First week with SG9665GC V3 - some questions

[nephilim]

Ok. maybe I am expecting too much then. So here is an example taken 10mins later. It's still significantly softer but you guys know better if that is normal.

https://www.dropbox.com/s/rikrke3jc2k0cil/2017_0518_150035_040-1039_042 - 20 second.jpeg?dl=0

 

[jokiin]

The trouble starts when direct sunlight heats the camera.
.

as a test can you try detaching the camera from the window, but leave it in the car, want to see if it's related to heat or direct sun

 

[nephilim]

I agree, I was too quick with that statement - the results of the hairdryer test I did on Sunday (results posted in this thread) show that heat is the cause - not sunlight. But I can test again tomorrow.

 

[jokiin]

I agree, I was too quick with that statement - the results of the hairdryer test I did on Sunday (results posted in this thread) show that heat is the cause - not sunlight. But I can test again tomorrow.

hairdryer heat is quick and hard to replicate real situations, it may be an issue with sunlight heating the CMOS sensor through the lens above working temps rather than the actual ambient temp, some tests we did last week in the UK on a 19°C day recorded interior temps up to 71°C, that's a much larger variation than expected

 

[nephilim]

Today was as hot as yesterday - too hot for me, hence I forgot to take the memory card out of the computer and put it back into the camera :-( So I have no samples from the same locations as above. I rushed home (without A/C to keep the camera warm), fetched the card and did a different comparison. The softness is there but -apparently- significantly less! I would really like to repeat that test in the following weeks when it is hot again. But your scenario with the CMOS getting heated through the lens is an interesting one. It would explain why I got soft videos on much cooler yet sunny days earlier this year. So what I will try is to keep the camera in the car and put a DIY lens shade (cardboard/aluminium foil/...) over the CPL and see if that has a beneficial effect.

 

[niko]

Yeah, it was only +19C and with non direct sunlight.
For the past 2 weeks I am conducting detailed tests with multiple SG9665GC daschams in different parking scenarios. I did get even much worse (higher) temperature readings lately with direct sunlight and at much lower outside temperature. There are still a lot of tests to be done and put all results together for proper detailed report along with video- and screenshots samples, measurements etc.

 

[jokiin]

Today was as hot as yesterday - too hot for me, hence I forgot to take the memory card out of the computer and put it back into the camera :-( So I have no samples from the same locations as above. I rushed home (without A/C to keep the camera warm), fetched the card and did a different comparison. The softness is there but -apparently- significantly less! I would really like to repeat that test in the following weeks when it is hot again. But your scenario with the CMOS getting heated through the lens is an interesting one. It would explain why I got soft videos on much cooler yet sunny days earlier this year. So what I will try is to keep the camera in the car and put a DIY lens shade (cardboard/aluminium foil/...) over the CPL and see if that has a beneficial effect.

ok, let us know how your experiments go

 

[Marc Onrust]

I'm having similar issues with my V2. When I leave it mounted in the car (cam not running) on a warm day, the result is a (much) softer video than in other circumstances.

 

[niko]

I'm having similar issues with my V2. When I leave it mounted in the car (cam not running) on a warm day, the result is a (much) softer video than in other circumstances.

This is normal behaviour. It does not depend so much on outside temperature but more on intensity of the sun.
Yesterday I measured dashboard being +87C， GC unit side (not recording while parked， just started ) at +61C with outside temperature ONLY +12.5C.

 

[Marc Onrust]

Well, in the end it is the unit heating up what seems to be causing the issue, either due to direct sunlight (and/)or due to high temperatures inside the car. Quite frankly, I would be disappointed if that's normal behaviour.

I'm at the point of buying a V3 for in our convertible Saab, but to be honest I'm not sure this is the best choice as this car is intended to be used on warm sunny days.

 

[niko]

Well, in the end it is the unit heating up what seems to be causing the issue, either due to direct sunlight (and/)or due to high temperatures inside the car. Quite frankly, I would be disappointed if that's normal behaviour.

I'm at the point of buying a V3 for in our convertible Saab, but to be honest I'm not sure this is the best choice as this car is intended to be used on warm sunny days.

I meant @ “normal behaviour“ as being what it is for any dashcam (even premium make/model with good lens and lens holder) tested within same sun/heat circumstances in parked and closed car left baking under intence sun or ambient high temp. if it reaches above rated operational temperature. After you turn on car AC and dashcam hardware reaches normal operational temperature the soft focus recovers to normal.

 

[Marc Onrust]

OK, misunderstood you there

And SG already is premium in my opinion, don't do yourselves short on that one

I'll think it over once more, but I guess you will find my order in your website within the next couple of days

 

[niko]

OK, misunderstood you there

And SG already is premium in my opinion, don't do yourselves short on that one

I'll think it over once more, but I guess you will find my order in your website within the next couple of days

I do understand concerns from customers in relation to SG (or other dashcams) having soft focus after car has been parked， therefore I am putting together few weeks of different tests made only while car has been parked in different light and temperature conditions. I hope to finish my observations soon， too many scenes still need to be re-tested.

For repeated orders you may send me PM before placing an order，some time I am able to provide some discount

 

[Marc Onrust]

Noted, thnx Niko!

If you need some movie files from me to supplement your material, just let me know.

 

[niko]

Noted, thnx Niko!

If you need some movie files from me to supplement your material, just let me know.

Thank you. Yours and other customers input， experience and observations are always welcome. This is helping us to understand dashcams behaviour in different circumstances and trying to find new and better solutions. You valuable tests will be useful and to keep all in one place I think if you can post your observations in the thread I will be creating after all my experiments will be finished， so we all can discuss soft focus and other parking related questions there.
Thanks again for your input and co-operation!

 

[nephilim]

OK, I did a similar check as @Leszek: car parked in sun from 8 to 16:41 with aluminium foil covering the lens. Removed the foil, saved the image, waited 5mins, saved the next snapshot. I then headed home and saved another snapshot while waiting at a traffic light 3mins later. The sequence clearly shows a decline in sharpness while parked (silver Opel, TF IA 22 license plate). In the third image it is virtually impossible to read the license plates of the cars on the left and right side.

Please help me to understand it: it's a fixed aperture, fixed focus, fixed focal length assembly. So more light means shorter "shutter time" means less motion blur. Correct? Usually, a lot of light would help a lens as one would stop it down to smaller aperture openings, which increases depth of field and increases contrast and sharpness. Here the aperture is fixed, so I would expect the same sharpness all from dashboard to horizon no matter how much light is available. With such a wide angle lens, DOF is usually never an issue. So there must be something moving inside the camera to get the system out of focus. But then I want to use a dashcam, not reverse engineer it :-/

 

[jokiin]

OK, I did a similar check

can you do me a favor and upload the videos these are from to a file host and send me the links please?

 

[nephilim]

PM sent

 

[nephilim]

Today was a hot day again. Car was parked in the sun for 7h. Driving with A/C for 50 minutes. Parked in the shadow of a tree for 20 minutes. The following images demonstrate that sharpness/focus is not fully restored even after almost 1h of driving. Just look at the license plate of the BMW Mini on the left side... I wonder if I should open the camera and play with the focus setting of the lens. It cannot get worse I suppose.

Cannot upload today (file too large). Here are the dropbox links:

https://www.dropbox.com/s/s2kz77k20oakh8e/2017_0602_152125_086-60625_101 - 91 second.jpeg?dl=0
https://www.dropbox.com/s/t9d62fn1wahv6rk/2017_0602_162755_102-70054_113 - 0 second.jpeg?dl=0

 

[jokiin]

I wonder if I should open the camera and play with the focus setting of the lens. It cannot get worse I suppose.

don't do that, not going to help, we're working on something now, will update you next week and give you a timeframe on sorting this out for you