Yes, will have another baseline coming out in later June. I will send to test shortly when i get it.
SMD 3.0 dahua
- Thread starter EMPIRETECANDY
- Start date
I talked with dahua engineer again, so will fix up he issue @Wildcat_1 mentioned on the 0-20 HLC washout issue, though less people use 0-20, but we still make it working out.,
Will have another BETA fw in 5hours, if test well, tomorrow we can release the Empiretech official version.
Thanks again for all guy's help here!
Will have another BETA fw in 5hours, if test well, tomorrow we can release the Empiretech official version.
Thanks again for all guy's help here!
IPC-T5442TM-AS , IPC-T5442T-ZE SMD 3.0 last testing version is ready, fix up the issue @Wildcat_1 talked about 0-20 HLC washout.
Any ones want to try, can make the uploading via this link.
we.tl
Tomorrow should have a real version. Will have an official releasing then.
Andy
Any ones want to try, can make the uploading via this link.
General_IPC-HX5XXX-Volt_MultiLang_NP_Stream3_V2.820.15OG001.0.R.210524.bin
1 file sent via WeTransfer, the simplest way to send your files around the world
we.tl
Tomorrow should have a real version. Will have an official releasing then.
Andy
Will be testing these final changes tonightIPC-T5442TM-AS , IPC-T5442T-ZE SMD 3.0 last testing version is ready, fix up the issue @Wildcat_1 talked about 0-20 HLC washout.
Any ones want to try, can make the uploading via this link.
General_IPC-HX5XXX-Volt_MultiLang_NP_Stream3_V2.820.15OG001.0.R.210524.bin
1 file sent via WeTransfer, the simplest way to send your files around the world
Tomorrow should have a real version. Will have an official releasing then.
Andy
All, I completed my testing this evening on 05-24 FW and am please to say that Dahua did take my additional feedback and have now applied changes across the entire HLC setting range from 0 - 100. I've tested all of these tonight with SmartIR on and each one adjusts correctly. I also repeated the flashlight glare test (to ensure HLC still adjusts for glare) and that works too.
I'm currently uploading the video and will post here as soon as complete.
My thanks again to Dahua for prioritizing these critical fixes and I hope now this code will be utilized on all SmartIR capable cameras across the Dahua portfolio as this bug I highlighted does exist there too.
I did notice that on this FW (05-24) a couple of recordings cut off prematurely (HLC 60, HLC 10, HLC 5 clips in the video when you see it) which was strange and had not seen that issue on 05-20 beta. However as SMD / AI detection algorithms were not touched on 05-24 this could just be an anomaly. Either way, I reported this back as SMD / AI detection is now being worked on by Dahua as they wanted to ensure SmartIR + HLC was fixed correctly first. This should now allow SMD 3.0 FW to continue being developed and hopefully in June a final release including the SmartIR + HLC + SMD / AI detection algorithms will be available.
Thanks again to Dahua for working with me directly, listening and acting on these bugs I reported and once again to @EMPIRETECANDY for also helping drive this forward.
VIDEO LINK
I'm currently uploading the video and will post here as soon as complete.
My thanks again to Dahua for prioritizing these critical fixes and I hope now this code will be utilized on all SmartIR capable cameras across the Dahua portfolio as this bug I highlighted does exist there too.
I did notice that on this FW (05-24) a couple of recordings cut off prematurely (HLC 60, HLC 10, HLC 5 clips in the video when you see it) which was strange and had not seen that issue on 05-20 beta. However as SMD / AI detection algorithms were not touched on 05-24 this could just be an anomaly. Either way, I reported this back as SMD / AI detection is now being worked on by Dahua as they wanted to ensure SmartIR + HLC was fixed correctly first. This should now allow SMD 3.0 FW to continue being developed and hopefully in June a final release including the SmartIR + HLC + SMD / AI detection algorithms will be available.
Thanks again to Dahua for working with me directly, listening and acting on these bugs I reported and once again to @EMPIRETECANDY for also helping drive this forward.
VIDEO LINK
Last edited:
They’d be mad not to have listened to such genuine and considered positive suggestions which only goes to enhance their products. Thanks to you @Wildcat_1 and @empiretechandy we all benefit.
yes, lastly get this firmware out, yes, a little hardworking on it. So the firmware we use now are bit higher level than dahua, i think we can post this firmware here later today.
Video link now posted above. YouTube processing is its usual mess unfortunately showing no thumbnail but the link should work, posting as well
WOW - great video and run through.
Some of the best video I have seen of zero/minimal IR washout up close.
Great job! Thanks for all of your work helping Dahua dial this in.
There is a little ghosting though
(just kidding...for those not watching the video, it is edit fade in/out of the different tests LOL)
Some of the best video I have seen of zero/minimal IR washout up close.
Great job! Thanks for all of your work helping Dahua dial this in.
There is a little ghosting though
(just kidding...for those not watching the video, it is edit fade in/out of the different tests LOL)
sebastiantombs
Known around here
@Wildcat_1 Thanks one more time for taking the time to work on this. I'll be trying this one out tonight to see how it works out for me based on the success you show.
Slightly off topic, but I have a question. I notice that you set exposure to a range from 1/60-1/100. I use a range as well but usually 0-1/100. My thought is to allow the camera more range of shutter speed to compensate for brighter lighting conditions, either steady or momentary, while keeping it long enough for good motion capture under less than ideal conditions. Why are you setting the range so narrowly?
Slightly off topic, but I have a question. I notice that you set exposure to a range from 1/60-1/100. I use a range as well but usually 0-1/100. My thought is to allow the camera more range of shutter speed to compensate for brighter lighting conditions, either steady or momentary, while keeping it long enough for good motion capture under less than ideal conditions. Why are you setting the range so narrowly?
looney2ns
IPCT Contributor
Awesome job @Wildcat_1 thanks for your efforts.
And thanks to @EMPIRETECANDY for acting on our behalf with Dahua to make the cams work great!
And thanks to @EMPIRETECANDY for acting on our behalf with Dahua to make the cams work great!
but...but....in my situation, cats/dogs/coyote do not give much of a IR washout at 3am when the Human AI gets triggered 
I'm glad this is getting fixed though. I have 2 x 5442 varifocals to replace 2 x 5231's and didn't want to slap them in yet due to this IR washout (gotta love the Human AI even though it has issues)
I'm glad this is getting fixed though. I have 2 x 5442 varifocals to replace 2 x 5231's and didn't want to slap them in yet due to this IR washout (gotta love the Human AI even though it has issues)Yes, the new version will out next month, so the smart IR on this fw is the best one we ever have. The new baseline will improve the working.
Yes remember as I said above, this release is only focused on the SmartIR + HLC (backlight program) bug fixes. SMD / AI detection fixes will be coming in the final release.
Sorry for the dumb question:
I see that you appreciate the HLC. I can also see the differences in the values of hlc.
For day light in the yard, I'm using WDR. looks like you guys don't like it. Sunny or shade, I can see it all and very clear.
Any reason for not using WDR in day light?
For night time, I can see the reason for using HLC. do you guys use WDR for day profile and HLC for night? makes a lot of sense to me anyway.
biggen
Known around here
- May 6, 2018
- 2,688
- 3,060
I use WDR during the day but not at night. It will add motion blur at night.
WDR certainly can be used during the day depending on needs BUT limit it’s value to just about what is needed. Just a small amount say 1-8 even as a value can help where needed. The key is don’t use at night and ONLY use backlight programs where absolutely needed and at a small value so you don’t negatively impact image.
@sebastiantombs, didn't forget about your question. To answer it, let me break it down into testing vs normal config:
- For testing I wanted a consistent set of fixed and ranged exposures. In testing SmartIR it made sense to focus the last few rounds of specific testing on ranges hence 1/60-1/100 is about the minimum anyone would want to run to get guaranteed blur free walking/jogging motion so settled on that
- With regards to your question on why not 0 - 1/100, that one is all about a) FOV, b) install location with key focus on lighting availability, c) target requirements and d) camera intelligence . Specifically whenever I dial a camera in I take into account a through c and then determine a range I know will work to guarantee an ideal (or as close as possible) video cap based on those inputs + experience I have with those parameters and the camera capabilities (amount and type of noise to mitigate, sensor type, aperture of lens, duties to be performed by the camera etc). Item D however is somewhat of a constant and that is that the intelligence on most cams are not good (not speaking about AI as much as intelligence of the cam to adjust per scene, correctly). Most cams operate in a relatively simple way in this regard. They will see a dark scene and then when left to auto or a more open, manual range (such as 0 to xxxxx) the first action they'll take is slower shutter, or full open IRIS, or increased gain etc etc. to make the scene brighter. They do this by sacrificing image quality in motion blur, frame streaking and noise. So to stop this 'non-intelligent' adjustment by the camera I prefer to reign in or put a sandbox around how much it can adjust and that is a long winded way of me stating why I never leave at 0 to xxxxxx and instead narrow its ability to adjust . This leads to a more consistent image and that means you are guaranteed something you can use.
In the current FOV I was testing on, I could push to 1/120 (comfortably) and for closer targets 1/250 but there really is no need on the latter due to speed of expected targets in the area. Similarly I never want to go to 1/30 because the risk of motion blur is greater there even for 'swift' moving humans. Therefore 1/60 becomes the absolute minimum and I dial in the faster shutter as needed. For daytime of course you have more flexibility BUT even then I narrow down a range that I know will accommodate a through c.
That certainly doesn't mean your FOV, requirements have to be the same BUT I would highly recommend that you keep a tight range so you know how the camera will react and give you a more consistent experience overall.
HTH
Just get the SMD 3.0 new baseline firmware for June, any guy want to test just let me know.
Attention: This firmware is without smart IR, we will desgin on it if the new one has some good improment.
Attention: This firmware is without smart IR, we will desgin on it if the new one has some good improment.
@EMPIRETECANDY what camera models are in the scope for the new FW? Is it still the 5442-series?
If so, count me in as I just installed a 5442T-ASE-NI to a great spot where I have horses and pigs (not together thought! )
And for you guys who wunder about the camera choice, simple, where the horses are it is A LOT of flying crap who simply love IR. The little f*ckers still like my blood more than IR though...
If so, count me in as I just installed a 5442T-ASE-NI to a great spot where I have horses and pigs (not together thought! )
And for you guys who wunder about the camera choice, simple, where the horses are it is A LOT of flying crap who simply love IR. The little f*ckers still like my blood more than IR though...