Pentax K-3 II Performance

Very good overall performance for its class, but slow buffer clearing.

The Pentax K-3 II's startup time was a little slow for a DSLR, but switching from Play to Record and taking a shot was quite fast.

The Pentax K-3 II's full autofocus shutter lag when shooting the same target multiple times was good for a prosumer DSLR. The K-3 II required only 0.140 second for full AF using the Pentax 35mm f/2.8 Limited lens, about the same speed as the K-3. Manual focus shutter lag was even faster than autofocus as expected, at 0.094 second. When prefocused, shutter lag dropped a little further to 0.090 second which is a little slower than average for a prosumer DSLR however still a bit faster than the K-3.

The Pentax K-3 II's prefocused shutter lag time in Live View mode was 0.274 second, noticeably slower than when using the optical viewfinder, but not bad for a DSLR. Note that we no longer test full AF lag in Live View mode for DSLRs, because the lens used makes such a huge difference that comparing is pointless. We'll try to comment on real-world Live View AF performance in our our field reports.

To minimize the effect of different lens' focusing speed, we test AF-active shutter lag with the lens already set to the correct focal distance.

Shot-to-shot cycle time was so fast when shooting Large Best Quality JPEGs that it was difficult to measure accurately. (Fast single-shot cycle times depend a lot on the tester's agility and rhythm, so your results may vary.) When shooting RAW+JPEG files, the K-3 II's cycle time was a bit slower but still quite fast, at about 0.6 second. We no longer test just RAW files in single-shot mode, as it's usually somewhere in between JPEG and RAW+JPEG cycle time.

Continuous H mode speed was very good for a DSLR at between about 8.1 and 8.2 frames per second depending on the file type, very close to Pentax's 8.3 fps spec and about a frame per second faster than the K-3. (Note that we test cycle times at ISO 200 to be able to read our electronic timer display, while Ricoh specifies burst speed at ISO 100.) The K-3 II also offers Continuous M and L modes rated at 4.5 and 3.0 frames per second respectively, but we did not test those modes.

Buffer depth in Continuous H mode was excellent when shooting just JPEGs at over 50 Large Best Quality JPEG frames with no signs of slowing (the manufacturer claims 60 frames). When shooting RAW files, buffer depths were quite good at 22 RAW frames and 23 RAW+JPEG frames before the camera started slowing down.

Buffer clearing was quite slow, though, even with our fast 95MB/s UHS-I SDHC card, taking 22 seconds to clear a burst of 50 JPEGs, 26 seconds after a max-length burst of RAW files, and 37 seconds after a max-length RAW+JPEG burst.

Bottom line, the Pentax K-3 II's performance was generally very good for its class in our tests, though startup was a bit slow for a DSLR, as was prefocused shutter lag. But if the K-3 II's performance has an Achilles' heel, we'd say it's the slow buffer clearing after long bursts. Be sure to buy the fastest UHS-I card you can afford!

Battery Life
Below average battery life for a prosumer DSLR.

The Pentax K-3 II uses a custom rechargeable D-LI90 lithium-ion battery for power, and ships with a dedicated charger. CIPA-rated battery life when using the optical viewfinder is well below average for a prosumer DSLR when you consider the K-3 II has no built-in flash, and Pentax does not specify battery life for Live View mode, which will certainly be a lot lower. We strongly recommend you pick up a spare battery and keep it freshly charged and on-hand for extended outings, or consider purchasing the optional D-BG5 portrait grip which doubles battery life with a second D-LI90 battery installed.

The table above shows the number of shots the camera is capable of (on either a fresh set of disposable batteries or a fully-charged rechargeable battery as appropriate), based on CIPA battery-life and/or manufacturer standard test conditions.

(Interested readers can find an English translation of the CIPA DC-002 standards document here. (180K PDF document))