Bosch 4K Tested

what was the effect to bit rate when using the various cameras IVA features?

Reactions:

Ethan Ace

In reply to dominic costello

When using IVA rules (cross line, object detection, etc.) there's no real change to the actual bitrate being transmitted. But you can possibly reduce storage requirements by using them to record on analytic events instead of simple motion detection, which is typically more accurate (thus less recording).

Bosch has a feature tied into their IVA, called CBIT (content based imaging technology) which claims to reduced bandwidth based on what is in the scene. But I can't tell you specifically what the reductions are, if any, because CBIT cannot simply be turned on or off. It a set of features built into the firmware which don't have an on/off switch.

Because of this and because Bosch recommends them (and the camera defaults to them), we test with these features on.

Reactions:

Undisclosed #1

IPVMU Certified

In reply to Ethan Ace

Here's a schlocky but believable video where somebody compares a Bosch camera with CBIT-less firmware to one with CBIT. Seeing maybe 30% reduction in bit rate.

Note his description of how it's done is very similar to one part of Axis Zipstream's method, though CBIT does not seem to have variable GOP.

Also this is from 2013. I'm not sure why Bosch would make it so hard to compare before and after. Axis definitely beat them marketing wise on this one.

Warning, this video is nowhere near IPVM production quality, viewer discretion is advised.

Reactions:

Ethan Ace

In reply to Undisclosed #1

That somebody is the local Bosch rep here, who's also an IPVM member. I saw it this morning.

Bosch does have variable GOP, but it's limited to 2 seconds max.

Reactions:

Undisclosed #1

IPVMU Certified

In reply to Ethan Ace

...the local Bosch rep here, who's also an IPVM member.

'schlocky'? I typed 'snazzy'. Damn Yiddish auto-correct.

Reactions:

(1)

Matt Golueke

Undisclosed, hope the "snazzy" video was useful or at least entertaining. Our rep team is a "try it before you believe it" group of people, so we capture these short videos of our discoveries from time to time. No match for IPVM productions, so we watch their evaluations closely as well. Cheers!

Reactions:

Undisclosed #1

IPVMU Certified

Well, it did the job, so no one is complaining. :)

Question, I had mentioned that CBIT didn't seem to have variable GOP, to which Ethan indicated it did.

More precisely what I meant to say was 'continuously variable GOP', i.e. one that dynamically adjusts on the fly based upon the image, like Zipstream. Does it have that?

Curious about CBIT, does Bosch hold any IP rights on the technology? Would you say it is similar to Zipstream or not?

cheers!

Reactions:

Matt Golueke

Undisclosed, while I'd love to engage in a Bosch chat on this forum it is out of place for the IPVM format unless Ethan or one of the Administrators wants to provide a technical explanation for you. I'd encourage you to connect with me offline or contact your local Bosch representative in your area. I'll be happy to locate him/her for you. matt@MidChes.com

Reactions:

Undisclosed #1

IPVMU Certified

In reply to Undisclosed #1

...a Bosch chat on this forum it is out of place for the IPVM format

Your respect for the forum decor is impressive!

However, I would say that a manufacturer being asked a direct technical question, e.g. does CIBT use a dynamic GOP, is certainly sufficient grounds for a manufacturer response.

From what I have seen sanctions are applied mostly in cases of undisclosed and unabashed self-promotion.

Indeed, many manufacturers perform an informative and helpful function by relaying authorative answers to member questions that otherwise might be time consuming to obtain. Representatives from Axis, Hikvision, Pelco, and Samsung, to name a few, are frequent contributors to the boards.

As long as the 'Bosch chat' is on topic and relevant, I would doubt there would be a concern. (John/Ethan, jump in any time :)

Reactions:

Michael Miller

•Jul 18, 2015

I say start a new thread and let her rip.

Reactions:

Undisclosed #1

•Jul 18, 2015

IPVMU Certified

In reply to Michael Miller

I would, but I'm not sure Matt is entirely comfortable with the IPVM discussion format, since he has expressed some reasonable concerns about engaging here.

And I'm not sure who else could provide low-level details on CBIT, so...

Though I believe he may be your local Bosch rep. as well, so if you run into him ask him for the lowdown between ZipStream and CBIT... :)

Reactions:

Matt Golueke

•Jul 18, 2015

Undisclosed, please feel free to reach me offline at matt@midches.com. Thanks!

Reactions:

Matt Golueke

•Jul 20, 2015

Undisclosed, over the weekend I posed your questions to a good friend of mine at Bosch who is far more technical. So as not to improperly use IPVM's forum for manufacturer specific conversation I have created a blog post on our site to begin answering your questions: http://blog.midches.com/blog/cbit-vs-zipstream

Reactions:

John Honovich

•Jul 20, 2015

IPVM

"It does not change the way the processor in the camera functions when it comes to FPS."

Nor does Zipstream. Zipstream can impact the I frame interval, changing the ratio of I to P frames but it does not change the total number of frames per second.

"[changing the GOP] .... from a forensic standpoint that could lead to calling evidentiary video into question …."

It could if you use lose the I frame. Are you going to lose the I frame? That's unlikely though certainly if you are real risk adverse you may reject it for that (ergo the small number of MJPEG absolutists).

The net/net is Axis has an advantage in that it allows adjusting the I frame interval, which can provide huge bandwidth savings while Bosch does not have this option. Agree/disagree?

Reactions:

Undisclosed #1

IPVMU Certified

Thx!

Reactions:

Matt Golueke

John, more information from Bosch regarding your follow-up questions/thoughts. Hope this helps.

Thanks, Matt

"The beginning of a GOP is the I Frame …. This is the base and foundation of viewable video, and without an I Frame in the proper place P Frames amount to usually useless video. Based on the ZIP Document from AXIS they are achieving low bit rates and storage by compressing. This is based on :

Scene motion
Scene content
Ambient light level
Configuration options:

This part basically equates to Bosch's CBIT and iDNR technology: since I don’t have this AXIS camera I can’t test it in my lab, all I have to go by is their documentation.

Configuration Options:

Compression parameter
Group of Pictures (GOP) length
Frame rate
Strength parameter
Dynamic GOP parameter
Dynamic GOP limitation parameter

After reading their documentation I am assuming the following:

They do a Dynamic ROI, this equates to our Encoder Regions which is a separate optional feature . Theirs does it automatically we set ours manually because it is a separate feature but works in concert with CBIT and IDNR. This is important due to the following: if their camera is auto detecting what it thinks is insignificant and reducing the resolution of that area on its own it could be detrimental from an evidentiary video standpoint. (I know I bring this up a lot, but from my perspective the sole purpose of video is evidence and its admissibility in court)

I appreciate the sentiment in the response, but here are my concerns: The GOP length and Dynamic Parameters coupled with compression. The end game of how they are doing things is encode the video, extend the length of the GOP by changing I Frame distance and crushing the P Frames with compression. The camera makes the decision to produce more I frames when needed based on Motion.

In a lab I am sure this works like a champ. Even my Bosch colleagues in Germany don’t like my answers sometimes.

In a real world deployment with 300 cams (equates to moderate network traffic), and I am assuming that we are in a NVR scenario, if you have 150 of these Axis cameras kick to high frame rate at one time:

o Do I drop an I frame because of network congestion: if you are doing UDP (not TCP) the I Frame is gone

o Does the NVR app corrupt the current video file because it only has P Frames (uselsss data with out an I Frame)

o If multiple cameras are all assigned to the NVR, does it cause the NVR to hiccup with the sudden change in traffic from all of its cameras? There are a ton of factors: are we doing Verint / Sec Center / Milestone…. What hardware is deployed, what is the network design, unicast or multicast, stacked switches, using Cisco or Brocade, how many streams per camera?

With the Bosch solution you can adjust the GOP length, we do set it standard for iSCSI recording, as we don’t not record to an NVR app with our solution…. Bosch doesn’t record video files we record raw data … I just wrote a paper for the NSA and the FBI on this topic.

So there are other mitigating factors that come into play with how we deploy and set our features up."

Reactions:

John Honovich

IPVM

"since I don’t have this AXIS camera I can’t test it in my lab, all I have to go by is their documentation"

Maybe Bosch should have bought their own camera or read our test results. It does not say much for a long winded response based on aggressively (mis)reading a competitor's docs.

What you are doing is FUD and is really is above Bosch as a company.

"extend the length of the GOP by changing I Frame distance and crushing the P Frames with compression."

Though crushing is not a technical term, the implication is that it is significantly hurting quality. Our test results show differently. Axis I frame 'extension' is occurring with low motion scenes where there is not much difference to encode between frames.

"if you have 150 of these Axis cameras kick to high frame rate at one time"

Again, the frame rate does NOT change. Is this really an engineer at Bosch who keeps on conflating frame rate with I-frame interval?

As for the 300 cam scenario, like any other recorder, it needs to be sized to handle max load / throughput. Recorders deal with this all the time, as most video is VBR.

"With the Bosch solution you can adjust the GOP length"

Ok, so can Axis. Axis just also gives the option of dynamically adjusting the I frame interval.

"Bosch doesn’t record video files we record raw data … I just wrote a paper for the NSA and the FBI on this topic."

Not sure what this has to do with Zipstream.

Reactions:

Undisclosed #1

IPVMU Certified