Sony 35mm Super Low Light Camera Tested

By: Ethan Ace, Published on Oct 31, 2016

Sony is making a big push for their SNCVB770 camera, touting its low light capabilities that they claim can deliver true color in pitch black darkness.

While the 'super low' light market segment is definitely growing, Sony's SNCVB770 has a dramatically larger 35mm imager than top low light cameras which generally are ~1/2". For example, here is the Sony SNCVB770 compared to one:

Does Sony's approach make a real difference? We tested the SNCVB770 vs top low light performers from previous tests, the Axis Q1635 and Hikvision Darkfighter as well as the Sony SNC-VM772R to see how well the SNCVB770 really performed head-to-head.

In this test report, we review our full results, including light levels from ~6.5 lux to below 0.01, with competitors in color, monochrome, and with IR on, as well as bitrate comparisons, configuration issues, and more.

*******

*** **** ***-***** ******** better *********** **** *** other ***** *** ***** camera ** **** ******, producing ****** ****** ** very *** *** ****** (below ~*.****), **** ***** detection ** *** **** subject ** ***'+, **** detection ********** ******* ** thermal ******* ** *** past *****. ********, ***** the***-***** ** **, ***** other *** *** ***** cameras *** ***** (** less), *** **** **** delivered **** ******* ****** the *** ** ****.

*******, ***** *** ******* key ********* ** ********, as ****:

**** *****:** **** $*,*** ********* street ***** (********* ***** lens *** ******* *******), the ***-***** ** ****** similarly ** ****** **** thermal ******* ******* ** visible ***** *******, **** $2000+ ****** **** ***********'* **-********* ****'* ******-******** ******* ** *** Hikvision *********** ****** (***** is *** $*,***).
************** ***********:******* *** ***** (*** VM772R) **** *** ******* any *********** ******* ** its *** ***** *****, coupled **** ******** ******* well ***** *** *** cap **** ** **** motion ******, *********** ********* were *********** ** *** tests, ******** ***** *******.
*** ***** **** ***** stream/WDR:**** ***** *****-****** ****, the ** ****** ** limited ** ** ***. Turning *** ** ******* FPS ** *. **** other ** ******* ** not **** ***** ***********, allowing ** *** ** the **** ****** ** both *****.
**** ***** **** *******:*** ***** **** * 12MP ****** *** *** be *** ** **** when ***** *****. *******, frame **** ** ******* to * ***, **** low **** *** **** open *****.
**** *********:*** *****'* ******** *** notably ****** **** **** common ** ******* (~* Mb/s ******* **. Mb/s ** ******), **** of ***** *** ******* smart ******.

*******, ** ******** **** low ***** ********* *********** is *** **** *** budget, *********, *** ********* are *** * *******, the ***** ** * strong ******.

*******

*** ***-***** *** ** MSRP ** $**** ***, not ********* **** ** housing. ***** ** **** price *** ****** *********, we ******** *** ****** price ** ** ~$*,*** USD ********* **** *** housing.

**** ** **** **** twice *** ***** ** even *** **** ********* standard ****** **** ** cameras, **** ** ************ **-******* (~$*,***)*******'* ***-****** (~$*,***).

Image ****** **** **********

*** ***** ***** ***** the **** ********** ******* the ***-*****'* **** ****** and * ******* ~*/*" sensor **** * ***** low ***** *****.

**** **** ****** **** is **** *** ****** in ******* ***********. **** comparison ** ***** **** for *********, ***** **** imager ** *********** ****** than ******* *******.

Physical ********

**** *******, *** ***** is ************* ****** **** typical *** *******, ********** with ***** * ***** lens. ***** **** **** and **** ***********, **** configuration ** ********* ******* to ***** *******, **** audio, */*, *****, *** PoE ** ******.

** **** ** *** physical ******** ** *** camera ** **** *****:

**** *******

*** ***** ** ********** with **** * ***** lenses. **** ****** ******** lens *******, *** *** E ***** **** *** be ****. **** **** the ****** **** *** support **** *********, **** as ********* *****, ****, stabilization, ***., **** **** back *****.

****** **** ****** ** size, **** ** *** comparison *****, ******* *** 35mm ***** **** *** 70-200mm *********. ***** ****** beware ** **** **** selecting ** ******* *******, to ****** *** ******** lens **** *** ** the ******** *******.

Test *****

** ****** ** ** outdoor ******* ***/***** **** exterior ****** ** *** off. ***** ****** ****** from ~*.*** **** ****** on **** *** ******* to ***** *.**** **** lights *** ** *** field. **** **** *** light **** ** *** right **** ** *** test *** ***** *** not **** **** *** exterior ** *** ********, but **** ****** ****** over **** * **** away. *** ************ **** this ***** ***** **** account.

Subject ******** (

** *** ******* *****, we ***** * ******* from **** ***' ** he ***** ******* *** cameras. ***** ****** ** this ***** **** ***, below *.**** ** *** field.

******** **** **** (~** MB .***)

** ********, **** ***** the ********** ** ** model, *** ******* *** not ******* ***** ***** ~200' *****, **** ******.

******** **** **** (~** MB .***)

Compared ** ******* *******

***** ***** **** *****, *** *****'* *** light ********* *********** ** similar ** ** ****** than ******* ******* ** longer ******. *** *******, the ********** ***** ***** both ******* **** ** side **** *** ******* at ~***' *****, **** clearly *******.

*******, ****** *** ******, at ~***' ***** *** closer, *** ***** ******** details *** **** ** the *******, **** ***** ideas ** ********, *****, etc.

Exterior ****** ***, *** ******* "**** ****"

** ***** ****** **** all ******* *** ** their "**** ****" ********: monochrome ** ***-** ******* and ** ** ** the ***-******. **** **** the ***-***** ** ***** only *** *** ** sensitive.

***** ** *** *******, with *** **** **** high (~***) *** ***-***** produces ********** ****** ******* of *** ******* **** other *******, ********* *** integrated ** ******. *******, the **** ***** *** more ******* ** ******, including *** ***** **** and ********* *******.

**** **** *** ** noise *** *********** *********, none ** *** ******* produced *** ********* ******** images ******** ** *** levels **** ****.

** ~** ***, *** VB770 ******** ** ******* line ** *** **** chart, *** **** ** any ***** *******. *** subject ** ********* ******** in **** *** **** 4K ******, *** ****** be **** ** *** 1080p ***** *** ***** models *** ** ********* noise.

*******, ** ~** *** (~100' ********), ** ******* are ********* ** *** camera, *** *** ***** still ****** ******* *** subject.

Exterior ****** ***, *** ******* *****

*** **********, ** ********* the **** **** **** all ******* ** ***** mode. *********** ** *** SNC-VM772R *** **** ******* reduced, **** *** ******* unseen *** ** ******* noise. *** **** *** Hikvision ***** */*" ****** provide ********* ** *** subject *** *-* ***** of **** ***** **********.

** ~** ***, *** subject ** ************ ** all ******* ****** *** SNC-VB770.

Exterior ****** **, ~*.*-*.***

** **** ****** **** exterior ****** **, **** lux ******* **** ~*.*** near *** ******** ** ~0.5lx ** ~***' ********.

** **** ***** *****, the ******* ** ****** washed *** ** *** SNC-VB770, ****** ******* ** the ********* *** ****** cameras. **** *** **** Q1635 **** *** **** out *** ******* ** test *****, ********* *** best *******.

****** ****, ** ~** PPF, *** ***-***** ******** the **** ******* ** the **** ***** *** rough ****** ******* ** the *******, *** **** in ***** *******.

*******, ** ~** ***, the ***** ***** ******** some ****** ******* (**** tone, ****** ****) *** one ******* **** ** the **** *****.

Daytime ***********

** * **** *** daytime *****, *** ***** provides *** ******** ******* of *** ****** ****** at *** ******.

******** ** ~*** ***, details *** ********* **** in *** *******, ********** both ** ******, ****** finder **** ***** **** is ******** **** ******* in *** *****.

** ~** ***, ******* in *** *** *** 772R **** ****, ***** they *** ******* ** the ***** ******, ********** the **** *****. ******* text ** *** ***** is ***** **** ******* in *** ***.

*******, ** ~** ***, the ***** ***** ******** recognizable ****** ******* ** the *******, ****** **** the ****. *** **** chart ** ***** ******* to **** *, ** well, ***** ** ** washed *** **** **** one ** *** ****.

Codec ********

** **** ***** ***** 4K *****, *** ***-******, the ***** **** *** include **** *** *********. Instead, ***** "***" ************** includes * "******" ******* and ***. *******, ** our *****, ****** ******* had **** ******* ****** on *********, **** ****** bitrates ********* ***** **** target, ****** ********* **** higher **** ***** ** cameras **** **** ** motion ** *** *****.

********, *** ****** ******* to ******** *********** ******* bitrates ********* ************* ***** than *** *** *** in *** *** **, resulting ** *********** *********** and ******* *******. ******* of ****, ** ********** images, "** ***" **** not **** *** **** as "** ***" ** typical *******.

** **** *** ******'* "VBR" ************** ** ****** in **** *****:

Bandwidth **********

*** ***-*****'* ******** *** slightly ****** **** ****'* older ** *****, *** SNC-VM772R, **** *** *** night. ***** ******** ******** decrease ******** ** *****, most ****** *** ** the ******* ** ****/****/***** details **** ****** *** day.

**** **** **** ** significantly ****** **** **** current ** *******, ********** as***** ************ **** ******. *** example, ******* ***** ** cameras ** ******** *.*** ******* ****, *** ******** **** under * **/* ** an ******* ******* *****, and **** *** ******* P *** ********* ******* were **** *.* **/* at *****.

*******, ** ****** ** noted **** ***** **** the ******* *** *** VB770 *** ******** ** see **** ******** ** bandwidth/storage ** * ***** concern, *** **** *** unlikely ** ** ******** in ***** **********, ****** less ********* ******.

Streaming ***********

*** ***-***** ******* *** streaming *********** ***** ** the ***-******, *** *** typical ** ***** ** models:

********* *****

*** ****** ****** *** 4K ********* *****, ****** and *****-******:

****** **** ******:** **** ****, *** camera ******* *** ** stream, ** ** ** FPS, **** ** ***-*******. This ***** *****-********* **********, requiring ***** ** ******* the **** ********** ****** regardless ** ****** ****, which ********* ********* *** CPU *****.
******** *******:** **** ****, ** to ***** ******* *** be **********, **** *** first ** ** ** resolution, *** ******* ** 10 ***. ****** *** third ******* *** ** 1080p *** ***, ************. No ** *** ********* is ******** ** **** mode.

*** **** *********

**** ******* *** **, framerate ** ******* ** 50%, ** * *** of ** *** ** single ****** **** *** only * **** *****-*********. Note **** *** **** not ******** ** **** mode (****** *** ******* is ***** ***** ** the *** **).

** ****** *** ****** above ** **** *****:

Test **********

******* **** ****** ***** default ********, **** *** following ******** ******* ** all *******:

******* ***** *** *** to * ******* ** 1/30s
*.***, ** ***
**** ******* **** *** to ****** ******* ** Mb/s *** *** ******* 32 **/*
***** ******* *** ** ~Q28, ****** ***** ****** were *******

*** ********* ******** ******** were ****:

**** ***-*****: *.*.*
**** *****: *.**.*.*
********* **-***********-*: **.*.*
**** ***-******: *.*.*

Comments (25)

Undisclosed Manufacturer #1

11/01/16 12:44pm

I´m pretty sure the mini usb on the side is actually HDMI.

When you test at 60ppf are the cameras at different ranges from the chart depending on resolution and lens.

Ethan Ace

IPVM | In reply to Undisclosed Manufacturer #1 | 11/01/16 01:17pm

Oh that was me purely misspeaking. It is mini HDMI.

And yes if we test multiple cameras with fixed fields of view, we stagger them at different distances. In this case, everything but the Sony 770 was varifocal, though, and we were able to match its FOV with the other cameras.

Undisclosed #2

In reply to Ethan Ace | 11/01/16 02:37pm

In this case, everything but the Sony 770 was varifocal....

Did you use the f1.4/35mm?

Undisclosed Manufacturer #1

In reply to Ethan Ace | 11/03/16 08:54pm

Ok, thanks Ethan for clarifying. that means that horizontal res is the same in the FHD cameras but they display a much smaller view or need to be at a closer distance compared to the 4k ones. I would say the main benefit is the resolution and the light sensitivity combined that you would not find in other 4k cameras from for example Axis and Hik.

Undisclosed #2

In reply to Undisclosed Manufacturer #1 | 11/03/16 10:00pm

...that means that horizontal res is the same in the FHD cameras but they display a much smaller view or need to be at a closer distance compared to the 4k ones.

IMHO, he is say that they are at the same distance and have the same FOV. He did this by adjusting the varifocal lenses of the other cameras to match the Sony 770.

Undisclosed Manufacturer #1

In reply to Undisclosed #2 | 12/22/16 04:48pm

Ok, but my point is that if you have a 4k camera and compare to a FHD either the fow is doubled in the 4k or the ppf should be doubled. So if all cameras are 60ppf when stated so the hfow must be larger meaning they cover 4 times the area of the FHD but not providing more pixels on the chart. So IMHO the VB770 is providing better detail and covering a large area at the same time. Unless my brain is totally fried.

Undisclosed #2

In reply to Undisclosed Manufacturer #1 | 12/22/16 04:57pm

Ok, but my point is that if you have a 4k camera and compare to a FHD either the fow is doubled in the 4k....

It may be, obviously you're seeing just a crop of the fov.

Undisclosed Manufacturer #1

In reply to Undisclosed #2 | 12/22/16 05:04pm

Obviously :)

Merry Christmas mate!

Ethan Ace

IPVM | In reply to Undisclosed Manufacturer #1 | 12/22/16 06:02pm

I realize the reason for the confusion now is due to the way the images are labeled. They give only the PPF for the 4K camera instead of showing it separately for 4K and 1080p both. We will fix that and show PPF for both. Sorry for the confusion.

The FOV width was the same in all cameras, so 4K PPF is double 1080p (so 90 instead of 180, 30 instead of 60, and 20 instead of 40). We always standardize FOV, not PPF, unless explicitly mentioned otherwise.

Undisclosed #2

In reply to Ethan Ace | 12/22/16 06:22pm

We don't use crops of the FOV for these comparisons.

Aren't these cropped to be the same size/aspect?

Marc Pichaud

In reply to Undisclosed Manufacturer #1 | 11/01/16 01:58pm

Hi "undisclosed" Not Sure USB is here for that reason.

Sony includes this in most new camera for a Wifi settings through Smarphone focal

Your put a Wifi dongle the time you need to Pan tilt zoom ...and then remove it

Ethan Ace

IPVM | In reply to Marc Pichaud | 11/01/16 02:39pm

There's a USB port on the back for that reason. The side port is definitely HDMI.

Jordi Ibero

Imago IP Solutions
11/02/16 08:03am

Can you make a comparison between Avigilon 8L-H4PRO, 24 or 30L-H4PRO (full frame) and this Sony?? It would be interesting

Ethan Ace

IPVM | In reply to Jordi Ibero | 11/02/16 03:47pm

We have an 8L-H4PRO on the way. I'd expect that test in the next couple of weeks.

We may look at the higher resolutions later on, but the direct comparison to this Sony camera and the upcoming Axis model next year is the priority.

Michael Miller

11/02/16 11:35am

What VMS platforms have you tested with the Sony? I would love to see on that can handle 30+ of these.

Undisclosed #2

In reply to Michael Miller | 11/02/16 02:24pm

...that can handle 30+ of these.

MSRP: $210,000

Ethan Ace

IPVM | In reply to Michael Miller | 11/02/16 03:53pm

Mainly Exacq. I added it to Genetec and Milestone just to see if it worked (it did).

When you say handle 30+, do you mean recording or viewing?

From a recording standpoint, the test Exacq machine recording this camera currently only has it enabled and the Exacq core service CPU usage is under 1%. So extending that out, it's possible it could manage 30.

Viewing is another story altogether. It does multistream (4 sub streams), so that's an option for viewing multiple cameras at a time.

Also curious, where would you see 30 of these in use at one site?

Michael Miller

In reply to Ethan Ace | 11/02/16 04:03pm

I should have been more specific. I met displaying in the client software. I would like to see the CPU, GPU and Memory usage from the clients. 1 camera and say 4 or more at the same time.

We have a number of customers with 10+ PRO cameras on one system one with ~30 PRO cameras. I can display all of them on my Surface PRO 3 without issue and without having to spend time setting up multiple streams.

Undisclosed #2

In reply to Michael Miller | 11/03/16 01:02am

...without having to spend time setting up multiple streams.

Some VMSes are a breeze to setup for multistreaming, and cutover automatically as needed.

On what VMSes have you had to spend a lot of time setting it up?

Michael Miller

In reply to Undisclosed #2 | 11/03/16 01:07am

What VMS's are a breeze? I would think the one with zero setup would be the least time-consuming.

Undisclosed #2

In reply to Michael Miller | 11/03/16 02:51am

I would think the one with zero setup would be the least time-consuming...

Yes, your right.

Avigilon's VMS is the clear winner when you need to handle 30+ streams coming from Sony 4k low-light cameras, because you don't have to spend time choosing secondary streams during camera setup.

good point.

Jordi Ibero

Imago IP Solutions
In reply to Ethan Ace | 11/04/16 06:17am

The Spanish football league are plenty of 30&16Mx cameras to monitor the stadium. In some cases more than 50. In this case I think the only option is Avigilon.

I think, the problem for this Sony will be the same that faced Avigilon when they first release a 12MPx camera and they don't have their own VMS. No one can handle thear cams, so they needed to develop his own VMS.

Undisclosed Manufacturer #3

11/02/16 04:17pm

Besides the price, it’s pretty dam impressive to see the lowlight performance.. woow

It is VERY dark at 0.01 lux.

The future is not looking so dark anymore, but pricy :-)

Nick Ames

11/04/16 03:28am

Is it correct that the Hikvision 6026 was tested with stock lens(~f1.4) and Sony VB770 tested with Sony E-mount 35mm f1.4? It would be interesting to add the f0.95 lens option for Hikvision, and for the Sony, using the Mitakon E-mount 50mm f0.95. The latter sells for approximately $820. There is a big difference of light transmission between f0.95 and f1.4.

Undisclosed #2

In reply to Nick Ames | 11/04/16 04:45am

They have tested the 6026 with the "darkeye" f0.95 lens before, so it may have been used here.

Read this IPVM report for free.

This article is part of IPVM's 6,534 reports, 880 tests and is only available to members. To get a one-time preview of our work, enter your work email to access the full article.