Camera quality C vs non-C model DS-2CD2386G2-ISU/SL

[benwillcox]

Hi all I have a couple of DS-2CD2386G2-ISU/SL cameras, bought last year. I recently bought another one which came marked on the box as 'AcuSense EXIR Fixed TURRET' and has the 'C' mark against the model number. The original version just said 'AcuSense Fixed Turret' and no C.

The cameras are next to each other with one facing left and one right. I was surprised at the difference in image between them, image settings on both are set to the same default settings
Original non-'C' model:

New 'C' model:

The new one seems to have much less vivid colours. Anyone else found this, and is it to be expected?

Ben

 

[codlord]

I have noticed this sort of difference between some HikVision cameras. I just adjust the image Saturation setting up or down to compensate.

 

[Phil]

@benwillcox
That's disappointing to see.
The B, C, D variants tend to occur when Hikvision make (are forced to make?) component changes within a product.
They declare a new suffix i.e. the same product model number moves from suffix B to C.
I have never known us to be able to establish any more detail on what actually changed within the product than that.
i.e. whether the sensor has changed, or memory, or processor, or lens, or what.

Broadly, I think these changes fall under the global heading of 'chip shortages'.

They sometimes have the further negative impact that the model that was current, let's say the B model before the C is launched, becomes by default no longer the current model and therefore firmware releases for improvements sake may stop. They should re-visit and update firmware if it is necessary for security issues.

It is only through real world / user feedback that we get to learn whether the new variant performs as well as the previous.
I believe that you are the first to raise this.
We will watch to see whether a consensus forms around the less vivid image that you have shown us.

Thanks for the feedback, explanation, and images.

 

[benwillcox]

Thanks for the replies. I tried adjusting saturation on the 'C' version but even on max is nowhere near as vivid as the original.

I had read about the 'C' variant previously on this forum and that it had 25fps vs the original 20fps and thus was the version of choice, but no mention of the image quality which I had assumed would have been the same.

I suspect that as you say, a different sensor type is being used due to component supply issues, but a bit of a shame that the picture quality doesn't seem as good as the older version (certainly as far as the colour saturation is concerned).

Ben

 

[JB1970]

I have both models and haven't noticed a difference between them, but to be fair both my G3 and G5 models cover driveways so perhaps it would be more noticeable if they were covering an expanse of lawn. I would be inclined to temporarily pan the new model around to give the same view and snapshot both images again. Although they are next to each other, the direction of light on the scene is different and may be affecting the colour. I have seen similar results with two cameras that are identical models.

Edit: I was about to post a pic from a G5 I installed but it's a ColorVu (2347) so the colour rendition could well be different. I just took a look at a G5 2346 and that looks fine (not washed out)

 

[johnfitzy]

I would be inclined to temporarily pan the new model around to give the same view and snapshot both images again. Although they are next to each other, the direction of light on the scene is different and may be affecting the colour. I have seen similar results with two cameras that are identical models.

Hi @benwillcox

I was just about to reply with a similar suggestion as @JB1970 above... If it's not too much trouble to pan the new camera to point in the same direction as the old one to test, you'll really take the big variable direction of sunlight from the equation.

I have DS-2CD2387G2-LU(C) and DS-2CD2087G2-L(non-C) cameras in the backyard that are pointing in the same direction but from opposite ends of the house and most of the time the images are similar... but when the sun hits one of the cameras and not the other, it looks just like your example pictures above.

@JB1970 it's ok to spell color as colour but please start driving on the right side of the road!

 

[benwillcox]

It's a good point and it had occurred to me it could be the different lighting affecting the picture. So I've done some experiments.
Pointing both cameras at the same angle gave me the same outcome, so I tried holding up a colour test card to see if I could see any colour variation:

Cameras both pointing at the same angle with testcard:

I the colours looked the same but I noticed that in the background the grass now was the same colour. However after moving the testcard and waiting a few seconds, the colour of the new camera slowly faded back to how it was:

Immediately after testcard:

10 seconds after testcard:

So it seemed to me that it must be some auto white balance related issue. Checking both cameras they were set to AWB1 which is the default, however I noticed a 'Natural light' setting which I imagine is better for outdoors to changed both to that. That gave me these results:

So I have the opposite problem! So I set the original camera to AWB1 and the new camera to Natural Light and here's the outcome:

Fixed! (Well, close enough). So I can only presume from this that there is potentially a bug in the selection of White Balance modes which causes the selected option to act differently in the different camera types.
Oh well, at least we now know that the image quality is as good as the non-C version with a minor setting change.

Firmware I'm running in the non-C version (B-R-G3-0) is V5.5.801 build 211105, and the C version (B-R-G5-0) is V5.5.113 build 210317.

Ben

 

[johnfitzy]

Wow! great job testing and figuring that out!

However after moving the testcard and waiting a few seconds, the colour of the new camera slowly faded back to how it was

I guess the grass is always greener on the other side

So I can only presume from this that there is potentially a bug in the selection of White Balance modes which causes the selected option to act differently in the different camera types.

It does seem like a "bug" to me... I image with the white balance set differently on the cameras you may see some differences throughout the day and night?

I assume from your spelling of color that you drive on the wrong side of the road also

 

[JB1970]

Good job with the diagnostics. So it's nothing major just a difference in the AWB setting.

Firmware I'm running in the non-C version (B-R-G3-0) is V5.5.801 build 211105, and the C version (B-R-G5-0) is V5.5.113 build 210317.

Just wondering why you are running such an old version on the G5? The latest version is 5.7.3_220216 SP:

HIKVISION UK PORTAL

www.hikvisioneurope.com

 

[benwillcox]

Ah I hadn't realised it was so far behind, this one is straight out of the box. I will upgrade!

 

[JB1970]

Ah I hadn't realised it was so far behind, this one is straight out of the box. I will upgrade!

Sometimes the updates are relatively frequent and sometimes months apart but it's worthwhile bookmarking the relevant pages on the portal, so you can check every now and again (though someone will usual post on this forum if they find a new version available)

 

[Phil]

@benwillcox Great testing!
Thanks for sharing

 

[benwillcox]

Good point, I should have checked the firmware straight away but for some reason totally forgot. I've now updated so both cameras are on their latest versions.

They still exhibit the same behaviour as before, and some of the other pre-sets on both cameras gives differing images too e.g.
Fluorescent, Natural Light, Warm light lamp, AWB1 all look different across the two types of camera.

The following do give matching images:
Incandescent lamp, AWB2 and MWB (at the same manual settings).

So for now I've chosen AWB2 for all the cameras which seems to give me a good consistent image.

Ben