One day in 1576 Christopher was tired of everyone calling him the slowest of all Belmonts,
so he left his son and his sub weapons at home and rushed out to prove the world wrong!
And he walked so fast that only lag would be able to slow him down...
this run was made with BizHawk v2.3.2 on the european version of the game
and it aims to defeat dracula with the fastest possible real time.
The only significant thing to add here would be that the whole run basically consists of reducing the loading times aka lag frames.
To do so, you have to get rid of as many objects on the screen as possible at any given time,
which in some cases looks like a slower option but in the end is the faster way to archive a good real time.
Maru: When looking for publishing different versions of games, we are looking for either approximately 50% of unique content and/or significant in-game differences (such as in the route or in the gameplay). Unfortunately, this submission does not seem to have that compared to the published GBC version. In addition, optimization is questionable due to the information provided in the thread and the fact that I was able to shave off 17 frames in the first screen of the first stage alone. That can be found here. For that reason, this TAS must be rejected.
i already sent it in pm, but GBC version cannot be compared to GB version, since GBC runs way faster than GB
sorry for repeating myself
ps: the 18:07~ min run that is submitted here, is the fastest that could be compared to my submission here, so i think i can argue that my time (17:41~) is faster by a large margin
i already sent it in pm, but GBC version cannot be compared to GB version, since GBC runs way faster than GB
sorry for repeating myself
ps: the 18:07~ min run that is submitted here, is the fastest that could be compared to my submission here, so i think i can argue that my time (17:41~) is faster by a large margin
Is GBC so much faster than GB that it really saves over 5 minutes in a 20 minute run?
Hopefully, somebody will provide a temp encode so that this run can be watched to see how optimized it is in terms of movement and lag reduction.
GBC version is a lot faster than GB version!
you can just watch a video of both and you will immediately see the difference,
there is no doubt whatsoever, those versions are just different games.
idk what the problem is here...
btw you can watch the video of my run on my youtube channel if you trust my encoding^^
We get it already. The GBC version from the Konami GB Collection lags less. (although the TAS that was done on that version was done on VBA-RR, which is notoriously inaccurate, but that's beside the point)
Anyway, watching the run, I noticed that during the second boss fight, the last fleaman spawns from the topmost hole, one you can't reach. This alone makes it sub-optimal, since it can spawn at any of the three other holes (and would inevitably be faster).
Therefore, I have doubts this can even be published, version differences aside.
Anyway, here's the same submission, but with about 2.5 minutes of extraneous input (including input just after Dracula is defeated) cut out.
i agree that the RNG on the second boss fight is not the best you can get, but i couldn't find out how to manipulate this RNG there. it most likely depends on the global timer of the game which means i would have to be significantly slower (or faster) in the run so i would reach the boss room at another time to get a better sequence.
however, i decided to take that trade off to showcase a more optimal run up until that boss fight (which is very boring to watch anyway), so i would say it is either just innevitable, because the global timer just hits an unfavourable value or it can be treated as entertainment trade off to show better movement before the boss.
also there is no perfect run anyway, what do you expect
actually, im highly optimistic that if my inputs from the GB version would be able to run on the GBC version, that it would even beat that current run there.
i've watched that video and i spotted a lot of flaws that my new run avoids.
idk if there is a way to translate inputs to other games, but if so, then someone should do that
however, i decided to take that trade off to showcase a more optimal run up until that boss fight (which is very boring to watch anyway), so i would say it is either just innevitable, because the global timer just hits an unfavourable value or it can be treated as entertainment trade off to show better movement before the boss.
also there is no perfect run anyway, what do you expect
That's not how it works. To be considered a speed vs. entertainment tradeoff, you have to compromise fastest time for something which is more entertaining. However, in your case, having 5 frames faster movement followed by an extra second and a half of waiting at the boss fight is much less entertaining than waiting 5 frames before starting the boss fight and hitting the boss immediately as the fight starts.
It's important to keep this in mind when making a TAS, especially one which is intended to go into the vault (which discourages speed vs. entertainment tradeoffs). If this run was accepted to the vault, then somebody could immediately obsolete it by just getting better RNG on the boss fight and then re-syncing the rest of your movie after that.
You wouldn't want your hard work to go to waste - so if you're interested in seeing your movie published, you should probably check if some combination of delaying your movement or pressing different buttons before the boss fight can change which hole the boss spawns out of in order to be sure that your run is truly optimal.
Anyways, I don't know if this is the video that Mazzin was referring to, but this is a link to an encode of A TAS of Castlevania on the Game Boy on his channel:
Link to video
Edit: edited to include the link to the correct video of this TAS.
did you even read the video name or look at the thumbnail?
it clearly states ingame time and not real time...
i was obviously refering to the video that matches my submission file:
https://www.youtube.com/watch?v=KvLYA-e2Wek
did you even read the video name or look at the thumbnail?
it clearly states ingame time and not real time...
i was obviously refering to the video that matches my submission file:
https://www.youtube.com/watch?v=KvLYA-e2Wek
Well excuse me for not being able to find the link to the encode of your TAS that you made over a month before submitting your movie to TASVideos which you also uploaded at the same time as another TAS of the same game... -_-
For future reference, you can edit your original submission text to include a link to your video. Doing so helps drive up interest in your submission, as people are less interested in downloading a Castlevania ROM and watching the run play back in real time for 20 minutes than they are watching a video online of the TAS which allows them to skip through autoscroller sections as they please.
Joined: 10/12/2011
Posts: 6438
Location: The land down under.
Did you know?
That instead of using the actual IGT you can use the value related to the IGT 0x436.
Since this TAS focuses on IGT apparently. Let's grab a table up to compare to the previous.
Stage 02 is the biggest pitfall in your TAS as you're too focused on clearing out the enemies behind you rather than focusing on the one's infront of you and letting the ones behind you roll off the bridge.
You can throw Dracula Phase 2 as a wash since there's no actual way to gather a time from the game's values, this is manual counting between frames.
However, the strategies are different, the 2010 TAS is using the brute force method, getting hit but returning damage whilst this one is staying safe below Drac.
WebNations/Sabih wrote:
+fsvgm777 never censoring anything.
Disables Comments and Ratings for the YouTube account.Something better for yourself and also others.
Did you know?
That instead of using the actual IGT you can use the value related to the IGT 0x436.
Since this TAS focuses on IGT apparently. Let's grab a table up to compare to the previous.
The TAS that Mazzin submitted here is supposed to focus on fastest real-time (from power on to last input to kill Dracula). The fastest IGT video was for a different TAS on his channel.
I've updated the link in my earlier post to reflect this.
Joined: 10/12/2011
Posts: 6438
Location: The land down under.
I've also compared with the submission that's here, and not the video as a note.
Even if IGT is not the focus, it's still usable enough to do comparisons such as these.
WebNations/Sabih wrote:
+fsvgm777 never censoring anything.
Disables Comments and Ratings for the YouTube account.Something better for yourself and also others.
I've also compared with the submission that's here, and not the video as a note.
Even if IGT is not the focus, it's still usable enough to do comparisons such as these.
In terms of total frame count minus lag frames, how does this TAS compare to the 2010 submission?
Did you know?
That instead of using the actual IGT you can use the value related to the IGT 0x436.
Since this TAS focuses on IGT apparently. Let's grab a table up to compare to the previous.
Stage 02 is the biggest pitfall in your TAS as you're too focused on clearing out the enemies behind you rather than focusing on the one's infront of you and letting the ones behind you roll off the bridge.
You can throw Dracula Phase 2 as a wash since there's no actual way to gather a time from the game's values, this is manual counting between frames.
However, the strategies are different, the 2010 TAS is using the brute force method, getting hit but returning damage whilst this one is staying safe below Drac.
very nice comparison! i really appreciate this, thank you for making it.
so this is my submission file vs the current best file here?
that makes me even more confident in my run, because i had to purposely sacrifice ingame time to combat lag and seeing that i only lost 12 frames (i assume it's frames in that list) in level 2, that is sick!
and you observed right, i always tried to get rid of the eyeballs behind me first since they roll in the same direction i have to go so they would be way longer on screen than the ones that come from the front, so the enemies behind me pose a bigger threat to my time than the others...
i recommend you to watch that sequence on the bridge in both videos and see the huge difference in lag.
oh and i believe you can actually cheat on IGT if you spam moves while still progressing as fast as you can, so you make the game lag way more and the timer runs slower, leading to a "better" ingame time.
also, i agree with lobsterzelda, if you could also do a comparison in terms of total time, that would be great.
Stage 02 is the biggest pitfall in your TAS as you're too focused on clearing out the enemies behind you rather than focusing on the one's infront of you and letting the ones behind you roll off the bridge.
You can throw Dracula Phase 2 as a wash since there's no actual way to gather a time from the game's values, this is manual counting between frames.
However, the strategies are different, the 2010 TAS is using the brute force method, getting hit but returning damage whilst this one is staying safe below Drac.
