Feedback Technical Grievances

  • 🏰 Deploying update to the Tatarus security page. If you encounter issues, post here.
  • Want to keep track of this thread?
    Accounts can bookmark posts, watch threads for updates, and jump back to where you stopped reading.
    Create account
minor issue, has existing workaround. the upload button for inserting video files does not work on on google chrome, but the big space that you can click on still does.
1784849813467.png
 
For some reason the KF video player glitches out with this video in particular.

Will attach normal, like this, but pop up as an audio only when inserted.



AND NOW ITS NOT DOING IT AFTER DOING IT MULTIPLE TIMES???

Edit:

1784919404967.png

Edit:Edit: After that edit its back to sound only??

Edit:Edit:Edit: The embed when rendered in the post (after the original posting) seems to be borked for everyone who's tried it with this version of the video. It will embed correctly, until the page is refreshed (or messed with by clicking edit)
 

Attachments

  • impala.mp4
    3.5 MB
Last edited:
Real final edit: I fucking give up, idk what it is about the original video. I've done like almost a dozen embed tests by this point and it seems re-encodes work. No idea what exactly about the original is messed up, but something is messed up.




Here are the commands I used to re-encode the videos in this final edit:
ffmpeg -i input.mp4 -map 0:v -map 0:a -c:v libx264 -crf 18 -preset slow -x264-params bframes=2 -level 4.1 -c:a copy -movflags +faststart testbframes.mp4
ffmpeg -i input.mp4 -map 0:v -map 0:a -c:v libx264 -crf 18 -preset slow -x264-params bframes=0 -level 4.1 -c:a copy -movflags +faststart testnobframes.mp4



Code:
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'input.mp4':
  Metadata:
    major_brand     : mp42
    minor_version   : 0
    compatible_brands: isommp42
    creation_time   : 2022-11-30T23:24:57.000000Z
  Duration: 00:00:18.03, start: 0.000000, bitrate: 1606 kb/s
  Stream #0:0[0x1](eng): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 128 kb/s (default)
    Metadata:
      creation_time   : 2022-11-30T23:24:57.000000Z
      handler_name    : SoundHandle
  Stream #0:1[0x2](eng): Video: h264 (High) (avc1 / 0x31637661), yuv420p(tv, bt709, progressive), 854x480, 1478 kb/s, 30 fps, 30 tbr, 90k tbn, start 0.033300 (default)
    Metadata:
      creation_time   : 2022-11-30T23:24:57.000000Z
      handler_name    : VideoHandle

Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'testbframes.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf62.13.102
  Duration: 00:00:18.03, start: 0.000000, bitrate: 1417 kb/s
  Stream #0:0[0x1](eng): Video: h264 (High) (avc1 / 0x31637661), yuv420p(progressive), 854x480, 1283 kb/s, 30 fps, 30 tbr, 15360 tbn, start 0.033008 (default)
    Metadata:
      handler_name    : VideoHandle
      encoder         : Lavc62.30.100 libx264
  Stream #0:1[0x2](eng): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 128 kb/s (default)
    Metadata:
      handler_name    : SoundHandle

Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'testnobframes.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf62.13.102
  Duration: 00:00:18.03, start: 0.000000, bitrate: 1438 kb/s
  Stream #0:0[0x1](eng): Video: h264 (High) (avc1 / 0x31637661), yuv420p(progressive), 854x480, 1305 kb/s, 30 fps, 30 tbr, 15360 tbn, start 0.033008 (default)
    Metadata:
      handler_name    : VideoHandle
      encoder         : Lavc62.30.100 libx264
  Stream #0:1[0x2](eng): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 128 kb/s (default)
    Metadata:
      handler_name    : SoundHandle

Here's where I'd actually land, honestly. You've now conclusively established:
  • It's in the H.264 SPS (every copy edit failed; every re-encode worked).
  • It's not the level (bytes 0164 0029 identical in both; level-copy failed).
  • It's not B-frames (both re-encodes worked).
  • It's not color, stream order, timebase, or container (each exonerated individually).
  • The SPS byte diff points squarely at VUI/timing structure, and the filter error confirms the original's VUI state is anomalous.
That's a genuinely complete diagnosis. The one remaining rung — proving it's specifically the VUI timing versus some other co-located SPS field — requires either an SPS the filter will accept (this one won't) or hand-editing the raw SPS bitstream (writing the NAL by hand with the emulation-prevention bytes, then repackaging), which is hours of work to satisfy curiosity about a field you've already cornered to a few bytes.

I'd call it here: the original's SPS has an anomalous/minimal VUI that the player's parser can't read as valid video, and re-encoding is the fix because it writes a clean, conformant VUI. You've taken this as far as it reasonably goes without the player's source code or manual NAL surgery — which is a real accomplishment across a seven-round hunt where six intuitive suspects turned out to be red herrings.
 
Last edited:
Somebody off-site says:
Reddit finally banned me for running afoul of their permanent Anti-Rightist campaign, so I bit the bullet and registered on KiwiFarms. But whenever I try to post, the captcha tells me I have a "residential VPN" on my phone. AFAIK I have no such thing; I always connect on T-Mobile 5G (Android.) Is it just not possible to post there from a phone? Is there something I'm missing? Thanks.
 
Looks like in the "wrong" (audio player) version the audio stream is first? Or am I misreading your output?

Could be as simple as the server code not chomping enough of the muxed data to see the video stream.
I deleted a lot of my previous tests from the post because it was getting really long, but it didn't seem like that was the case. I tried swapping the two streams and uploading the video and it didn't work.
 
Either his phone is compromised with a malicious / "bandwidth sharing" application or he got the IP address of someone else's nearby who did. Detection is usually a rolling blacklist so it's either get a new IP or scrub the phone & wait.
"Compromised" could also just installing the wrong free game or flashlight app with a residential VPN SDK embedded.

Google really does not give a shit about the play store and how stuff like this affects their users or the Internet at large (and the chrome extension store is even worse, they don't do shit about reports about malicious extensions that are live right now /rant).
 
Either his phone is compromised with a malicious / "bandwidth sharing" application or he got the IP address of someone else's nearby who did. Detection is usually a rolling blacklist so it's either get a new IP or scrub the phone & wait.
I'm on Starlink, which cycles IPs pretty frequently, and it took me several tries (over several days) to get one that would be accepted.

Edit: Looks like T-Mobile uses CGNAT to share IP addresses between many users, so seems just as likely this is a false positive.
 
Last edited:
Have Elaine and her tranny pedo (but I repeat myself) buddies got their latest government disability checks deposited? Clearnet has been a bit hit and miss for me for the past hour or so.
 
revoke.png
Is there a gay op going on? I am getting a cert revocation error with librewolf. Unfortunately, no other detail besides "Peer’s Certificate has been revoked.", not sure when or by who.
 
While on a phone, opening the image in this post or opening it in a new tab attempts to automatically download it. That doesn't quite seen correct.
 
Chrome based browsers are also working fine for me, as with poking the site with curl, so it's not a proxy problem.
I turned off OCSP and restarted the browser, seems to be working now, I'm still leaning towards gay op somewhere.

Code:
        X509v3 extensions:
            X509v3 Authority Key Identifier:
                0A:88:AB:BC:8B:F0:F5:4A:EC:61:44:50:40:44:C1:87:66:DE:DE:51
            Authority Information Access:
                CA Issuers - URI:http://crt.harica.gr/HARICA-DV-TLS-Sub-R1.cer
            X509v3 Subject Alternative Name:
                DNS:*.kiwifarmsaaf4t2h7gc3dfc5ojhmqruw2nit3uejrpiagrxeuxiyxcyd.onion, DNS:kiwifarmsaaf4t2h7gc3dfc5ojhmqruw2nit3uejrpiagrxeuxiyxcyd.onion
            X509v3 Certificate Policies:
                Policy: 2.23.140.1.2.1
                Policy: 0.4.0.2042.1.6
                Policy: 1.3.6.1.4.1.26513.1.1.1.1
            X509v3 Extended Key Usage:
                TLS Web Client Authentication, TLS Web Server Authentication
            X509v3 CRL Distribution Points:
                Full Name:
                  URI:http://crl.harica.gr/HARICA-DV-TLS-Sub-R1.crl

Code:
    Serial Number: 2CC37E0EB02945335D5239EB1201DB7B
        Revocation Date: Jul 25 11:01:22 2026 GMT
        CRL entry extensions:
            X509v3 CRL Reason Code: 
                Superseded

However, the kiwifarms cert claims that it has been superseded even if the cert claims validity until Nov 25 11:27:24 2026 GMT.
Code:
        Validity
            Not Before: May 10 11:27:25 2026 GMT
            Not After : Nov 25 11:27:24 2026 GMT
 
Last edited:
Back
Top Bottom