Dorico 4 Performance

So after working on a project over the last few days I have observed that subjectively Dorico4 feels less responsive than 3.5 considering performance, in spite of running it on a very powerful computer. Especially bigger scores get considerably more laggy than they were in 3.5 on my end.

As this hasn’t been discussed here, I just want to check whether this is a personal issue on my end or whether people feel the same?


On the three computers I’ve been using - Mac and PC, it’s felt subjectively faster. They’re high end machines and I run big orchestral templates with chunky libraries, FWIW.

1 Like

Some edits in Dorico 4 should be considerably faster than they were in Dorico 3.5, for example any time you apply options from one of the five main options dialogs, control will be returned to you much more quickly, and in general there are a few classes of edits where Dorico would have to essentially regenerate the entire layout you’re working on that are now much quicker.

I would expect edits in general, e.g. note input, moving things rhythmically in Write mode, etc. to have broadly the same performance as in Dorico 3.5.

Dorico 4 will use more RAM than Dorico 3.5, partly due to the new user interface toolkit we are using and the fact that, at the moment, because we have not yet transitioned the whole of the project window to use this new toolkit, we are having to use it in a manner that is sub-optimal from a RAM consumption point of view. And it could be that the increased RAM pressure is causing paging, though I imagine you will have at least 32GB RAM and even with Dorico’s increased memory usage it shouldn’t be putting pressure on a machine with plenty of RAM headroom.

Hard data is more actionable than a feeling, so perhaps you could get more of a sense of which particular edits you feel are noticeably slower in 4.0 than they are in 3.5, and we can look into it further.

Thanks @dspreadbury for your answer. Yes, of course, subjectiveness is not the greatest source to target specific issues. I will see if I can isolate a few cases where timings got actually slower on my end. RAM is definitely not an issue with this machine.

From the recent project I had the impression that the finishing touches on the score with condensing activated were slower than they used to be in 3.5, like creating frame breaks, hiding extra player labels, adding custom bracketing on specific pages etc. I’m aware that activated condensing is not the greatest way to start when expecting performance, but I will try and see if I can come up with some hard numbers in this regard.


If it helps, a few of us were posting some CPU utilization, RAM, and GPU numbers for D4 in this thread. I’d be curious how your setup compares.

This is really great information to have, Daniel, thank you.