Low Throw Frame Data

Core Information Summary

• Typical low throws are 10 frame startup with 1 active frame and 36 total frames
  
  
• When a player is forced into a crouch position from a hit, they are still registered as crouching for another 4 frames. This means a +6 normal hit forced crouch move will allow the attacker to attempt a low throw on the opponent if the opponent does not attack or evade.
  
  
• The forced crouch position from a hit also allows the defender to use from crouch moves without the need to buffer a crouch.

Detailed Information Summary

• From neutral, holding 2+G for a minimum of 5 frames will register the player as fully crouched for a single frame's duration for throw purposes. If 2+G is held for exactly 5 frames, the player will be registered as fully crouched for throw purposes for exactly 1 frame. EDIT: Specifically, I am stating it requires holding the input for 5 frames to register a crouch for a duration of 1 frame. This does not state anything about when the player enters a crouching state.
  
  
• A forward crouch dash will be considered crouching for throw purposes for 8-22 frames (depending on if and when G is pressed). EDIT: Specifically, I am stating the duration of a crouch state, not when the crouch state occurs.
  
  
• A backward crouch dash will be considered crouching for throw purposes for 19-21 frames (depending on if and when G is pressed). EDIT: Specifically, I am stating the duration of a crouch state, not when the crouch state occurs.

To Do List

• Test low throws from characters other than Aoi
  
  
• Test more forced crouching moves other than Aoi's 6KP, 64P, and 3P moves.

Recreating Testing Data

• Tests were done only with Aoi low throws, so data may be skewed.
  
  
• I suspected a 10 frame execution, with 1 active frame for low throws based on side roll fall recovery windows and verified it with further testing in various circumstances. Low throws needed to be done 4 frames later than Aoi's 6P to hit the same frame, implying a 10 frame startup. Active frames longer than 1 frame would have resulted in much larger low throw windows. The nail in the coffin was in neutral testing where the low throw window could be restricted to only 1 frame.
  
  
• Testing scenarios were considered from the DIY frames analysis thread, however some of the data was either incorrect or changed for REVO. Attempting to recreate some of the scenarios resulted in different outcomes from what was stated in the thread. One of the first tests stating low throws may have a 0 frame startup with 10 active frames was from an incorrectly applied 1P > 33G scenario with Kage and Aoi. Strictly applying the inputs causes the low throw to whiff instead of connect. Further testing showed a crouch dash can have 8-22 frames of being fully crouched depending on when G is pressed.
  
  
• Testing data can be recreated from the attached Eddienput scripts. They assume the vf5revo_keyboard.json is used from the Testing Frame Data thread.
  
  
• To easily sync the attacker and defender actions, the recorded commands should not enable the delay option. The non-recorded commands should enable the delay option. A 2-frame delay is usually consistent on my setup. If the recorded commands do not sync properly to the 2-frame delay, re-recording the commands normally does the trick instead of needing to change the 2-frame delay. Your mileage may vary.

 

I was very specific with my wording. I did not say at what frame either input enters a crouching state. I was stating the durations. The durations were stated since I did not see readily available information on the duration of the crouching states. I'll edit the original post to emphasize those points. If you want a video demonstration of at least the first point, I just uploaded a video showing holding 2+G for 5 frames registers a character as fully crouched for at least a single frame. It is exactly a single frame, but I did not bother showing how shifting the timing would result in the high throw connecting unless you want further video proof. Holding the input for more frames results in more frames of crouch state, as well. The setups are notated in the Eddieinput scripts (2g attacker/defender).

This post was specifically to determine what the frame data was on low throws since searching for information on low throw frame data was mostly missing. The closest I found was in the DIY Frames Analysis thread stating low throws were likely 0 frame startup with 10 active frames. This did not seem to match up with my experience of how the rules work in VF5, so I set about trying to test it as rigorously as I could.