Analog vs HD Night Shootout

By: Benros Emata, Published on Mar 26, 2011

The whole point of megapixel is better image quality but what if megapixel was actually worse than analog at night? In the consumer world, maybe this is acceptable but in surveillance, most cameras need to 'see' 24 / 7.

[WARNING: This report is OUT OF DATE and should only be used for historical review. Megapixel / HD camera low light performance has improved significantly since 2011. See: Ranking IP Camera Low Light Performance]

In this report, we share our test results of head to head comparisons between analog, SD IP and Megapixel cameras. We asked a number of analog centric integrators, including the folks at CCTVForum.com, for their recommendations on their favorite analog cameras for use in low light scenes. Based on their recommendations, we bought a Bosch Dinion [link no longer available] and CNB Mona Lisa and set them up side by side with megapixel cameras from our previous tests. We then tested them in a series of real world scenes.

Here is a preview of our test :

Inside the Pro section, we dig through the tradeoffs and key findings revealed in the tests. For background, make sure you understand the issues in exposure setting adjustments that we explained in detail in our 'How Exposure Impacts Low Light Surveillance.'

Key Findings

We note 3 key findings:

  • In a parking lot lit with street lamps (3 to 5 lux), megapixel cameras provided only minimal improvements in image detail over analog - far less than the number of pixels would expect to provide.
  • In a field with no direct street lamps and only ambient light (~0.5 lux), analog cameras provided far more detail than the megapixel ones. The analog cameras provided outlines of human subjects at a 32 ft width FoV while the megapixel cameras were so dark that it was almost impossible to make out the human subjects.
  • In under 1 lux scenes, the only way the megapixel cameras could provide similar image detail as the analog cameras was to significantly lengthen the exposure setting. However, of course, this increase motion blur - a major negative tradeoff.

In the video below, we show the key findings by analyzing the recorded test samples:

Get Notified of Video Surveillance Breaking News
Get Notified of Video Surveillance Breaking News

Recommendations

With these key findings recognized, here are our recommendations:

  • Using megapixel in an outdoor scene with street lighting is likely still superior to analog (or SD IP). At night, minimal benefits are gained and during the day, significant benefits can be achieved. However, at night, megapixel cameras essentially waste bandwidth. If possible with one's VMS system, consider dropping the resolution level during the night.
  • Using megapixel in a scene with no direct lighting at night is problematic. If the scene must be surveilled in these conditions, either (1) lengthen the shutter, (2) add IR illumination or (3) use SD instead of megapixel. All of these have important drawbacks. Do not lengthen the shutter beyond 1/8s if you need to capture clear details of moving subjects (i.e., if you are not confident that an intruder will stop in the scene). Use integrated IR only if the subject is close to the camera (less than 40 feet). Otherwise, use add on IR (though expect to pay an additional $300+). [Note: for more, see our IR comparison report.] Finally, use SD only if you can accept less detail or the use of more cameras during the day.

Video Clips Download (ZIP File)

Review our Analog vs Megapixel Sample videos zip file [link no longer available] (~230 MB) containing all the video clips and images generated during our testing. The clips are embedded inside an ExacqVision player and simply needs to be double clicked to view. In the Exacq player's Options menu we recommend enabling 'Show Camera Names' to show the names of the cameras on-screen.

The clips and images are from our Outdoor test (presented in this report) and Indoor (preliminary) tests as well as alternative setting tests.

Methodology

Three (3) Standard Definition, and two (2) Megapixel cameras are included in our low-light performance testing.

Standard Definition Cameras:

  • CNB BBM-24F - 1/3" CCD; F1.0 Computar; 600TVL (650TVL B/W); 0.005 Lux (B/W); online $120
  • Bosch LTC0498 [link no longer available] - 1/3" CCD; F1.0 Computar; 540TVL; 0.98 Lux (100 IRE, B/W); online $400
  • Bosch NBN-498-21P [link no longer available] - 1/3" CCD; F1.3 Bosch, 4CIF (704x480); 0.98 Lux (100 IRE, B/W); online $700

Megapixel Cameras (IP):

  • Sanyo HD4000 [link no longer available] - 1/2.5" CMOS; F1.8 integrated lens; 1920x1080; 0.1 Lux (F1.8, b/w, gain: High); online $950
  • Pelco Sarix IXE20DN [link no longer available] - 1/3" CMOS; F1.2 Pelco; 1280x720; 0.25 Lux (Mono, 33 ms); online $1,300

All SD cameras were analog, except for the Bosch NBN-498. A Sony SNTEP104 encoder was used to stream the analog cameras to our ExacqVision VMS. During testing default settings were used on the Sony encoder including a 720x480 resolution. The two MP cameras were configured for 720p resolution during all tests.

 

Here are the key default settings relevant to low light performance for each camera:

  • CNB BBM-24F - 1/60s shutter; AGC on '15'
  • Bosch LTC0498 - 1/60s shutter; SensUp 4x; AGC on '15'
  • Bosch NBN-498-21P -  1/60s shutter; SensUp 4x; AGC on '15'
  • Sanyo HD4000 - 1/30s shutter; AGC on 'normal'
  • Pelco Sarix IXE20DN - 1/8s shutter; (no gain setting in interface)

In the video below we examine each camera's admin interface focusing on exposure and gain settings, showing the defaults and available ranges for these low-light optimizations.

Scenarios/Applications

Two (2) nighttime/low-light test scenarios were devised that correspond to 'real-world' video surveillance applications:

  • Artificially Lit Area (Parking Lot); 3 to 5 Lux
  • No Light (Open Field) Area; ~0.5 Lux

In both primary scenarios two human subjects provide various examples of motion. The human subjects are positioned at a location within scene corresponding to a 32 foot wide FoV. As such the SD cameras provided roughly 20 pixels/ft. and the MP cameras provided roughly double the SD or 40 pix/ft. at the human subjects.

Tests

In the first Parking Lot scenario two (2) key tests were performed:

  • All 5 test cameras at default shutter and gain settings. This meant that the SD cameras had either 1/30s or 1/60s shutter speeds, the MP Sanyo had 1/30s and the MP Pelco had 1/8s shutter speeds.
  • MP Pelco camera configured for 1/30s shutter speed to normalize with the default 1/30s or 1/60s shutter speeds of the SD cameras.

The Open Field scenario featured four (4) key tests:

  • All 5 test cameras at default shutter and gain settings. This meant that the SD cameras had either 1/30s or 1/60s shutter speeds, the MP Sanyo had 1/30s and the MP Pelco had 1/8s shutter speeds.
  • MP Pelco camera configured for 1/30s shutter speed to normalize with the default 1/30s or 1/60s shutter speeds of the SD cameras.
  • MP Pelco camera configured for 1/250s shutter speed
  • MP Pelco camera configured for 1/500s shutter speed

We considered other MP cameras as well. For example, we considered the Axis 5MP P1347, and the Sony HD CH140. Due to various technical issues we decided to exclude them from our main study. However, it is key to note that their low-light performance levels were similar to the Sanyo and Pelco MP cameras included in the 'main' study.

Testing Issues

The Sanyo HD4000 featured in this report presented several issues during our testing and analysis. The first issue involved an inability to set the correct time on the camera. This became an issue when searching its archived video on our Exacqvision VMS. According to Exacq this is a potential problem when cameras and VMSes are not synced up, since the VMS will try to 'honor' the camera's time setting.  Because of this issue we could not export clips from Sanyo along with the other cameras in synchronous mode.

Another issue we faced with the Sanyo HD4000 was the inability to configure a slow shutter setting. Sanyo calls this a 'long' shutter, and it is a feature of the HD4000. However, we had difficulty in configuring the 'long' shutter setting reliably and excluded it from the study. Thus, the default setting of 1/30s was used throughout all tests.

If cameras shipped with lenses we used them in testing as it represents what users are likely to use in real world scenarios. The three IP cameras shipped with lenses, while the two analog cameras not coming with lenses were fitted with separately purchased Computar (F1.0, AI, varifocal) lenses.

5 reports cite this report:

Multi-Megapixel Deployment Statistics on Jan 01, 2012
Megapixel camera resolution increases regularly, with manufacturers pushing...
Directory of Camera Shootout Series on Dec 27, 2011
The following directory lists all of our camera shootouts. These shootouts...
CNB Network Camera Review on Sep 08, 2011
Korean surveillance manufacturer CNB has historically been one of the more...
ACTi's 4MP Price/Performance Analysis on Aug 30, 2011
Over the past year, Taiwanese megapixel manufacturer ACTi released a line of...
Color Versus Day/Night Shootout on Apr 16, 2011
Most people prefer color over black and white but when it gets dark out,...
Comments : Members only. Login. or Join.

Related Reports

Video Surveillance Cameras 101 on Feb 25, 2020
Cameras come in many shapes, sizes and specifications. This 101 examines the...
Video Analytics 101 on Mar 16, 2020
This guide teaches the fundamentals of video surveillance...
Integrated IR Camera Shootout 2020 - Avigilon, Axis, Bosch, Dahua, Hanwha, Hikvision, Panasonic, Uniview, Vivotek on Jan 30, 2020
The best and worst cameras tested in this IPVM shootout showed major...
Video Surveillance History on May 06, 2020
The video surveillance market has changed significantly since 2000, going...
New Axis M30 Cameras Tested on Mar 26, 2020
Axis has released a new generation of, for them, relatively low cost M30...
Cheap Camera Problems at Night on Feb 19, 2020
Cheap cameras generally have problems at night, despite the common perception...
AI/Smart Camera Tutorial on Feb 20, 2020
Cameras with video analytics, sometimes called 'Smart' camera or 'AI'...
Uniview Deep Learning Camera Tested on Jul 14, 2020
Uniview's intrusion analytics have performed poorly in our shootouts. Now,...
Mobotix 7 Line Camera Tested on Mar 12, 2020
Mobotix is attempting a turn-around, struggling for years, then releasing the...
TVT / InVid White Light Camera Tested Vs Hikvision ColorVu on Mar 18, 2020
With mega China manufacturers Dahua and Hikvision facing both bans and human...
Vivotek LPR Camera Tested on Apr 15, 2020
Vivotek has historically sold license plate capture cameras but not LPR. Now,...
Sony Launches AI Camera Sensors on May 18, 2020
Weeks after exiting the branded video surveillance business, Sony is making a...
Multi-Imagers Added to IPVM Camera Calculator on Jun 15, 2020
IPVM has added multi-imager cameras to the Camera Calculator, supporting this...
Dahua, Hikvision, ZKTeco Face Mask Detection Shootout on Jun 19, 2020
Temperature tablets with face mask detection are one of the hottest trends in...
Video Surveillance Trends 101 on Apr 01, 2020
This report examines major industry factors and how they could impact video...

Recent Reports

VSaaS Will Hurt Integrators on Aug 06, 2020
VSaaS will hurt integrators, there is no question about that. How much...
Dogs For Coronavirus Screening Examined on Aug 06, 2020
While thermal temperature screening is the surveillance industry's most...
ADT Slides Back, Disappointing Results, Poor Commercial Performance on Aug 06, 2020
While ADT had an incredible start to the week, driven by the Google...
AHJ / Authority Having Jurisdiction Tutorial on Aug 06, 2020
One of the most powerful yet often underappreciated characters in all of the...
SIA Coaches Sellers on NDAA 889B Blacklist Workarounds on Aug 05, 2020
Last month SIA demanded that NDAA 899B "must be delayed". Now that they have...
ADI Returns To Growth, Back To 'Pre-COVID Levels' on Aug 05, 2020
While ADI was hit hard in April, with revenue declining 21%, the company's...
Exposing Fever Tablet Suppliers and 40+ Relabelers on Aug 05, 2020
IPVM has found 40+ USA and EU companies relabeling fever tablets designed,...
Indian Government Restricts PRC Manufacturers From Public Projects on Aug 04, 2020
In a move that mirrors the U.S. government’s ban on Dahua and Hikvision...
Directory of 201 "Fever" Camera Suppliers on Aug 04, 2020
This directory provides a list of "Fever" scanning thermal camera providers...
Face Masks Increase Face Recognition Errors Says NIST on Aug 04, 2020
COVID-19 has led to widespread facemask use, which as IPVM testing has shown...
Dahua Loses Australian Medical Device Approval on Aug 04, 2020
Dahua has cancelled its medical device registration after "discussions" with...
Google Invests in ADT, ADT Stock Soars on Aug 03, 2020
Google has announced a $450 million investment in the Florida-based security...
US Startup Fever Inspect Examined on Aug 03, 2020
Undoubtedly late to fever cameras, this US company, Fever Inspect, led by a...
Motorola Solutions Acquires Pelco on Aug 03, 2020
Motorola Solutions has acquired Pelco, pledging to bring blue back and make...
False: Verkada: "If You Want To Remote View Your Cameras You Need To Punch Holes In Your Firewall" on Jul 31, 2020
Verkada falsely declared to “3,000+ customers”, “300 school districts”, and...