KronTalk

Chronos => Chronos User Discussion => Topic started by: Nikon1 on July 02, 2021, 04:29:24 PM

Title: BUG REPORT --- HDMI output Problems and Softlocking the Camera
Post by: Nikon1 on July 02, 2021, 04:29:24 PM
As has been reported by multiple People here, there seems to still be some Serious issues with the HDMI Output (Chronos 2.1).
 For me It usually worked reasonable well, and recently i didnt have any issues at all. But today i had serious Problems getting the Camera to even output a HDMI signal once, when testing out some Exposure Settings.
 So today i decided it was finally time to have a detailed Look at this Problem and Do some Testing. 
 
[...]
 2. Camera doesn´t freeze everytime whenever i Plug HDMI in. I would say on average its somewhere between 1 out of 15 and 1 out of 20 times it would freeze as an rough estimate, but havent taken any records about it to give accurate numbers so far. Similar to what SergeyKashin is reporting, when it freezes, it most of the time does that also on the next 1 or 2 reboots also, untill it starts working normally again.
 [...]

 .
 Some References to previous / other discussions about this topic, before i get into detail of my tests:
 https://forum.krontech.ca/index.php?topic=670.new#new (https://forum.krontech.ca/index.php?topic=670.new#new)
 https://forum.krontech.ca/index.php?topic=596.0 (https://forum.krontech.ca/index.php?topic=596.0)
 .
 .
 So i started with the following Test:
 (Exposure Settings for all of the Tests listed below where 1000fps at full Resolution at 357° and 0dB /0dB Gain also using Firmware 0.6.0 for all tests)
 .
 Test Nr.1
 Monitor used is the SmallHD 502 Bright via HDMI Input. This Monitor was used for all tests below.
 Test was done as Following: Monitor is Powered on, but HDMI is unplugged. Camera is Powered up and allowed to boot up, after the Camera first displays a Live Image from the Sensor, i would wait at least 3 Seconds, then Plug HDMI in. In case HDMI wouldnt output anything, i would wait 3 Seconds, and if nothing else happens, i would unplug it again, wait for another 3 Seconds, and Plug it back in. In case of a "Freeze" of the Camera/ Softlock Situation, i would have to Force power the Camera down. To give it a fair and Reproduce-able result, i have booted the Camera back up after any Freezes and Power it down normally again.
 Sample size was 20 reboots (reboots to "reset" the Camera after Freezes are not counted in this). Results where 20 times no HDMI Output at all (there is a signal, but the Only output is a static Solid black Frame), and after removing HDMI and Plugging it back in a Freeze on all 20 tries, after which the camera wouldnt respond to any inputs anymore, and the Displayed image on screen would go dark.
 .
 .
 Test Nr.2
 After this i did another test.
 This time, i Had the Monitor Powered off, but Plugged in already while the camera was still powered off. I would then boot the Camera up, allow the Camera to fully boot, and also wait for at least 3 Seconds after the Camera outputs the first live Image Output on the Internal Screen, bevore powering the External Monitor on. After each test, i would first power down the Camera, wait 3 Seconds to Power down the Monitor, wait another 3 Seconds for the Next Boot /Reboot. If i would run into a Freeze, i would do the same as on Test Nr.1, and reboot the Camera to make sure i was powered down correctly bevore the next try.
 Sample size on this Test was 104 Reboots (also not counting The Reset-Boot-Cycles for the Freezes).
 Result was 100 "normal" boot-Cycles, where HDMI would work as intended, and 4 Freezes.
 Those Freezes Happened on Boot Nr. 3; 44; 45 and 73 out of 104 Total Reboots.
 .
 .
 Test Nr.3
 As i feel like i didnt quite wait the Full 3 Seconds (about the Time the Camera still shows "No Batt" on screen and needs to switch to the Actual Voltage and Percentage after first Image Output is Displayed), especially on Boot Nr. 3 and 73 of Test Nr.2, i did a Control-Test, where i used basically the Same Testing rules as for Test Nr.2, but would Power up my external Monitor shortly before, while, or very shortly after the Camera would display the First Live Image Output from the Sensor.
 Sample Size was 5 Reboots on this, all 5 of them just worked normally, and displayed Proper HDMI Output Image.
 .
 .
 Conclusion/ Summary:
 While i still dont really understand whatever is going on here, i would claim to have found a reasonable reliable way to get a Freeze and a somewhat reliable Way to get an HDMI output working on >>90% of Boots.
 There is still a lot more going on, and a lot of more testing to be done, because it can be really inconsistent sometimes, but for the described order of operation it seems very reproduceable in terms of Results, at least for my Unit.
 .
 .
 Further Notes:
 .
 There is some other weird stuff going on, i noticed.
 For this series of tests i just Rebooted the Camera a total of 158 Times (a bunch more before i started systematically testing, but didnt take any notes about this).
 During all those Reboots, i noticed that the Camera would just randomly allready be recording right after Boot, while the Option "Auto Record" is not enabled on my camera, and also never was during all the Testing, i checked multiple times after i noticed. This happened 18 times out of the 158 Reboots. While this isnt really an issue (at least for me...), i think this is also not supposed to happen, and i couldnt figure out, why it would do that. Never Rebooted that many times in a Row, so i never really noticed or questioned that, and just assumed i would have accidentally hit the Record Button whenever it happened, but on those tests i was sure i had not, also happened quite often (>10% of Boots).
 .
 On 2 of the Boots, the camera took like a full minute to then decide it couldnt boot, powering down completely, and then booting normally, without me needing to do anything more than wait. One was right after a Freeze, so i would consider that "out of normal Conditions", but the other one happened just randomly between two of the Working Tries on Test Nr.2, so that was somewhat surprising.
 .
 On one of the Reboots, the camera just showed a really glitched out Image on screen, i didnt even bother trying the HDMI, and just restarted, but took an Image of if, see attachment. Never had this happen since i have the camera, but seen someone else post something just like this on the Forum somewhere. Worked just fine the Next after a Restart of the Camera.
Title: Re: BUG REPORT --- HDMI output Problems and Softlocking the Camera
Post by: clkdiv on July 05, 2021, 03:30:29 PM
Maybe this randomness is the same randomness that appears with other features of the camera? As posted here in the forum I recently had some issues with the cameras IO settings. Seems this happens to others too now and then.

Since the camera remembers its settings when being turned off and booted up again, there has to be a file or a place where the settings are stored. If the values are messed up somehow, maybe this might be a reason?

However it would be nice to know where settings are stored...
Title: Re: BUG REPORT --- HDMI output Problems and Softlocking the Camera
Post by: BobbyKurtz on July 08, 2021, 09:12:04 AM
Did a test shoot last night, hooked up the NinjaV hoping I might get it work using some of the methods described above.  Was curious if it was lighting levels as I had done some tests with monitors without a fully lit scene.  Had 2 Forza 500's and Nanlite FS300 so plenty of light. 

Still unable to get anything but black but monitor is sending 1080p60.
Title: Re: BUG REPORT --- HDMI output Problems and Softlocking the Camera
Post by: Nikon1 on July 08, 2021, 09:21:32 AM
i did use the monitor/ HDMI quite a bit with Sensor Blocked (so just Black Level Noise) output as an image, and about half of the Test Which i posted here was done at/ after sunset, with just a Single Lamp in the Distance even visible in Frame, everything else basically black. So really bad light for Highspeed. Barely enough to see that there was a valid Live image Displayed on the External Monitor.
 .
 While i never particularly tested it, i would still have to say that i doubt that Image Brightness does have anything to do with this Bug at all. Could be wrong, but never had any experience that would even make me consider that this would be the case.
 But let me know if you make any observations that have different results.