Sentry Sucks USB Space

[SMAlset]

I’m not positive but I think when TeslaCam was first introduced it was only on during driving and Park would disable it. Also it was only the front camera recording. With the update to TeslaCam and addition of Sentry Mode for parking, TeslaCam now records 24 hours straight (over writes in the 1-hr buffer file “RecentClips) and they’ve included footage from the two side repeaters to TeslaCam, so the buffer went from 1.8GB to 5.4GB. Plus any saved driving videos are now coming from 3 cameras. Same applies to Sentry Mode parked footage. Any SavedClip files are not overwritten.

So yes, unless manually turned off in your garage for example TeslaCam is recording all the time. You might also notice the gaps in time from the buffered 1-hr of files. If something is manually Saved or Saved from Sentry Mode, it will pull those 10 minutes from the live buffer, create a dated/time stamped folder in the SavedClips folder and deposit the 10-minutes there.

 

[BobAbooey]

That should work well. Here's a guide:
MicroSD Cards We Recommend for Dash Cameras

Thank you

 

[Knightshade]

I’m not positive but I think when TeslaCam was first introduced it was only on during driving and Park would disable it. Also it was only the front camera recording. With the update to TeslaCam and addition of Sentry Mode for parking, TeslaCam now records 24 hours straight (over writes in the 1-hr buffer file “RecentClips) and they’ve included footage from the two side repeaters to TeslaCam, so the buffer went from 1.8GB to 5.4GB. Plus any saved driving videos are now coming from 3 cameras. Same applies to Sentry Mode parked footage. Any SavedClip files are not overwritten.

So yes, unless manually turned off in your garage for example TeslaCam is recording all the time. You might also notice the gaps in time from the buffered 1-hr of files. If something is manually Saved or Saved from Sentry Mode, it will pull those 10 minutes from the live buffer, create a dated/time stamped folder in the SavedClips folder and deposit the 10-minutes there.

The cameras don't record when the car is sleeping- same as before.

The only change to the dashcam feature is it now also records the 2 side cameras (increasing the buffer size 3x)

Then they also added a different feature, which is Sentry mode, which is only "on" when you turn it on. This mode uses the same 3 cameras, but only a 10 minute rolling buffer.... if the car 'alerts' to someone coming near it, it saves the buffer permanently.... so how much space this will eat if left on will depend entirely on how busy the area near the car is while it's on.

One user filled a 32GB key in 8 hours at work because it was a busy area for example.

When the car has sentry mode on it never goes into deep sleep (which makes connecting with the app virtually instant, but it does eat about 1 mile of range per hour in this mode)

 

[muleferg]

These work for me. Model X and Model 3P. 256.

 

[PoitNarf]

View attachment 391154 These work for me. Model X and Model 3P. 256.

This is the drive I'm using. Zero issues so far.

 

[GTIceman]

View attachment 391154 These work for me. Model X and Model 3P. 256.

I had this drive in a 64GB that I have been using, likely will jump to the 256. My only issue was it filled and didn't seem to erase for some reason. Happened only once.

 

[japhule]

I would continue to not recommend any flash based USB drives. I think it'll be a matter of short time before the drive goes bad with the amount of constant writing that's going on. My Samsung FIT 32GB usb flash drive worked for a bit before sentry mode. As soon as sentry mode upgrade came, the drive died quickly afterwards. Use a dashcam friendly microsd cards with a usb adapter.

Samsung PRO Endurance
https://www.amazon.com/gp/product/B07B9KTLJZ/ref=oh_aui_search_asin_title?ie=UTF8&psc=1

Micro SD USB adapter
https://www.amazon.com/gp/product/B01FRR86QU/ref=oh_aui_search_asin_title?ie=UTF8&psc=1

 

[Knightshade]

I would continue to not recommend any flash based USB drives. I think it'll be a matter of short time before the drive goes bad with the amount of constant writing that's going on. My Samsung FIT 32GB usb flash drive worked for a bit before sentry mode. As soon as sentry mode upgrade came, the drive died quickly afterwards. Use a dashcam friendly microsd cards with a usb adapter.

Samsung PRO Endurance
https://www.amazon.com/gp/product/B07B9KTLJZ/ref=oh_aui_search_asin_title?ie=UTF8&psc=1

Micro SD USB adapter
https://www.amazon.com/gp/product/B01FRR86QU/ref=oh_aui_search_asin_title?ie=UTF8&psc=1

....you realize SDcards are also flash based, right?

 

[japhule]

....you realize SDcards are also flash based, right?

Yes. I'm also not recommending just any sd card either. Dashcam friendly cards like the one above are ideal for dashcam and constant video recording uses. I'm not aware of any USB flash drives that are built for dashcam uses, maybe b/c most dashcams use sd cards. As linked earlier in the thread, there's more detail in the link below.

MicroSD Cards We Recommend for Dash Cameras

 

[japhule]

And to clarify, modern SSDs are also flash based, I'm not saying don't use large external USB SSD drives (multi/parallel flash). I'm talking more about the cheaper flash drives (single flash memory). Trying to save tesla owners from buying usb flash cards that they'll regret later.

Recommended:
SSD
High Endurance micro sd cards

Not Recommended:
single memory USB flash drives (ie, most small usb flash drives)

 

[DB-Cooper]

When the car has sentry mode on it never goes into deep sleep (which makes connecting with the app virtually instant, but it does eat about 1 mile of range per hour in this mode)

I left my car at the airport, it was plugged in at 90% before someone removed the plug (shared chargers). It was removed on Wednesday about mid-day, it's now Friday and as of 6am PDT, my car is at 78%. Let's call it 12% in 40 hours. 12% of 310 is 37-38 miles, so your 1 mile per hour of battery drain appears to be pretty spot-on. Granted, even with Sentry disabled, I'd suspect I'd lose 1-2% or 3-5 miles per day, but it definitely increases drain by about 5X.

 

[Kmartyn]

128 GB Sandisk acquired. $17.99.

Windows 10 can't format the drive with more than 32GB, but luckily it comes from the factory formatted to 128- just have to create the TeslaCam folder- seems to be working just fine.

https://www.amazon.com/gp/product/B00TKFCYP0/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1

Be aware that your selected USB drive is not rated to work below freezing. That’s why I ordered the Samsung, to get a -40 rating. If you live up north better to get one with the cold temperature rating.

 

[astrowunder]

I suspect that's just a safety rating or something. Usually solid-state electronics don't mind colder temps. I probably won't find out now until next winter.

 

[SMAlset]

Summer heat inside the car in a closed compartment like where the USB connections are...I’m not sure even my Blackvue microSD card will do well then. Should be an interesting test. Might be relying more on Tesla’s temporary storage of video in accident or vandalism cases during that time.

When we had our Blackvue 650S in our Model S and pretty hot temps in 90s-100s (and I think this was before cabin overheat protection but might have been even after it), we’d get video files during hotter hours of the day that would be pink and green. As soon as the temp got cooler the video would be fine so we knew it was temperature related and same was happening to other owners like in Australia who put up videos of their video corruption on a dashcam site. Now was this a result of media or components overheating in the camera I don’t know.

 

[SMAlset]

This is the drive I'm using. Zero issues so far.

Your seeing great videos across the board? No green screen, no blurry or pixelated videos or zero size files?

 

[srs5694]

When we had our Blackvue 650S in our Model S and pretty hot temps in 90s-100s (and I think this was before cabin overheat protection but might have been even after it), we’d get video files during hotter hours of the day that would be pink and green. As soon as the temp got cooler the video would be fine so we knew it was temperature related and same was happening to other owners like in Australia who put up videos of their video corruption on a dashcam site. Now was this a result of media or components overheating in the camera I don’t know.

Digital video files are stored numbers, usually with compression algorithms applied to them. It's hard to imagine how the numbers could be corrupted by heat in such a way as to create a consistent change in color to give them an overall pink or green hue. Corrupted media would more likely produce completely unviewable videos, stuttering, pixellation, or something like that -- these are symptoms of damage to video files.

An overall change in the color of a video is more likely the result of problems with the video sensor. Digital camera/video sensors detect millions of red, green, and blue point (pixel) sensors. If the green sensors are consistently over-firing, the images would acquire a green hue. A magenta cast (perhaps close enough to pink that you'd describe it that way) could be produced by under-firing of those same green sensors. Once the image is processed into video, these sensor errors would stay locked in, and if the files stored on the media were not corrupted, there'd be no problem with playback -- just the inaccurate color cast. You could get similar results if the monitor were malfunctioning, too, but from your description, this seems unlikely.

 

[PoitNarf]

Your seeing great videos across the board? No green screen, no blurry or pixelated videos or zero size files?

To be honest I haven’t checked the video files in over a week. I probably should pop the flash drive into my laptop this weekend to confirm the quality of the video isn’t getting messed up...

 

[SMAlset]

Digital video files are stored numbers, usually with compression algorithms applied to them. It's hard to imagine how the numbers could be corrupted by heat in such a way as to create a consistent change in color to give them an overall pink or green hue. Corrupted media would more likely produce completely unviewable videos, stuttering, pixellation, or something like that -- these are symptoms of damage to video files.

An overall change in the color of a video is more likely the result of problems with the video sensor. Digital camera/video sensors detect millions of red, green, and blue point (pixel) sensors. If the green sensors are consistently over-firing, the images would acquire a green hue. A magenta cast (perhaps close enough to pink that you'd describe it that way) could be produced by under-firing of those same green sensors. Once the image is processed into video, these sensor errors would stay locked in, and if the files stored on the media were not corrupted, there'd be no problem with playback -- just the inaccurate color cast. You could get similar results if the monitor were malfunctioning, too, but from your description, this seems unlikely.

I don't know if I saved any images or video from back then, would have been 2017, but will look. Also know I saw footage from people in Australia on a dashcam forum and it's pretty much what our camera was doing and only on really hot days and it "fixed" itself when the temps came back down that same day.

 

[SMAlset]

Digital video files are stored numbers, usually with compression algorithms applied to them. It's hard to imagine how the numbers could be corrupted by heat in such a way as to create a consistent change in color to give them an overall pink or green hue. Corrupted media would more likely produce completely unviewable videos, stuttering, pixellation, or something like that -- these are symptoms of damage to video files.

An overall change in the color of a video is more likely the result of problems with the video sensor. Digital camera/video sensors detect millions of red, green, and blue point (pixel) sensors. If the green sensors are consistently over-firing, the images would acquire a green hue. A magenta cast (perhaps close enough to pink that you'd describe it that way) could be produced by under-firing of those same green sensors. Once the image is processed into video, these sensor errors would stay locked in, and if the files stored on the media were not corrupted, there'd be no problem with playback -- just the inaccurate color cast. You could get similar results if the monitor were malfunctioning, too, but from your description, this seems unlikely.

I don't know if I saved any images or video from back then, would have been 2017, but will look. Also know I saw footage from people in Australia on a dashcam forum and it's pretty much what our camera was doing and only on really hot days and it "fixed" itself when the temps came back down that same day.

ah, found that I saved an image of our issue back in 2017 and posted to the forum here. What do you think?

Blackvue or Thinkware dash cam, and installer in San Diego

My husband has a friend that fabricates things and he had him design a heat sink to go around our Blackvue DR900-S 2-ch since he knew it would run hotter than our DR650-S that it replaced. Saw that other people started having problems in the heat with their 900s but ours has been running fine. Heat really can be the nemesis of dashcams. We'll run cabin overheat protection on when interior temps start reaching that threshold.

 

[srs5694]

ah, found that I saved an image of our issue back in 2017 and posted to the forum here. What do you think?

Blackvue or Thinkware dash cam, and installer in San Diego

That image looks like file corruption to me. The digital corruption could be anywhere in the system -- storage or something before or after the storage (like data getting garbled between the sensor and the CPU that does the compression). I'm not sure why most of the image is green and pink in this case; I'm not enough of an expert on digital imaging technology to know if that could be an artifact of the data corruption or if it indicates a sensor problem in addition to the data corruption.