Woo, that’s exactly what I was looking for! Thank you guys for the detailed process descriptions.
Also, I could take a photo or something of the music and put it on a monitor, I just don’t want to worry about OCR and such yet, plus if it’s handwrittend, etc…
Thanks again to all the great replies!
If one starts with a first pass to set note durations, which sounds like a good idea, my thought would be to use a MIDI keyboard and Lock Duration to enter the pitches on a second pass–but then I am a lot more secure on a MIDI keyboard than on a typewriter-style keyboard.
I think Alan Silvestri does something like this, if I remember from one of the Dorico videos. I haven’t tried it myself but it sounds like a good idea.
I do use Lock Durations sometimes with Dorico for homorhythmic passages, but you do have to be accurate with your MIDI input and watch the screen for errors. I’ve mostly gotten away from accepting straight-up copy work except for projects I want to be involved with, but from 1996-2015 or so I did a lot of work for composers from their pencil scores. (A quick search for *.mus* in my main Finale Files folder shows 14,420 files!)
I input fast, so with pitch-first input, I don’t even need to look at the screen, the numpad, or my MIDI keyboard. I can just keep my eyes on the manuscript and I will hear any MIDI errors audibly from my MIDI keyboard. If I mis-play a MIDI key it doesn’t matter as the note won’t be input until I press the value on the numpad, so I simply play the correct one and input it without looking. I’m hitting the duration just milliseconds after I hear the note, so it’s not really any slower than duration-first, and of course all it takes is one misplayed MIDI key per system that has to be edited to make duration-first substantially slower. If you can input error-free without looking away from the manuscript, then I’ll grant that duration-first may be faster, but for me pitch-first is faster due to the fact that MIDI errors won’t be input as I won’t hit the duration. At the end of each system/staff of the manuscript I glance up to make sure I’m in the correct place, and start with note input on the next.
Once the notes are in, I move on to other elements such as dynamics, slurs, chord symbols, etc. As I’m doing this I’m also proofreading and double-checking my previous work against the manuscript. Playback is a useful proofreading tool too. For most of my clients, I really got to know their style, so if anything seemed odd and particularly out of character, I’d make a note of it and add it to the question sheet, so I wouldn’t be constantly bugging them. I had a good reputation for catching more mistakes than I made so I always had work.
At the end of each page I do a quick proof before moving on to the next page. I know the type of errors I’m personally likely to make so obviously I look more for those and focus less on other elements. I had been trained on the Schirmer method of proofreading, so even though I didn’t literally follow it, it’s always sort of in the back of my mind:
While proofreading is important, you can’t bill for it will all clients, and often there isn’t a ton of extra time anyway, so being as accurate as possible first pass while working quickly without additional editing steps is important for me anyway. That’s why I usually use pitch-first even for jobs from a manuscript. (Obviously, I think pitch-first is much preferable for comp/arr work.)
Thanks for thisinteresting post…I will certainly try to adopt the “pitch first” technique and then put in “insert nmode” the duration of the notes (I had already asked the question but perhaps I did not put it well expressed)
To go from (pitch first entry):
To the final result :
I must do in insert mode :
5→5→6→5→5→6→4→4→→4→
so the question:
how to do so as not to need to press the arrow key → between each duration value entry
Hi @Dup , maybe some naming confusion arose… the Input method that @FredGUnn is referring to is this:
For the other (very different) method to write all notes with one note value first, and then adjust the rhythm later with Insert mode on, to which I was referring to here:
if you go linear than you use the arrow key to move to the next note (as you said), but you can also select not adjacent notes and set their long at once.
(This is by no means the fastest method, but can be convenient to use sometimes, and funny 
):
I was referring to your comment and you made my mouth water! but therefore it is not possible to advance automatically without using the arrow key: too bad, that is great for certain cases
Input errors with a job that size are of course impossible to completely avoid. Everyone makes mistakes. But compare the amount of time it takes to edit a mistake with the amount of time it takes to input correctly the first pass. You have to look away from the manuscript, arrow back to the mistake (or mouse over to it), select the error (or restart the input caret), and fix the error. I could easily have input 10-12 more notes correctly in the time it took to fix a single mistake. Editing input errors will completely tank the amount of time it will take to input a job this size, so I would design your workflow around being able to input correctly on the first pass without having to look away from the manuscript.
Mouse input is ridiculously slow and inefficient. Real-time MIDI input is too error prone unless the music is very simple and you are a rhythmically accurate pianist. OCR is not an option with handwritten music, and honestly usually requires too much editing with engraved music as well, unless very simple. MIDI input is much more natural (and faster) for me than QWERTY, although you will have fewer enharmonics to flip with QWERTY. So that leaves step-time MIDI input and you have to choose between pitch-first and duration-first. I use pitch-first because it minimizes input errors as the note isn’t even input if I hear the wrong pitch, and therefore is faster for me as I have fewer edits.
I personally avoid the use of Insert mode if at all possible and even disabled the keycommand for it. Placing the Score Self-Destruct Button (I) right next to the Force Duration command (O) where it could easily be hit accidentally seems like a very questionable decision IMO.
Totally agree with @FredGUnn. It’s almost always better to input from scratch. Little quirks and hidden oddities can cost you a lot of time (and sometimes be a major problem later). Personally, I think some of the other methods might be helpful for special cases, but they can be gimmicks too, that don’t always translate to real efficiency. Every time you do something like that, your brain has to change directions. It’s hard to get into a rhythm.
My right hand is on the number pad and my left hand is on the MIDI keyboard. I’ve remapped the NumEnter to advance the cursor by the selected value. Index, middle, and ring fingers on 4, 5, and 6. Fingers move up to 7,8,9 when needed. Notes in one pass, slurs in the next, then articulations, then dynamics.
I’ve seen some videos of people inputting incredibly fast. By page 50 or so, you’ll be an input phenom. 
I’m curious, can you link those videos?
Sorry, I can’t remember what to search for. It was somebody who had mapped all sorts of note values (and even tuplets) to MIDI keys. It was really quite impressive. But like advertised download speeds, I wonder if it was a burst, or if it was a sustained pace.
I can input music pretty quickly using the posture I described above. Not sure I want to foist videos on anyone though.
@Dup Ahaha, sorry for that.
But if you like this method, it is easy enough to program a macro that sets the note value and advances the selection, and set this macro to a single shortcut, or use the Stream Deck macro possibilities…or other methods. The fantasy is the limit
Aaha!!!. Yes, for complex score it can be indeed “dangerous” to use Insert Mode, especially if one forgets to deactivate it after use (as I always do, I mean that I always forget to deactivate it…)
I would like to have the Insert Mode icon red coloured, so that is clear when it is activated.
If I may… I relink an older Post referring to the Hyperscribe functionality with tapping the tempo for live recording (with a link to an explication video for whom may be interested) that I loved in Finale: this have been the absolutely fastest method of note inputting that I personally experienced, and I dream that something similar may be implemented in the future into Dorico:
Dorico’s version of Hyperscribe is quite usable, and superior to Finale in some ways. There’s still room for improvement, but if you’re a keyboard player, it works great.
However, I only use it for real-time sketching. For note input, step-time is faster and requires no cleanup.
Hi Dan, would you share more about the “Dorico’s version of Hyperscribe”? Is that a macro/plug-in or some sort of workflow for similar input method like Hyperscribe?
No, it’s just simple MIDI entry to a metronome. Recording.
To add to this. I do actually use Insert mode to correct little rhythmical errors after having input a line of music.
If I look up to Dorico and see the music I have just input to be out if synch, I’ll enable Insert mode and correct the rhythmical errors.
They can happen, if one is really fast on the MIDI keyboard, whilst changing note values on the num pad.
@FredGUnn I’m curious, if you don’t look at the screen, do you correct enharmonic spellings in a separate pass after entry?
Enharmonic spellings just got a lot better in 5.1.40! Especially when inputting from transposed scores! Yeah, when I’m adding slurs, dynamics, articulations, etc., I’m also proofing enharmonics at the same time. (Finale is typically very predicable with enharmonics and has favor sharps and favor flats settings in addition to spelling tables, so it wasn’t much of an issue back in my Finale days.)
