Axis vs Sony Super Size Shootout (Q1659 vs SNC-VB770)

By Ethan Ace, Published Sep 11, 2017, 10:04am EDT

Super low light, super sized sensor cameras are a growing trend. In the past year, 2 of the most notable entrants for these IP cameras have been Axis and Sony.

IPVM has individually tested the Axis Q1659 and tested the Sony SNC-VB770.

Now, we are pitting the 2 cameras against each other, as shown below:

We shot these cameras out, testing:

  • Daytime image quality
  • Low light (<1lx) image quality
  • VMS support
  • Multi-streaming support
  • Bandwidth/smart codecs
  • Camera application support

Executive *******

 

**** ******* ********* *** better *** ***** *********** than **** *** ************ 1/3" ** */*" *** light *******. *******, ** darkness, *** **** ********* ************* ******* details **** *** ****.

***** ******* *** ***** performance ** *** *** factor, *** **** ***-***** ** * better ****** **** *** Axis *****.

*******, ** *** ***** has ******** ***** **** lighting, *** **** ***** has ******* ********** ********* ******** increases ** ******* ** longer ****** ** **** as ***** ******, **** lower ******** ** **** scenes (******* * ~**** increase ** **********).

************, *** ***** ****** some **** ****** ************ to ** *** ** board, **** ******* **** *********,******* ************, ***., *** ******** on *** **** *****. 

*** ******* ******** ** both ******* ** ****** to ** *****, ** both ******* *** ~** more ********* **** ************ high-end *******, **** *** price ******* ********* * combination ** ****** ********** and ******** *** ***** performance **** ** *** available ** ***** ************ cameras.

*******

*** **** ***** ** priced ********** ***** **** the ***** ** **** lens *******, *** ********* on ****** *** *********/*** F-stop *******. **** ***** pricing ****** ********* ** which **** ** ****, from ~$*,*** *** *** the ****/**.* ***** ******* ~$7,200 *** *** **-*****/**.* lens. *** ***-***** ***** for ***** $*,*** *** ******, plus ** ********** $*,***-*,*** depending ** **** ******.

******** ********

**** *** **** ***** and **** *** *** much larger **** ******* *** cameras. *** ***** ** slightly ********, *** ****** than *** *****, ****** it ********** ****** *******. Note **** **** **** and **** ********* ****** outdoor ******** *** ***** cameras, ** ******* *** camera ******** *** *** fit *** ****** *** lens ** **** * large ****** ***** ****** to *** *** ****** field ** ****.

*** **** ***** ** the ***** ******** ******* connects *** ********, *****, SD ****, *****, ***. The **** ******* ******** is ** *** (***** form ****** *********) **** which *** ** **** for ***** ***********.

Sony ****** ******

*** ***** **** * larger **** ****** **** the ****' ***-*, ********** for ** ***** **** of *** ****** *** light ***********. *** *** are ***** ***** **** by ****.

20MP **. **

*** **** ***** ******* at *********/****, **. ** in *** *****. **** results ** ~*,*** ********** pixels ************, ** **** as * ****** *:* aspect ***** ***** *** better *** **** ****** than *** ***** **:* ratio. **** **** *** Q1659 ** **** ******* of ****, ** ** the *****.

Full ***** *********** *******

** ****** ******* *********** in ** ******* ******* lot/field *****, ***** *****:

** ** ******* ******* test *****, *** ***** provided image ******* ******* ** the ***** ** * ~60' ***** ** ****, shown *****. *** ***** was ******** **** *****, with **** ********/***** *********, but ******* ******* *** practically *** ****, ******* a ~** *** ********* in *** *****.

** ****** ******** (~***'/~***' HFOV), *** ***** ******** a ******** ******** ** details ******** ** *** VB770, **** *** ******* slightly *******, *** **** legible ** **** *, vs. **** * ****** totally ******* **** ** the *****.

Daytime **** *******: **** *********

****** ** * **** intersection **** *****, *** Q1659 ******** ********** ******** in ******* *********. *** example, ******* ****** **** clearly ******* ** **** in *** ****** ** the ************ ** *** Q1659, *** ***** ** the *****. ** **** longer ***** (~***'), * banner ** ******* ******* in *** *****, **** truer ******, *** ****** be **** ** *** VB770, ***** ******* ********** desaturated.

*ony ****** *** ***** ***********

** *** *****, *** SNC-VB770 *** **** ******** in ********* ****** **** the **** *****. *** example, ** ***** **** no ******** ******** (~*.** lux), *** ***-***** ******** rought ******* ** *** subject ** **** ** 3 ******* ***** ** the **** *****, ***** the ***** ******** **** detection, ** *******.

*******, ** ~**', *** SNC-VB770 ******** ********* ** the ******* ***** *** he ****** **** ******* noise/background ** *** *****.

Both ******* ****** **** ******* ***** *** *****

***** *** ***** ******** worse ** *** ***** than *** **** *****, not **** ** ** superior ** ******* ***** low ***** ******, **** using */*" *******. *** example, *** ***** ***** show ******** **** **** tests ** *** **** Q1659 *** **** ***** against */*" *** */*" super *** ***** ******.

******** *** ******* ******* in **** *** **** and **** ****** **** rough ******* ** *** VB770, but ** ******* ** detection ** *** ******** super *** ***** *****.

Axis ****** ************

*** ***** ** ******* of ******* **** **** camera ************ (****) ** board, **** ********* ************************ *********. *** ***** **** also ** ********** **** the **** ***** ***** (Fence *****, ****** *****, etc.) **** ******** ** updated ** *.**+ ** the **** *** ******. The ***-***** ******** ****'* DEPA ******** ********* (*** *** ****) *** ***** ************ cannot ** *****.

Bandwidth **********

******* ** *** ********* of *********, *** *****'* bitrates **** ***** ** all ****** ****** ********* still than *** ***-*****, ******* its **** ****** ***** count. **** **** ******-***** **** *** *** a **** *** **************, *** ******* **** target/max ********, **** ** user ******* ** ***********. This **** **** * target ******* ** ** Mb/s *** *** ** 32 **/* ***** ** recommendations **** ****.

VMS/Multistreaming *******

*****-********* *** ****** **** Avigilon ******* ******, *****, Genetec ******** ******, *** Milestone ********.

*** ***** ********* ** least *** ******* ** most ***** ****** (*** but ********), *** ** many ** * *.*** streams ** ******* (***** below) *** *********. *** VB770 ******** ** ** five *******, *** *** support ******, **** ***** and ********* ****** *** five, *** ******* *** Avigilon **** *********** *** main ******. ********* ** a ******* ** ** FPS ** *** ** stream. ******* *** ** further **** **** ********* in ****.

*** **** ******* ** specifics ** *****-****** *******, see *** ********** ***** of *** **** ***** *** **** *****.

Test **********

*** ****** **** *** images ** **** ****** were:

  • **** *****:***** **** **.**
  • **** *****:***** **** **.*

******* *** **** ********* with ********* ******, ** well, *** ****** ** not ****** **** ******:

  • **** *****:***** ** **-*****, **.**
  • **** *****:***** ** **-*****, **

*** ********* ******** ******** were **** ** **** test:

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

Comments (23)

Michael Miller

09/11/17 02:58pm

Any reason that the Avigilon 4K PRO camera wasn't included in this test?  It's significantly cheaper than either option and delivers similar performance. 

John Honovich

IPVM | In reply to Michael Miller | 09/11/17 03:02pm

We already tested the Sony VB770 against the Avigilon 4K PRO camera here.

Undisclosed Integrator #6

In reply to John Honovich | 09/14/17 01:00am

From the axis test:

"

Further Testing Coming Up

We plan to test the Q1659 camera vs. competitive large format models from Avigilon, Sony, etc. in upcoming tests."

Undisclosed Manufacturer #1

09/11/17 03:26pm

8 MP vs 20 MP.....why? One has almost 50% more horizontal pixels than the other. OF COURSE one image looks incredibly better than the other.  Like what you see here? Head over to Motor Trend to see their comparison between the Honda Accord Sport and the Porsche 911. 

What about the lenses? They're different models, yeah? Can we get some more info on them to determine if we should care about the low light image comparisons? If the F-stops are completely different, then they too are pretty meaningless apples to coconuts comparisons.

John Honovich

IPVM | In reply to Undisclosed Manufacturer #1 | 09/11/17 03:38pm

8 MP vs 20 MP.....why? ... Head over to Motor Trend to see their comparison between the Honda Accord Sport and the Porsche 911.

That's an inaccurate comparison. Both of these cameras are the respective manufacturer's highest end offerings.

Ethan will respond with details about lensing and shutter speeds.

Undisclosed #2

09/11/17 03:34pm

When testing in low light, what shutter speed is used?  If you mentioned this in previous tests, sorry for the repeat question.

Avatar

Ethan Ace

IPVM | 09/11/17 04:25pm

In low light, 1/30s (33ms) shutter is used. It's generally always used, unless otherwise specified.

The lenses used for the bulk of testing were:

  • Axis Q1659: Canon 24mm F1.4L
  • Sony VB770: Zeiss 35mm F1.4

So the same F-stop was used.

We did some testing with 70-200mm varifocal models to verify things weren't simply due to the fixed lens choices (note that images using these do not appear in the report).

  • Axis Q1659: Canon EF 70-200mm, F2.8L
  • Sony VB770: Sigma FE 70-200mm, F4

Undisclosed Manufacturer #3

09/11/17 04:27pm

Its an interesting comparison, but why did you test the Sony at 4K rather than the 12MP resolution vs the 20MP of the Axis ?  That really scues the distance results.

It would seem reasonable to test at the same resolution 4K and then at max resolution perhaps.

Similarly if you are testing 20MP cameras, it may be more reasonable to compare the Axis Q1659 to the Sony VM772R 20MP camera.
Also both Sony cameras support 5 streams, not just 2 as was stated in the report.

Avatar

Ethan Ace

IPVM | In reply to Undisclosed Manufacturer #3 | 09/11/17 04:56pm

Its an interesting comparison, but why did you test the Sony at 4K rather than the 12MP resolution vs the 20MP of the Axis ? That really scues the distance results.

Similarly if you are testing 20MP cameras, it may be more reasonable to compare the Axis Q1659 to the Sony VM772R 20MP camera.

We didn't test the VB770 at 4K or the 772R at 20MP because they are MJPEG 2.5 FPS streams, and in our testing, have had really high bitrates, which would further skew the bitrates since Axis is already much lower bandwidth for a higher resolution. 

Thanks for pointing out the 5 stream spec. They do appear to be supported in Milestone and Exacq when the camera is in multi-stream mode, but not Avigilon or Genetec. I'll update that in the report.

 

Avatar

Franky Lam

Stan Group (Holding) Limited
In reply to Undisclosed Manufacturer #3 | 09/12/17 02:39am

You should be well known of SONY product...GREAT JOB

Undisclosed #4

IPVMU Certified | 09/12/17 02:35am

   

Situation reversed when looking at physical sensor size:

Less pixels and bigger sensor make bigger pixels.  

Undisclosed #5

09/12/17 08:49pm

Soon as the lights went out I was like WTF.

Should show a night comparison for all the day comparison photos, such as the intersection. This would give another perspective of those comparisons provided. Just a thought.

I mean, if you want to make it look like you captured bigfoot at night then definitely use the Axis camera LOL.

Add, what ISO ranges were used in the night comparisons?

james cordell

IPVMU Certified | 09/13/17 03:15pm

Have you tested the Arecont 10MP camera or the Avigilon 7K (30mp) camera.  We tested both of the cameras at work location with good results.  3 10mp or 2 7K  to cover parking lot, with decent details at night, since there is lighting for the parking lot.

V/r

Kelly

Jordi Ibero

Imago IP Solutions
09/13/17 07:02pm

All these cameras have the same issue. The stream. No VMS capable to display more than 2 or 3 units at the same time.

Speaking about performance, only Avigilon is the ONLY chance above the 4K. IMHO sure.

Michael Miller

In reply to Jordi Ibero | 09/14/17 07:49pm

I have both Axis and Avigilon PRO cameras.   I also have customers with 30+ Avigilon PRO cameras and I have no issues streaming all of them at the same time on my Surface PRO.  I only have one Axis camera but without multi-streams and the time to configure all the streams this will be a change to display multiple cameras on one client. 

Undisclosed Manufacturer #8

In reply to Jordi Ibero | 09/15/17 07:54pm

[Note: Avigilon Employee]

Your knowledge is definitely not up to speed as mentioned by Michael below. You can have 32x 7K AVIGILON cameras on a single display, LIVE or Playback. Try it. 

Undisclosed #4

IPVMU Certified | In reply to Undisclosed Manufacturer #8 | 09/15/17 09:57pm

You can have 32x 7K AVIGILON on a single display...

Would that be at ~640x480 resolution per stream?

What about 4 cameras on 4 4k monitors on one client, have you tried that?

 

Undisclosed Integrator #7

09/14/17 07:35pm

Have been really impressed with the Avigilon H4A line of cameras and would love to have you compare Avigilon (Made In Canada) solution to the Axis and Sony.

 

Any chance of that?

Avatar

Ethan Ace

IPVM | In reply to Undisclosed Integrator #7 | 09/15/17 03:42pm

We tested the H4A line 4K and 5MP cameras against others, as well as the Avigilon H4 4K Pro camera versus Sony. We've looked at testing one of the higher resolution Pro cameras as well, but haven't set anything up yet.

Undisclosed #4

IPVMU Certified | In reply to Undisclosed Integrator #7 | 09/15/17 10:01pm

...compare Avigilon (Made In Canada)...

Are those preferable to Avigilon (Made in America) ones?

james cordell

IPVMU Certified | 09/15/17 08:02pm

My limited understanding of some VMS, the # of camera you display is linked to # of CPU/ you to push the monitors.  I planned a project with 78 imagers, a combination of multi imagers and single image all IP, and required 4 additional CPU's to drive all the displays.

Maybe I am lost in the weeks.

 

V/r

Kelly

 

Chris Chambers

03/16/19 02:01am

I know this is a year and a half ago now, but I would suggest using the Sony, if it is still the only full frame 35 mm sensor out there in a security camera, as a low light reference in all tests.  The 0.05 lux scene in particular shows the absurdity of some of the 0.00n lux claims for much smaller sensors, for one.

I would also note that given full frame DSLR prices, it seems to me somebody ought to be offering APS-C at around $2,000 with no lens, and full frame 35 mm at around $4,000 or a bit less with no lens.  I guess the issue is even lower volume potential on full frame security cameras than photographer cameras.

Michael Miller

In reply to Chris Chambers | 03/16/19 02:18am

Avigilon's 4K APS-C PRO camera is $2K without a lens. 

Read this IPVM report for free.

This article is part of IPVM's 6,736 reports, 909 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.

Already a member? Login here | Join now
Loading Related Reports