Testing I did for how safe sidestep/backdash are.

Harry

[10] Knight
Today I decided to test how safe sidestep and backdash are because I haven't seen hard numbers on it. Only people saying it's safer than it was in SCV which isn't that helpful. How I tested was with Sophi and Sophi since there is frame data available for her. I set the cpu to 1st action guard and 2nd action AA since Sophi's AA is i12. Then with the Sophi I controlled I would do 44K because frame data says (and my own test) says it's neutral on block. Attempting to sidestep after a blocked 44K would get me hit by the AA. I then set the cpu Sophi to do 3A which is a i18 move, and I did 4KK again. That was safe. I could block before 3A connected. I then tested it with 4K which is -2 on block. That was also safe. Then I tested it against Jump B which is -4 on block and that was safe as well. So my test gives me that sidestep is -12. A member in SC Competitive FB group told me that his testing found backdash to me -14, so I tested that with Sophi Jump B on block against Sophi 3A and that checked out. So TL:DR my test have found that sidestep is -12 and backdash is -14. UPDATE- Slade has found that backdash is -15 not -14.
 
Last edited:
Confirming that sidestep is -12:
  • Groh B (+2) -> A (i12) = F10, not blockable
  • Groh AVN A+B6 (+9) -> AVN A+B (i20) = F11, not blockable
  • Groh 6A (+0) -> AA (i12) = F12, not blockable
  • Groh 4A6 (+3) -> AVN K (i16) = F13, blockable
However, I think that backstep is -15:
  • Groh 6A (+0) -> BB (i14) = F14, not blockable
  • Groh A+B6 (+9) -> AVN A (i24) = F15, not blockable
  • Groh 6A (+0) -> 6B (i16) = F16, blockable
To verify that the F15 setup is in fact F15: BB (i14) beats AVN A (i24) but 6B (i16) loses.
 
Whoever told you was wrong.
SCV had Quickstep and the infamous quickstepG glitch. The fastest and safest step of the Whole series.
The frames are right there. -12 and -15 is safer than -20 and -20.

If by "the infamous quickstepG glitch" you're referring to 665G, that only negated step counters and allowed one to block CEs. 5 didn't have actual step G. And as for the most evasive step, that title probably goes to SC1. Step canceling in that game was nuts.

5's is certainly more evasive than 6's, though. Movement in 6 feels like SC4 movement minus step G. Nightmare can't even backstep Talim's AA now. It's awful. It's if the designers thought it would better to force slow characters to counter mashing by mashing armored moves, instead of treating the game like an actual 3D fighter where spacing matters.
 
What darkfender is referring to is the pre-patch SCV QS4G, which wasn't a glitch. A glitch implies something that isn't normally possible or supposed to be possible. QS4G was something they've known you could do since SCIV (although in SCIV is was just sidestep 4G since it didn't have QS). Backdashing in pre-1.03 patch SCV was safe because you could cancel into guard on reaction to attacks. So, people would sidestep or QS and cancel into 4G to make movement safe. It's why a lot of Tekken players liked SCV before the 1.03 patch. @darkfender however in the 1.03 patch they changed backdash to -20, so that any movement other than forward movement was unsafe. QS4G is far from being the most OP movement option in SC. Step G will forever be the best movement glitch/OS SC has ever had because unlike SCI, SCIV, and SCV (and I think SCIII) Step G didn't cancel your movement. You only stopped to block if you're opponent did a horizontal attack. @Slade oh okay. I only tested this with Sophi mirrors, and she doesn't seem to any attacks that can give me odd number frame data, so the best I could do was approximate how minus backdash was.
 
Last edited:
I forgot to mention that Run Counter state is active past the unsafe window of movement. I believe it's for 20 frames after you sidestep/backstep, but not entirely sure. It's around 18-20 frames that Run Counter state is active.
 
What darkfender is referring to is the pre-patch SCV QS4G, which wasn't a glitch. A glitch implies something that isn't normally possible or supposed to be possible. QS4G was something they've known you could do since SCIV (although in SCIV is was just sidestep 4G since it didn't have QS). Backdashing in pre-1.03 patch SCV was safe because you could cancel into guard on reaction to attacks. So, people would sidestep or QS and cancel into 4G to make movement safe. It's why a lot of Tekken players liked SCV before the 1.03 patch. @darkfender however in the 1.03 patch they changed backdash to -20, so that any movement other than forward movement was unsafe. QS4G is far from being the most OP movement option in SC. Step G will forever be the best movement glitch/OS SC has ever had because unlike SCI, SCIV, and SCV (and I think SCIII) Step G didn't cancel your movement. You only stopped to block if you're opponent did a horizontal attack. @Slade oh okay. I only tested this with Sophi mirrors, and she doesn't seem to any attacks that can give me odd number frame data, so the best I could do was approximate how minus backdash was.

Oh as a Raphael main i quit after 1.02
It totally made raph 100% unviable even on Hit along the other nerfs. (infact became bottom tier with danpierre)
QS4G didn t work so well in scIV mostly because there was no QS in scIV.

I seriously doubt it was intended, expecially since it was removed with 1.03 (and i guess appeared with 1.01 or 1.02).
 
Sidestep may be less safe than I thought. Apparently 2G might have a step-G bug that allows you to block low moves or duck high moves when you shouldn’t be able to.
 
@Slade could you elaborate more on that?
It seems that Grøh AVN K → AVN K can’t be blocked if you try to step with 8 and block with 2G, but if you step and block low with 2G it’s possible. This also seems to be the case when it comes to ducking highs in these frame-tight scenarios. I’m not sure if it’s because 2G is special in some way or if 8~2G is the issue.

I can’t think of a situation that tests i13 besides that Grøh one, though.
 
It seems that Grøh AVN K → AVN K can’t be blocked if you try to step with 8 and block with 2G, but if you step and block low with 2G it’s possible. This also seems to be the case when it comes to ducking highs in these frame-tight scenarios. I’m not sure if it’s because 2G is special in some way or if 8~2G is the issue.

I can’t think of a situation that tests i13 besides that Grøh one, though.
Ok, I figured that's what you might have meant, so I went ahead and managed to find two scenarios to test. Groh 4B is -7 on block, and Sophitia 1K is i20. I was able to step and then duck guard with both 2G and 8G, although 2G was more consistent. I think this is due to 2G-crouch guard being easier to execute, and also stepping in the same direction that Groh's 4B steps in. With 8G, you're stepping in the opposite direction. Although I did manage it a couple of times.
The other scenario I tested was Mitsu's SC 3BB. According to the Scuffle Frame data doc, his SC 3BB is +1 on block, and Mitsu BB clashes with Sophi AA after SC 3BB on block, so that seems to check out. With both 2G and 8G I was able to step and block.
I did have to test both situations against the wall/edge due to the pushback both moves have, so that might have affected the results.
Oh I also tested with Asta 4KA (-1 on block) against Azwel 2A (i14). That one worked for both sides as well.
 
Last edited:
Ok, I figured that's what you might have meant, so I went ahead and managed to find two scenarios to test. Groh 4B is -7 on block, and Sophitia 1K is i20. I was able to step and then duck guard with both 2G and 8G, although 2G was more consistent. I think this is due to 2G-crouch guard being easier to execute, and also stepping in the same direction that Groh's 4B steps in. With 8G, you're stepping in the opposite direction. Although I did manage it a couple of times.
The other scenario I tested was Mitsu's SC 3BB. According to the Scuffle Frame data doc, his SC 3BB is +1 on block, and Mitsu BB clashes with Sophi AA after SC 3BB on block, so that seems to check out. With both 2G and 8G I was able to step and block.
I did have to test both situations against the wall/edge due to the pushback both moves have, so that might have affected the results.
Oh I also tested with Asta 4KA (-1 on block) against Azwel 2A (i14). That one worked for both sides as well.

There is a bug on step 2G similar to old scV 22G bug (and yes it was unintended to have an unfair advantage on all vertical Attacks).
 
Step 2G lets you interrupt step to block lows and evade high..Thougt you already knew.
Yesterday i had the source today i lost it… maybe google will help i will try later.
Look at the above post. I've done 3 different test scenarios with that theory already. My own test have debunked that idea. Sidestep is the same minus frames whether you do it as 2G or 8G. I think the reason some people think that is that many people play on either pad or stick, which can make it difficult to consistently execute a crouch guard after an 8 sidestep. Going into crouch from a 2 sidestep is simply a matter of holding 2G. I play on a Hitbox, so it is considerably easier for me to transition from 8 to 2G to crouch guard, so I can more consistently execute that.
 
Last edited:
can I just say that I appreciate you guys doing all the lab work for everyone else. It's a selfless act. After all, you could easily just keep the secrets to yourselves, but you don't, for the good of all SC players. if no one else thanks you guys, at least I will.

I don't have the time/desire to 'lab' like that. I just wanna 'play' the game without doing boring stuff in training mode. You guys should earn a paycheck from namco. It was because of you guys that voldo and tira are being patched. I love you guys :)
 
Look at the above post. I've done 3 different test scenarios with that theory already. My own test have debunked that idea. Sidestep is the same minus frames whether you do it as 2G or 8G. I think the reason some people think that is that many people play on either pad or stick, which can make it difficult to consistently execute a crouch guard after an 8 sidestep. Going into crouch from a 2 sidestep is simply a matter of holding 2G. I play on a Hitbox, so it is considerably easier for me to transition from 8 to 2G to crouch guard, so I can more consistently execute that.
Will post mine as soon as i have time since i definitely had some raph 66(B+K),B crouch stepped. (should not be possible).
 
can I just say that I appreciate you guys doing all the lab work for everyone else. It's a selfless act. After all, you could easily just keep the secrets to yourselves, but you don't, for the good of all SC players. if no one else thanks you guys, at least I will.

I don't have the time/desire to 'lab' like that. I just wanna 'play' the game without doing boring stuff in training mode. You guys should earn a paycheck from namco. It was because of you guys that voldo and tira are being patched. I love you guys :)
We wouldn't have to if Namco would stop being stupid about their ideas of "allowing" people to discover the game themselves. There's a bunch of modern fighting games that have their frame data available in one form or another. Namco is one the few companies that refuses to change their minds on that. All because they think that it somehow helps to extend the lifespan of the game.
@darkfender assuming you mean on block, I tested it with Sophitia. She can't step it to either side as long as you do not delay the B from stance at all. You have to do it as fast as possible. However, I don't really feel like testing that right now with the entire cast, but what you're referring to may simply be an issue of some characters having better sidesteps than others, along with 66(B+K),B having weaker tracking to one side compared to the other. That's not an uncommon case if that is the issue you're speaking of.
 
Last edited:
Back