Does anybody have an idea why Dracula is only called "The Count" in the manual of the original "Castlevania" as well as any additional material?
I mean, was this really supposed to be because of trademarks? Is this the reason the developers named? For a character from
1897?
And what happened when they released "Castlevania II" and used the name Dracula everywhere? Did they now believe they had aquired the rights?
Also, what was the point of not using the name Dracula,
if he's called like that in the game itself?
Why do you think it has anything to do with rights? "The Count" seems like a perfectly good name to use.
rainwarrior wrote:
Why do you think it has anything to do with rights?
Because that's probably the bullshit that the developers would tell you if you asked them.
rainwarrior wrote:
"The Count" seems like a perfectly good name to use.
Instead of, you know, "Count Dracula".
I'm wondering why the name Dracula
never ever appears anywhere in materials concerning the first "Castlevania".
DRW wrote:
Does anybody have an idea why Dracula is only called "The Count" in the manual of the original "Castlevania" as well as any additional material?
I mean, was this really supposed to be because of trademarks? Is this the reason the developers named? For a character from 1897?
It could have been a fear of ambush litigation. Perhaps Konami considered Stoker's estate litigious, as it had had all but one print of the film
Nosferatu destroyed. The only reason copies of
Nosferatu survive is widespread piracy of that one print. The copyright in
the first authorized film adaptation from 1931 still subsists, and Konami would have had to work carefully around similarities between the in-game appearance of the character and the appearance of the character in the 1931 film. This same restriction made production of Disney's
Oz the Great and Powerful more difficult, as a lawyer had to be present to ensure that costume and set design carefully stepped around any details introduced in MGM's film.
tepples wrote:
It could have been a fear of ambush litigation. Perhaps Konami considered Stoker's estate litigious, as it had had all but one print of the film Nosferatu destroyed.
So, they feared that the copyright that had "Nosferatu" from 1922 banned was still in effect in 1987?
But they included his name into the game nevertheless.
And Frankenstein and Igor are still fair game? Even though, while Dracula just has a generic design (i.e. he doesn't particularly look like Bela Lugosi or Christoper Lee), these two characters
are specifically based on the movie version.
Very strange.
Oh, yeah: And the MSX version "Vampire Killer" calls him Dracula again in the materials. Only this time they changed Transylvania to Dransylvania. Are they all a bunch of idiots?
What does the Japanese manual say? This could be the result of NOA censorship.
The Japanese manual probably says Dracula.
But the question still remains: Why censorship? Why does that name need to be censored?
I mean, I get that they renamed "Demon Castle Dracula" to something not including the word "demon". But why is the name Dracula off-limits, especially when Frankenstein is called by name in the manual?
Nintendo of America was very strict with their localization process. There were very worried about how fragile the market was (after the video game crash) and wanted to do everything they could to make sure they didn't suffer the same fate as Atari. The name could have been cut just to be extra careful.
I figure if they were worried about rights, "The Count" is a much worse pick than "Dracula", since that name was actually being used contemporarily with the game's development:
And this Count ended up getting
his own game using the same voice codec as
Big Bird's Hide and Speak.
"Got it!"
I would say that the voice samples are very impressive for the NES, but really, is it just that pretty much all others are bad? Doesn't the NES just have a standard DPCM channel? I'm guessing the main reason so many other voice samples in other games on the NES are so bad is due to memory constraints? Has anyone ever actually tried to stream "CD quality" music on the NES?
(tangent) The NES only has a 7-bit DAC; while one could reasonably stream (at the cost of all CPU time) audio at ~100kHz, it's harder to compensate for the lesser bit depth.
The DPCM compression format imposes a fairly large quality hit. And is a really lousy compressor too, especially for speech (which compresses very nicely with
LPC or similar)
So, I take it the voice samples in the game actually are quite impressive, even if there's almost nothing going on when they are being played.
You know, I found this just now and I just couldn't resist showing it:
Espozo wrote:
I would say that the voice samples are very impressive for the NES, but really, is it just that pretty much all others are bad? Doesn't the NES just have a standard DPCM channel? I'm guessing the main reason so many other voice samples in other games on the NES are so bad is due to memory constraints? Has anyone ever actually tried to stream "CD quality" music on the NES?
I disassembled Big Bird's sample decompression here:
http://forums.nesdev.com/viewtopic.php?t=8675In my opinion it is the best sounding speech in an NES game. It's got an unusual compression format but it's basically a little more than 5 bits per sample, raw PCM.
There are 3 factors that I think make NES samples suffer the most:
1. Memory constraints.
2. CPU constraints.
3. DPCM quality constraints.
In Big Bird's case, #1 was solved by compressing to ~5bps, and dedicating most of the cartridge's data to it (including some of the CHR). #2 was solved by dedicating the CPU almost entirely to the sound while it's playing (with small animations carefully placed at split-points within the sample). #3 was solved by not using DPCM.
For other games, DPCM is a solution to #2, and a partial solution to #1 (both at a huge quality loss), but I think #1 is the biggest reason so many games don't even use DPCM samples. They probably thought it was better to use that room for other things than crummy audio samples. With Big Bird, though, they made the game
about the audio.
lidnariq wrote:
(tangent) The NES only has a 7-bit DAC; while one could reasonably stream (at the cost of all CPU time) audio at ~100kHz, it's harder to compensate for the lesser bit depth.
I actually think the bit-depth isn't the biggest concern. 7 bits allows a signal-to-noise ratio of about 42 decibels, I think, which isn't too bad. I saw a claim that cassette tapes usually had an SNR equivalent to about 5 or 6 bits (
video). The quantization noise for the NES is a factor for sure, but I think it really pales in comparison to the problems of memory/CPU/DPCM.
Of course, when you've got continuous music going everything becomes more apparent. I don't like listening to Amiga or GBA hardware recordings much because of the 8-bit audio depth. There's plenty of SuperNSF NSF files that sound reasonably well with 7-bit mixing though. E.g.
bananasp.nsf
rainwarrior wrote:
I don't like listening to Amiga or GBA hardware recordings much because of the 8-bit audio depth.
I don't know about the Amiga, but I thought the main problem for the GBA was
cheap developers limited memory for voice samples.
Espozo wrote:
I would say that the voice samples are very impressive for the NES, but really, is it just that pretty much all others are bad? Doesn't the NES just have a standard DPCM channel?
Yes, and this isn't using it. It's using a codec with higher quality than 1-bit DPCM, which requires the CPU's full attention.
Quote:
I'm guessing the main reason so many other voice samples in other games on the NES are so bad is due to memory constraints?
That and the 100% CPU use.
lidnariq mentioned linear predictive coding (LPC), which uses about 10 multiplies per sample. It's fine on a DSP, not so fine on a 6502. The closest thing to LPC on 8-bit microprocessors is ESS MX/DigiTalker compression, which uses several techniques to reduce data rate: 2-bit DPCM, repeating a waveform at the fundamental voice period within a frame (4:1 savings), cosinization of the waveform's phase to allow ping-pong looping (2:1 savings), and silencing of the quiet half of the cosinized phase (another 2:1). It was used in several C64 games (including
Impossible Mission and
Ghostbusters) and one Famicom game. Perhaps ESS charged too high a royalty for its use in other games.
ESS also developed the 5-bit linear PCM codec used in the Sesame Street games, which is far less space-efficient but far clearer, and clarity is of utmost importance in this sort of edutainment.
Quote:
Has anyone ever actually tried to stream "CD quality" music on the NES?
Not exactly CD quality, but there's an
NES demo that plays part of "Never Gonna Give You Up" at full 7-bit quality. It uses 99% of CPU time.
My
Action 53 menu plays me saying "Make your selection now" at 4-bit DPCM and sounds at least as good as the ESS 5-bit samples. One trick I do is sibilant inversion, which shifts some frames of audio up into a higher frequency band to fake 16 kHz sampling with only 8 kHz of actual samples.
tepples wrote:
One trick I do is S-enhancement, which shifts some frames of audio up into a higher frequency band to fake 16 kHz sampling with only 8 kHz of actual samples.
Could you explain what this means? I'm curious.
I'd guess it's approximately the same as Vorbis's noise floor encoding; effectively encoding a sibilance=white noise channel.
While you were composing your reply, I edited it from "S-enhancement" to "sibilant inversion", which is a more accurate term.
My QuaDPCM (quadratic-delta pulse code modulation) codec takes 4-bit quadratic deltas, accumulates them, and produces an output of 7-bit samples at 8 kHz. This is fed into a 2x linear interpolator stage, which expands it to 16 kHz where most of the energy is in 0-4 kHz. But if sibilant inversion is enabled for a frame, the interpolated samples are negated, which has the effect of inverting the frequency spectrum of the overall output: 0-8 kHz become 8-0 kHz. So 0-4 kHz out of the DPCM stage represent 8-4 kHz out of the speakers, which is exactly where 's' sounds have most of their energy. This takes almost no time because the unsigned interpolated value is XOR'd with either $00 (if off) or $7F (if on).
The encoder takes 16 kHz input and decides whether to use sibilant inversion on a particular frame by calculating the frame's autocorrelation at lag 1; it's enabled if the correlation between adjacent samples is negative. Then it applies sibilant inversion to the source audio (moving the S sounds down to the lower frequencies) where needed, applies a low-pass filter, decimates by a factor of two, and encodes the result with QuaDPCM.
"Quadratic" means the nibbles are indexes into the table [-64, -49, -36, -25, -16, -9, -4, -1, 0, 1, 4, 9, 16, 25, 36, 49].
"Linear interpolator" translates [0, 16, 41, 5] into [0, 8, 16, 28, 41, 23, 5].
"Sibilant inversion", if enabled, produces [0, -8, 16, -28, 41, -23, 5].
"Autocorrelation at lag n" is defined in any DSP book
OK, got it. So for an "inverted" frame, you're replacing the interpolator with its opposite function, which is a bit weird to think about the consequences of. I guess it would add a strong component at the nyquist freuqency, but with subharmonics below it falling off gradually? Is the goal just to create something that approaches noise, like lidnariq is suggesting, or does it satisfy some other desirable property?
Is this your technique, or is it based on prior research? If there's a whitepaper on it, I'd like to read it. Kinda curious if there are some non-obvious properties of abusing a linear interpolator like this.
rainwarrior wrote:
OK, got it. So for an "inverted" frame, you're replacing the interpolator with its opposite function, which is a bit weird to think about the consequences of. I guess it would add a strong component at the nyquist freuqency, but with subharmonics below it falling off gradually? Is the goal just to create something that approaches noise, like lidnariq is suggesting, or does it satisfy some other desirable property?
The other desirable property is that it is a more-or-less accurate representation of the part of the spectrum between half-Nyquist and Nyquist. The spectral inversion effect is exactly the same as the difference between upper and lower sideband modulation. But yes, several wideband voice codecs do represent upper bands with inversion in this manner, using traditional LPC for 0-4 kHz and some inverted thing for 4-8 kHz.
Quote:
Is this your technique, or is it based on prior research? If there's a whitepaper on it, I'd like to read it.
The terms to Google are
spectral inversion and
heterodyne. It's the result of multiplying a Nyquist-frequency cosine wave with a signal. Performing the multiplication by negating only interpolated samples just makes it more efficient to implement at 1.8 MHz. And I'm not the only one to have thought of it;
other people are inverting the spectrum in MATLAB. But interpolation followed by switchable inversion, where the output can contain only low frequencies or only high frequencies but never both at once, isn't something I've seen done elsewhere.
Oh, you're literally inverting the spectrum of the signal. That makes sense. Pretty interesting, thanks.
tepples wrote:
It could have been a fear of ambush litigation.
Movax12 wrote:
What does the Japanese manual say? This could be the result of NOA censorship.
Having it in the game itself but not in the manual, especially when the game was explicitly localized? (・~・)
It's possible that Dracula's name being used in the localized game was a last minute thing and nobody else was told it was OK to use the name. Or maybe they were changing everything to The Count at last minute and had simply overlooked that instance.
Dracula is mentioned several times through the
Japanese manual (FDS version) not to mention the game's title.
Maybe the name change to "Castlevania" was partly because they wasn't sure if they where allowed to use Dracula's name.
I am once again sorry I wanted to avoid this thread but I cannot resist due to the ridiculous turn it has taken.
All this thread is entierely based on consipration theories and speculations. The character we're talking about is part of the free domain, and the mentions of it as "Draclua", "Vlad Tepeş","Count Dracula" or "The Count" are strictly equivalent and have nothing to do with copyright, intelectual property, censorship or god knowns what.
Usually when someone cannot explain something they'll automatically say it's because of censorship, copy protection or copyrights. But no this isn't always the case.
If the name was changed, it's just because it sounded cooler, or because "Akumajo" doesn't mean anything in english.
Yeah it's speculations, but copyright stuff could be serious trouble so I wouldn't be surprised if they avoided it at first to be safe. Even though as you say it's not copyrightable they may not have been sure about it.
akuma = demon, demon related
-jou = castle name sufix I think
Akumajou = Demon Castle or Castle Demon or something like that.
Castlevania maybe sounds cooler. It's probably more memorable at least.
Bregalad wrote:
The character we're talking about is part of the free domain
This was my initial thought, but then I remembered this was only because of a screw up in the US print and doesn't have to be necessarily the case elsewhere (even though, incidentally, we're talking about the US release).
Bregalad wrote:
I am once again sorry I wanted to avoid this thread but I cannot resist due to the ridiculous turn it has taken.
All this thread is entierely based on consipration theories and speculations. The character we're talking about is part of the free domain, and the mentions of it as "Draclua", "Vlad Tepeş","Count Dracula" or "The Count" are strictly equivalent and have nothing to do with copyright, intelectual property, censorship or god knowns what.
Usually when someone cannot explain something they'll automatically say it's because of censorship, copy protection or copyrights. But no this isn't always the case.
That's exactly the thing that I mean: Ask these people and all they'll tell you is bullshit:
"Bla bla bla copyright."
"Bla bla bla we weren't able to animate Mario's hair." (As if short hair woluld have to be
animated in the first place.)
"Bla bla bla we made Mega Man blue because there are more shades of blue than of any other color." (Even though he consists of only two shades of blue anyway, and one of them is the same color as the sky in two levels.)
The only question is: Why did they
really decide to use just the name "The Count"?
I mean, it's not like that name is just one of many titles that are used. The story doesn't read: "Count Dracula has been resurrected. ... Can you defeat the Count?"
The thing is: They
only call him The Count and
never call him Dracula. Even in the list of characters, you have Medusa, Frankenstein, Igor. And then: The Count.
But no, using
just "The Count" is not equivalent to using "Dracula".
If I ask someone: "How do you like movies with the Count?", people wouldn't necessarily make a connection to Dracula without further context. "The Count" is not an official title for Dracula that would be comparable to a proper name.
So, they
did avoid using his name. The question is: Why?
Bregalad wrote:
If the name was changed, it's just because it sounded cooler, or because "Akumajo" doesn't mean anything in english.
The question is not why the game's title was changed, by why the character of Dracula is never mentioned by name in the materials of the first game, despite his name appearing inside the game and they also calling the Frankenstein monster by name.
I don't think they put as much thought into localization and translating games of this era as you're giving them credit for. Even games by fantastic developers like Konami and Capcom were not except.
"You now prossess Dracula's rib."
"I feel asleep!!"
"Congraturation this story is happy end. Thank you."
@darryl.revok:
That's not the thing we are talking about. Your examples are simple in-game translation errors made by the Japanese.
We are talking about supplementary material like the manual, which was clearly written my someone who speaks English natively. And it's not about any errors either, but about a name choice.
The point I'm getting at is that game companies at the time generally did not feel that the localizations needed that much effort. I don't wanna get wrapped up in this I promise.
(Although the audio stuff was interesting)
Lots of different reasons one person could decide or be told one day to call him something different. The perception of parents who would be more likely to see the manual than the entire game is a possibility.
Game companies didn't care. Nintendo cared. Some things could have been removed arbitrarily without a solid reason.
darryl.revok wrote:
I don't think they put as much thought into localization and translating games of this era as you're giving them credit for.
Exactly, they didn't seem to take the localization seriously. Just look in to the introduction of the
English manual ("good eeeeeevening, are you ready to go a round against the count? bla bla bla"), just a half page with a bunch of rubbish. While the Japanese manual has a full and detailed backstory over several pages, telling that Dracula is being resurrected once every hundred years by men with evil hearts when the powers of Christ is at its weakest. Then during an Easter carnival, evil men once again gathers in the outskirts of town in a black mass with human sacrifices to resurrect Dracula again. Christopher Belmond is also mentioned as a hero that defeated Dracula a hundred years before.
It's clear that they are deliberately avoiding Dracula's name in the English manual. And the only explanation I can think of is that they simply wasn't sure about the rules, and didn't want to take any risks at the moment so it was easier to just change to "The Count". Sneaky.
Pokun wrote:
Exactly, they didn't seem to take the localization seriously. Just look in to the introduction of the
English manual ("good eeeeeevening, are you ready to go a round against the count? bla bla bla"), just a half page with a bunch of rubbish.
These are two different things though: Gibberish in-game texts were done by the Japanese developers who just didn't give a shit about correct translation. The cheesy manual story on the other hand was done by Americans who tried to be funny. Ergo: Two different kinds of not taking it seriously.
Pokun wrote:
It's clear that they are deliberately avoiding Dracula's name in the English manual. And the only explanation I can think of is that they simply wasn't sure about the rules, and didn't want to take any risks at the moment so it was easier to just change to "The Count". Sneaky.
I still find it ironic that they did use the name Frankenstein.
Also, this might actually be the correct explanation. But what I hate is that you don't hear this when you ask them. Ask anybody at Konami USA and I'm pretty sure they won't tell you: "We were not sure and avoided the name because we didn't want to take any risks. But in the end, yeah, we could have used the name Dracula right from the beginning." Instead, they will tell you: "Yeah, that had to do with copyright" and they're done with it.
Pokun wrote:
the Japanese manual has a full and detailed backstory over several pages, telling that Dracula is being resurrected once every hundred years by men with evil hearts when the powers of Christ is at its weakest.
Bingo. NOA didn't want religion on its platform and wouldn't let them say "Christ" in the manual.
tepples wrote:
Bingo. NOA didn't want religion on its platform and wouldn't let them say "Christ" in the manual.
Which, again, still doesn't have anything to do with the original question. I'm pretty much aware that
religion was not allowed to be named. (I never asked why "Demon Castle"/"Devil Castle" was changed to "Castlevania".) But not using the name
Dracula, that's a totally different issue which probably has nothing to do with Nintendo's censorship politics. (Dracula is a literary character. Unlike, for example, the devil, he doesn't come from religious stories.)
Also, this doesn't even explain why the backstory was changed. You don't need to mention Christ to tell the story that Dracula gets resurrected every 100 years and was defeated by Christopher Belmont the last time. I guess the backstory in the American manual was just the Americans trying to be funny. They also changed "Contra"'s backstory from a futuristic setting to a present-day setting, without there being any need at all.
You should really keep them apart:
Religious censorship != fear of copyright infringement != Americans getting creative with the backstory in the manual != Japanese not giving a crap about correct English in in-game texts
Pokun wrote:
It's clear that they are deliberately avoiding Dracula's name in the English manual. And the only explanation I can think of is that they simply wasn't sure about the rules, and didn't want to take any risks at the moment so it was easier to just change to "The Count". Sneaky.
So what ? Let's admit that they deliberatedly avoided to use the name Dracula, and you're asking us why :
Quote:
Does anybody have an idea why Dracula is only called "The Count" in the manual of the original "Castlevania" as well as any additional material?
How could we have any clue ? There is no way we'll never know *if* they deliberatedly avoided the name Dracula, even less *why* that did it. We can only suppose things at this point, so this whole thread is completely useless because you're asking a question that cannot be answered. Thus, this thread should have stayed with 0 answers.
Also using the name "Frankenstein" for the monster is incorrect, it's name is "The Creature" and "Frankenstein" is the name of the doctor who created The Creature.
Bregalad wrote:
How could we have any clue ?
Maybe an official statement does exist (that is not just a bullshitty "Meh, copyrights") and somebody knows about it.
Bregalad wrote:
Also using the name "Frankenstein" for the monster is incorrect, it's name is "The Creature" and "Frankenstein" is the name of the doctor who created The Creature.
Which everybody already knows and which is completely irrelevant in the current context.
Bregalad wrote:
Also using the name "Frankenstein" for the monster is incorrect, it's name is "The Creature" and "Frankenstein" is the name of the doctor who created The Creature.
The scientist is Victor Frankenstein, and he hasn't quite finished his doctorate yet. The creature names himself Adam, and he is Victor's "son" in the same way that Pinocchio is Geppetto's "son". Hence
Adam Frankenstein is as close as we get to a name for the creature.
As for the original question of omission of "Dracula" from the manual, it's unlikely that it'll receive a public, definitive answer. We've covered most of the hypotheses, and none of these is quite a complete explanation.
DRW wrote:
Bregalad wrote:
How could we have any clue ?
Maybe an official statement does exist (that is not just a bullshitty "Meh, copyrights") and somebody knows about it.
I thought you had pre-emptively rejected any and all official statements?
DRW wrote:
Ask these people and all they'll tell you is bullshit.
tepples wrote:
Bregalad wrote:
Also using the name "Frankenstein" for the monster is incorrect, it's name is "The Creature" and "Frankenstein" is the name of the doctor who created The Creature.
The scientist is Victor Frankenstein, and he hasn't quite finished his doctorate yet.
I love these kind of super detailed corrections. lol
BTW Igor isn't mentioned in the Japanese manual (gasp)! He is just referred to as semushi otoko (hunchback man) just like the normal hunchback enemy. Also several enemies got new names in English:
Bat - Phantom Bat
Medusa - Queen Medusa (in Japanese both Medusa and medusa heads have the same name "Medusa")
Armor - Black Knight
Axe Armor - Axe-Man
White Dragon - Skele-Dragon
Bone Pillar - Dragon Skull Cannon
Frankenstein and Hunchback - Frankenstein and Igor
Also I noticed that in the Japanese 1993 cartridge manual the hunchback seems to have changed name to nomi otoko (flea-man). That explains why they are called Flea-Man in some later instalments of the series. Maybe this is widely known but it was news to me.
rainwarrior wrote:
I thought you had pre-emptively rejected any and all official statements?
DRW wrote:
Ask these people and all they'll tell you is bullshit.
My statement wasn't supposed to be a dogma, but my experience with answers to those kinds of questions. This doesn't exclude the possibility that someone of them might give a convincing explanation.
For example, if they say:
"We couldn't name him Dracula because the name was protected under copyright law"then I know that they're talking bullshit:
Oh really? So, when "Castlevania II" came out, you had somehow aquired the license? Please tell me this story as well. Where did you get the license from? With whom did you have to speak to get the license of the name Dracula? Was there an official meeting between Konami USA and the owners of the novel? Hm? You said you omitted the name in the first part because of copyright reasons. This means you must have bought the rights before you localized part 2. No? No recollection of that? Well, then maybe your story was bullshit to begin with.
But if they say:
"We were told to write the manual for the game, but one guy of our team pointed out that Dracula is protected and that Universal Studios owns the name. We weren't really sure if that was true. Obviously, it wasn't. But nobody of us cared enough to fact-check, so we just played it safe and didn't use the name.
The manual for the next game was written by other people and they obviously checked it since they did use the name Dracula. Not to mention the fact that the name appears right in the first game anyway, but that detail went over our heads because we hadn't thoroughly played the game.
For the gameplay stuff, we just summarized what was written in the Japanese manual. We barely knew Japanese, by the way.
For example, in the controls, we made a distinction between:
- "activating" the whip (left, right or down plus B)/"activating" the weapons (up plus B)
- and "cracking" the whip/"firing" the weapons (just B).
Which is of course total nonsense, so you might imagine how much we actually knew about the game"then I would of course believe it.
So then... confirmation bias?
Nope. A confirmation bias can only happen if you don't what's actually true.
But we objectively know for a fact that the novel "Dracula" was in the public domain at the time "Castlevania" was created.
So, whenever some developer just says "It was because of copyright issues", then we objectively know for a fact that this cannot be true. It cannot be true that the copyright holder approached Konami. Because there was no copyright holder.
On the other hand, if they just say "It was because we thought that we may run into copyright issues and didn't check", then this is a totally believable statement. But just saying "We didn't have the rights" is objectively false because everybody has the rights to use Dracula in his works.
So basically you are angry with them because you believe that if you asked, which you haven't, they would probably lie about it?
DRW wrote:
Not to mention the fact that the name appears right in the first game anyway, but that detail went over our heads because we hadn't thoroughly played the game.
For the gameplay stuff, we just summarized what was written in the Japanese manual. We barely knew Japanese, by the way.
For example, in the controls, we made a distinction between:
- "activating" the whip (left, right or down plus B)/"activating" the weapons (up plus B)
- and "cracking" the whip/"firing" the weapons (just B).
Which is of course total nonsense, so you might imagine how much we actually knew about the game"[/i]
They did make changes to the ending, I believe "BELMONDO" was changed to "BELMONT" (why didn't they just go with Belmond? Maybe they overlooked the diacritical marks over the "do"-syllable). And of course they had it converted to UNROM.
Those kind of manual translation errors seems to be quite common. I suspect the people translating them have often never played the games and thus makes their own assumptions of how it works. The Japanese manual simply say:
Whip/Item Operation
Whip... B button + Joystick (Left, Right, Down)
Item... B button + Joystick (Up)
Then it's just says while standing (B button), while crouching (↓ + B button) and while jumping (A + B). I see why it could be confusing for someone that never played the game.
DRW wrote:
My statement wasn't supposed to be a dogma.
I know, I was just teasing.
If I may offer a practical approach to solving your problem:
You're asking a journalistic question. The way to get an answer is through research, e-mails, phone calls, travel, translators, etc. If you care about the answer, this is how to get it for yourself. Complaining to us that we don't know anything won't inspire anybody here to do it for you.
Ah, but obviously this question is too trivial for someone to want to give you their time, isn't it? Well, maybe you should spend some time thinking of enough meaningful things to ask that might make a full interview worthwhile. To have an interesting article or interview, usually you must do a lot of research to find many things that are interesting. Bits of trivia like this turn up in research like that, but you don't get to choose what you find, you just dig until you find enough interesting things. Interviews and articles often have lots of little details like this stuff in them, but it's very random what details actually appear; somebody had to find them. They didn't go looking for the answer to an insignificant question, they went looking for anything they could find, and shared anything that seemed interesting.
It's easy to ask a question about a trivial thing in the middle of a larger interview. It's much harder to demand someone's time simply to answer your insignificant bit of trivia. Even if you find the person responsible, they might not remember, or
even care to answer.
Every now and then, you find someone who's very accessible on twitter or reddit or wherever and happy to answer this kind of crap, but again, it is a very random selection. Only a few people will be like this, and you don't get to pick which. You can't attack a specific question this way unless you get lucky. Similarly, someone looking to get an interview may have to approach a lot of people before they find someone who will give them helpful information. It takes a lot of time and persistence to get this stuff answered.
Pokun wrote:
They did make changes to the ending, I believe "BELMONDO" was changed to "BELMONT" (why didn't they just go with Belmond? Maybe they overlooked the diacritical marks over the "do"-syllable).
"
Belmont" and "
Belmondo" are both real surnames; I assume the former was just considered more accessible to English-speaking audiences, since it's fairly common in North America and the UK, whereas "Belmondo" seems much more common in mainland Europe.
(Then again, when you consider that mainland Europe is where the Castlevania games take place to begin with...)
I see, I thought Belmond sounded like a real surname but maybe it isn't.
DRW wrote:
Also, this might actually be the correct explanation. But what I hate is that you don't hear this when you ask them. Ask anybody at Konami USA and I'm pretty sure they won't tell you: "We were not sure and avoided the name because we didn't want to take any risks. But in the end, yeah, we could have used the name Dracula right from the beginning." Instead, they will tell you: "Yeah, that had to do with copyright" and they're done with it.
Which is why you never have to ask them about any possible speculations (no matter how minor of a hint) when asking, they may not remember clearly and assume that what you said is the correct thing.
Both
Belmont and
Belmond are real French surnames (and are pronounced exactly the same - the final "t" or "d" is never pronounced), and Belmondo is the italian variant but also exists in
France, mostly in south eastern regions close to Italy.
In all case this name has probably nothing to do with Transylvania as the region is populated by Maygars (aka Hungarians), Romanians and Germans. However I might be wrong, Romania and Italia are actually close in language (no joke) so it might happen that some Belmondo families do exist in Romania, but then I'd bet it'd be spelled "Belmondu" or something like that - Romanian names almost always ends in -u.
Interesting. If they wanted the French pronunciation they would have written it as ベルモン (berumon). But they might just have taken a random European name. I don't think the clan's origins are ever mentioned.
Mathias Cronqvist however (without spoiling Lament of Innocence) have a very Swedish sounding name. Cronqvist (or Kronkvist in modern spelling) is a common surname meaning Crown Branch (i.e. the top branch of a tree) in modern Swedish.
Pokun wrote:
ベルモン (berumon)
That's not a Digimon, but
Cerberumon is.
Quote:
Cronqvist (or Kronkvist in modern spelling) is a common surname meaning Crown Branch (i.e. the top branch of a tree) in modern Swedish.
Which meshes nicely with Harry Nyquist helping to establish a "new branch" of science, namely signal processing theory.
Haha yeah.
(For people not getting the joke: ny = new in Swedish.)
I just looked at how the DPCM channel works and apparently, it has 128 amplitude levels but can only go up or down 1 level at a time, meaning that in order to have a somewhat realistic sample, you'd need to filter it at 125 Hz.
At least you can have a very gradual frequency roll off.
DPCM samples actually apply +2 /-2 at a time, so it's basically 6-bit. You only get 7-bits if you're doing raw PCM.
Up or down 2; and you don't usually encode samples to full scale (because that's noticeably louder than the pulse wave channels).
The summator (integrator) in the DPCM playback is effectively a lowpass filter at f=0, so the lower the frequency, the louder it can be represented. Correspondingly, a naïve DPCM encoder will end up just encoding the lower frequencies at the cost of the higher ones. It produces not-unreasonable results if you 1- encode relatively narrowband signals, and 2- encode at a reduced and comparatively constant volume such that the two different sources of noise are balanced for the duration of the sample. (Noise source 1: slew rate overload distortion, which has strong non-linear and enharmonic components. Noise source 2: "white" noise from quantization error)
Has anybody tried compressing audio by performing RLE and LZSS on the slope data?
RLE would help in frames dominated by slew rate distortion, not so much in frames dominated by dither noise. LZSS helps only with patterns that repeat on a byte boundary. Very few mappers support RAM at $C000-$FFFF anyway, which would be needed to play decompressed waves without taking 100% of CPU writing to $4011. And if you're decoding to $4011, you might as well use QuaDPCM which offers higher quality per bit than 1-bit delta modulation.
I meant for any system in general.
psycopathicteen wrote:
I meant for any system in general.
There are plenty of cases of games with generically compressed packfiles that included WAV or other raw PCM data, but it typically doesn't offer much compression of PCM at all (ZIP up a WAV and see how ineffective it is). It's just not a very good way of compressing PCM.
For example, there are many MAME ROMs that are ZIP files containing ROM data, some of which happens to be raw PCM data. It gets used just because they apply the same generic compression to everything, but it doesn't mean that it's doing it effectively.
I've actually seen RLE used to trivially compress silent segments of PCM data. I suppose it could do some good on samples that are heavily clipped and with much low frequency content (before clipping), but again it's just not something that's well suited to audio in the general case.
Has anybody tried 1 bit brr/adpcm?
Most NES games that have samples already use 1-bit DPCM with a fixed step size: 1/63 of full scale.
Or by "1 bit brr/adpcm" do you mean
continuously variable slope delta modulation (CVSD)? That adjusts the step size by the length of a run of 0 or 1 bits.
I meant like what the SNES has only instead of 4 bit samples, you have 1 bit samples.
Why is doing DPCM so friggin convoluted?
Sogona wrote:
Why is doing DPCM so friggin convoluted?
It's not. I'm just curious about other crazy ways to compress audio. I want to know how much redbook music could you cram into a commercial game from the 8-bit to 16-bit era.
psycopathicteen wrote:
Sogona wrote:
Why is doing DPCM so friggin convoluted?
It's not. I'm just curious about other crazy ways to compress audio. I want to know how much redbook music could you cram into a commercial game from the 8-bit to 16-bit era.
Did you try
my "Max 300" demo for NES?