How Long Does Other DAWs Take To Save 20,000 Midi Items?

[bazwillrun]

he obviously is severely lacking in the brains department....make music !? hahahaha...does he shit, spends all his time pissing about "auditioning" midi ?...hes a complete tool, look at his style of writing these posts...at best a troll at worst a nightmare for the care in the community people....

[Guenon]

Regarding the 20,000 colours analogy, sure a painter might only need 8-12 primary colours and the painter can combine those colours to get 20,000 colours or whatever the maximum combination is but if you take away one of the colours and only have 11 colours to work with then suddenly the painter don't have access to the maximum combinations anymore.

If you reduce that 12 colours to one colour, let's say black, the painter is reduced to making sketches? That may be okay to some people but I'd rather have access to the full spectrum. Sure, 20,000 midi drum patterns is nowhere near the full spectrum but I'd rather have that than say 400 midi drum patterns.

You are confusing a pallette of colours/tones/flavors with pre-generated semantics/structure/form. Those 20,000 MIDI drum clips contain structural information, they are actual rhythm patterns to feed into which ever drum sound source(s) you are using. Instead of comparing this to a painter using a "spectrum of 20,000 colours", it's like a painter having stored 20,000 random forms of trees and bushes for example, beforehand, independent of the rest of the hypothetical landscape. Randomly generating different patterns and forms for all the pieces of a landscape might be a thing, but it's very different from a painter just using the full spectrum of available colours to, you know, paint. You are comparing your 20,000 random drum patterns to the latter, when you are actually trying to accomplish the former. I've seen you deal with this before, elsewhere, and even though many of the comments you are receiving might sound harsh and inappropriate, I think you might also benefit from honestly assessing why so many people are telling you there's something strange in the way you are approaching and describing this

[harryupbabble]

It depends what your definition of "music skill" is. I think selecting by ear what to keep and what to throw out is a musical skill. Sure, it might not be an instrumental skill but it's a listening skill? Okay so I never automatically listen to everybody, but still.

Also as stated before, music is subjective. If I were to put up 100 random drum patterns for people to listen to... what's the chance that all people will say "Random drum pattern #12 is what I would keep"?

That's why everybody's music collection differs? My music collection, for example, is probably 60 percent British, 30 percent American, and 10 percent other. Now if your music collection is 100 percent Swedish or 100 percent New Yorker then... I don't know.

Regarding speed scrabble, I really enjoy doing that but it's not productive, nothing is made. I only do it to trigger adrenaline rushes. Plus no one earns a living playing speed scrabble. Some people do earn a living winning slow scrabble championships but I think the money made is equivalent to what a minimum wage worker earns, or even less. So sad.

Music-making on the other hand, to me, is although a lot less exciting compared to playing speed scrabble, music-making is still fun and when a song is completed I get that feeling of having been productive. It's a great feeling even if the songs made is totally sucky compared to the music in my music collection. But my music collection has Pink Floyd in it. And the Cure. And David Bowie. And the Beatles. And Led Zeppelin. And Sex Pistols. And...

And compared to other scrabble players who also make music... it's sometimes less sucky.

Sometimes I think "Why am I bothering trying to write songs for? I am only really only trying to be as good as my musical heroes and if one day I equal them, so what, they've already done it". And what if all the musicians in my musical collection are songwriting freaks and I'm not? Then it's even more futile?

And what if one day I'm like "Dude, listen to this great song I made". And dude says "It sounds "Just Like Heaven".

But having stated the above, I'm still off to audition some more random drum patterns. What's the chance that any of those 20,000 drum pattern sounds like a Cure drum pattern? Zero chance I hope. Okay that's that then, goodnight humans. And vurt. Hahaha. I'm Just kidding.

[Hink]

please leave the personal attacks out of it, if you dont agree with or understand his approach so be it but there's no reason to attack the person's intelligence.

[yellowmix]

Yes, as far as I know, batch midi export in REAPER could only be done from the Project Bay.
On my computer, I do that by clicking the "Media Items" tab in the Project Bay, then selecting all the midi items, then I press the Ctrl + Alt keys and drag and drop the midi items to the chosen folder.

I batch import the 20,000 midi files in to one track in REAPER. Batch importing them in to 20,000 tracks is much much much more harder on the CPU, and probably harder to edit.

The "File > Export project MIDI" function won't do because the 20,000 midi items gets saved as one midi file. I want the 20,000 midi items to be saved as 20,000 midi files so that later I can just drag and drop them one by one in to REAPER to audition them.

Okay, yes, it takes time because it is rendering each MIDI file. It's not the file write time, it is transforming what seems like a simple bunch of notes into an individual MIDI file, which can be complex. Think about batch audio rendering, it's the same deal; it's not necessarily the file write time that makes it slow, it's the CPU necessary to process everything.

With that said, I'm not sure what other DAW allows an equally easy batch MIDI import and export at all. Cubase, you need a track for each file, from what I understand. I don't have experience with other DAWs but TBH, I think you should abandon trying to use a DAW, and look for MIDI file utilities to programmatically make the changes for you. Something like http://www.gnmidi.com/

Or use this: http://www.goodeveca.net/Midifile_utils.html - basically the ones that convert to and from a text file. That way you can use a text editor to manipulate it. You can use search and replace or more advanced stuff like regular expressions.

[EvilDragon]

There's one more thing you could do, instead of using project bay.

There's a preference to load MIDI files as references, instead of in-project objects (Preferences->Media->MIDI ->Import existing MIDI files as (*) .MID file reference).

This means that the MIDI files that you load are edited "live", so to speak. Well, not 100% live, but you get save and revert buttons in MIDI editor toolbar, so that you can explicitly save your edits to the MIDI file that is already on the hard drive. This means no batch exporting of 20000 files from project bay.

Then you need to make a custom action that looks like this:

* File: Save file (MIDI file mode only)
* Activate next visible MIDI item

And after that you can make a quick Lua script that will run THAT custom action 20000 times.

It should probably work faster than render from project bay...

[jancivil]

please leave the personal attacks out of it, if you dont agree with or understand his approach so be it but there's no reason to attack the person's intelligence.

Evidently he's clever enough for speed scrabble.
I wonder how in the world you come away with anyone failing to understand the approach. It's 20,000 one bar MIDI items downloaded off the net, and the best he's done is the <average the kit pieces velocity> bit, and just go with that result randomly rather than "approach" drum parts as music in any way.

AND he's using a lot of space with this. Some people are enabling this behavior. If that's what he wants to do, so be it but as an idea of what to do, he's opened the idea up for discussion. It seems like this is not the first time we've seen the 20,000 one bar clips as a topic here, but I could just be losing it.

[Guenon]

It depends what your definition of "music skill" is.

If that was a direct answer to what I wrote, just in case: no it doesn't . I didn't touch the subject of having "music skill" at all, or tell you that you have it or don't have it. By all means experiment and see where it gets you. Just be open to considering what might be funny in the way you are rationalizing this.

My comment was pointing out that the comparison you are drawing between pre-generated structures (like 20,000 random drum patterns) and a painter's pallette of colours is confusing different categories, and repeatedly using that discrepancy to rationalize what you are doing. Only one of these categories is actually dealing with predetermined structural elements and form.

This doesn't depend on your definition of skill. If you like, compare your method to visual artists using loads of pre-generated structures as elements in their visual works if you want to make a more valid rationale for what you are doing, but the argument of doing this in order to be a "painter with a wider spectrum of colours" -- and not even leaving it at that but actually describing it in such a literal sense... Well, that doesn't quite fit, and it also doesn't depend on your definitions.

Edit: oh, and I also googled an article on how to play speed scrabble, thanks for that

[jancivil]

Interest of some kind in what a drum part does is a definite prerequisite to dealing in 'music skill' for discussion.
It is passing hard to imagine someone with that interest needing 20k MIDI items out of the aether; the sheer number of that tells us of the randomness of it all.

The ideation averaging of velocity of someone else's version of a supposed part in a MIDI one-bar clip shows a clear avoidance of getting one's hands dirty. Now there is just rationalizing it like you can't look at people steering you to something more like doing music. It's absurd to argue 'it depends on what your definition of music skill is' in the context you provided: mechanical and as thought-free as you can come up with; by definition.

20k MIDI clips is not a palette of colors. Painters *make decisions* according to something in their very own actual mind as to a painting they're about to do and they apply them to a palette for mixing, for decisions based in their experience of color. It's not random; it's not a vast array of colors which they never examined before.

Stop bullshitting us or be called out for bullshit. It's not that you're not bright, but here you're applying yourself to layer upon layer of bullshit. It's intellectually dishonest; no one is particularly deceived by your rationalizations.
Look around, note the responses. It's not about being mean to you, it's only fair.

[jancivil]

It depends what your definition of "music skill" is.

If you like, compare your method to visual artists using loads of pre-generated structures as elements in their visual works if you want to make a more valid rationale...

But he wants to do pop music. Songs. Ringo Starr parts was mentioned; this stuff is not abstract in the least.
That's all a nice deconstruction of the analogy, but he's just bullshitting us with it.

[Guenon]

It depends what your definition of "music skill" is.

If you like, compare your method to visual artists using loads of pre-generated structures as elements in their visual works if you want to make a more valid rationale...

But he wants to do pop music. Songs. Ringo Starr parts was mentioned; this stuff is not abstract in the least.

I know, that's my point . The fact that comparing this methodology to a painter and taking these categorical issues into account... also makes it pretty hard to find a painter who works like that.

Earlier I mentioned that this is more like a landscape artist having stored 20,000 patterns/forms for different kinds of trees and bushes, for example, independent of the rest of the varying elements that make up the whole hypothetical finished landscape. And then producing the landscape by going through all those stored patterns and randomly trying them out against each other. Comparing to such a visual method is of course possible, but it's also very different from comparing to a pallette of colours

[Hink]

please leave the personal attacks out of it, if you dont agree with or understand his approach so be it but there's no reason to attack the person's intelligence.

Evidently he's clever enough for speed scrabble.
I wonder how in the world you come away with anyone failing to understand the approach. It's 20,000 one bar MIDI items downloaded off the net, and the best he's done is the <average the kit pieces velocity> bit, and just go with that result randomly rather than "approach" drum parts as music in any way.

AND he's using a lot of space with this. Some people are enabling this behavior. If that's what he wants to do, so be it but as an idea of what to do, he's opened the idea up for discussion. It seems like this is not the first time we've seen the 20,000 one bar clips as a topic here, but I could just be losing it.

be that as it may, I'm not going to judge whether or not this person's question is worthy of KvR. I get people questioning why he does this, I get people suggesting he cold speed up his process by learning more, but personal attacks are unwarranted, opening up an idea for discussion does not mean one needs to be subjected to personal attacks.

[Dunbar]

Not sure what's more interesting, OP's analytical approach to "music" making or the vitriolic responses he's received! I've never witnessed a process get such a pummeling before; surely we can't really make a judgement until we hear the results?

I understand OP sees himself as a 'curator' rather than a composer or musician. This state of affairs (love it or loath it) has been made possible by music technology; I'm certain it will be widely embraced as a legitimate approach, eventually. Composers already use 'generative' software, like Super-Collider, for music making where they adopt the role of 'editor'. OP's approach isn't that different.

Personally, i find the idea of an average Richard Starkey snare velocity hilarious. Assuming OP's using standard midi velocities I'd like to know the number, I'll forever call it: the 'Magic Starkey'.

[whyterabbyt]

Not sure whats more interesting, OP's analytical approach to "music" making

analytical? seriously?

I think there are far more appropriate adjectives.

Synonyms for random
adj haphazard, chance

accidental
aimless
arbitrary
incidental
indiscriminate
irregular
odd
unplanned
adventitious
casual
contingent
designless
desultory
driftless
fluky
fortuitous
hit-or-miss
objectless
promiscuous
purposeless
slapdash
spot
stray
unaimed

[jancivil]

"surely we can't really make a judgement until we hear the results?"
I can.