Image quality

The E-M1X appears to share the same 20MP Four Thirds sensor as the E-M1 II and offers similar Raw and JPEG performance. Raw output is also nearly identical to the Panasonic G9, though we continue to prefer Olympus JPEG color. And the camera's 80MP tripod-based, high-res mode offers both resolution and noise benefits.

Key takeaways:

  • Raw image quality is on par with the best Micro Four Third sensors
  • Olympus JPEG color continues to be a DPReview favorite
  • While JPEG noise is well-controlled, sharpening is a tad aggressive
  • High-res mode can bring impressive resolution along with noise benefit of combining multiple images

Studio scene

Our test scene is designed to simulate a variety of textures, colors and detail types you'll encounter in the real world. It also has two illumination modes to see the effect of different lighting conditions.

The E-M1X captures an impressive amount of detail displaying similar levels of aliasing in the text portion of our scene to its 20MP Micro Four Third peers. If anything, it seems to capture more detail than the E-M1 II, though we're almost certain they share the same sensor. The discrepancy here is due to an upgrade in our studio lens from the Olympus 45 F1.8 to the sharper Panasonic 42.5mm F1.2.

At base ISO the amount of detail capture from the E-M1X looks very similar to the larger-sensor Fujifilm X-T3. However the E-M1X is more susceptible to moiré due to its use of a Bayer-type sensor with no AA filter. As light levels drop, noise performance from the E-M1X matches the E-M1 II and G9 but, as you might expect, falls behind its better larger-sensor rivals. This difference is more noticeable the higher the ISO, but is rarely a huge difference, suggesting the sensor is out-performing some of those in its larger sensor peers.

We've long been fans of Olympus JPEG color science and the E-M1X continues the tradition of pleasing color. Reds are nicely saturated and yellows avoid a dreaded green shift. It compares well with popular rivals such as Canon's JPEG color, which is a good thing.

The E-M1X shows slightly-aggressive JPEG sharpening at base ISO, but it's better than the Panasonic's sharpening, which renders the straight lines of the playing cards jagged. At high ISO, the E-M1X does a good job keeping noise at bay. Noise reduction tends to be more intelligent than the E-M1 II. However sharpening at high ISOs can feel a little crunchy. on the wholethe X-T3 does a better job than the E-M1X at balancing noise suppression and detail retention.

Dynamic Range

Exposure Latitude

Our Exposure Latitude test looks at how noisy the shadows of an image look when lifted. We use progressively darker exposures to simulate deeper and deeper shadows. The E-M1X appears to perform slightly better than the E-M1 II, but this could be a difference in Adobe Camera Raw noise reduction (the X's images look less saturated). Compared to the Nikon D500, the E-M1X looks noisier when given the same exposure and more similar when the Nikon has been given 1EV less light (which is what you should expect from the roughly 2/3EV difference that the D500's larger sensor should make). This, along with the Nikon's lower base ISO (greater ability to tolerate light), means it'll produce cleaner, more flexible images when given the same exposure. But the E-M1X's sensor compares extremely well, in terms of read noise to its larger sensor peers, which appear to pay a noise cost for their rapid readout.

The multi-shot mode, which combines eight images, is much cleaner and more tolerant of 'pushing,' as you might expect. Visually it gives a result comparable with a good full frame camera given twice as much light, again, this fits with what you'd expect based on sensor size: combine eight images from a sensor 1/4 the size of a large one and you'll collect twice as much light overall.

ISO Invariance

Our ISO Invariance test tries to cancel-out the effect of photon shot noise by maintaining the same exposure settings, to see how much or little benefit there is to applying amplification to the signal. There's a slight noise cost to shooting at ISO 200 and brightening, compared with natively shooting at ISO 1600. This indicates that the camera is adding very little noise to its images (it's mainly the randomness of the light). This opens up the option of shooting at low ISO settings in low light, then brightening the image later, since this approach won't be much noisier but will preserve up to 1 stop of highlight data for each ISO step you're dropped.