Axis 33MP Camera Tested (Q3709)

Published Nov 30, 2015 05:00 AM

In 2012, Axis declared the megapixel race over.

Now, Axis has released a 33MP camera, the Q3709-PVE, one of the highest resolution cameras in the entire industry.

This release also is a major marker in the new race for multi-imager cameras, a high growth segment of the industry.

We bought a Q3709 and put it up against Arecont's SurroundVideo and Pelco Optera to see how it performs in full light and low light in the wide outdoor FOV shown below, along with VMS integration, bandwidth performance, and more.

Inside, we see how Axis stacks up and what impact this new camera will have on Arecont's historic dominance of the multi-imager space.

Summary and Market Impact

The Axis Q3709-PVE suffers from a number of significant flaws that limit its competitive positioning. The two most important are wasted pixels and tilted images coming from its 3 imager format, providing minimal increased details and coverage despite its much higher pixel count. The other key issue is the camera's 3 IP addresses which will require having to buy an additional 2 VMS licenses in most VMSes compared to other multi-imagers.

On the positive side, the image quality (night and day) is solid and bandwidth, despite the much higher pixel count, is, overall, not that far higher than competitors. Given this is an Axis product, and their much stronger reputation and track record than Arecont and Pelco, many will find this attractive, though they will have to temper that given the above noted limitations.

Key Findings

In our tests, the Axis Q3709-PVE delivered better details in full light and similar performance in low light to the Pelco 12MP Optera. Performance was all around better than Arecont's 12MP SurroundVideo models. However, relative PPF is not as high as one might expect given its 33MP total resolution, due to the use of three imagers instead of four. Horizontal PPF is only 30% higher than Arecont and Pelco, with a taller vertical FOV.

Additionally, the use of three imagers places the left and right imagers at a harsher angle than in four imager models, resulting in tilted images and more sky in the FOV, making images less usable and somewhat strange to view, as subjects appear tilted on the left and right sides.

For bandwidth, the Q3709-PVE had lower total bitrates than Arecont and Pelco multi imager cameras during the day, but significantly higher bandwidth at night, about double other models. Given the nearly 3x higher higher pixel count, this was competitively strong.

Finally, the Q3709 is expensive relative to other multi-imagers on the market, both on product price and VMS licensing cost. This is compounded by its use of three IP addresses means three camera licenses are required by VMSes (aside from Genetec), further adding cost compared to Arecont's SurroundVideo, which typically uses only one license.

Pricing

The Q3709-PVE sells for about $2,400 USD online. This is about 50% higher than Arecont and Pelco's 12MP multi-imager models, both ~$1,500-1,600 USD.

Field of View Comparison - Wasted Area / Tilted Images

Since it uses three imagers instead of four to cover the same width, the Q3709's vertical angle of view is wider/taller than the Arecont and Pelco models. This results in more area covered near the camera, as can be seen below. The Axis Q3709 is able to capture almost the entire length of the parking space in the near FOV, while Arecont is only able to capture ~1/2. 

Drawback: Off Level Images

However, using three imagers instead of two also requires the outer imagers to be tilted more harshly than in four camera models. This results in off-level images produced by the left and right imagers, even when the camera uses only shallow downtilt.

In this example, the center imager is set to ~20° downtilt, but the outer imagers are significantly off level, with a substantial portion of the image taken up by the sky, not seen in the outer Arecont imagers (second row).

We review these issues in more detail, using video examples in this video:

Physical Overview

The Q3709-PVE is large, even for an outdoor dome model, about the diameter of Axis' speeddomes, though not as tall. Additionally, it uses larger 1.5" mounts (the same mounts as Axis PTZs), adding to its overall size.

Note that all imagers move together, and cannot be angled differently, with degree markings provided next to the center imager for setting downtilt.

We review the physical features of the camera in this video:

Configuration Overview

For the most part, the Q3709's web interface is similar to other Axis models. However, there are a few things worth noting:

  • Three IP addresses used: The Q3709 uses an IP address for each imager, unlike other multi imager models we have tested. Users may connect to and configure the camera via any of these addresses. In order to add it to VMSes, it must be added three separate times, instead of only once as in other multi imager models. Here is an excerpt showing that from the camera's configuration:

  • Global/individual configuration: Most image settings are configurable either globally (for all three imagers) or on a per imager basis. This may be useful for tuning the image if the imagers are viewing drastically different scenes, but overall, settings are generally performed globally.
  • Alignment calibration: Users may manual adjust image alignment of the three imagers. We found that little, if any alignment was needed from the factory, only a few pixels tweaking to best line the FOV up.

We review these issues and others in this screencast:

VMS Integration

The number of VMS licenses required for the Q3709-PVE varies. Of the four major VMSes tested, only Genetec Security used a single license for all three channels, similar to the Arecont SurroundVideo line, and VMSes which fully integrate Pelco's Optera.

Most require one license per channel. For example, Milestone explicitly shows three licenses required for the camera in their supported devices page, which we confirmed in testing.

Exacq shows this as well, specifying one channel per license on their supported devices list. Other multi-channel devices, such as four channel encoders or the Arecont SurroundVideo, show 4 channels per license.

Image Quality Comparison

We tested the cameras outdoors during the day and night to see how performance compares. this image shows the full FOV of the Axis Q3709-PVE.

At close range, ~45' from the cameras, PPF is ~85 in the Q3709-PVE, with strong details of the subject and the chart legible to line 6/7, better than both the Arecont and Pelco cameras, which deliver ~60 PPF.

Moving back from the cameras, ~65', our subject is still fairly recognizable in the Q3709, though with a reduction in detail.

 Finally, at ~95' range, subject and chart details are unrecognizable.

Low Light Testing 

Using the same ranges at night, performance of the Q3709-PVE drops to near that of the Pelco Optera, though with moderately increased test chart legibility. Both perform much better than the Areont SurroundVideo, which essentially displays only digital noise.

Moving back from the cameras, to ~65' range, Axis/Pelco performance equalizes further, while the subject is nearly impossible to see in the Arecont camera.

 Finally, at ~95', both Axis and Pelco provide similar detection level video, while the Arecont camera does not capture the subject or chart at all.

Dark Testing

Turning exterior lights off, lux levels drop to ~0.1. None of the cameras were able to produce more than faint outlines of the subject/chart at this light level.

 

Video Samples

Users may download a sample from our nighttime testing (~260 MB .zip file) showing our subject walking through the scene:

 

Bandwidth Consumption

In well lit scenes, the Q3709-PVE had the lowest total bandwidth of cameras tested (all imagers combined), under 6 Mb/s, while Pelco and Arecont were near 8 Mb/s. 

However, in low light, bandwidth spiked drastically, to nearly 27 Mb/s, about twice as high as the Arecont SurroundVideo, and a 11 Mb/s increase over Optera. 

 

Test Parameters

The following settings were applied to all cameras:

  • H.264, 10 FPS, ~28 quantization was used (except Arecont, due a max 6 FPS framerate)
  • 1/30s maximum shutter speed

Other settings were left at defaults unless otherwise noted.

Comments are shown for subscribers only. Login or Join