Why nobody made pixel game with orchestral score?

Or "hyperrealistic" 3D game with only chiptunes and bleeps for music/sound. Personally would be interested in playing second outcome.

But why no one has made it?

 

It doesn't match with the expectation of the player. It would be irritating at first, but of course it can work.

 

There is a match 3 game out now that has a score done by some large symphony orchestra.

Also, one of the unlocks in Lego Jurassic World allows you to play with a chiptune score. It's neat at first but gets grating.

 

Possibly because some people would see chiptunes in a 3D game as a mismatch and amateurish.

Probably because most people making pixel games these days really have no clue that many classic 2D pixel art games did not have chiptunes (like they probably don't know what an Amiga or Neo Geo is and never actually played coin-op arcade games). I do believe a lot of the people making this stuff don't really even understand what they are making exactly. So they create a sort of bland facsimile of their impression (aka "best guess") of the games.

 

I don't think it's that uncommon:

As long as the tone doesn't clash, it's fine.

 

I thought he meant as a standard practice. There will always be some exceptions. Actually I am surprised the retro pixel art rpg games (assuming there are some and with all the games being made I am sure there must be) don't have orchestral music. There were some pretty impressive non-chiptune-sounding tunes in some of those even when they really were chiptunes back in the 16-bit days.

 

He said "But why no one has made it?". I took it as he's never seen it happen.

I wouldn't call it "an exception" though. I've seen other games do it too. It's just not as common as the pixel art+chiptune combo.

 

Good point! I agree with you. Of course it has been done. There are just so many games of some will certainly have what he is describing!

 

You haven't played Pond Wars yet? I found some free orchestral music on the asset store. But I couldn't find free boat images.

 

I'm going for a "realistic" retro pixel game and I'm leaning towards doing an orchestral/synth hybrid soundtrack. But I'm postponing doing sound and music till the game is done so that I can better judge at that point what would seem fitting.

 

I think orchestral music in games and movies usually sounds kind of generic and boring. But I guess it is what people expect, like piano music in silent movies.

 

I'm fairly sure it has been used in couple of titles I played, although I can't exactly remember which ones. You most likely will hear chiptune in 3d if it is hacker-related or game-related stuff.

As for orchestral music in sprite games...

Check out Beyond Oasis for sega genesis. It is still played by game chip, but it is pretty much proper orchestral music.

Square Enix pretty much always used high grade music in their titles (Final Fantasy 6, for example).

Keep in mind that recording orchestra is gonna cost you. That's one of the reasons why it isn't used often.

 

In my opinion the Lego games are using this kind of effect as well. Having a huge orchestral track with Lego characters gives a unique feel. Of course it is not exactly the same as with pixel art, but the effect that can be achieved with it, may be similar.

 

Recording an orchestra is really expensive, but for ~1k-3k and upwards (only software, not including the time to learn composing music) you can get pretty good results with sample libraries. Imho most people are not able to tell the difference and for game soundtracks I'd say most on the market are done with sample libraries. Or you hire a composer that works with his own software.

 

I could live with everything having some essence of Rock music to it. In fact, this reminds me I need to track down some free Christmas rock music or hire someone to make rock versions of traditional Christmas tunes.

 

I've never heard a single software synth that could properly do grand piano with sustain pedal pressed. Live music instruments resonate with each other, synths usually omit this effect. Some "instruments" (like choir) work best when recorded.

But, as you said, most people won't notice and synth/sample based music has its own advantages. You'll have hard time performing most of Akira Yamaoka's music live, for example.

 

Probably, I have to shamefully admit I never watched that movie ^^.
You can check out some of the recent native instruments libraries if you want:

http://www.native-instruments.com/en/products/komplete/orchestral-cinematic/action-strings/

http://www.native-instruments.com/e...al-cinematic/symphony-series-string-ensemble/

I don't know about the second one but Action Strings follows an interesting concept that gives a good ratio of realism and flexibility imho. They've recorded an ensemble playing different melodies or rythms on the same note across a partial range of the note spectrum and when you play it with a keyboard or sequence it in your DAW you can change the patterns and the root note of the patterns and the sampler blends between the recordings in the different pitches. That way you can avoid that dreaded "machine gun effect" that often is a dead giveaway of sampled music if you have a lot of repetitions of the same note without any other notes in between.

I don't know how the second library was made but it sounds great to me and I assume they also work with prerecorded phrases to some extend. I have the feeling that NI go for better sound quality, realism and ease of use at the cost of flexibility and uniqueness. With libraries like Damage or Action Strikes you can basically hold down one key on the keyboard and epic cinematic drum loops come out out the speaker, zero skill required. It's a bit like using art assets but less recognizable for the end-user if done right.



Oh and by the way, I have a semi related question that I didn't want to open a new thread for if it's not necessary: how possible and practical is it to have music run in a separate audio bus and apply realtime convolution reverb to it? I was wondering if there are good ways to piece together orchestral music at runtime from different prerecorded audio files. applying reverb to the mix might help make it sound better and save memory in the audio assets because reverb tails can be kept shorter.

 

Akira Yamaoka is Silent Hill series composer.

Using those kinds of industrial sounds in live orchestra is problematic.

Here's how you can quickly test piano synth. That's quick and dirty test.

Find a real piano, press and hold right pedal (sustain), then place elbow on keyboard pressing multiple keys at once. Now listen to the sound and remember it. Or just play a chord.

If you find any synth that can do the same kind of sound, I'll be very surprised.
With right pedal down piano strings sing in response to any sound that reaches piano, that includes sound produced by piano itself.
That results in very rich sound. Basically, every piano string has about 17 frequencies (well, that's the ones you can actually hear by ear) it can resonate with and with few 88 keys and hundred strings in it, you have a system that is going to be computationally expensive to simulate properly.

Strings, percussion and instruments like accordion would be easier to simulate, although you'll have very hard time replicating unique instrument-specific playing techniques using synth. Still, if you don't have an orchestra available, that can work as cheap replacement.

There are decent synths that can replicate standard piano sound WITHOUT sustain pedal, but once you press the pedal, illusion shatters.

I wouldn't expect good results when echoes are automatically applied to live recording.

Either way, you may want to look at FMod studio and ELIAS software for sound-related ideas.

 

This is the best I could come up with on short notice:
http://www.keinebilder.de/temp/pianoresonancetest.wav

What do you think?

Thanks for the link, I'll check it out!

The Silent Hill music is interesting, but a little hard to listen too for me. Might be because I never played the games. About the most dissonant that I occasionally listen to are the soundtracks to Dead Space 1 and 2 by Jason Graves. If you are into dissonant horror game music I highly recommend them.

It's actually quite common in sounddesign to record things with a close microphone position to get a dry sound without much reverb and a position further away that captures more of the reverb. Many Sound libraries offer "dry/close" and "stage" recording as a choice. When you combine different libraries it can make sense to use them dry and then use impulse responses from real locations to virtually put the different recordings "into the same room" by simulating how sound in that room would be changed. The quality of convolution reverbs that use impulse responses from sampled real locations is quite good imho.

 

I'm 95% certain it is synthesizer.

For example (not the best examples, just what I could find quickly):

(see 0:45 in this one)

If you press too many keys at once (with sustain pedal on), real piano will roar.
Pressing low keys with sustain will result in very rich sound, because the rest of the keyboard will resonate. That doesn't happen with synths.

 

Guess I can't fool you then . I don't have access to a real piano and I couldn't play it anyway. I don't think I would be able to tell if this was real or fake. I am sure though that, if this was pure synthesis, we both would be 100% sure it's fake. This was is a sample based instrument that uses ~13gb of prerecorded sounds from a real piano (Native Instruments "The Maverick" if anyone is curious). If any synthesis at all is going on it'll be a very very minor contributor to the final sound. Synthesis and Sampling are fundamentally different and I don't think any of the high end virtual piano solutions is 100% synth based because it just doesn't get nearly close enough to the real sound. They did sample overtones and what they call "sympathetic string resonance" on this one, but it is very possible that there is a fundamental difference of the sound that the strings produce if different keys are hit at the same time compared to playing the resonance samples from only those individual keys being pressed at the same time. That might be a level of behaviour that could no longer be sampled because the different possible combinations are too many.