The Sony a7 V’s Sensor Is Even More Impressive Than it Seemed

Another day, another Sony a7 V dynamic range update. Our initial speculation has proven correct, and the new a7 V is employing Dual Gain Output (DGO) to achieve outstanding dynamic range when using its mechanical shutter at base and low ISO settings. Further, building upon research by Adam Horshack, William “Bill” Claff of the excellent Photons to Photos has revised his a7 V testing to indicate that the camera is not using baked-in noise reduction with its RAW files. What initially appeared to be evidence of noise reduction is, in fact, DGO.

What Is Dual Gain Output?

There is a lot to unpack here. Before doing so, it is worth briefly recapping what Dual Gain Output (DGO) is, how it works, and why it matters.

Back in August, Adam Horshack determined that the Lumix S1 II’s surprisingly strong dynamic range performance relative to the Nikon Z6 III, which, like the S1 II, also uses a 24-megapixel partially stacked sensor, is the result of Panasonic’s implementation of Dual Gain Output.

A camera utilizing Dual Gain Output, DGO, captures two readouts of a single exposure with two different ISO settings. In the case of the S1 II, the camera at base ISO combines an image shot at ISO 100 and a second at ISO 800 and merges the results into a single RAW file. This means the camera can take advantage of the higher saturation capacity at low ISO and the improved deep shadow performance at a higher ISO.

As one would expect, this comes with some limitations. There’s really no such thing as free lunch with image sensor technology. In the case of the S1 II (and a7 V), DGO is not available when using the camera’s electronic shutter, which is required for it to shoot at its fastest continuous shooting speeds. Based on our experience with the a7 V and Claff’s electronic shutter dynamic range results, this same limitation is clearly in effect with the a7 V. DGO increases sensor readout times, which are not an issue when using a mechanical shutter but are problematic when using an electronic shutter, increasing the risk of rolling shutter problems.

When Horshack discussed the S1 II’s Dual Gain Output with PetaPixel in August, he said he expected “most future cameras to implement some form of sensor-based HDR or DGO,” in large part due to its immense dynamic range advantages in best-case scenarios, like when the photographer can shoot with the mechanical shutter at base ISO.

Revising the Narrative of the a7 V

“After consulting the work on Dual Gain Output (DGO) by Horshack, and re-examining the data, I have decided that what appears as Noise Reduction (NR) is actually a form of DGO (it seems there is no single implementation going by this name),” Bill Claff wrote on DPReview‘s forums.

This is extremely new territory here. Dual Gain Output remains relatively rarely utilized with full-frame cameras, for starters. But beyond that, every company takes different approaches to balancing image quality and sensor speed, performance at different ISO settings, sensor architecture, and processor designs, to name just a few of the important variables that impact how each camera performs in terms of dynamic range.

Add in the fact that companies themselves say very little about how their image sensors work, for obvious reasons, and almost never divulge any significant details. It took significant effort and time to get Panasonic to acknowledge that the S1 II has Dual Gain Output. That isn’t because Panasonic is particularly cagey, either; it’s just how it is.

All that is to say that Claff took a closer look at his results in view of Horshack’s work on DGO, and what initially appeared to be baked-in noise reduction is, instead, evidence of the a7 V employing a form of DGO. Determining precisely what form and how it works will require more digging.

What Does This Mean for the Sony a7 V?

There are two very important takeaways here with Claff’s revised results. The first is that the Sony a7 V is not employing baked-in noise reduction in its RAW files, which is, generally speaking, good news. We have criticized many a camera for baking noise reduction into its RAW files to improve dynamic range results. While the real-world impact of noise reduction varies from negligible to very noticeable, we’d rather not see cameras implementing this approach.

The second is that, as we speculated (hoped), the a7 V is utilizing Dual Gain Output to achieve improved dynamic range results when using its mechanical shutter. This is a smart way to achieve better image quality and comes with essentially no drawbacks. As sensors become faster, like the partially stacked chip inside the a7 V, and processors more powerful, ditto with the a7 V and its brand-new Bionz XR2, DGO becomes an increasingly compelling option for camera manufacturers seeking to deliver excellent dynamic range from partially stacked and stacked image sensors.

Excellent dynamic range is the name of the game with the a7 V, at least with its mechanical shutter. Typically, the price photographers must pay when a sensor is partially stacked or stacked is lower maximum dynamic range. However, DGO can overcome this limitation without affecting the sensor’s desired speed or performance.

But DGO does not yet make sense for use with an electronic shutter for numerous reasons, so the a7 V’s performance there still leaves something to be desired. As Claff explains on DPReview‘s forums, since DGO only happens with the a7 V’s mechanical shutter, the electronic shutter values look that much worse by comparison.

“So, is the glass half full or is it half empty?” Claff asks. The answer is, of course, yes.

As of yet, at the price point the a7 V targets, it is not feasible to deliver class-leading dynamic range at the camera’s top-end speed of 30 frames per second. However, photographers can still have both of those things, just not at the same time. For many, that’s a worthwhile compromise.

Buy the Sony a7 V new on B&HBuy the Sony a7 V used on KEH.com

Image credits: Sony. Dynamic range testing and charts by William J. Claff (Photons to Photos)