Testing Arecont's 10MP Camera 2010

BE
Benros Emata
Published Nov 09, 2010 05:00 AM

With megapixel adoption surging, manufacturers are expanding their product offerings to even higher resolution levels. Up until this year, the highest resolution 'normal' megapixel cameras were 5MP. Beyond that, only a few multi-imager cameras existed at 8MP and a few super high-end 10MP+ cameras (at $8,000+).

Earlier this year, Arecont introduced a 10MP box camera that has an online price under $1,000 (without a lens). For the pixels and compared to what's previously available, that's pretty inexpensive.

In this report, we test the image quality and bandwidth requirements of the AV10005DN in various outdoor, nighttime, indoor, low-light and daytime scenarios. Furthermore, we perform the testing side by side against two 5MP cameras (the Axis P1347 and the StarDot SDH500BN) to understand what improvements are possible.

On a spec basis, a 5MP jump is really significant but what does that translate to in real world benefits? How much further can you see? How much more will this cost in terms of storage and bandwidth? Does it make sense to move to 10MP?

Additionally, in a Pro member note, we examine Arecont's QA and Reliability including a response from Arecont's CEO.

Recommendations

We found the best fit for the Arecont 10MP to be in evenly, well lit scenes of more than 30 foot wide FoV. In those areas, a modest increased in visible detail is possible. This will allow the 10MP to cover a wider area (up to 50%) than a 5MP.

However, the camera's ability to deliver improved visible detail was reduced by variances and drops in light levels. Moving from very evenly lit conditions to modestly sunny, somewhat reduced visible details (though still better than 5MP). However, night time scenes with street lights showed no quality benefits over 5MP and low lux scenes (1 lux or lower) were considerably worse).

Be careful about the increase in bandwidth/storage consumption. The increase was significant, especially outdoors or in high motion scenes where H.264 bandwidth consumption at 10MP was about 15 Mb/s. This could pose a serious network and storage cost premium (especially relative to the somewhat modest visible detail improvements).

Additionally, we had a number of hardware and firmware problems. We used beta firmware in this test to improve an imaging / iris opening problem. Despite using beta firmware (65151) (that improved the situation), a modest issue still exist. Finally, make sure your VMS supports the camera. As of today, according to Arecont, only Exacq supports the camera and you need the Oct 8th Exacq release (4.3.6.23123) to fix a streaming/video quality issue.

Key Findings

  • 10MP provides no appreciable increase in benefits under 20 feet wide) in comparsion with 5MP cameras
  • In ideal even lighting conditions, 10MP provides modest gains in visual clarity of objects in wider (e.g., 30 feet +) field of view applications in comparsion with 5MP cameras. This is likely to produce greater details of people and license plates at 10 - 20 feet farther than a 5MP camera.
  • During nighttime (outdoor) scenes, the 10MP provided little to no improvements in visible details.
  • In day outdoor high scene, Arecont's 10MP H.264 bandwidth averaged approximately 14 Mb/s or twice that of a 5MP camera's average (roughly 7 Mb/s)
  • In non-complex indoor scenes Arecont's 10MP H.264 bandwidth averaged 1.5 mbps - twice that of the 5MP camera's average rate of 750 kbps
  • Arecont's AV10005DN iris could not be fully opened or optimized for both day and night - on a bright clear day fully opening the iris produced highly over-exposed images (this occurred even after loading beta firmware that appreciably reduced the problem)
  • Under low-light conditions, even with the iris fully opened the AV10005DN performed poorly compared with the Axis P1347 5MP camera

Product Overview

The AV10005DN [link no longer available] is a 10 megapixel (3648x2752) D/N indoor IP camera that retails for $1390 (online price: $952). Note that the camera does not include a lens, and must be purchased as a separate item - CS mount.

Like many Arecont cameras the AV10005DN feature set is limited (e.g., no audio, no WDR, no auto-back-focus, etc.). The camera supports both H.264 and MJPEG compression modes and can achieve a maximum 6fps at full 10MP resolution. The AV10005DN provides an IR cut-filter, while the color only version, AV10005, does not.

The MSRP pricing for the two 5MP cameras used for comparison in this report:

For normalized price comparisons, note that the 5MP comparison models include a megapixel rated lens while the Arecont offering requires a separate purchase.

Download and review video clips [link no longer available] of the testing scenarios.

The AV10005DN was outfitted with the manufacturer recommended manual-iris Tamron F1.6 / 4.5 - 10mm lens for all tests in this report.

Daytime Image Quality Comparison (Outdoor)

In this video we provide a comparison between the AV10005DN, and two 5MP cameras by Stardot and Axis.  The modest resolution benefits of the 10MP Arecont camera are clearly demonstrated while digitally zooming into a far-away sign. The AV10005DN provides the best resolution amongst the 3 competitors.

Nighttime Image Quality Comparison (Outdoor)

This AV10005DN camera did not allow us to set the iris completely open during the bright daytime shot. In this video you'll see that the highly closed daytime iris setting results in poor exposure levels for the nighttime scene. We then manually open up the iris completely to show 'ideal' settings for the nighttime environment. The modest benefits in visual clarity observed during the daytime tests essentially 'disappear' during the night scene.

Daytime 'Overcast' Image Quality Comparison

In an overcast scene, visible details were maximized as lighting variations were muted. This short video demonstrates a quick image quality analysis during this overcast daytime scenario.  

Daytime Indoor Bandwidth Comparison

During MJPEG codec usage the AV10005 and Axis P1347 exhibited similar bandwidth profiles of ~15 mbps. However, at H.264 streaming modes the Arecont 10MP camera averaged 1.5 mbps, rougly twice that of the 5MP camera at 750 kbps.

Daytime Outdoor Bandwidth Comparison

In this busy outdoor daytime scenario, the AV10005 camera averaged 14 mbps. The 5MP camera averaged roughly half that at 7 mbps.

Nighttime Outdoor Bandwidth Comparison

In this video, we examine the bandwidth usage of the AV10005DN and Axis P1347 in an outdoor nighttime scenario. We discovered that the Arecont 10MP camera's bandwidth usage did not materially differ from its daytime usage upon the same outdoor intersection scene (~14 mbps at H.264), while the 5MP H.264 camera approximately halved its bandwidth usage at night (from ~7 mbps to ~3mbps).

Low Light Performance (1.0 Lux and 0.3 Lux)

In this video, we examine the low-light performance of the AV10005DN. Specification sheet for the camera lists 0.0 Lux night-mode operation, however, we observed a very under-exposed image at the 0.3 Lux level in our test.