscaled, 8MP images..."

that is of course not BClaff-like test, but they have the actual camera

[Jim Kasson]

https://www.dpreview.com/articles/3979321923/sony-a7r-iv-what-you-need-to-know?slide=7

"...that '15 stops of DR' figure Sony is advertising is with downscaled, 8MP images..."

that is of course not BClaff-like test, but they have the actual camera

And what is the denominator?

[Martin Kristiansen]

exactly the point - before moving above 24mp on А9 line Sony needs to match or best everything that top line (D* & 1D*) dSLRs do ... 36mp will be only an obstacle to AF and FPS - there is no point to have 36mp on a sports camera if 20-24mp dSLR still does better in some aspects

I like the split in capability. I used to use 4x5 and 35mm. Was never a fan of MF. A very high res body and a high performance body gives me that split again but with huge convenience.

As a thought experiment imagine Sony manages to produce a global shutter at 24MP, I’m sure they will one day. I would rather have that than a 36MP camera without the global shutter.

Incidentally the ability to switch to APSC format and still get 26MP is attractive. Turn my 400 into a 600. That could be useful.

[shadowblade]

I like the split in capability. I used to use 4x5 and 35mm. Was never a fan of MF. A very high res body and a high performance body gives me that split again but with huge convenience.

As a thought experiment imagine Sony manages to produce a global shutter at 24MP, I’m sure they will one day. I would rather have that than a 36MP camera without the global shutter.

Incidentally the ability to switch to APSC format and still get 26MP is attractive. Turn my 400 into a 600. That could be useful.

A7S line might need 8K ... A9 line does not

The A9 is a photojournalist's camera, not just a sports camera. Increasingly, photojournalists shoot both stills and video, for print, online and TV publication, mixing the two formats at the same event. This requires a camera capable of the AF needed for fast action,  as well as the video capability needed for the high-resolution TV standard of its day. Going forwards, this means 8k.

36-42MP will be the new standard for action cameras, either in this generation or the next. This is needed for 8k output. The A7s series is likely dead in its current form, as a dedicated low-resolution, low-light camera with a strong video focus. Future video standards will also call for 8k, which turns the A7s series more into a cut-down A9 (possibly more optimised for video, at the expense of some stills capability) rather than a dedicated low-light, low-resolution monster.

A camera which is good for still sports photography and not much else won't have much of an audience. Even wedding and event photographers shoot video clips, which, in future, will be 8k. With the latest firmware update, Sony has already beaten both the 1Dx2 and D5 in terms of AF capability, with a 24MP sensor (vs the Canon/Nikon 20MP sensors). In the last two -and-a-bit years, there's no reason they won't have, or can't have, built the same capability into a 36MP sensor.

Totally agree about the ability to crop. That's also why 36-42MP makes sense for a wildlife/sports body. Shooting distant subjects from a fixed location with a long prime often calls for cropping (or switching to a shorter lens, when they get too close). Crop a 36-42MP image to 1.5x and you still have 16-18MP - reasonable for an APS-C camera. Crop a 24MP image to 1.5x and you're not left with much at all.

[shadowblade]

Regarding the dynamic range claim, maybe this news from 7 months ago reveals a bit more:

https://www.sonyalpharumors.com/sr5-more-specs-from-the-five-new-sony-full-frame-sensors/

The A7r4 uses the IMX 551 sensor, with a 14.9 stop DR. This sensor was not made available to other companies (at least not yet).

The less-capable IMX 455 was made available to third parties, and is the sensor Nikon has purchased for its own cameras. This will likely show up in the Z8. Note that the IMX 455 has less DR (14.4 stops), but uses a base ISO of 64 (with another base at 400) - the same setup as has appeared in recent Nikon bodies.

[BernardLanguillier]

The A7r4 uses the IMX 551 sensor, with a 14.9 stop DR. This sensor was not made available to other companies (at least not yet).

What is your source of information?

Cheers,
Bernard

[shadowblade]

What is your source of information?

Cheers,
Bernard

Mentioned sporadically on FredMiranda. Also seems to be mentioned a lot on Chinese-language websites (such as this article), accessible using Google Translate. Type in "A7r4" or "A7r IV" and "IMX551" into Google and you'll get a few hits from FredMiranda and other English websites, but a ton of articles in Chinese about it.

Also, it's the only 61MP sensor that Sony itself has announced manages close to 15 stops of DR (14.9 stops). The other 61MP sensor only manages 14.4 stops and also has a base ISO of 64, not 100 like the A7r4.