Soul Calibur 6: Technical Gameplay discussion

So there are Impact Counters. GI, RE and SC can be IC'd.
You can Impact Counter an opponent with all attack types.
While it's often a tag along condition, at ring out: B:L IC'd Blueboy twice, both resulting in hard knock downs.
& Blueboy is again hit with an IC at Ring out, thanks to Signia punishing SC with a CE.
& that last example is what really bakes my noodle on the properties of this counter.

I recall a Yoshi IC CE punishing a Taki's SC twice, will link once found.

Examples below.
Sophie IC CE Mitsu's Re @ 2:46. Thanks Frayhua.
Ivy IC @ 2:40 with a post parry RE on Sophi's clutch GI.
Sophie IC @ 1:47 post parry RE on Kilik's very early RE response.
Groh IC @ 1:49 swinging some SC lightning at Sophi's late(?) GI.
The first RO example above @ 2:54:50
The last example is at 3:20:10

Yoshi IC @ 10:38 with an unblockable
X low kick ICs Sieg's GI @ 12:34
 
Last edited:
Revenge!
@ 13:28. What is this?
Edit: The Silent Joel Youtube channel says,
"I don't know the exact parameters in this one, but it's an attack that will absorb certain attacks that are less than a set amount of HP"
 
Last edited:
guys question can you cancel the RE animation ! i saw Taki and Talim do it so i think its a universal thing !!
i tried to figure out how to cancel it but came out with nothing

another question
can you cancel the backdash or sidestep with g during the animation like the old games before SCV change it later
 
i feel like you can't guard cancel backdash. I get caught a lot trying. Im unsure about guard cancelling side step, sometimes i feel i can but other times nope.
 
Just a GIF that proves custom created characters (CaS) have a different hurtbox/hitbox. Female Revenant with the smallest body parts was used for the test, Ivy on both sides. Also tool-assisted for better syncing.
PaleCloudyHamadryas-size_restricted.gif

higher quality > https://gfycat.com/PaleCloudyHamadryas

Aris was complaining about it too youtu.be/wADBEyVpQqE

Somebody expanded on my test using different sizes HERE.
I don't have proper video editing software, and so can't place them next to each other for frame-by-frame comparison, so you'll have to have a little faith in me.
TallSkele: https://gfycat.com/RemarkableOccasionalGreatdane - Success!
ShortSkele: https://gfycat.com/AlarmedWelloffBlackfootedferret - Failure!
WideSkele (also short!): https://gfycat.com/GrippingUnitedBernesemountaindog - Failure!
MicroIvy: https://gfycat.com/OblongEvenLarva - Failure!
MicroIvy vs. MicroIvy: https://gfycat.com/AppropriateSpotlessCanine - Failure!
MicroIvy vs. Ivy: https://gfycat.com/FrighteningRareCoelacanth - Success! (Fun note: 37 damage compared to Ivy's 33 damage!)
As was the case in V and possibly throughout the entirity of Soulcalibur history, **knockback is affected by your character's height**. The example I always used was Viola's basic A+G 8A+B combo failing on characters that were just slightly taller than Astaroth.
Here is a comparison between Talim and Astaroth, scaled to remove differences in the position of the camera, having been hit by Azwel's K B (which requires no directional input to ensure no differences in my position: imgur.com/a/5IEpxRi
As you can see, Astaroth's center is slightly further away than Talim's.

And HERE.
So i just did a bunch of testing. The results are interesting.

The combo in this gif (https://gfycat.com/PaleCloudyHamadryas) does the following.

Does NOT work on Talim (who's size 1).
Does NOT work on Nightmares skeletons 1, 2, 3, 4, and 5.
Does NOT work on Nightmare (who's size 5).
Does NOT work on Ivy skeletons 1, 2.
DOES work on Ivy skeleton 3 (and presumably 4 and 5).
DOES work on Ivy (she's size 5 for reference).
Does NOT work on Seong Mina (who's female size 3).
Does NOT work on Seong Mina skeleton 3 (male).

Do you see the weird thing there? It's that it doesn't work on any of the Nightmares, CaS or not, but DOES sometimes work on Ivy's of the same height as those nightmare skeletons but are still shorter than normal Ivy. But it also works on Ivy size 3 when Seong Mina size 3 blocks it every time.

Try it on Nightmare 3, he blocks it every time.
Try it on Ivy 3, he eats it every time. (Ivy herself is size 5, so she also will eat it every time (without Ukemi) btw.)
But try it on Ivy 1 or 2 and they block it every time.
But try it on Seong Mina (or a size 3 skeleton with her weapon) and they block it every time.

So character size + fighting style DOES matter, but at least in terms of being combo'd (by this one specific string) it doesn't provide an advantage that the normal character wouldn't have. I'm gonna go try it on a Talim size 5 skeleton and see if that one eats the hit or not.

Jk I brain farted. If Ivy 5 eats it but Ivy's 1 and 2 don't then clearly they have a defensive advantage over Ivy, but they pay for it by having less range and (presumably) different combo timings/etc on others.

Talim skeleton size 5 doesn't eat the combo, just like Talim herself won't.

And just to confirm, gender doesn't affect heights. I made a male and female Astaroth CaS and set them to the max height for both and they look each other in the eye in training mode.
 
Last edited:
Mabye this is old news but I'm gonna share this here anyways since I did the testing. Post-guard-break damage scales different depending on what move you broke guard with, and Soul Charge makes it weird, with some moves actually INCREASING the amount of damage over what it would be raw:


Okay, I'm not sure where to put this, but since it's relevant to post-GB combos, I guess I'll put this here.
I was gonna go to bed but while I was messing around in training mode I noticed that scaling of my CE changed depending on what move I used to guard break, so I tested her whole movelist and it's downright bizarre. CE Raw does 75 on normal hit. Some moves do not scale this at all after a guard break and you get the whole shebang (2B+K, SXS A+B, bK, others). SXS B in particularly scales it worse than anything else, whether it's charged or uncharged, giving you a measly 45 damage for the same move.

So I figured each move was manually placed into one of a few set categories. Most of them fall into two bunches: Most moves do 59, 10 moves do 75, and there are some outliers that do 45, 52, and 67. So I wanted to test out what the Soul Charge exclusive moves did, and realized I'd have to compare them to other moves in SC in order to get a relevant comparison.

And then things got -weird-.

Not only are the numbers much less consistent, but several moves actually cause CE to do more damage post-guard-break than it does -raw-. If you break their guard with the last hit of SC 66A+B B, it does 90 damage. This is 8 more damage than CE does raw while in SC and 5 more damage than it does on counterhit. It actually scales up.

Here's my data, this should also include all of the moves X can guard break with. It's more or less ordered from best-scaling to worst-scaling:
https://docs.google.com/spreadsheets/d/10m3GFJ4ZyPMj7rYxTzVjrbCI4JOYnfA6oyYtBHHEEto/edit?usp=sharing

This was all done on a Nightmare training dummy set to "Guard All" and 2P Guard Crush settings set to "Just Before Guard Crush" on Kunpaetku Temple. The numbers have remained consistent in all variations I've tested them, though I have not been anywhere near exhaustive.

So uh, the tl;dr is that it really, really matters which move you GB with, and SC probably gives some sort of addition-based buff to damage rather than the numbers being manually adjusted for balance. I think. I'm not a math guy.
 
another question
can you cancel the backdash or sidestep with g during the animation like the old games before SCV change it later
Thanks god NO!!!
That broke scV completely and was possibly a bug.
And you couldn t in scIV nor previous afaik
 
GI activates on frame 2.

RE activates on frame 6.

Testing scenarios with Groh:

F4: AVN B+K [+8 on block] -> 2A [i12]
F3: A+B6 [+9 on hit] -> AVN B [i12]
F2: CH 3K [+10 on CH] -> 2A [i12]
F1: 44A [+11] -> AVN B [i12]
 
How are Ring Out moves overall here compared to other games? I remember back in the day being able to RO with almost everything yet here it seems that only very specific moves in the minority can RO unless you're Siegfried or Nightmare.
 
The patch today (on PC) addresses customization and online, but apparently it's unknown how yet. Maybe it'll address the CaS hurtboxes just for online?

https://gearnuke.com/soulcalibur-6-update-version-1-01-pc-patch-notes/
nah man this is the actual changelog:
https://twitter.com/Flying_Wonkey/status/1057300055944658944
https://www.reddit.com/r/Kappa/comments/9spos4/funny_tuesdays_remember_that_6gb_patch/e8r0zxn/

TL;DR
6 GB Steam-only patch to "download" 5 player bans lmaoaoaoao
 
Mabye this is old news but I'm gonna share this here anyways since I did the testing. Post-guard-break damage scales different depending on what move you broke guard with, and Soul Charge makes it weird, with some moves actually INCREASING the amount of damage over what it would be raw:
I tested with Taki and there seem to be 2 categories:

1: Attacks with scaling, for Taki i found 4:
PO B+K, PO B, WRO K and the worst is 3B.
After which her CE does 64 instead of 90 damage.

2: Attacks without scaling.
Here CE does the normal 90 damage.


BUT in both cases, if the crushing attacks do chip damage, because of Mekki-Maru or Soul Charge, this is added to the total damage.
With Stalker B crush (10 damage on block, no scaling) into CE doing 100 damage for example.

I assume this is the same with your X in Soul Charge.
 
Last edited:
How negative is step? I've confirmed it's at least -10 but wasn’t able to think of a scenario to test -11. It’s definitely not -14 or worse so my guess is that it’s an even -10.

EDIT: Seems to be -12 for sidestep and -15 for backstep. See here for the tests.
 
Last edited:
Testing odd-numbered frame data is much harder in this game, since there are very few impact frames or advantage/disadvantage frames that aren’t even. Here’s a trick that can help. If you’re trying to prove that a move is in and you believe n is odd, construct a scenario where an in move would lose to a move that’s i(n - 1) and beat a move that’s i(n + 1). If it turns out that this is the case and that neither move trades with the move being tested, you can infer that the move is in.

For example, to prove that Grøh’s 4A6 is +3 into his Avenger stance on block, make a recording of Grøh doing 4A6 into AVN A+B, which is i20. Then see if AVN A+B loses to Grøh’s 6K (i16) but beats 3B (i18). This turns out to be the case, and because neither move trades, 4A6 must be +3. If it were +2 it would trade with 3B, and if it were +4 it would trade with 6K.

(To verify that this technique works, you can also prove it directly by having Grøh do 4A6 into SCh AVN 6A, which is i33, and showing that it trades with 4A, which is i30.)
 
What settings do you put the training dummy on to be sure that they will block non-combos? I've been setting Action 1 to "Stand / Guard after first hit" and Action 2 to "Stand / Guard All" with Ukemi and Air Control usually on random unless I'm trying a specific direction. Even with this it still seems like the computer gets hit by things that end up not really comboing that should be blocked. Should I change up Action 1 and 2?
 
What settings do you put the training dummy on to be sure that they will block non-combos? I've been setting Action 1 to "Stand / Guard after first hit" and Action 2 to "Stand / Guard All" with Ukemi and Air Control usually on random unless I'm trying a specific direction. Even with this it still seems like the computer gets hit by things that end up not really comboing that should be blocked. Should I change up Action 1 and 2?
I usually use those settings when experimenting to find combos, then verify that it's a true combo by recording the dummy doing it and making sure I can't escape it. For testing CH combos you just have to do a slow move that'll get CH when you start the playback.
 
Back