Concerns about the "glitchless" category
6 years ago
United States

Right now, there are two categories of the game:

  • RNG Manipulation + RNG freeze + Memory Corruption (dead party glitch)
  • RNG Manipulation + RNG freeze, but without Memory Corruption

The problem is that RNG freeze itself is a glitch. While it doesn't outright corrupt any memory, it does freeze a Linear Feedback Shift Register to no longer produce random values. So despite being called "glitchless", it still relies on the RNG freeze bug in the game.

Then there's Basic RNG Manipulation, where you use Change Message Speed on another save file to reset the RNG to something desirable. Should runs that forgo that get their own category?

Michigan, USA

The RNG is not technically frozen, it is in a extremely short sequence. Short loops are commonplace in a LFSR and not a bug, glitch, or programming error/oversight.

A category with no seedlock but allow other savegame checksum manipulations could be created but until there are runs of it or runners showing interest in it, there is no reason to create more categories.

Any% - glitchless arguably might not be the most accurate and appropriate name but is done for the sake of simplicity. Until a new category is made that could be confused with the category in question, I see no reason to change it unless it is requested by others in the community.

Thanks for your interest.

From Vaxherd's RNG research: "there are two initial seeds which give an LFSR sequence length of just 2: $5FEA and $AFF5. This property itself is unavoidable in any LFSR equation.... Since the RNG algorithm clocks the LFSR 16 times per random n umber, this effectively means that if the seed is ever set to one of these two values, the RNG will simply return the counter plus a constant value."

編集者 投稿者 6 years ago
ゲームスタッツ
フォロワー
63
走行
47
プレイヤー
13
最新スレッド
モデレーター