Page 3 of 4

Re: 2010 Roadmap

Posted: 09-19-10 3:49 am
by Arbpoca
I would like to add that I, too, would give my extra two stars to manual note fingering if I could.

I've played at least 50 hours of (Learning Pack) Synthesia in the last month alone, and inability to remember correct fingerings is one of the only things keeping me from playing many pieces perfectly. I'd compare it to driving in an unfamiliar place without street signs. It's usually not so bad when the piece has only two voices, but playing a fugue feels like driving in DC.

Re: 2010 Roadmap

Posted: 09-19-10 4:37 am
by Nicholas
Arbpoca wrote:I've played at least 50 hours of (Learning Pack) Synthesia in the last month alone...
!!! :shock:

Re: 2010 Roadmap

Posted: 09-19-10 7:03 am
by Arbpoca
Welllll, now that I think about it, I played piano for at least 50 hours this month. However, Synthesia was at least 30 of that.

Re: 2010 Roadmap

Posted: 09-19-10 4:49 pm
by Nicholas
That's still a lot. I wish I had the time to do that. That would be awesome.

Re: 2010 Roadmap

Posted: 09-20-10 11:56 am
by aria1121
Nicholas wrote:
Arbpoca wrote:I've played at least 50 hours of (Learning Pack) Synthesia in the last month alone...
!!! :shock:

Re: 2010 Roadmap

Posted: 09-20-10 4:57 pm
by vicentefer31
Manual Track Split: 96 Stars
Song Library Feeds: 20 Stars
Live Video Feed: 31 Stars
Track Volume Control: 75 Stars

These four features have 222 Stars

Sheet-only Song Display : 360 stars
Manual Note Fingerings :331 stars
Chord Library: 313 stars

Re: 2010 Roadmap

Posted: 09-20-10 5:44 pm
by Nicholas
Stars are also just a helpful community suggestion. I'm happy with the direction things are going. Hopefully you guys will be too, once we get there.

And in the end, the stars don't even matter. All of those things are going to be finished eventually. We're on a journey... and that journey doesn't end in December.

Re: 2010 Roadmap

Posted: 10-10-10 4:45 pm
by vicentefer31
My best wishes for Online Scoreboard.

Re: 2010 Roadmap

Posted: 10-21-10 4:42 pm
by vicentefer31
In 2010 Roadmap there is a very useless feature: Manual Track Split. The idea about this feature is that some midis are composed of a single track that contains both the left and right hand parts and with a feature like we could split it.
Ok, this sound very good, but it's just an utopia. Take any midi you have with two tracks (one for left hand and the other for right) and choose the same colour for both tracks for the falling notes; go to the play window, do you really think you could split it correctly if you had the "Manual Track Split"?
Also, Synthesia is not an MIDI editor and this is a tool for a Midi Editor. So, this feature is not for Synthesia.

Re: 2010 Roadmap

Posted: 10-21-10 6:04 pm
by Nicholas
vicentefer31 wrote:... there is a very useless feature: Manual Track Split.
It's also become the very useless feature I receive the most email about. Not only does it make Synthesia more turn-key and not require outside software, but something that I have to keep in mind (constantly!) is that the larger I want to scale the distribution and use of Synthesia, the more it has to stand on its own.

When a new user downloads the game, there isn't suddenly a new Nicholas that appears ready to answer their questions via the forums or email. If the game doesn't solve problems for the user, I simply continue to spread myself thinner until I don't have a chance to interact with anyone on a meaningful level. (To some extent, the 250+ emails in my inbox sort of prove this has already happened. :( )

You could argue changing instruments was a MIDI editing feature as well. It is trivial to change them in a separate editor. And yet, I continue to get thank-you email every couple days from users upgrading from older versions where they're now able to do it.

In general, you forum users have a much greater expertise dealing with outside programs and MIDI. The average Synthesia user (that I hear from in email) is excited to learn they can connect a keyboard to their computer! They need solutions that work in a couple clicks. Not a couple downloads followed by a couple steep editor learning curves.

Also, outside of getting one or two notes wrong when they're close together, I think splitting accurately along the boundary isn't that difficult. It is certainly a much easier task than throwing someone manual note fingering and hoping they can guess the right numbers to assign.

Re: 2010 Roadmap

Posted: 10-21-10 9:19 pm
by Pianotehead
Yes, I suppose you have to scale down or define the program, what it is and what it should be able to do. I myself would like to see the sheet feature improved and being able to enter finger numbers. MIDI-editing or related features are less important. Assume by doing too much, you can easily break the $25 barrier!

I use Synthesia with Notation Composer and they work very well together. The first is useful in seeing how the song looks on the piano and which reduntant notes I can remove, for example shrinking a five note chord to three or four notes, so it can be played by one hand. Then I remove the notes in Composer, look again in Synthesia and the process repeated as necessary.

Re: 2010 Roadmap

Posted: 10-27-10 3:18 pm
by jimhenry
Brand new to Synthesia, so please forgive whatever follows that has already been discussed and decided. Nicholas, I think you are absolutely on the right track when you guess that the introduction of a keyboard in Rock Band 3 is going to create a big crop of new keyboard players. I am the creator of a free virtual theatre organ, the Miditzer at In the 8 or so years that I have been distributing that program, one constant challenge has been how do we create more keyboard players so that we have more potential theatre organists. (Theatre organs are the fun pipe organ cousins of church pipe organs that were put into theatres in the 1920s to accompany silent films and do other entertainment duties. I found Synthesia because I was looking for ways of building on the expected interest in keyboards by Rock Band 3 players.

My suggestions for increasing the appeal to RB3ers based on my first few hours with Synthesia are as follows:
  • First and foremost, view Synthesia as a content platform. I realize that every MIDI file is a potential piece of content for Synthesia. However, the RB3er is not going to want to hunt around for content initially. They are going to want to see specific songs they are interested in with the selection information available at a glance. The RB song finder screen should be the model for this. If you want to pull in the RB3ers, you need to make your web site a place that will feel familiar. I know you have already seen the effect web site design can have on the acceptance of Synthesia.
    RB song finder.png
    RB song finder.png (150.95 KiB) Viewed 11420 times
  • Plan for a mechanism that will support providing paid content. We all know that really good content will drive the use of Synthesia and good content is hard to make. Plus, [ugh]there are license fees that will need to be paid for current copyrighted content[/ugh]. I know that there are very messy rights issues involved with providing content. (IAAL, no N, I am a lawyer, in the area of intellectual property no less, so I really know how messy it is.) But you are going to have to wrestle with the rights demons and provide good content in an easily accessible form if you intend to win the hearts and minds (and ears and fingers) of the RB3ers.
  • I slipped good content in the above at every chance. Now here it is with its own bullet item. What has made Guitar Hero (a trademark of Activision) and Rock Band successful is that they deliver a fun experience. They make you feel like a Rock Star and that sells almost as well as some the kind of stuff we've agreed not to discuss here. Rock Band has now pushed this to the point where they can teach you how to play a real instrument to some degree without losing the fun. Synthesia is an opportunity to take the "teach you how to play a real instrument" part of that further for the keyboard player. That is possible because Synthesia has the whole screen just for the keyboard and the program isn't limited to the 25 keys of an RB3 keyboard. Synthesia already has the dropping notes on lanes that are familiar to the RB3er. It lacks the animated eye candy backgrounds but I think they can be left out. I would like to see the opportunity to provide custom still images as backgrounds but I view that as a low priority. What is sort of lacking for Synthesia is graded content that makes the keyboard player part of a band (or orchestra or combo or duet). Yes, I know that you can do this with the right choice of available MIDI files but this needs to be readily available right off the bat. A big part of the success of the first Guitar Hero game was that you were playing with other (better) musicians. Even if you weren't good, you had other musicians (non-complaining computerized ones, the best kind) who would carry you so that it was still fun to play. It just got more fun as you got better at playing. Even though I sort of know how to play a keyboard, I needed to start with the 0.0 level tunes. Playing "A Tisket A Tasket" is pretty grim as a piano solo. Really grim if you are bad, not much better even if you are really good. Then, just to see how Synthesia handled complex MIDI files, I tried the Overture from Phantom of the Opera. I selected the easiest part I could find in that arrangement. It was a real kick to play that even though I was really, really bad at it. My part was small and not that easy to hear. Even though my performance would have had me thrown off the stage, it was fun because all the other stuff still sounded good. So, to get to the point (a little late I guess) Synthesia needs content that sounds good and provides an easy part that a beginner can't do too much damage with so they can learn and have fun at the same time. That has always been the missing link in music education. No one HAS to learn to play an instrument. It takes real dedication to stick to it if it isn't fun. Rock Band makes the process fun and, hopefully, will be able to teach a bit of musical performance along with the fun. I think Synthesia's mission is to teach a bit more musical performance without losing the fun of Rock Band. And I think the content is a huge part of that.
Hope this is helpful, perhaps to spark some better ideas if nothing else. Anyway, thanks for your patience if you are still reading this.

Re: 2010 Roadmap

Posted: 10-28-10 2:38 am
by Nicholas
So, first of all, great first post. ;)

Going bullet-by-bullet:
  • Synthesia as a content platform: this has been coalescing (very) slowly. The "Song Library Feeds" item (that is deceptively low) on the voting list will be a great first step in that direction. The gist of it is that you'll be able to add a URL to the song library instead of a local folder. That URL points to something very much like an RSS-feed for songs. That lets anyone be a content provider and I get to just aggregate a few of the best ones as suggestions.

    The other real strong content platform feature (not on the voting list) will be synthesia:// links. This is much like torrent programs register a URI handler for "magnet:" or AOL's old instant messaging program registered "aim:" as a protocol. A more recent example are iTunes App links. If you have iTunes installed, clicking one in your browser opens iTunes right to that App's page.

    With Synthesia installed, you'll be able to click a synthesia: link on a site (that may or may not look very similar to the song list image you posted) and it will open Synthesia immediately with that song already loaded. With any luck, you'll be able to (optionally) embed track and other mode/settings right into the URL so it really can be one-click playing. An example that isn't just a straight content list might be a virtual piano lesson. There might be some block of text, maybe a link to a YouTube video, and then a "now try it yourself" link, first for just one hand and then another link (to the same song) for both hands. I'm very excited about this one.
  • Paid content: Man, those messy rights have me scared away. This feels like a really strong direction... it's just incredibly challenging. Also, unlike the expertise it sounds like you have, I have literally zero experience with any of that. So, I'm likely even more sheepish because of it. (By the way, I found what "IAAL" meant -- it helps you also said it just afterward -- but what does "no N" mean? I couldn't find that anywhere.)
  • (Lots of good insight I agree with ;) and) better graded-difficulty content: Yes. At some point I'm going to have to out-source this. There is no way I have the bandwidth (or ability!) to do it, but a great set of pieces -- original, to avoid many of the licensing/copyright issues? -- would be a huge boon.

    This next bit can become a pretty heated argument, but gating that content and requiring users to unlock it through progression (like the progressively-harder set lists in GH/RB) is something I'm thinking about. It gives players a really great incentive to stick around. If I were to jump into a list of 150+ songs like Synthesia has now as a brand new user, there are a few things that might happen:
    1. I might become totally overwhelmed.
    2. I might have the same experience as you and play Three Blind Mice or whatever and just stop there.
    3. I might recognize some cooler (read: crazy hard for a first-time player) piece and get totally discouraged.
    Gated content solves all three of those problems. OK, maybe not the second one, but good content does.
I totally read to the end. ;) Sorry for the wait on my response. I wanted to have enough time to give it the proper response it deserved.

Re: 2010 Roadmap

Posted: 10-28-10 5:08 am
by jimhenry
Starting with the easy stuff, IANAL (I am not a lawyer) is sometimes tacked onto opinions about legal matters expressed by non-lawyers. I probably notice that stuff more being a lawyer. I've never seen IAAL, probably because lawyers are usually smart enough not to express opinions about legal matters, at least not without being paid.

Now onto to the hard stuff, dealing with rights. My comments are based on the laws in the United States. This can be different in other countries. Figuring out what law applies can be an undertaking. However, if you are completely US based, it is fairly safe to concentrate on complying with US law.

Anything first published before 1923 is in the public domain. So anyone can base a MIDI arrangement on a pre-1923 publication and the only rights at issue are those of the arranger. Most of what is worthwhile music in this category is the classical library, but there is some popular music and that plus classical works might provide some raw material for starting to build a library of MIDI files specifically aimed toward Synthesia.

The approach I would take to getting current popular music in the library is to create an infrastructure capable of delivering paid content and build up a customer base using whatever early content you can scrounge up. Once you have a viable market channel you could approach vendors who are already selling licensed MIDI backing tracks. The going price for such things seems to be $5-10 per song. My hope is that you could present a market that could sell in sufficient volume and with sufficient limits on the use of the files sold that these vendors might consider selling a Synthesia file for something like $1. It's a chicken and egg problem. You have to build a hen house and hope that if you build it, they will come. (Field of Dreams mixed metaphor)

I think you can copy most of the ideas for creating a library of DLC from Rock Band Network, particularly their legal agreements: and If you decide to go this route, I can help you find your way through the legal jungle.

Your ideas for seamless delivery of on-line content sound exciting. I would humbly suggest that you not let the technical whiz bang get ahead of the business issues. I think figuring out how to pay content providers will prove to be the most important part of delivering good content. If you can make Synthesia a convenient storefront for the advertising, monetization, and delivery of the content so much the better. Just don't put yourself in the position of having a great delivery mechanism except that there is no way to charge for what you are delivering. You are getting into some ideas that seem to go beyond what RB and GH have done so far. If you come up with something really effective, selling the methodology to RB and/or GH might be a possibility. Do know that Harmonix is ambivalent about DLC as a profit center. So far, they haven't been able to sell it in sufficient quantities to make the DLC library an iTunes type success. Yes it makes money and it sells games. But it doesn't make enough money that they aren't cautious about the amount of money they spend creating DLC. For example, they reportedly will not be preparing additional Beatles DLC because of the cost. It's a bit scary to think that Beatles DLC wouldn't be a surefire money maker. A mechanism that doubles the amount of DLC sold (analogous to your web site redesign) would be a game changer for them.

Re: 2010 Roadmap

Posted: 10-28-10 12:38 pm
by jimhenry
After thinking about your idea of linking to URLs in addition to files, I do think this may provide an opportunity to jump start a DLC mechanism. What I am visualizing is a sortable, searchable database of MIDI songs that includes information like difficulty and user ratings. I am pretty sure that what makes a "good" MIDI file for use in Synthesia is different from what makes a "good" MIDI file for listening. A big problem for a Synthesia user is sorting the wheat from the chaff amongst the many thousands of MIDI files that are available.

Pulling in another idea that has been discussed, perhaps the Synthesia community could create metafiles that are associated with the listed content. The metafile would include additional information for adapting the MIDI file to Synthesia use. This would include things like fingering and chord symbols. It could also include things like track assignments to make the playing and back track seperations.

Re: 2010 Roadmap

Posted: 11-09-10 7:57 pm
by jimhenry
Nicholas are the goals for the remainder of 2010. Some of these are things that aren't on the voting list and some I've never even mentioned before. Of course, all of these are subject to change...

Absolutely must make it in before 2010 is over!
  • A simplified way to choose track settings in the one or two piano track case.
  • Online scoreboard.
  • Some specific visualization improvements and a couple new special effects.
  • 0.7.4: Video mode, resolution, and windowing options. These are standard with EVERY game except Synthesia at the moment. ;)
With Online scoreboard well on its way to being checked off, are simplified track settings and visualization improvements now the final musts for 2010? Do you know what you want to do or are you still sifting through possibilities?

Re: 2010 Roadmap

Posted: 11-09-10 8:00 pm
by Nicholas
I hope so. The simplified track settings should be easy at this point. The online scoreboard was really the elephant in the room... and while there is a tiny bit more polish left for it, the rest of the really important ones for RB3 users should be easy to clean up.

The last six months have certainly been a journey!

Re: 2010 Roadmap

Posted: 11-30-10 11:23 pm
by Kasper

What about this?
For example, In the beginning of the song and every time you pause you get these options.

If you could make even some shortcuts, like I donĀ“t know:
[ is left hand
] is both
\ is right hand
- is wacht and listen

than this would save really much time.

By the way, unbelievable what you have accomplished these six months.
With this online scoreboard it is even fun to play this game with people who don't really play piano.
I mean everybody can play one hand of most songs 0.0 and 0.1 in ten minutes.

Re: 2010 Roadmap

Posted: 12-01-10 4:45 pm
by DC64
you should show the full picture instead of the half left side of the "played by you L, R, or both" image

Re: 2010 Roadmap

Posted: 12-05-10 2:45 pm
by Nicholas
DC64 wrote:you should show the full picture...
Was your browser sized a little small? It looks like the whole picture is there to me.

Regarding the suggestion, changing track settings right there on the play screen is still a little tricky in the code. Unless you meant that was just a choice you would get to make before the song started only, and then after that you couldn't change it again?