Author Topic: SpaceX Falcon 9 v1.1 CRS-3 Splashdown Video Repair Task Thread  (Read 791414 times)

Offline princess

  • Member
  • Posts: 65
  • Liked: 106
  • Likes Given: 25
Re: SpaceX Falcon 9 v1.1 CRS-3 Splashdown Video Repair Task Thread
« Reply #1880 on: 07/16/2015 12:41 PM »
I had a look at the Lightsail data and I can confirm what saliva_sweet found - it's not JPG data, it's some other data in 1024-byte packets. Other things I found are:

1. The packets seem to be repeating data with bitflips between each packet. I can't see any values that change numerically, and the bitflips seem fairly randomly distributed through the packet, so I don't think it's telemetry. It seems more like the computer is attempting to transmit the same packet again and again, with corruption occurring.

2. I tried splitting the packets at different boundaries - the lowest difference between consecutive packets occurs if you start at 928 bytes into the file.

3. The packets go out of alignment in two different places, by two bytes each time (both are a two byte deletion). If you take the first line of each packet's hexdump and display them one after another, the misalignment places are easy to see.

4. There seem to be two different types of packet. The file initially only contains the first type of packet, and then the second type starts appearing interleaved with the first. It's not a perfect interleave, you sometimes get multiple instances of one type before it flips to the other type.

In summary, it looks more like the repeated transmission of two different 1K packets of data, with random corruption. I'm not sure if a radio fault or a CPU fault would do this, but it's certainly intriguing!

Offline meetsitaram

I was looking back at the video that has been improved here, at http://www.spacex.com/news/2014/04/29/first-stage-landing-video but that video no longer works. Had that been removed?

Offline moralec

A little bit of our video made it here.
Very happy to know that NSF played a role in this amazing project.


Offline princess

  • Member
  • Posts: 65
  • Liked: 106
  • Likes Given: 25
Re: SpaceX Falcon 9 v1.1 CRS-3 Splashdown Video Repair Task Thread
« Reply #1883 on: 10/19/2016 04:25 PM »
In case anyone is interested, I've posted my MPEG-TS repair software to Github. The repository contains the latest code that I used to make the final version of the TS file - the one that went into the main MPEG4 macroblock repair effort and the fixed TS file that I used to make the pretty error pictures.

Link is here: https://github.com/pmillerchip/spacex-tsrepair

Over the next few days I'll try to get together all my fix commands, which will allow someone to download the raw.ts file from SpaceX's page and produce a fixed TS file in one go. Pull requests are always welcome!

Offline Req

  • Full Member
  • ***
  • Posts: 355
  • Liked: 277
  • Likes Given: 2499
Re: SpaceX Falcon 9 v1.1 CRS-3 Splashdown Video Repair Task Thread
« Reply #1884 on: 09/14/2017 07:34 AM »
Our video made it into the crashes video  ;D

https://twitter.com/elonmusk/status/908229340139323392
@elonmusk
Just posted a video https://www.instagram.com/p/BZA0s7EAmF1/

Edit:
https://twitter.com/elonmusk/status/908254739825025025
@elonmusk
High res version at
« Last Edit: 09/14/2017 09:19 AM by Req »

Offline Lar

  • Fan boy at large
  • Global Moderator
  • Senior Member
  • *****
  • Posts: 8160
  • Saw Gemini live on TV
  • A large LEGO storage facility ... in Michigan
  • Liked: 4913
  • Likes Given: 3334
Re: SpaceX Falcon 9 v1.1 CRS-3 Splashdown Video Repair Task Thread
« Reply #1885 on: 09/14/2017 10:45 PM »
Our video made it into the crashes video  ;D
Awesome, even if uncredited (I tweeted Elon about it, I'm sure he'll get right on fixing that)
"I think it would be great to be born on Earth and to die on Mars. Just hopefully not at the point of impact." -Elon Musk
"We're a little bit like the dog who caught the bus" - Musk after CRS-8 S1 successfully landed on ASDS OCISLY

Tags: