SG9665GC firmware updates and pre release access

I drove around in circles with the sun blasting directly into the camera. AE is behaving nicely, so far so good.

Considering my long time discouraging experiences with the GC I want to see how it performs in the challenging scenarios I’m so familiar with in my daily driving and that were really the primary impetus for this firmware update in the first place such as traveling in and out from under canopies of trees, so I’m sure you’ll understand why I’m reserving my judgement. I think customers should be afforded the opportunity to evaluate firmware updates in their own time and manner rather than have impressive looking, cherry picked images presented to us by a retailer affiliated with a manufacturer.

Curiously, considering the actual reasons for the firmware AE fix-its in the first place, why didn't you post the "sun blasting directly into the camera" shots rather than the above one? Come to think about it, the "sun blasting directly into the camera" was not really the problem, per se. It is high contrast/high dynamic range scenarios with dark and light areas within the FOV as well as transitions from flat lighting into brighter lighting. That's what we need to see.
 
The last test I performed was to check last GPS MPH bug was fixed.

AE table hasn't changed the last few updates and was already proven good a few pages back. I didn't feel it needed to be shown again. I can perform it again if you want.
 
The last test I performed was to check last GPS MPH bug was fixed.

AE table hasn't changed the last few updates and was already proven good a few pages back. I didn't feel it needed to be shown again. I can perform it again if you want.

I understand. I guess I was reacting to your "sun blasting directly into the camera" comment when you responded to my earlier post about how the image you posted doesn't tell us how the new firmware AE tables will perform to resolve the major issue. We'll all need to see samples from a variety of conditions in order to make a proper evaluation. I can't wait to test everything out myself when my new replacement camera finally arrives tomorrow. Glad the bugs are getting squashed too but in some ways the GPS glitches feel like a distraction from the initial exposure issues, so for me that's what I am most focused on.

Things are looking good at any rate.
 
V2 beta 19 issue: Setting the "carno" is difficult (as you can't see what character your setting as you set it.)


Follow the "firmware update process" text file... After resetting to default, I hit "menu" twice, and scrolled down to "carno", and hit OK. As expected, I had a blue box with an arrow above/below. When I hit "MIC" or "REC", the blue box (cursor or current char) didn't change to reflect the currently selected character. Instead, it stayed as a "empty" blue box.

If I hit "MENU" again (to move to the next space in the string), I could see what I had previously selected. If I hit "MENU" continuously to cycle back to the first character, the blue "cursor" box would again cover (and obscure) the current letter/character.

EXPECTED: I expected that the currently selected character would show through the "cursor" -- either white text on the blue cursor background, or black text... or something so that it was visible.

NOTE: I was doing this while in my car in sunlight. It's possible that there is some way to see what is "under" the blue cursor, but that lighting conditions made it impossible for me. If that's the case, I'd think this is still a bug, as most people would be setting the carno while the GC device is installed in their car.

Edit: This is, in fact, an issue of just not being able to see the characters under the cursor in bright sunlight and while the GC is mounted. In my garage (outside of the bright light) things are more visible (if still somewhat hard to see due to the angle.)
 
Last edited:
I just installed the new firmware, following the instructions. Took the cam into the house for that, but didn't experience the issue described above by Gary D. However, I think with lots of light and sitting in front of a tilted screen you might get this symptom.
 
However, I think with lots of light and sitting in front of a tilted screen you might get this symptom.
Okay, that makes me feel better. If there isn't much contrast between the blue cursor and the text beneath it, I could certainly understand why I couldn't see it. (Perhaps it's a white character under the light blue box?) Doing it in bright sunlight probably wasn't my best choice. I'll try it again when I get home tonight (and am parked in the garage.)

If not lighting conditions, Is it possible color blindness is at play?
No.. I can clearly see that the "28" in your avatar is green and the dot over the "i" is blue. :rolleyes:
 
I'm seeing something odd with b19, but I haven't quite pinned it down yet. I have the speed displayed turned off. I started my car and there were hyphens where the speed used to be. A few seconds later, but before lat/long data started to be embedded in the video, the hyphens disappeared. (See attached images. The timestamps are in UTC if that isn't obvious.)

Edit: My issue reported a few posts above concerning setting CARNO is, in fact, an issue of just not being able to see the characters under the cursor in bright sunlight and while the GC is mounted. In my garage (outside of the bright light) things are more visible (if still somewhat hard to see due to the angle.)
 

Attachments

  • Frame 2016.06.15 20-27-15 (0-00-01.508).jpg
    Frame 2016.06.15 20-27-15 (0-00-01.508).jpg
    298.2 KB · Views: 24
  • Frame 2016.06.15 20-28-28 (0-01-14.888).jpg
    Frame 2016.06.15 20-28-28 (0-01-14.888).jpg
    409.5 KB · Views: 23
Last edited:
I'm seeing something odd with b19, but I haven't quite pinned it down yet. I have the speed displayed turned off. I started my car and there were hyphens where the speed used to be. A few seconds later, but before lat/long data started to be embedded in the video, the hyphens disappeared. (See attached images. The timestamps are in UTC if that isn't obvious.)
when speed display is on the hyphens indicate that the GPS doesn't have signal yet, they probably shouldn't show this when speed is turned off so will have a look if we can hide that when speed display is turned off, thanks for the report
 
We'll all need to see samples from a variety of conditions in order to make a proper evaluation. I can't wait to test everything out myself when my new replacement camera finally arrives tomorrow. Glad the bugs are getting squashed too but in some ways the GPS glitches feel like a distraction from the initial exposure issues, so for me that's what I am most focused on.

the GPS issues (amongst other things) have stopped us getting this out to a wider test group, the AE table changes have fixed a basic flaw but we still have IQ tuning to do, we'll really need a lot of source material to judge what needs to be done so getting the bugs squashed is important so that we can get this version out there and start getting feedback that will help us decide on the next steps, what changes we make and do we need to just make the adjustment fixed for all, or can we set it up as a user option etc

Things are looking good at any rate.

improving, but in no way are we finished, still plenty to do
 
the GPS issues (amongst other things) have stopped us getting this out to a wider test group, the AE table changes have fixed a basic flaw but we still have IQ tuning to do, we'll really need a lot of source material to judge what needs to be done so getting the bugs squashed is important so that we can get this version out there and start getting feedback that will help us decide on the next steps, what changes we make and do we need to just make the adjustment fixed for all, or can we set it up as a user option etc



improving, but in no way are we finished, still plenty to do

A very welcome response, thank you!
 
A very welcome response, thank you!

just to give you (and others) a bit of insight for the first 12 months at least we had zero support from Novatek (to be fair not many do get direct support) and have had to do everything we've achieved until very recently without any outside help, some stuff is reliant on fixes happening in the SDK before we can do what we want, we finally reached a stage where we are able to get vendor support, some of that is to do with increased volume but I think in part it was because we never gave up asking and just kept pestering for fixes, in other respects I think because they worked out that what we were asking for was correct it became harder to keep ignoring

say what you like about us but we've never stop trying to improve things, even if it takes longer than what we would like, it's not for the lack of effort I can assure you
 
Last edited:
without tyre smoke I believe, very hard temptation to resist, did well.

It is slightly challenging keeping 444 HP, in a slow 5 MPH tight circle.

But yeah, as the direct low sun shined in and out of field of view, sky remained blue. (no more exposure blow outs)
 
But yeah, as the direct low sun shined in and out of field of view, sky remained blue. (no more exposure blow outs)

it is definitely improved but we still need to look at how it handles those situations where there are lots of dark areas and small areas of really bright light, I expect we'll need to look at contrast and peak brightness levels, it's going to take a while to get the fine tuning right, getting this version stable and out to a wider audience will help so we can get more sample video to study to work out the changes
 
just to give you (and others) a bit of insight for the first 12 months at least we had zero support from Novatek (to be fair not many do get direct support) and have had to do everything we've achieved until very recently without any outside help, some stuff is reliant on fixes happening in the SDK before we can do what we want, we finally reached a stage where we are able to get vendor support, some of that is to do with increased volume but I think in part it was because we never gave up asking and just kept pestering for fixes, in other respects I think because they worked out that what we were asking for was correct it became harder to keep ignoring

say what you like about us but we've never stop trying to improve things, even if it takes longer than what we would like, it's not for the lack of effort I can assure you

That explains alot of why features weren't being added.
 
Hmm, 444hp. lemme guess. Could be a beemer m3 or 4. Could be a cummins truck (which I'm somewhat partial to). But I'm guessing it's a Ford 302. umm, Mustang? (judging by the bonnet which doesn't seem to resemble any of the others, unless of course they've done a massive Cummins upgrade while I wasn't looking.

:D
 
Back
Top