Mirrorless


Panasonic GH5s Review

August 22, 2018

By Greg Scoblete

Camera lines are a bit like animals, evolving toward ever-greater fitness. And, as in biological evolution, the progression of a camera line doesn’t necessarily follow a straight path. It can branch off in different directions.

Perhaps no other camera maker embodies this evolutionary branching quite like Panasonic. As you ascend in price and capabilities, the company’s camera lines fork. Follow the road to still image optimization and you’ll land on the G9. Take the path that emphasizes video prowess and you’ll find the GH5 and, just a bit further up the road, the new GH5S.

Panasonic GH5s Camera

Features

Where the GH5 offered a 20-megapixel sensor, the GH5S features a 10-megapixel sensor—a similar tradeoff made by Sony’s a7S/II to prioritize low-light sensitivity and dynamic range over absolute resolution.

According to Panasonic, the sensor offers Dual Native ISO Technology with a pair of readout circuits on the sensor. One circuit prioritizes dynamic range at lower ISO values while the second improves noise reduction when shooting at higher ISOs (800). The GH5s is also optimized for low-light focusing, capable of locking focus in incredibly low light down to -5EV. A “Live Boost” function lets you boost the brightness on the display to confirm your composition without impacting your footage.

Beyond its low-light prowess, the camera also delivers 4K/60p recording (that’s cinema 4k, too–4096 x 2160), a new benchmark for mirrorless cameras. 4k/60p footage is captured at 8-bit 4:2:0 in camera. You can also record 4K/30p at 10-bit, 4:2:2 in camera–another mirrorless milestone. Full HD video can be captured at a motion-slowing 240 fps, faster than the GH5 and indeed, any other camera at this price point. The GH5S is one of the few cameras that can record video in the new HEVC codec, a more efficient but more computationally onerous successor to H.264.

The new sensor also offers the same angle of view when shooting in a variety of aspect ratios, including 4:3, 17:9, 16:9 and 3:2. This so-called Multi Aspect sensor is actually physically larger than Micro Four Thirds so that each crop of the sensor generates a 10-megapixel image. The sensor also gives the GH5S the ability to capture 14-bit RAW still images, whereas previous Panasonic cameras were capped at 12-bit RAW files. What the sensor lacks, however, is image stabilization.

The Time Code functionality has also been enhanced in the camera. The GH5S is compatible with Time Code In and Out to sync audio and video across multiple cameras and the GH5S can be used as a time code generator for other cameras. The GH5S comes pre-installed with the V-Log color profile and a Rec.709 Look Up Table (if you want V-Log on the GH5, you have pay extra). You’ll be able to load up to four additional LUTs in the camera, too.

Design

The GH5S may be a crop sensor mirrorless, but it’s built like a DSLR and heavier than other mirrorless models in its class. It’s weather sealed and bulky, though not at all uncomfortable to handle.

The body design conforms to the original GH5, with a nice selection of programmable custom buttons, a full-sized HDMI port and a speedy USB-C (3.1 Gen 1) connection. New for the GH5S is a faster display that now refreshes at 120 fps, making framing through the 3-inch monitor that much smoother.

Sample images from the Panasonic GH5s Camera

While its 10-megapixel sensor may not send hearts aflutter, straight out of the camera JPEGs were quite pleasing.

Image Quality

A 10-megapixel sensor won’t set many hearts aflutter but for what it’s worth, the GH5S performs quite admirably in the stills department. JPEGs straight from the camera were well saturated and crisp. RAW files showed excellent dynamic range.

The central promise of the GH5S is better performance in low light and at higher ISOs. Shooting DCI 4K video at both of the dual native ISOs (saved direct to card) produced very clean footage, delivering extremely impressive results even above ISO 2500. We captured both V-Log, Standard and Vivid profile footage with very little artifacting or visible noise.

A sample image from the Panasonic GH5s Camera

A sample image from a video recorded with the Panasonic GH5s Camera.

Performance

One virtue of fewer megapixels is less data to manage, which usually translates into faster speed. The GH5S can burst at up to 12 fps with focus fixed on the first frame or at 9 fps with continuous autofocusing for up to 60 RAW images. This isn’t the fastest burst shooting you can find at this price point (Olympus’ E-M1 Mark II gets the nod) but it is quite brisk. What’s more, Panasonic has improved the 4K Photo Mode on the GH5S so that it now captures bursts at 60fps instead of the original 30fps, bringing a whole new meaning to “spray and pray.”

You’ll enjoy a decent battery life of 410 shots per charge, equivalent to the performance on the GH5 and on par with most high-end mirrorless cameras with the exception of the a7 III. Panasonic has been improving battery life on its cameras but still has work to do to catch Sony.

Bottom Line

If you want a video-centric camera in a conventional camera body, you don’t really have a ton of options. There’s the forthcoming (and cheaper) Blackmagic Pocket Cinema Camera 4K, Sony’s aging a7S II and Canon’s quirky fixed-lens XC15. While the GH5S is the priciest of the bunch, it also delivers the most video bang for your buck. It’s the best low/available light video camera you’ll find at the price. Though the lack of image stabilization and low-res sensor makes it less attractive as a stills camera, it’s a formidable filmmaking machine.

Panasonic GH5S

www.shop.panasonic.com

PROS: Class-leading video features; great low light performance and autofocusing; strong dynamic range; rapid burst shooting.
CONS: Low resolution; lacks image stabilization; pricey.
PRICE: $2,499

Notes from the TIPA Test Bench

PDN is a member of the Technical Image Press Association which has contracted with Image Engineering to perform detailed lab tests of digital cameras. See here for a full methodological rundown of how Image Engineering puts cameras through their paces. Full res files of every visual in this review are available to download for your pixel-peeping pleasure here


Resolution

  • At ISO 160, records 1304 line pairs per picture height (LP/PH), 94 percent of the theoretical maximum of its 10-megapixel sensor.
  • The lowest four ISOs show similar resolution curves. Resolution recorded decreases slowly as ISO rises, with 90 percent (1247 LP/PH) at ISO 1600.
  • At ISO 3200, 1221 LP/PH are recorded (89 percent of the theoretical maximum).
  • At the highest non-extended ISO, ISO 51,200, 1027 LP/PH are recorded (74 percent of the theoretical maximum).
  • The predecessor, the GH5, had a 20-megapixel sensor and so recorded more line pairs per picture height: e.g. 1917 at ISO 400, or 99 percent of the theoretical maximum and 1799 line pairs per picture height at ISO 3200 (93 percent of the sensor).

This graph shows the loss of contrast (y-axis) as a function of the spatial frequency in line pairs per picture height (x-axis) for different ISO-sensitivities (colored lines). The further to the right a curve stretches before descending, the better the resolution at that ISO. The limiting resolution for each ISO can be found by identifying to the highest spatial frequency which results in a contrast of 0.1, or where the ISO curve crosses the thicker horizontal thicker black line marking 0.1. The vertical pink line is a reference representing half the number of pixels in the sensor height (the Nyquist frequency).

Texture loss

  • Texture measurements show a moderate amount of artifacts at lower ISOs in high-contrast parts of the image (e.g. 26.9 percent at ISO160, 27.7 percent at ISO800).
  • At ISO6400, the proportion of artifacts in high-contrast parts of the scene is 34.4 percent.
  •  In high-contrast portions of the scene, MTF50 is 960 LP/PH at ISO160 and 950 LP/PH at ISO400.
  • MTF50 is 675 LP/PH (high contrast) at ISO6400.
  • In low-contrast parts of the scene, MTF50 is 932 LP/PH at ISO160, and 908 LP/PH at ISO400.
  • Proportion of artifacts in low-contrast parts of the scene is 24.8 percent at ISO 160 and 30.7 percent at ISO800.
  •  40.5 percent artifacts were measured at ISO6400 in the low-contrast scene area.

An artifact is an alteration in a digital image due to technology or technique of processing. Artifacts stem from noise, compression, and sharpening. This graph plots the calculated difference in digital signal between two methods (DeadLeavesCross & DeadLeavesDirect). The colored lines represent response at different ISOs and in reference to a high-contrast target and a low-contrast target. Values plotted are the Dead Leaves SFR difference against the spatial frequency. The larger the area under the curve, the more artifacts are present.

Edge contrast / sharpening

  • Mild to no sharpening along high-contrast edges: undershoot is 0.0 percent at many ISOs, except ISO160, where undershoot is 0.7 percent; at ISO1600, where it is 0.3 percent; and at the two highest ISOs (0.6 percent at ISO25600, 1.1 percent at ISO51200).
  • Overshoot along high-contrast edges is somewhat stronger: for example, at ISO160, it is 6.9 percent.
  • Overshoot at high-contrast edges is 4.8 percent at ISO1600, and 6.7 percent at ISO3200.
  • Along low-contrast edges, sharpening is a little more pronounced. For example, overshoot and undershoot are 10.0 percent and 5.8 percent, respectively, at ISO160, and 7.7 percent and 3.7 percent at ISO1600.
  • At ISO12800, both overshoot and undershoot are 0.0 percent along low-contrast edges and also along high-contrast edges.

This graph shows the degree of sharpening in the image by representing an over- and undershoot along contrasted edges. The colored lines represent measurements at different ISOs and in high- and low-contrast situations. The size of the dip before the edge (in both depth and breadth) indicates the degree of undershoot; similarly, the amount overshoot is indicated by the height and breadth of the peak. Thus, larger dips and/or peaks indicate that a sharpening effect is visible.

OECF VN/visual noise

  • Visual noise is visible even at the lowest ISOs in Viewing Condition 1 (e.g., score of 1.1 at ISO160, 1.5 at ISO800, 2.1 at ISO6400).
  •  Similar results to the assessments of visual noise modelled from images produced by the GH5: 1.1 at ISO200 and 1.7 at ISO6400 in Viewing Condition 1.
  • In a postcard-sized print or mobile screen (Viewing Condition 2), noise would not be noticeable at the lower ISOs. Above ISO3200 (score 1.0), noise would become apparent to the observer (e.g. score 1.2 at ISO6400).
  • The GH5 produced noise scores of 0.8 at ISO3200 and 1.9 at ISO6400 (VN2).
  • In a large print (Viewing Condition 3), noise would not be visible in images shot at lower ISOs. At ISO3200, the visual noise would become apparent (score 1.3).
  • The GH5 also produced observable visual noise (score 1.0) at ISO3200 (VN3).
  • Visual noise produced by the GH5S would be most apparent in the darker tones.

This chart shows the noise behavior at various ISO-sensitivities (colored lines) as a function of the brightness of the target image, which is indicated by the relative darkness of the circle on the outer edge of the diagram (noise in shadowed areas are above, and in highlights below). The larger the area inside a curve, the stronger the noise. The degree to which noise disturbs the appreciation of an image, depends on the image size and the viewing condition. The right-hand side of the chart shows the visibility of the noise in an image that is displayed 100% on a monitor (VN1). The left-hand half shows the visibility of noise in a 40-cm tall print (VN3).

This chart shows the noise behavior at various ISO-sensitivities (colored lines) as a function of the brightness of the target image. The perception of noise is represented by the area that is encircled by the curve. The larger the area, the stronger the noise. How much the noise disturbs the viewing of an image, depends on the image size and the viewing distance. This chart shows the noise visibility for an image that is displayed 100% on a monitor (VN1).

This chart shows the noise behavior at various ISO-sensitivities (colored lines) as a function of the brightness of the target image. The perception of noise is represented by the area that is encircled by the curve. The larger the area, the stronger the noise. How much the noise disturbs the viewing of an image, depends on the image size and the viewing distance. The chart shows the noise visibility for an image that is about postcard size (scaled to a height of 10cm) viewed at a distance of 25cm.

Dynamic Range

  • Dynamic range is above 8 f-stops at all non-extended ISOs measured.
  • The best dynamic range is at ISO800, where it is 10.0 f-stops.
  • At other ISOs from ISO160 to ISO 3200, the Dynamic Range is more than 9 f-stops.
  • The GH5 also showed a dynamic range of more than 9 f-stops from ISO400 to ISO12800, and 10.0 at the lowest ISO of 200.

Color Reproduction

  • Color reproduction is fairly good, with very few strong deviations from the original color.
  •  ∆E is very consistent and a little over 10 for nearly all ISOs tested. Only at the highest non-extended ISO of ISO51200, is ∆E 11.3.

Color reproduction is shown here in two ways. The upper figure is a chart comparing a reference color (right-hand half of each color patch) directly with the color reproduced by the camera (left-hand half of the color patch). Below is a table that lists the DeltaE of each color patch. Red cells indicated strong color deviations, light green cells represent colors with noticeable deviations, and a dark green field represents a moderate deviation.

**See how the GH5s Compares to Other Cameras in Color Reproduction**

Automatic white balance

  • Automatic white balance at the lowest ISO of ISO160 is fairly good (1.7).
  • White balance in ISOs from ISO400 to ISO6400 is similar and ranges from 2.5 (ISO1600) up to 2.9 (ISO800).
  • At ISO12800 and ISO25600, the automatic white balance improves again slightly to 1.6 and 1.3, respectively.
  • At the highest end of the ISOs, the automatic white balance performs fairly poorly (e.g. 3.3 at the highest non-extended ISO of ISO51200).
  • The predecessor GH5 showed better scores for automatic white balance: 0.8 at the lowest ISO of 200, 0.6 at ISO400, increasing to 1.0 at ISO1600 and 1.4 at ISO6400.

Video

  • Resolution in video stills is 1020 LP/PH (94 percent of the theoretical maximum) at low ISO and 982 LP/PH (91 percent of the theoretical maximum) at high ISO.
  • MTF50 in high-contrast scenes is 787 LP/PH at low ISO, and 599 LP/PH at high ISO. Artifacts: 27.9 percent at low ISO and 35.5 at high ISO.
  •  In low-contrast parts of the scene, low-ISO video stills show 24.7 percent artifacts, while high-ISO frames contained 37.8 percent artifacts. MTF50 was 675 LP/PH at low ISO and 590 at high ISO.
  • Sharpening in video is mild: 4.4 percent overshoot and 2.6 percent undershoot at low ISO along high-contrast edges, and 1.2 percent and 0.7 percent, respectively, at high ISO.
  • Along low-contrast edges, sharpening was 6.2 percent (overshoot) and 3.7 percent undershoot at low ISO, and 1.0 percent overshoot combined with 1.3 percent undershoot at high ISO. Visual noise was below the threshold to be noticeable for Viewing Conditions 2 and 3 at both low and high ISO. The noise would be slightly noticeable at both ISOs in Viewing Condition 1 (100 percent; scores of 1.4 (low ISO) and 1.6 (high ISO)).
  • Automatic white balance in video is excellent, measuring 0.1 at both low and high ISO.
  • Dynamic range is also excellent: ∆E measured 11.0 f-stops at low ISO and 10.9 at high ISO.

This chart shows the noise behavior at two ISO-sensitivities (ISO100 and ISO1600) as a function of the brightness of the target image. The amount of noise perceived is reflected in the size of the area encircled by the curves. The larger the area, the stronger the noise and its perception. The degree to which the noise disturbs the viewer, depends on the image size and the viewing distance. This chart shows the noise visibility for a video frame that is displayed 100% on a monitor (VN1).

This graph shows the sharpening in the image due to an over- and undershoot along edges. Depending on the size (based on width and height) of the additional emerging area, a lower (shallower additional area) or stronger (higher and narrower additional area) sharpening effect is visible.

This graph shows the loss of contrast (y-axis) as a function of the spatial frequency in line pairs per picture height (x-axis) for two ISO-sensitivities in video mode (colored lines). The further to the right a curve stretches before descending, the better the resolution at that ISO. The limiting resolution for each ISO can be found by identifying to the highest spatial frequency which results in a contrast of 0.1, or where the ISO curve crosses the thicker horizontal thicker black line marking 0.1. The vertical pink line is a reference representing half the number of pixels in the sensor height (the Nyquist frequency).

Speed                                                                    

  • Takes 0.9 seconds to start up, compared to 1.2 seconds in the GH5.

Continuous shooting

  • Shoots 9.2 frames per second in the best JPEG format until the card is full. As there are two cards and the cards can be set to relay from one to another, it seems that there are many options for creating a lot of frames with this camera.
  • Performs nearly as well in RAW, shooting 9.1 frames per second until the card is full.
  • Compared to the GH5: 10.0 frames per second in both RAW and JPEG, up to 71 frames (RAW) or the card becoming full (JPEG).

Autofocus (300lx) Live View

  • The total shooting time including autofocus in bright light takes 0.15 seconds, of which the autofocus takes 0.08 seconds.
  • Better than the predecessor GH5, which took 0.18 second to focus and shoot in bright light.

Autofocus (30lx) Live View

  • In low light, the autofocus in Live View took 0.12 second, for a total shooting time of 0.19 seconds.
  • Better than the GH5, which took 0.32 seconds to focus and shoot in low light.


In Depth

With the Lumix DC-GH5S, Panasonic took a relatively unusual step that went against the trend of development in consumer-level mirrorless digital cameras: they removed some features in comparison n with the predecessor camera. So, for example, Panasonic reduced the number of megapixels in its Micro Four-Thirds sensor from 20 in the previous camera (the Lumix DC-GH5) to 10 megapixels in the GH5S.

Resolution measurements show that the GH5S uses 94 percent of its sensor at the lowest ISO of 160 (1304 line pairs per picture height (LP/PH)). The camera continues to use most of its sensor at all lower ISO s. Resolution in images recorded then decreases slowly as ISO  rises, with, for example, 90 percent of the theoretical maximum (1247 LP/PH) being used at ISO 1600. At ISO 3200, 1221 LP/PH are recorded (89 percent of the theoretical maximum). The higher ISO  s are less good: at the highest non-extended ISO, ISO 51200, 1027 LP/PH are used (74 percent of the theoretical maximum).

The DC-GH5 recorded more line pairs per picture height, which is not surprising as its sensor was 20 megapixels instead of 10: e.g. 1917 LP/PH at ISO  400 (99 percent of the theoretical maximum), and 1799 line pairs per picture height at ISO  3200 (93 percent of the theoretical maximum).

The effect of in-camera processing on the detail and texture in images captured was assessed using the ‘dead leaves’ pattern. Texture measurements show consistency in artifact proportion in high-contrast parts of images shot at lower ISO  s (e.g. 26.9 percent at ISO  160, 27.7 percent at ISO  800). Looking at the graph (*.-_Artifacts.pdf ), one can see the low proportion of artifacts produced at ISO  400, in both high- and low-contrast parts of the scene. At ISO  6400, the proportion of artifacts in high-contrast parts of the scene is 34.4 percent, with the proportion increasing with higher ISO  s. In high-contrast parts of the scene, MTF50 is 960 LP/PH at ISO  160 and 950 LP/PH at ISO  400; at ISO  6400, MTF50 is 675 LP/PH (high contrast).

In low-contrast portions of the scene, MTF50 is 932 LP/PH at ISO  160, and 908 LP/PH at ISO  400. The proportion of artifacts in low-contrast parts of the scene is 24.8 percent at ISO  160 and 30.7 percent at ISO  800. 40.5 percent artifacts were measured at ISO  6400 (low contrast).

The GH5S produces very mild to no sharpening along high-contrast edges, with undershoot being measured at many ISO  s at 0.0 percent. Exceptions are at ISO  160, where undershoot is 0.7 percent; at ISO  1600, where it is 0.3 percent; and at the two highest ISO  s (0.6 percent at ISO  25600, 1.1 percent at ISO  51200). The overshoot along high-contrast edges is stronger: for example, at ISO  160, it is 6.9 percent, 4.8 percent at ISO  1600, and 6.7 percent at ISO  3200.

Sharpening is more pronounced along low-contrast edges. For example, overshoot and undershoot (low contrast) are 10.0 percent and 5.8 percent, respectively, at ISO  160, and 7.7 percent and 3.7 percent at ISO  1600. At ISO  12800, both overshoot and undershoot are 0.0 percent along low contrast edges and also along high-contrast edges.

Visual noise is moderate. In Viewing Condition 1 (a modelled condition simulating enlargement to 100 percent on a monitor), the noise is visible even at the lowest ISO  s (e.g., score of 1.1 at ISO  160, 1.5 at ISO  800, 2.1 at ISO  6400). Visual noise would be most apparent in the darker tones.

In a postcard-sized print or on a mobile screen (Viewing Condition 2), noise would not be noticeable at the lower ISO  s. Only above ISO  3200 (score 1.0), would visual noise become apparent to an observer (e.g. score 1.2 at ISO  6400). The GH5 has slightly better visual noise production, with scores of 0.8 at ISO  3200 and 1.0 at ISO  6400.

A large print (simulated in Viewing Condition 3) made from an image shot by the GH5S at lower ISO  s, would not show visible noise, although the visual noise would become apparent at ISO  3200 (score 1.3). In comparISO  n, the GH5 also crossed the threshold to visual noise being noticeable (score 1.0) at ISO  3200 in Viewing Condition 3.

Dynamic range is above 8 f-stops at all non-extended ISO  s measured in the GH5S. The best dynamic range is at ISO  800, where it is 10.0 f-stops. At other ISO  s between the lowest, ISO  160, and ISO   3200, the dynamic range was measured at more than 9 f-stops. The GH5 also showed a dynamic range of more than 9 f-stops from ISO  400 to ISO  12800, and 10.0 at the lowest ISO   of 200.

Color reproduction is fairly good, with very few strong deviations from the original color: nearly all are reds. ∆E is very consistent and a little over 10 for nearly all ISO  s tested. Only at the highest non-extended ISO   of ISO  51200, is ∆E 11.3.

The automatic white balance at the lowest ISO   of ISO  160 is fairly good (1.7). White balance in ISO  s from ISO  400 to ISO  6400 is ranges from 2.5 (ISO  1600) up to 2.9 (ISO  800). At ISO  12800 and ISO  25600, the automatic white balance improves again slightly to 1.6 and 1.3, respectively. At the highest end of the ISO  s, the automatic white balance performs fairly poorly (e.g. 3.3 at the highest non-extended ISO   of ISO  51200). The predecessor GH5 showed better scores for automatic white balance: 0.8 at the lowest ISO   of 200, 0.6 at ISO  400, increasing to 1.0 at ISO  1600 and 1.4 at ISO  6400.

The Panasonic DC-GH5S takes 0.9 seconds to start up, compared to 1.2 seconds in the GH5.

The GH5S shoots 9.2 frames per second in the best JPEG format until the card is full. As there are two cards and the cards can be set to relay from one to another, it seems that there are many options for creating a lot of frames with this camera. The GH5S performs nearly as well in RAW, shooting 9.1 frames per second until the card is full. The GH5 was a tiny bit faster: it could shoot 10.0 frames per second in both RAW and JPEG, up to 71 frames (RAW) or the card becoming full (JPEG).

The total shooting time including autofocus in bright light takes 0.15 seconds, of which the autofocus takes 0.08 seconds. This is an improvement over the predecessor, which took 0.18 second to focus and shoot in bright light.

In low light, the autofocus in Live View took 0.12 second, for a total shooting time of 0.19 seconds. This is an improvement over the GH5, which took 0.32 seconds to focus and shoot in low light.

Video

It is said that the GH5S has been designed primarily for video, and this is borne out by the image quality results of frames grabbed from video clips filmed at low and high ISO  . The camera uses as much of its 10-megapixel sensor for video as it does for still images. Higher ISO   videos are very similar in quality to videos filmed at a lower ISO  . So, for example, resolution measurements of frames grabbed from a video captured at low ISO  , show a resolution of 1020 LP/PH (94 percent of the theoretical maximum). At high ISO  , the resolution is very similar at 982 LP/PH (91 percent of the theoretical maximum).

There is relatively little problem with texture: in high-contrast parts of the scene filmed, MTF50 was measured at 787 LP/PH at low ISO  , and 599 LP/PH at high ISO  . Artifacts were 27.9 percent at low ISO   and 35.5 percent at high ISO  . In low-contrast parts of the scene, low ISO   video stills showed 24.7 percent artifacts, while high ISO   frames contained 37.8 percent artifacts. MTF50 was 675 LP/PH at low ISO   and 590 at high ISO  .

Sharpening in video is mild: 4.4 percent overshoot and 2.6 percent undershoot at low ISO   along high-contrast edges, and 1.2 percent and 0.7 percent, respectively, at high ISO  . Along low-contrast edges, sharpening was 6.2 percent (overshoot) and 3.7 percent undershoot at low ISO, and 1.0 percent overshoot combined with 1.3 percent undershoot at high ISO.

Visual noise was below the threshold to be noticeable for Viewing Conditions 2 and 3 at both low and high ISO  . The noise would be slightly noticeable at both ISO s in Viewing Condition 1 (100 percent; scores of 1.4 (low ISO) and 1.6 (high ISO)).

Automatic white balance in video is excellent, measuring 0.1 at both low and high ISO  . Dynamic range is also excellent: ∆E measured 11.0 f-stops at low ISO   and 10.9 at high ISO.