Super Castlevania IV

Bablo

Edit history:

Bablo: 2012-04-01 10:31:17 am

Bablo: 2012-04-01 10:31:11 am

Bablo: 2012-04-01 10:31:04 am

Bablo: 2012-04-01 10:29:59 am

2012-04-01 09:36:56 am

I'm a flippin' ninja

Felt like updating you guyz of my progress. I just beat the SDA run by 7 seconds (35:07 was my time).
Obviously that's not exactly my goal, but at least I got the world record and a little confidence boost.

Here's a disgusting table of my splits:

Rowdain | +0.9
2-3 | +0.9
Orphic Vipers | -3.9
Koranot | -4.3
5-2 | -9.1
Dancing Specters | -10.7
Sir Grakul | -12.2
Frankenstein | -4.6
Zapf Bat | -9.1
Mummy | -14.1
Death | -13.0
Dracula orb | -7.0

As you'll see this run was quite inconsistent in a lot of places. IIRC my best split at Koranot has been -13.0. But this is not too bad after only a few weeks of practice and optimal route finding. I'm going to stream my attempts once I get the capture card I've ordered several weeks ago...

Edit: actually 35:06.81 was my time. wsplit likes to round things up. sda run is high 35:13.81 according to my split so exactly 7 seconds saved.

Felt like updating you guyz of my progress. I just beat the SDA run by 7 seconds (35:07 was my time).
Obviously that's not exactly my goal, but at least I got the world record and a little confidence boost.

Here's a disgusting table of my splits:

As you'll see this run was quite inconsistent in a lot of places. IIRC my best split at Koranot has been -13.0. But this is not too bad after only a few weeks of practice and optimal route finding. I'm going to stream my attempts once I get the capture card I've ordered several weeks ago...

Edit: actually 35:06.81 was my time. wsplit likes to round things up. sda run is high 35:13.81 according to my split so exactly 7 seconds saved.

lxx4xNx6xxl

2012-04-01 10:33:57 am

My name is 4N6/Forensics.

Congratulations, Bablo!!!

I can't wait till you start streaming some real console attempts. I've been watching you practice lately and your strategies are excellent. I can't wait to see how much time you'll save over all should be very interesting .

DK28

2012-04-03 11:40:45 am

General Kong - Bullets and Bananas

Go, go, go!

Riskbreaker Y

2012-04-04 08:58:49 pm

berserker status

Exxxxcellent!!

Bablo

2012-04-16 08:38:58 pm

I'm a flippin' ninja

I made a theory speedrun on snes9x (v1.51) which clocked at ~33:47. It'd be kind of an utopistic time for a console run, but I guess somewhere around 34:00 is what I'd want to get. The lag on snes9x is inaccurate to its favor too.

As of late I've been practicing the rest of the game starting from stage 7, due to them not only being at the end, but some of the hardest and most deciding stages. Obviously I'd want to get them consistent since good runs really don't get there that often... And every run that makes it past stage 8's disappearing platforms is precious.

I also had a problem setting up my living room so that I could stream without having to breath through wires.
But I'm happy to say this is now settled, and I should be ready to attempt a lower time and stream the action starting tomorrow!

Join the fun @ twitch.tv/babloization

puwexil

Edit history:

puwexil: 2012-04-16 09:41:59 pm

2012-04-16 09:41:32 pm

Professional Second Banana

Looking forward to seeing some CV4 stream action (and senseless slaughtering of my namesake)!

DK28

2012-04-17 01:45:21 pm

General Kong - Bullets and Bananas

Quote from puwexil:

Looking forward to seeing some CV4 stream action (and senseless slaughtering of my namesake)!

+1. Minus the puwexil joke of course...

Bablo

2012-04-29 04:57:46 pm

I'm a flippin' ninja

A brief update.

I got a 34:41 this weekend down from previous 35:06. I started finishing all my attempts even if they have some mistakes since 35:06 was kinda easily improvable with my strategies. There's still a loooong way to go but I'm slowly getting there.

I have timed a lot of stuff on emulator to be sure what I'm doing is optimal. Which led me to a bunch of different strategies that have added up to 7 seconds or so from today's testing alone. I still need to time a lot of damage boosts and rooms where the game lags to find out the optimal solution. (This is actually something I'd like to encourage everyone to do for whatever game they're running. We did a lot of this type of very precise timing with Seth Glass for Mega Man 2 and it was very rewarding.)

By the looks of it stage 3 is the hardest stage to do optimally. Ideally I'd get hit in 3-1 (mudman section) only once and do a bunch of crazy damage boosts. In addition this level involves two ring glitches.

I finally found a way to do damage boosts in stage 5 without relying on randomness. This stage is especially good for damage boosts because you don't catch an orb to refill life at the end, which actually costs a bit of time (iirc 4 frame rule per 4 health to refill). Some of them aren't really a big deal but look cool and are reliable.

Mummy can be beaten in one cycle if it spawns in the right corner...but it's frustratingly tight. Obviously this would save a lot of time but likely won't ever happen in an attempt. You also have to correctly guess if it's going to attack you with flames or bandages to correctly position yourself beforehand.

I'm able to reduce a good amount of lag and grab some hearts at the bat spawning, collapsing bridge in stage B. Apparently the game lags less when you jump instead of walking. A combination of walking, killing a few bats and destroying candles (which add lag too) reduced a second or two of lag, although I'm not sure how that translates from snes9x v1.51 to console. At least they are generally similar when it comes to lag.

There's still a lot of stuff I want to optimize. I could totally start doing serious sets of attempts already but these little things tend to add up to several seconds. Also the fight against randomness is a big part of running this game. There are some parts where I kind of rely on something specific to happen. What I generally need to do is have a strategy that works against several cases of RNG. Consistency is especially important past stage 8 where a lot of runs end because of the disappearing platforms. Fortunately majority of the hard tricks happen before that though.

I hope I won't later regret saying this, but my goal time for this game is definitely sub-34. It's getting more and more reachable. And it'd be cool.

So much for being brief. But this run is getting quite deep.

A brief update.

I got a [b]34:41[/b] this weekend down from previous 35:06. I started finishing all my attempts even if they have some mistakes since 35:06 was kinda easily improvable with my strategies. There's still a loooong way to go but I'm slowly getting there.

I have timed a lot of stuff on emulator to be sure what I'm doing is optimal. Which led me to a bunch of different strategies that have added up to 7 seconds or so from today's testing alone. I still need to time a lot of damage boosts and rooms where the game lags to find out the optimal solution. (This is actually something I'd like to encourage everyone to do for whatever game they're running. We did a lot of this type of very precise timing with Seth Glass for Mega Man 2 and it was very rewarding.)

By the looks of it stage 3 is the hardest stage to do optimally. Ideally I'd get hit in 3-1 (mudman section) only once and do a bunch of crazy damage boosts. In addition this level involves [i]two[/i] ring glitches.

I finally found a way to do damage boosts in stage 5 without relying on randomness. This stage is especially good for damage boosts because you don't catch an orb to refill life at the end, which actually costs a bit of time (iirc 4 frame rule per 4 health to refill). Some of them aren't really a big deal but look cool and are reliable.

Mummy can be beaten in one cycle if it spawns in the right corner...but it's frustratingly tight. Obviously this would save [i]a lot[/i] of time but likely won't ever happen in an attempt. You also have to correctly guess if it's going to attack you with flames or bandages to correctly position yourself beforehand.

I'm able to reduce a good amount of lag and grab some hearts at the bat spawning, collapsing bridge in stage B. Apparently the game lags less when you jump instead of walking. A combination of walking, killing a few bats and destroying candles (which add lag too) reduced a second or two of lag, although I'm not sure how that translates from snes9x v1.51 to console. At least they are generally similar when it comes to lag.

There's still a lot of stuff I want to optimize. I could totally start doing serious sets of attempts already but these little things tend to add up to several seconds. Also the fight against randomness is a big part of running this game. There are some parts where I kind of rely on something specific to happen. What I generally need to do is have a strategy that works against several cases of RNG. Consistency is especially important past stage 8 where a lot of runs end because of the disappearing platforms. Fortunately majority of the hard tricks happen before that though.

I hope I won't later regret saying this, but my goal time for this game is definitely sub-34. It's getting more and more reachable. And it'd be cool.

So much for being brief. But this run is getting quite deep.

ozzy88

Edit history:

ozzy88: 2012-04-30 08:26:46 am

2012-04-30 08:26:18 am

I guess this game has gotten some well needed babloization! I can't feel anything other then greatness and hype!

Also kinda jelly since I did try to get decent at the route that Josh took for the current SDA run with some modification to make it easier but couldn't succeed. (although only on emulator)

Bablo

Edit history:

Bablo: 2012-05-01 02:42:43 am

Bablo: 2012-04-30 11:43:15 am

2012-04-30 11:33:44 am

I'm a flippin' ninja

Alright, made a theory run on emulator. Very ideal RNG, no mistakes, clean execution. I didn't involve 1-cycle mummy though since it's just stupidly tight and most likely will never happen.

33:05

These strategies should be sort of final. Some of the stuff you see I probably won't do in a speedrun, just because they're not very forgiving if you mess up, and don't save much time. But everything you see there is possible in a real time speedrun.

Edit: I should note that none of the strategies expect a certain type of RNG to happen eventhough it's made on emulator. All of this is doable on console, and if undesired RNG happens I'm able to easily adapt, often without wasting time. Some of the damage boosts (and boss strategies) do require something specific RNG related to happen, all of those which I hit in the theory run. If the desired RNG doesn't happen it's not like I'm screwed though, just going to miss out on a timesaver.

Edit2: Also console lags more, which adds up to several seconds.

Alright, made a theory run on emulator. Very ideal RNG, no mistakes, clean execution. I didn't involve 1-cycle mummy though since it's just stupidly tight and most likely will never happen.

[url=http://www.twitch.tv/babloization/b/316690309]33:05[/url]

These strategies should be sort of final. Some of the stuff you see I probably won't do in a speedrun, just because they're not very forgiving if you mess up, and don't save much time. But everything you see there is possible in a real time speedrun.

Edit: I should note that none of the strategies expect a certain type of RNG to happen eventhough it's made on emulator. All of this is doable on console, and if undesired RNG happens I'm able to easily adapt, often without wasting time. Some of the damage boosts (and boss strategies) do require something specific RNG related to happen, all of those which I hit in the theory run. If the desired RNG doesn't happen it's not like I'm screwed though, just going to miss out on a timesaver.

Edit2: Also console lags more, which adds up to several seconds.

Bablo

Edit history:

Bablo: 2012-08-10 12:11:19 pm

2012-05-03 11:22:28 am

I'm a flippin' ninja

34:29! This time completed while streaming.

The splits are still kinda inconsistent but at least I'm getting closer. Quite a few mistakes, some bad luck, but overall an OK run. Going to continue getting more attempts in though.

PJ

2012-05-03 12:12:00 pm

Is PJ

Nice job! I'm so mad that I keep missing your streams. I'm going to have to give you my phone number and have you call me when you begin. ^_^

DK28

2012-05-04 11:29:38 am

General Kong - Bullets and Bananas

That run was ridiculous Bablo. I definitely look forward to the real deal. Your boosting was great. Although, I think some of the stuff may not be in the cards given this game's RNG, cause some things I can't see happening in the same run. But your spider boosting in stage 2 was fantastic and creative. Smiley

Bablo

2012-05-12 12:17:57 pm

I'm a flippin' ninja

I just lost a run at the last stage because a platform didn't spawn for whatever reason, which I've never seen happen before. I was 27 seconds ahead my 34:29 run at that point, even with some crappy luck in the earlier mistakes, and some big mistakes were made too. I did get the 1-cycle Mummy fight though.

So eventhough I lost the run to something very stupid, I was quite glad to see the -27 seconds split. With perfect luck it could've been -40 or so. I've had a pretty hard time gaining any good amount of time over the 34:29 run so this gave me some well needed confidence. I'm getting there, and it is going to happen eventually. I am going to have a break when D3 comes out (in like 50 hours) but I intend to play it 24/7 for the first week, then continue speedrunning.

It is going to happen.

Bad

2012-05-14 04:53:13 pm

Good luck man! Funkdoc needed some challenge!

Reeve

2012-05-14 07:45:39 pm

That's a shame, Bablo. It's really frustrating when something like that happen and you're not even prepared because it's something that you didn't know that could happen. It's good to know you saw the good side of the situation. The big improvement you got means that the your consistency is growing and soon you'll get the run you want. Good luck!

ConHuevos

Edit history:

ConHuevos: 2012-05-15 06:43:36 am

2012-05-15 06:42:21 am

Ciento Dos Huevos

It has begun!

ConHuevos

Edit history:

ConHuevos: 2012-05-15 10:21:23 pm

ConHuevos: 2012-05-15 10:18:20 pm

2012-05-15 09:00:17 pm

Ciento Dos Huevos

Castlevania 4 Hitbox lua script:

http://pastebin.com/fLxjFqmh

NOTE: snes9x 1.43 or 1.51 re-record required:

1.43: http://code.google.com/p/snes9x-rr/downloads/detail?name=snes9x-1.43-rerecording-v17-win32-lua51dll.zip&can=2&q=

1.51: http://code.google.com/p/snes9x-rr/downloads/detail?name=snes9x-1.51-rerecording-v6-win32-lua51dll.zip&can=2&q=

You can toggle the center axis (crosshairs in the pics) off. One thing to note, although the box shows up on the whip behind simon, it can't hit anything behind him. I could not find what makes the whips box active to hide that. You can also toggle which boxes to draw and hp display, they're at the top of the script, set to 'false' if you do not wish them to be drawn (or true if you do).

Another interesting note, is the link next to simon's hand on his whip has a wider hitbox than all the other links on his whip.

BLUE = Simons vulnerability box
RED = Enemy boxes/Pickups (money bags, sub weapons, hearts, etc...)/Other objects (destructible floors etc...)
WHITE = Simon's Whip/Projectiles
GOLD = Candles

[b]Castlevania 4 Hitbox lua script:[/b]

http://pastebin.com/fLxjFqmh

[b]NOTE: snes9x 1.43 or 1.51 re-record required:[/b]

1.43:  http://code.google.com/p/snes9x-rr/downloads/detail?name=snes9x-1.43-rerecording-v17-win32-lua51dll.zip&can=2&q=

1.51:  http://code.google.com/p/snes9x-rr/downloads/detail?name=snes9x-1.51-rerecording-v6-win32-lua51dll.zip&can=2&q=

[img]http://i50.tinypic.com/esssnd.png[/img]    [img]http://i48.tinypic.com/6rmuy0.png[/img]

[img]http://i47.tinypic.com/dhebu1.png[/img]    [img]http://i47.tinypic.com/29cqfti.png[/img]

[img]http://i48.tinypic.com/2z4fql3.png[/img]

You can toggle the center axis (crosshairs in the pics) off.  One thing to note, although the box shows up on the whip behind simon, it can't hit anything behind him.  I could not find what makes the whips box active to hide that.  You can also toggle which boxes to draw and hp display, they're at the top of the script, set to 'false' if you do not wish them to be drawn (or true if you do).

Another interesting note, is the link next to simon's hand on his whip has a wider hitbox than all the other links on his whip.

[b][color=blue]BLUE[/color][/b] = Simons vulnerability box
[b][color=red]RED[/color][/b] = Enemy boxes/Pickups (money bags, sub weapons, hearts, etc...)/Other objects (destructible floors etc...)
[b][color=grey]WHITE[/color][/b] = Simon's Whip/Projectiles
[b][color=gold]GOLD[/color][/b] = Candles

PJ

2012-05-15 09:03:07 pm

Is PJ

Sick! Great work as always! Cheesy

Bablo

2012-08-08 12:44:03 pm

I'm a flippin' ninja

Can't believe I hadn't noticed ConHuevos's post. I definetely need to play around with the lua script. Cool stuff.

So yea I got busy with something else, and lost focus on speedrunning. However I'm also moving soon, which means I'll likely get a better setup for streaming (current one is very awkward). I'll be unemployed for at least a month, so I've got some time on my hands to get back into this.

I won't set any dates though, but I do feel like I'm not quite done with this game. Although I'm considering dropping the risky hope-for-the-best strategy for stage 8's disappearing platforms. It's just dumb, and there are quite a few risky strategies I'm going for anyway. A couple of seconds for consistency isn't a huge deal due to the RNG theme going on for this game. Even if I get a well executed run done with risky strategies, there's a possible ~20 seconds to lose on randomness, especially in the bossfights. Hard stuff I'm not dropping though, I'm quite okay with the tight jumps and damage boosts.

Can't believe I hadn't noticed ConHuevos's post. I definetely need to play around with the lua script. Cool stuff.

So yea I got busy with something else, and lost focus on speedrunning. However I'm also moving soon, which means I'll likely get a better setup for streaming (current one is very awkward). I'll be unemployed for at least a month, so I've got some time on my hands to get back into [i]this[/i].

I won't set any dates though, but I do feel like I'm not quite done with this game. Although I'm considering dropping the risky hope-for-the-best strategy for stage 8's disappearing platforms. It's just dumb, and there are quite a few risky strategies I'm going for anyway. A couple of seconds for consistency isn't a huge deal due to the RNG theme going on for this game. Even if I get a well executed run done with risky strategies, there's a possible ~20 seconds to lose on randomness, especially in the bossfights. Hard stuff I'm not dropping though, I'm quite okay with the tight jumps and damage boosts.

ozzy88

2012-08-08 11:47:11 pm

Great to have you back on this...well soon. I keep coming back to this game and my silly attempts to get through it quickly. I think it's because I've been playing this game since I could play video games at all and just love the series overall.

Last time I played it the jump button broke on my controller, lol, pressing so hard cuz you never want to miss a jump in this game Smiley

P.S.
I already got the controller replaced so no worries.

Bablo

2012-08-09 06:09:19 am

I'm a flippin' ninja

Oh yeah I've also had lots of trouble with snes controllers. Still haven't got a good one, only four average ones with their own little quirks.

Miyamoto Musashi

2012-08-10 07:14:36 am

Shoryuken?

this run is insane......NEXT! =)

JoeDamillio

Edit history:

JoeDamillio: 2013-12-01 12:58:38 pm

JoeDamillio: 2013-12-01 12:57:18 pm

2013-12-01 12:56:32 pm

Hi. It seems this thread has been revived before, so I will assume it is ok to dig it back up once again.

It was about a year ago now that I watched Kareshi's 39:38 run of Super Ghouls 'n Ghosts. It was the first time I've ever seen a speedrun, let alone knew speedrunning existed. Castlevania IV has always been one of my favorite games since I was very little. I was interested to see if a speedrun was done, and I found Funkdoc's 37:30 run from agdq2012. This is when I was gracefully entered into the world of speedrunning. I absolutely love this game, and even though the early stages over and over and over and over and over (etc.) get irritating, I was surprised to see it so low key.

Anyway, I have been able to improve my previous PB's tremendously over the past months (37:17, 35:23, 34:27). The current WR holder, Hanage_Belmondo, taught me the stairs in Stage 8 after my 34:27 run. He is an awesome Japanese player, who has a time of 33:41. Since he taught me that, I've learned the chandelier strat in Stage 6, which I'm pretty consistent with now, and am able to break his WR time. Actually, I can improve this time by a lot if RNG treats me great (especially Mummy one cycle) + great execution. I have plenty of room to improve his run regardless, though there is a problem. His run from Stage B to the end is very, very good. I will inevitably lose some time here, so I will need to be ahead by at least a good margin before Stage B. This is still possible with slight leniency, but not much. I've been going for attempts for a little while now and have found myself about 7-8 seconds ahead after Stage 5 multiple times. I've stopped for now and am practicing until the 4th, and I'll go back to runs. For now I will leave you guys with my PB video, and WR video. Hopefully my next post is WR. I'll be cool with a sub34 to have better splits against, since that run is really bad by now. Also I will be doing a full game in-depth tutorial when I am deemed worthy of such a thing.

Sorry about extended post, first one outside of tech support. Take care fellow Belmonts.

/watch?v=JG69Z1_elHo 33:41 WR
/watch?v=Syw-hO4P0qI 34:27 PB

Hi. It seems this thread has been revived before, so I will assume it is ok to dig it back up once again.

It was about a year ago now that I watched Kareshi's 39:38 run of Super Ghouls 'n Ghosts. It was the first time I've ever seen a speedrun, let alone knew speedrunning existed. Castlevania IV has always been one of my favorite games since I was very little. I was interested to see if a speedrun was done, and I found Funkdoc's 37:30 run from agdq2012. This is when I was gracefully entered into the world of speedrunning. I absolutely love this game, and even though the early stages over and over and over and over and over (etc.) get irritating, I was surprised to see it so low key.

Anyway, I have been able to improve my previous PB's tremendously over the past months (37:17, 35:23, 34:27). The current WR holder, Hanage_Belmondo, taught me the stairs in Stage 8 after my 34:27 run. He is an awesome Japanese player, who has a time of 33:41. Since he taught me that, I've learned the chandelier strat in Stage 6, which I'm pretty consistent with now, and am able to break his WR time. Actually, I can improve this time by a lot if RNG treats me great (especially Mummy one cycle) + great execution. I have plenty of room to improve his run regardless, though there is a problem. His run from Stage B to the end is very, very good. I will inevitably lose some time here, so I will need to be ahead by at least a good margin before Stage B. This is still possible with slight leniency, but not much. I've been going for attempts for a little while now and have found myself about 7-8 seconds ahead after Stage 5 multiple times. I've stopped for now and am practicing until the 4th, and I'll go back to runs. For now I will leave you guys with my PB video, and WR video. Hopefully my next post is WR. I'll be cool with a sub34 to have better splits against, since that run is really bad by now. Also I will be doing a full game in-depth tutorial when I am deemed worthy of such a thing.

Sorry about extended post, first one outside of tech support. Take care fellow Belmonts.

/watch?v=JG69Z1_elHo 33:41 WR
/watch?v=Syw-hO4P0qI 34:27 PB

PJ

2013-12-01 01:08:18 pm

Is PJ

Keep up the great work, and good luck trying to beat Hanage's run! He is a really talented runner and a super nice guy. ^_^ It's nice to see that he's helping out.

I'm looking forward to seeing your progress with this. Cheesy