Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


When's Melty on Steam?
ahaha that's no--wait, what?

Messages - Viskar

Pages: [1] 2
1
Melty Blood Auditorium / Re: CCCaster v3.0 Rollback Edition
« on: March 10, 2015, 01:59:33 PM »
Is it possible to get a deadzone adjustment as part of the in-game button config? Perhaps adjusted via left-right when highlighted.

Right now when I play on pad, I usually need to tweak with the deadzone for the L2/R2 trigger buttons to make it way more sensitive. When playing without cccaster, it's just a light touch on the trigger but the default in cccaster requires me to push down pretty far to register an an input. Making this adjustable in-game would be nice when hot-swapping controllers at a setup.

2
Regional Community / Re: SoCal Melty: Bringing out the Dead
« on: July 23, 2014, 07:50:27 PM »
Any plans this weekend? I don't mind hosting if people wanna do some UNIEL/MB Sat/Sunday.

3
Aoko Aozaki / Re: [MBAACC 1.07] F-Aoko Discussion Thread
« on: July 21, 2014, 11:04:01 PM »
This thread needs some love too. Been in the lab a bit and working on a few things.

Here is one of them: corner oxi/mixup string that leads into a safe throw mixup. We might not have Kicking Ass tech punishes, but this isn't so bad either. I present to you the F-moon version of corner gimmicks:

No F'ing Escape
http://youtu.be/-mqK8ufZbkg

1) EX orb corner combo into a High/Low mixup.
2) If they get hit, it leads into a special situation where the proxy orb is timed to detonate right after their wakeup, making a dash in grab unmashable. Similar to Akiha ribbon setup.
- Mashing and DPs get CH by orb.
- EX wakeups can be blocked on reaction, and usually CH by orb too (grab is not input immediately)
- Orb catches jump outs and pushes them back near Aoko. Potential to guard break here.
- Jump+attack gets CH (e.g. Nero j.b on wakeup)
- Blocking, Shielding, Dodging, or Bunkering gets thrown
- Backdash gets hit by orb

If you think they aren't going to sit there and let themselves get grabbed, you can just do a meaty or meaty with backdash punish OS to catch a jump outs and backdashes, and the orb keeps you safe from most danger.

Note: After grab, you can combo EX DP against W.Len, Len, and Miyako in the corner.

edit: New video uploaded (youtube broke my old one)

4
edit: changed my idea a bit.

Yeah after thinking about it, I think I'll add in a hotkey so if you press [Tab] it will swap what sides P1 and P2 are in.  So it would ask (P2) to press start first, or if the player(s) already have keys mapped, everything will swap sides. Might also be useful when steamers/TOs already setup the P1/P2 names and you want to switch chairs.

I'll have to clean up a few things to make it work, but seems like it would make things simple.

5
Welp, I am an idiot and forgot to uncomment the code that makes it work. It was turned off briefly when I was isolating an unrelated bug and I forgot to re-enable it. Updating OP with 0.11.2 in a sec.

6
Well I stayed up way too late the night before the EVO trip to get the next version out, but the OP is updated with the new build.

I will be at EVO and at the MBAACC side tourney, so I'm going to bring this on a USB stick since most people probably won't have a chance to download it.

edit: There was a bug in the released version, new one is uploaded (0.11.1)

7
Aoko Aozaki / Re: [MBAACC 1.07] H-Aoko Discussion Thread
« on: July 09, 2014, 12:13:21 AM »
I think I'll get on that after Evo and before Super Arcade / Rebel Up re-opens.

Looking at my F-moon notes remindmed me that H-Aoko doesn't need to safejump Roa. Every moon can beat his stupid fast DP's with any non-held orb for free. So you can just do tk. 214[a] after 421a.

Examples:
https://www.youtube.com/watch?v=aA8jGlLRxV0

Some follow up ideas:
* 66 2b
* 66 BE4C 214a (release orb) 66 2a 2b 5b combo
* IAD j.a j.a (land) 2a (true blockstring, whifs if he crouch blocks oki orb though). I think you can OS ground throw on whiff, throw them into held orb. This can link into 2c 5c 4c 623a (Kicking Ass!) or hard knockdown.
* IAD j.a j.a dnj/iad j.b.
* IAD j.b /j.c into fuzzy high dj.b or low 2b (very small gap in blockstun, Shield/DP is tricky to time)

8
Aoko Aozaki / Re: [MBAACC 1.07] H-Aoko Discussion Thread
« on: July 08, 2014, 01:13:37 AM »
Got inspired to record some of my H-Aoko character-specific notes. Some of this overlaps with the wiki, but I also found that positioning, timing, and number of loops all affect whether or not you can get a well positioned orb from 4[c] xx 421a enders. If I finish recording things and get more notes, I think I'll re-edit it and correct things.

The recordings use j.c on the safejump, but j.b is better if you plan to 2b after. Otherwise you risk getting pushed out too far from j.c pushback on block.

Characters with 4 frame DPs can theoretically be safejumped, but the timing is really tight. Looking for alternate safejup setups that are more reliable (Ries has one I found).

Common setup followups:
* High-Low - j.b (land) 2b 5b 5c 214a ReleasOrb 66 combo...
* Fuzzy High - j.b 8 j.b 66+ReleaseOrb j.c (If they EX reversal, you should be able to land and block. With perfect input you may be able to block non-EX flash reversals. Short characters can't get fuzzied, but you can air dash instead of immediately landing for a mixup (release orb to meaty while air dashing).
* Empty jump - (whiff j.b) (land) 2a/2b or Throw when landing, or release orb if you think they will mash.

-- First 1/3 of cast --

0:00 Aoko (nothing special)
* 2b2c.
DP: 6 frames, pretty lenient safe jump. Aoko is a slow riser need to delay 421a cancel more than usual.
0 loops xx 421a (Cancel on BE4C frame ~30 near the end)
-or-
1 loops xx 421a (Cancel on BE4C frame ~35 near the end)

0:13 Tohno
* 2a2c
* 5 Frame DP, 6 Frame EX DP. H-moon DP's even slower
0 loops xx Orb (Cancel instantly after BE4C)
-or-
1 loop + t.k. EX Orb (Cancel on BE4C ~frames 28-29 foot touches floor)

1:15 Hime
* 2a2c
0 loop xx Orb (Cancel instantly after BE4C)
1 loop xx Orb (Cancel on BE4C ~frames 28-29)

1:47 Nanaya
* 2a2c
* DP: 6 Frame / 4 Frame EX. Difficult to safe jump EX
0 loop xx Orb (Cancel instantly after BE4C). Loses to EX.
1 loop tk EX Orb (Cancel on BE4C~frame 30)
0-1 loop 2a2c 421a (Jump in at a higher angle to go over DPs. Kind of a gimmick the j.c won't be meaty unless you release orb. This is a whole other setup though you can j.cb, j.8 j.b double overhead into fuzzy, or j.cb 2b for double overhead into low.)

2:50 Kouma
* 2b2c
0 loop xx 421a (Instant cancel)
1 loop xx 421a (Won't hit during fuzzy)
1 loop tk EX Orb (Cancel BE4C around ~frame 30-35)

3:13 Miyako
* 2b2c works but not ideal in corner
* Back up 2a2c so the orb can hit outside of block/hitstun. Skip  BE5C if you back up or it will launch.
* Too short to fuzzy, can instead release orb for meaty, then j.66 j.B for late overhead.
1 loop xx 421a (Cancel BE4C around ~frame 26-28)

3:36 Ciel
* 2a2c
0 loop xx 421a (Cancel instantly after BE4C)
1 loop tk EX Orb (Cancel BE4C around ~frame 28)

4:26 Sion
* 2b2c
* DP: 4 frame / 4 frame EX. Safe jumps unreliable.
Cannot do 0 loop version safe
1 loop xx 421a (Cancel BE4C around frame ~26)

4:49 Ries
* 2a2c
* DP: 4 frame / 6 frame EX. Hard to safe jump from loop.
* BE5C launches unless you step forward after BE4C
* 0 loop xx 421a (Cancel BE4C around frame ~27 early but not quite instant). Pretty difficult.
* (not pictured, but useful) 1 loop + 2a2c (recover) sj9 j.C is safejump. Remains safe even with high gravity situations. No orb in this setup though.
* 2 loop xx 421a (Cancel BE4C around ~28). Pretty difficult.

5:31 Wara
* 2a2c
* Can't BE5C
0 loop xx 421a (Cancel BE4C immediately)
1 loop xx 421a (Cancel BE4C around ~frame 30)

6:08 Roa
* 2a2c
* DP: 4 frame, hard to safe jump.
* 0 loop xx 421a is unsafe
* 1 loop xx 421a can work but difficulty to time.

9
Damn that's pretty weird/unfortunate. It's written in Auto-It which is just a generic windows scripting utility that can compile to .exe for easy distribution. I guess I'm not surprised that it might be picked up as some blanket thing. I scanned with Security Essentials / Malwarebytes and wasn't getting any false positives.

If anyone wants to see the source code I don't mind providing it as well. I just haven't setup a public repo for it but figure I could at some point.

edit: Also, I think I'm going to try and squeeze out one more release before EVO to address a couple edge case things. If I'll do, I'll also bring a USB stick in case people haven't updated (though the current version is pretty much OK too).

10
Regional Community / Re: SoCal Melty: Bringing out the Dead
« on: July 04, 2014, 06:25:25 PM »
Doing another small get-together Saturday afternoon/night. Last chance for some pre-evo matches. Anyone is welcome again.

Here's the directions if you take 5 South to Jeffery exit.
http://puu.sh/9Xjgt/98f64c52bd.jpg

For food and stuff, there's places to eat nearby open 'til 10pm.

11
Regional Community / Re: SoCal Melty: Bringing out the Dead
« on: June 20, 2014, 09:28:30 PM »
Sorry for the late post, just got off work.

Anyway, talking after last Rebel Up we had some interest in some casuals this weekend. I was willing to have people over so I think that's the plan on Saturday. My place will be open all day/night so any time is OK (just don't come before 12:00 noon).

Message me or post if you want to go and need my directions (it's in Irvine).

I believe right now there's about 4 attendees interested including KFL/iPhail.

Setup Status:
Viskar PC + Monitor
Viskar Laptop + Monitor (KFL)
TV(LCD) w/ Melee

I think we have room to fit 2-3 more setups. If anyone wants to bring stuff, it wouldn't hurt to have a small folding table, power strips, or extension cables too.

12
Which keyboard configs are people using? Mauve's mbaacctool? Or MBAACC Keyboard Config?

To implement keyboard config in MeltyLauncher I either need to run the 3rd-party tool behind my UI, or re-implement what they are doing to achieve keyboard config (which I believe is overwriting parts of the MBAA.exe file). I think I may be able to accomplish this if I can work out some of the details.

Also (from what I've seen from other tools), even if you have two keyboards, the button mappings cannot overlap. Has anyone gotten around this limitation? I imagine it would suck when two keyboard players are against each other and they can't use their preferred controls.

13
The next best part was 1 hour of Kouma vs Kouma casuals after the tourney.

14
For sure bringing my laptop for Melty. I think I will be able to bring an Asus Monitor as well.

15
Aoko Aozaki / Re: [MBAACC 1.07] F-Aoko Discussion Thread
« on: March 17, 2014, 11:35:12 AM »
That's not the case here. I am dashing while akiha is still in wake up frames so she cannot crouch yet. I should have had the input display on but I am holding crouch the whole time. Kusa also made use of this in his HAoko kicking ass video when he shows the midscreen fuzzy hi/low combo.

16
Aoko Aozaki / Re: [MBAACC 1.07] F-Aoko Discussion Thread
« on: March 15, 2014, 11:11:45 PM »
Was playing around with some midscreen three-way setups today:
http://www.youtube.com/watch?v=CoPkVtAfoFE

After slider it's a back jump (j.7), immediately cancelled in neutral jump (j.8) and then (6) to drift forward. The cross-up low looks very ambiguous. If you are going to be on the same side, you can air dash in and do a high, even if they are holding crouch.

edit: Switch video to a HQ version. Fixed bad execution on some dummies, so they all look the same now. Fixed the Air Dash -> j.B timing so it is meaty instead of chicken blockable.

17
Made a slightly modified version that changes how it waits for MBAACC to close, give it a shot if you can:
http://puu.sh/7r5gn.zip

Also, if you have a chance to try any previous versions please do. I'm curious if they all have the same behavior or not. Also try running them under administrator to see if that changes anything.

Thanks for all the help/feedback so far. It's hard to test all these cases myself but I want to make sure this works for everyone.

edit: Found an even better solution (I think) to the problem, which I put as v0.10 in the OP
http://puu.sh/7rdF0.zip

18
I think right now it waits for up to ~5 seconds after presisng space for the MBAACC config window to load. If the standard config window doesn't load in that 5 seconds, it won't auto-close it and start the game. I wasn't sure how much this varied between people's setups, so I can increase that value since it's pretty simple change to make.

So to be clear high-ish CPU usage you are seeing is while MBAACC is running then? And only for Win8?

19
I has quite a bit of activity while it's running button configs, but once MBAACC starts it up should drop down to 0 CPU, since it stops what it's doing and just waits for MBAACC process to close again. I am mostly concerned if it's having any sort of CPU impact while the game is running, since it should be 100% idle. If there is CPU usage while the game is running, there must be a problem with the way it's waiting for the process  to end.

That being said, seeing 50% CPU usage on Win8 is really weird, since I only get about 1% CPU usage on Win7 while running it. I don't want to cook anyone's hardware so I will see about adjusting the delays between scans to be a bit less aggressive. I wish I had a Win8 machine handy to see what might be going on though.

20
Just released 0.9 which has some fixes for what you mentioned. The .zip also contains a debug version of the tool that logs which buttons/directions are being detected.

21
I think I have an idea of what the problem is. The tool is thinking that a button/direction is in the "pressed" state even though the stick is neutral, which prevents any inputs from happening. I was able to reproduce this on a MadCatz stick when the stick is in "RS" mode. MeltyLauncher is getting a constant reading of the Z+ state.

I'm going to bring the launcher to the Rebel Up event tonight where I will have a chance to try out a wide variety of sticks and debug any of these issues. Should be a good chance to clear up any final problems before FR.



22
Made a quick release that works with AppLocale if you edit your MeltyLauncher.exe shortcut as noted in the OP patch notes. I did my best at auto-closing the nag popup that AppLocale shows on each run (at least on Win7).

When I get around to making the settings window (later this week), MeltyLauncher.exe shortcuts won't be needed.

23
Thanks for the tip, I got AppLocale installed now so I can look into that. Now that I'm done tweaking the UI for a bit I'm going to switch over to working on the "Settings" section, which will provide ways to configure things like AppLocale.

My plans for settings include:
* Allow me to save data between runs of the launcher
* Remember/Edit path to MBAA installation location, so you don't need to mess with shortcuts. It should auto-save after first run.
* Set location of AppLocale executable if you want to launch MB under AppLocale
* Maybe provide other config from the vanilla Melty config screen, like resolution / fullscreen



24
Great ideas, I think I'll be able to work in some changes later today. I totally forgot about regional/locale settings, so thanks for the reminder. I'll see why that's acting up (probably the way I'm doing filesystem operations).

edit: New version is up that has tweaks for A+B, FN1, FN2.

edit2: I wasn't able to test AppLocale since it is unsupported on Windows 7. The best I could do was verify that setting windows locale to Japanese worked without issues. I'm still interested in fixing it, but it will have to be at a later time and I would need to get some details on the problem.

25
Current Version (0.11.2): http://puu.sh/aakf6.zip

Hey guys, I'm here to share a brand new tool I just created which greatly reduces the frustration of doing button configs during tournament/casual play when people are constantly cycling through the setup.

I created it because we've found that maybe 25% of our time in these events is spent painstakingly assigning buttons using our awkwardly outstretched arms on a tiny laptop trackpad. To make things worse, the laptop is more often than not tucked away behind some monitor, on the floor, or just not facing you at all. It might have even falling into the trash can you were playing on in the back alley.



Overview:
* Run the program and it will ask for your MBAA.exe file the first time you run it.
* 1 or 2 players can map their buttons simultaneously in the button config GUI. P1 presses start first, and then P2 can press start while P1 is in the process of mapping buttons.
* Press SPACE to launch the game when all players are done setting buttons.
* Melty Blood should launch directly into the game
* After closing Melty Blood, the launcher will pop back up and allow for players to re-configure their sticks
* If any sticks are added or removed, the launcher will immediately update as necessary.
* ESC will quit from the launcher

This program was written with AutoIt3 and compiled to an .exe file. I believe one version got a false positive virus warning by Avast, so you may have to whitelist the app depending on what antivirus you use. I have had no problems with Microsoft Security Essentials / Malwarebytes. In any case, I'm open to people looking at the source code if you want to message me (right now I don't have it in a public repo).

-- Releases --
v0.11.2 - http://puu.sh/aakf6.zip
* UI Updated. Added several new features which can be enabled via simple buttons.
* All settings are now stored  in a config file (AppData\Roaming\MeltyLauncher\config.ini)
* Remembers path to MBAA.exe after you select it. Using the shortcut parameter no longer works.
* Remembers path to AppLocale (if using) after you select it. Using the shortcut parameter no longer works.
* Added ability to enable the "Keyboard" flag for the game. So now you can launch the game via MeltyLauncher and play on keyboard.
* Added support for turning on the secret stages (thanks to MadScientist/Erkz for the info on that).
* (Bugfix: 0.11.1) App wouldn't startup correctly
* (Bugfix: 0.11.1) Made the "[Space] Start Game" button clickable
* (Bugfix: 0.11.2) "Keyboard: ON" wasn't doing anything. Should work now.

v0.10 - http://puu.sh/7rdF0.zip
* No longer uses polling to determine when MBAACC finishes running, which should eliminate any CPU problems on Win8. Due to the new way the MBAACC process is monitored, MeltyLauncher will immediately show when MBAACC is closed, instead of having a small delay.
* The above improvement is unable to be used in conjunction with AppLocale, so running MBACC under AppLocale will still use the old polling mechanism.
* A few untested tweaks to the polling mechanism hopefully reduce CPU usage bugs on Windows 8.


v0.9 - http://puu.sh/7oeZt.zip
* Added some fixes for certain joysticks by detecting which buttons/directions are constantly held.
* Now includes a MeltyLauncher_debug.exe file that can be used to see log output. If a stick isn't working, try to capture some information with the debug tool and I can try to fix it.

v0.8 - http://puu.sh/7nbjs.zip
* Internal clean up that fixes some confusion caused by adding/removing controllers when MB is running. The app should now always be up to date with any controllers that are connected.
* Adding/removing controllers does not reset player configurations, except for the player that removed the controller.

v0.7 - http://puu.sh/7kESh.zip
* Basic support for AppLocale. Requires making a shortcut to MeltyLauncher.exe like this:
"C:\Games\MeltyLauncher.exe" "C:\Games\MBAACC\MBAA.exe" "C:\Windows\AppPatch\AppLoc.exe"

v0.6 - http://puu.sh/7imaR.zip
* Increase overall size of the text and UI
* Added small tips along the bottom

v0.5b - http://puu.sh/7i4DO.zip
* The GUI will scale correctly for people that use the windows settings to enlarge text ("Make text and other items larger or smaller")
* (0.5 ->0.5b) Fixed FN1/FN2. The bindings were incorrect for Player 1

v0.4 - http://puu.sh/7hi48.zip
* A bad attempt at a logo

v0.3b - http://puu.sh/7hh1r.zip
* Pressing 3 buttons at once will dissociate the controller
* (0.3 -> 0.3b) Textual bugfixes

v0.2 - http://puu.sh/7gY7J.zip
* Added support for FN1 / FN2 (FN1/FN2 were bugged, fixed in 0.5b)
* A+B, FN1, FN2 are all considered optional buttons
* To unbind a button, press a button that is already bound in a higher slot.
* Removed text borders since it was done in a hacky/laggy way.

v0.1 - http://puu.sh/7g1XY.zip
* Original version

-- Future features --
* Auto-saved stick configurations and profiles tied to which button is pressed to select a stick.

Pages: [1] 2