Mac D1 sounds contain clicks

[Yarn]

no sarcasm im serious ~ i upscaled the D1 sounds to 22k 8bit mono and they sound much better. upscaling to 44k 8bit mono is just padding, it sounds the same as 22k. this upscaling sounds much better than SDL_mixer's upscaling.

If you're using Windows Sound Recorder for upscaling, then understand that (as far as I remember) that program uses "nearest neighbor" scaling, probably the most primitive kind that exists. It's not actually adding detail, even though it may sound like it is. SDL_mixer most likely uses a decent upscaler that doesn't add that false "detail."

you do notice that D2 sounds better than D1 right?

Yes, but most (I think) of those sounds (the ones also in D1) either were remastered (using more sophisticated methods than simple upscaling), were re-recorded, or existed in a higher quality in the first place. These three methods, especially the latter two, produce better results than the upscaling that I think you're describing.

[ThugsRook]

ive done an extreme amount of testing on this ~ D1 (files) can sound better, easily as good as D2's.

you overestimate Parallax, their upscaled files sound no better than mine. (which is still damn good compared to D1's 11k)

[Yarn]

ive done an extreme amount of testing on this ~ D1 (files) can sound better, easily as good as D2's.

Can you provide a few of your upscaled files?

you overestimate Parallax, their upscaled files sound no better than mine. (which is still damn good compared to D1's 11k)

Actually, upon examination, many of those files are obviously not "upscaled"; rather, they most likely came from the higher-quality masters that the D1 sound files were also made from.

You know how a Blu-ray version of an older movie still looks better than the DVD version, despite the fact that HDTV and Blu-ray didn't exist back then? It's the same deal there: The Blu-ray version is made from the higher-quality master; it is not upscaled from the DVD version. (The effective resolution of film is quite high.)

[ThugsRook]

what i did was i ripped a wav file out of D1. (11k)
made a copy of it and upscaled it to 22k 8bit mono.
then i ripped the same file out of D2. (22k)
...and compared them. (on a raw player, no eq no fx)
theres a clear difference between the D1's 11k file and D2's 22k file. my upscaled file sounded as good as D2's. i couldnt tell a difference.

i think the "master copy" was 22k. they had no forseeable need for anything higher.

D1 can sound better. i think making D1 up to par with D2 is something we should be doing. no?
a simple 22k (or 44k) 8bit mono pack and Zico flips the enable switch for us (the 44k files are no bigger, they are all padding, but it might be easier for Zico) your current SFX pack for MAC D1 doesnt even need SDL_mixer to run.


ill get you those files if you cant make them yourself. (yes i used plain old sndrec32)

[Yarn]

ill get you those files if you cant make them yourself. (yes i used plain old sndrec32)

Yes, I would like some of them, especially the countdown ones. I'm running Windows 7, so I no longer have that version of Sound Recorder. (The newer one can't open files or do any kind of processing, and it only saves WMA files.) I'd like to examine how it really upscales files.

[ThugsRook]

wav test files: 3 versions of each file
'self destruct sequence activated'
'cheater'

http://www.mediafire.com/download/6d68f7...avtest.zip

[Yarn]

Yup, just what I thought: nearest-neighbor scaling, which, again, is the absolute most primitive form of scaling. The result really isn't as good as having true higher-resolution material. For example, in the "self-destruct" message, the "s" sounds are clearer in the native 22k version than in your upscaled version. (You probably need something better than laptop speakers to hear the difference, though. Those things almost always suck.)

If an audio pack consisting entirely of these upscaled files were made, calling it "higher resolution" (or anything along that line) would be a complete lie. At least that's my opinion. I'm sure that audio experts worth their salt would agree with me on this. Perhaps you don't.

[ThugsRook]

as you mentioned sndrec32 is the lowest form of upscaling. im not asking to use my files, im asking to use yours

90% of the files we'd need are already 22k in D2.

[jihnsius]

I don't want to say it's impossible to upscale and achieve better results because I haven't done enough research on it to give you a definitive answer, but from my audio engineering experience I can tell you it's theoretically impossible and any effects are more than likely coincidental or placebo.

[ThugsRook]

^ kinda like the SDL_mixer? 

as i said most of the 22k files are already at our disposal, in D2.