That was awesome! NOW I'm really interested. :)
Did you have any desync problems or something like that? And, does it have an AVI dump function or are you using an external tool to capture those videos?
Anyway, keep up the great work. I'd love to see more of this game.
You're just fucking stupid, everyone hates you, sorry to tell you the truth. no one likes you, you're someone pretentious and TASes only to be on speed game, but don't have any hope, you won't get there.
As far as I can tell, Mednafen's emulation is very good. The only problems I've had are due to bugs in my own code.
I've been using screen capturing software, I don't think it has AVI dumping. Perhaps it could be ported from FECU? I haven't looked into this issue.
Some development assistance would be awesome if that interests you. :)
I believe that it'd be faster to backflip instead of walking.
But anyway this is very cool. It's nice to see progress on so many platforms in this day.
there doesn't seem to be a difference between jumping speed and walking speed. at least, not one as profound as in the SNES game. i don't believe backflipping will help.
Level 2
Backflips seem to be slower than normal walking.
I added individual hotkeys for saving and loading savestates. I also added hotkeys that can speed up and slow down the rate of the frame advance key when it is held. I think this makes Mednafen about advanced enough to release a windows binary, but even after a lot of attempts I can't seem to put together a working toolchain for compiling one.
Amazing music; but why do the voices sound so funny? :o)
It's great to see more progress. Are you compiling in Linux or...? I remember I tried once to compile Mednafen in Windows, but I couldn't find some of the required libraries...
Also, I'll probably try to help you at some point, but I need to finish some projects first, before I start getting into other ones. It's kind of depressing to have so many unfinished projects floating around.
You're just fucking stupid, everyone hates you, sorry to tell you the truth. no one likes you, you're someone pretentious and TASes only to be on speed game, but don't have any hope, you won't get there.
Amazing music; but why do the voices sound so funny? :o)
Cause it's running too fast.
Criticisms:
1:14 - You can jump through the wall crash, just as long as you're in the air before you get to the spot. That'll put you farther ahead when you can continue.
1:44 - Is falling there faster then jumping down? Just make sure - your decision may be correct, but I want to see.
1:55 - I think you should jump into the gate and use the key, which should be faster then stopping to use it.
Yeah, I fast-forwarded the cutscene with the intention of cutting it out but I thought it was funny. :) It looks like this normally.
mz: I'm currently working on it in an Ubuntu VM running in Windows due to the difficulty of getting it to compile any other way. I've tried several configurations of MinGW in Windows, MinGW cross-compiling in Linux, and Cygwin and I couldn't get anything to work. Fortunately there is massively less work to be done on Mednafen than what you had to do on PCSX. It's already very usable for my purposes.
Sir VG:
1 - Thanks, I will look into that.
2 - Falling was faster when I tested it, but I'll check again when I redo that part.
3 - I think you're right, I'll look into that as well.
EDIT: Level 3
There is now a recording/playback/stopped indicator by the frame counter. Also, savestates made during recording and playback are now separate from ones made while stopped. This should prevent potentially ruining a movie by saving states while stopped.
For those interrested to watch WIPs of rondo, theres a svn here :
http://code.google.com/p/rondoofblood/
(me, adelikat and paul_t are co-author)
Just get turtoise or whatever svn client you want, be sure to use FM2 version of mednafen (r226+)
Compared to the old run from paul_t theres is tonnes of improvements, also whe are aiming at multi-movie, namely any%,100% and richter runs
The point of the richter run is just entertaining value, obviously as soon as you reach maria, the games goes pretty faster with her, even by considering the needed "dialogue", this is why any% and 100% will use her mostly, so whe dont need a "maria" movie (that would just do stage1/2 differently)
The runs are going to be high quality i believe, im not going to post on youtube wips coz its too tedious, theres changes/improvements everyday
Edit: ah yeah, theres also a bot advertising updates made on the run available at #arukAdo on freenode
Agreed. While somebody did a speed run video some time back of Maria only while saving everybody (So they started in Stage 1 and took the alternate route the entire way, sans 5'), there's no point for TAS purposes.
As a note for restarts for the any%/100%, you have to wait until at least the black screen, maybe even until the next stage appears. Sometime during the black screen is when the save file gets written to.
Also remember that you can fast reset by Select + Run.
As for Richter only, I think that should be done low%, which would be more of a variety then any%. So to do that, you do Stage 0, Stage 1 (alternate boss), Stage 2 (Oarsman Route), Stage 3' (alternate boss), Stage 4' (Normal Boss would likely be faster), Stage 5, Stage 6, Stage 7, Stage 8.
Good luck. I REALLY want to see a TAS of this, being a fan of the game (and a speed runner of it too).
Also remember that you can fast reset by Select + Run.
Yes i remember theres a soft reset, whe didnt finished stage2 for any% yet, so i didnt tested so far if its faster or slower than hard reset (both are available with fm2 anyway)
Edit : tested and result is that you cant soft reset in black screen, you can hard reset tought, so hard reset win in the end :p
Sir VG wrote:
As for Richter only, I think that should be done low%, which would be more of a variety then any%. So to do that, you do Stage 0, Stage 1 (alternate boss), Stage 2 (Oarsman Route), Stage 3' (alternate boss), Stage 4' (Normal Boss would likely be faster), Stage 5, Stage 6, Stage 7, Stage 8.
The only trouble i see for this is: the variety WILL come from 100%, since it visit all stages (also, whe plan on reseting at stage6 and not after dracula like you, might need the stage select code, but at the least whe finish with dracula, not in stage5')
What i was whilling for richter is to fight werewolf and minotaur, since they are epic bosses reused in many later castlevania
Also i dont realy have a choice for stage2, i WANT the key, its going to be a key-run, key does twice damage of whip and remove the heart-management-hassle
Need to consider that hydrostorm will need 99 hearts at shaft to be realy efficient, witch i want to avoid coz its slower to catch hearts (and boring to optimize too), the difference is realy noticable in stage2 between heartfull or heartless run
Sir VG wrote:
Good luck. I REALLY want to see a TAS of this, being a fan of the game (and a speed runner of it too).
Thanks, well, the progress can be monitored easly ;)
Also, of course whe know your (sda) speedruns, that basicly our reference to work on
One thing i was wondering yesterday, it cost like 6000 frames to save maria (1min30), considering the "new" way richter is moving around, i had a doubt that shes realy faster in any%, but well, i guess even with such a big handicap she can overcome richter
Well, it's faster for speed running purposes, but while Richter has faster movement, Maria CREAMS bosses with Guardian Fist. Richter TASed may attack bosses faster, but question is...how much time will Maria make up on all the other enemies because she can attack better?
Just to give you a concrete compare, richter kill werewolf 150 frames faster with key than with 2 hydrostorm (40 hearts) that was only about 450 frames long... :p
He die in 6 hit with the key, witch is ridiculously fast
But in stage3 apparently maria goes much faster than richter, so more likely richter will not beat her after all
Its just weird to see richter WAY ahead so far (hes still like 6000 frames faster than maria at boss of stage3)
The dialog with maria is realy gigantic and troublesome to make estimations
Wait wait...you can kill him with the fricken KEY?! Damn that's hilarious!
BTW, for the any% run you are going to Dogether in Stage 3? Stage 4' is much quicker then Stage 4 (especially with Maria).
That something whe are not sure about yet (for the any%), it require some testing before whe realy choose witch stage whe are going to pick up
And yeah, KEY = best weapon in game ;)
(twice damage of whip!)
Joined: 12/29/2012
Posts: 2
Location: Pittsburgh, PA
I'm not sure if this would actually save time or not, but I ran into an interesting glitch while doing a few runs of this game this morning.
http://www.twitch.tv/skiffain/c/1803437
you normally have to walk to the edge of the screen, but getting hit here after sliding backwards into the enemy and scrolling the screen, changed it up
Joined: 12/29/2012
Posts: 2
Location: Pittsburgh, PA
I found a floor clip that could save a lot of time in a 100% run.
http://www.twitch.tv/skiffain/c/1866002
The clip works by jumping right before you get hit, I'm guessing the bridge floor moves up just as you move below it, and therefore you fall down through it. I had this happen a slightly different way earlier, on wii VC, but this was the most consistent way I could reproduce it to record it.
This saves around 15 seconds just in realtime, not sure what the optimal time would be for a tas.
Reminder that this TAS is completely obsolete and has been since its inception because the TASers wanted to be epic and use the key. The RTA record is faster than the fastest Richter TAS.
Reminder that this TAS is completely obsolete and has been since its inception because the TASers wanted to be epic and use the key. The RTA record is faster than the fastest Richter TAS.
Joined: 11/13/2006
Posts: 2792
Location: Northern California
Countdown wrote:
Reminder that this TAS is completely obsolete and has been since its inception because the TASers wanted to be epic and use the key. The RTA record is faster than the fastest Richter TAS.
In RTA timing, the TAS is sub-20 minutes. TASes and RTA runs are not timed the same way, as TASes are always timed from power on to the last input needed to finish the game, while this particular RTA is timed from taking control of Richter.
In the RTA WR, Richter begins moving at roughly 35 seconds and the final orb is collected at roughly 22:20. Roughly 21:45. In the TAS encode, Richter begins moving at roughly 3:44 and the orb is collected at 23:33. Roughly 19:49.
So that's an insanely egotistical way to be completely wrong. Congratulations!
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Reminder that this TAS is completely obsolete and has been since its inception because the TASers wanted to be epic and use the key. The RTA record is faster than the fastest Richter TAS.
