Yes, that’s it. A slow attack with a swelling to volume, with sometimes a decay at the end. This is classic “blooming” in younger brass players, and something we try to break them of as soon as we hear it. It’s not terrible, just annoying. 
Oh, and ignore the wrong notes. It’s an XML import from Finale, and I haven’t gotten past just listening for problems.
Thanks for jumping in Arne.
The indicated dynamic of the excerpt was “p”.
And because of what you said further in your answer, I took a closer look at the piece of mine that I had chosen to try out the 5.1.1 update.
It appeared to have an unusual (well, not for me, I prefer to write quiet dynamics for the sax section in general) high amount of soft dynamics for the saxophones.
That way, I had kind of “missed” that at louder dynamics, the saxophones in 5.0.1 and 5.1.1 indeed already sounded almost the same.
So because my pieces have the softer dynamics for saxophones for the majority of the time, I was also getting the softer saxophone sound the majority of the time with 5.0.1.
Now that you’ve extended the “hard” sound towards the softer dynamics in 5.1.1, I was tricked into believing the whole saxophone sound got a makeover.
I made 2 more audio exports of the (same but shortened) excerpt, but now played p - mf - ff, in that order.
There’s only a very big difference in sound when played at “p”, but you’re right about it using the same sounds, the difference is negligible at mf and ff.
5.0.1:
5.1.1:
So I humbly stand corrected about my statement that the saxophone sound sample got worse between 5.0.1 and 5.1.1, although I still don’t like the “hard” sound that is now present across the whole dynamic range.
If there is any way by changing some setting that I somehow can get the “soft” sound back at any dynamic, please let me now. And I think fellow wind band writer @Estigy will also be interested in that answer.
Oh yeah, and if there’s a way to get the old clarinet back, I’d be happy to hear!
Thanks, @James_Bondjas for doing all these tests, this is really helpful!
Some background information: I just happen to work on a choral piece right now around Christmas that starts out quite soft and only goes to higher dynamics for a very short time, so getting the “hard” saxophone sound also for the quiet parts really makes it worse for me.
(As a sidenote, I also somehow feel that the players are quite out of tune more often, but this could be just me listening more carefully now for changes after the update…?)
Agreed on both counts. 
For my part, yes it’s really important to be able to assess how the dynamics of your sections will sound together, and how articulation will come across in a real performance.
However I also wish to make as high-quality mockups as possible without having to resort to a DAW and spending enormous amounts of time learning how to be effectively a mix engineer.
This is why I loved 4.5.1, and in fact I am still using it. There’s no decent alternative to it, and having spent many thousands on sample libraries, I am loathe to dump them, or spend the time to learn DAW mockups when I would rather spend that time writing.
I think this is a fairly common goal too. It’s all well to say what the primary goal of a product is, but in the end it’s how the users use the product that determines the primary use of the product. When the goals don’t match the use, then companies tend to not do so well. But with no competition in this segment, there’s nothing to drive it in any direction or another. But still I am very grateful to Arne for everything he does for us all, and I feel sometimes people underestimate his deep understanding of all these things.
The saxophone sound sources are identical to those in 5.0.1, but the dynamics have been adjusted to use a higher range of the same instrument. They were previously configured to use a lower dynamic range and didn’t approach the highest dynamics, even at FFF. I’m sure there are further tweaks that can and will be made, but we need to find a middle ground where the instruments fit more than one style. With version 5.0.1, we had many users indicating that the saxophones were too soft. Suppose this audio example was notated PPP-PP-P. In that case, I agree they sound too skewed towards high dynamics in 5.1, but they may still be too soft in 5.0.1, and need some middle ground. Even if we implement that middle ground, there’s a real chance you will still prefer 5.0.1 because it’s your frame of reference for this music.
Regarding the saxophone sound, have you considered creating separate “Classical” and “Jazz” saxophone sounds, or perhaps adding a CC that can switch between such modes? Out of all instruments, the saxophone is probably the instrument whose sound changes the most depending on which ensemble or style of music it’s playing, making a one-size-fit-all approach very difficult to satisfy all genres.
For example, I’d love to be able to get a classical saxophone playback sound that sounds similar to this performance: https://youtu.be/4FiaRL1Kdnc?si=1V_6Cddw6k7Gy4bh&t=47. There’s no way that a playback instrument suitable for this style of saxophone playing could also be used for modern jazz and pop. Even a performance with louder dynamics like this one would preferably use a classical sax instrument, rather than a jazz/pop style of instrument which has the sound of a sax player using a metal mouthpiece with jazz reeds.
I’m sure I’m not the first to suggest this, and I understand that if you only have one set of saxophone base sounds to work from, making two instrument presets with different dynamic curves might not fully capture enough of a tonal difference to warrant multiple saxophone styles in NP.
Thanks for the link. Gorgeous!
We’ve just released the 5.1.2 update. Can those of you suffering from “slow brass attacks” please try if the update resolves the problem?
The saxophone ranges have also been tweaked. The dynamic ranges are now between that of versions 5.0 and 5.1.
Our update link:
I can confirm the brass attacks are much better. I can still hear it, but I have to strain, and may even be imagining it, it’s that subtle. Also, saxophone is much better at low volumes, but everything mf and higher still has a bite to it. I’d love to hear a Donald Sinta style of sound, but this is a good compromise. It also sounds like your average high school band sax section, and that’s what I mostly write for, so there’s that.
Yes, I also can confirm that this version is much (much!!) nicer, especially at low volums. Notably the Tenor Saxophone now sounds much friendlier and less like my mum 
Thanks, @Wallander !
NP 5.1.1:
NP 5.1.2:
(Those files are pure Dorico with NotePerformer. The the only manual adjustment being the overall volume in NP reduced to 90 in order to prevent clipping.)
What I’m not hearing in your recording is any vibrato at all. I haven’t checked that yet with my files, but vibrato is part of a good saxophone sound and should be present most of the time. Did you do something to change that?
You are right, those saxophones play completely without vibrato. But I did not actively deactivate that. I guess I would have to set a certain CC to activate it… Will try, thanks for the idea!
(Interestingly, Flutes play with a very nice vibrato…)
I think there’s a way to create an articulation that activates NP’s vibrato. At least there was in Finale, but I don’t remember if this was needed or created for Dorico.
Can confirm you can just add the vib playing technique to a passage, and NP interprets it correctly. The sound, while synchronized among players, is much warmer and more traditional.
Good to know NP can do that. Sax vibrato-whether it’s used and/or how much- is contextually dependent, though.
That is a very nice march. Well done! 
Reverb for me is better. I was getting a hint of a double hit in double bass staccatos (low F and thereabouts) but it’s cleaner now.
I completely second this.
Although the 5.1.2 update brought a slight improvement, the saxes still sound way too sharp for standard wind band usage, especially starting from mf and all louder dynamics.
It would be great if we could indeed switch the sax sound, just like you can do with the drum kit, where you can choose for instance between standard sticks or brushes in the Noteperformer mixer.
But a sincere thank you anyway for already hearing our cries, @Wallander
And if 5.1.3 could bring the old clarinet sound back, I’ll buy you a Belgian beer some day.
If you want to enforce a rounder or sharper tone, you can manually change the dynamics and use the instrument’s volume to compensate. It mirrors the difference between these NotePerformer versions.
Example: set our MIDI CC110 (secondary expression) to a value of 1, and increase the instrument’s volume in our mixer until you’re satisfied.
Here’s “secondary expression” in our user’s guide:
https://noteperformer.com/?page=support_dorico&h=561971509#561971509&a
Arne, it would be very useful if NotePerformer Expression map had plaback techniques for some of those CC controls. If one could somehow import custom playing techniques for notepeformer that would match those CC controls.
I have edited the stock noteperformer expression map, so that it has playback techniques for a1 … a12 section sizes. These work very well for string divisis and actually make a very big difference in playback.
Editing CC lanes is just so slow and cumbersome that it is out of question for me: in a large project with constantly changing divisis it would just take way too much time. If there were more custom playing techniques like this, that can be quickly added to the score and hidden, at least I would surely use them a lot.
Using my own edited noteperformer expression map is a bit sketchy, it has to have the same name as the original and always reimported for new projects. Also, for some reason these a1..a12 add-on techniques only work if they exist for the first note in the flow.
Do you think this kind of approach with custom playing/playback techniques could work?
Of course, if Dorico had midi-trigger-region-like CC messages, those could be used for the same purpose.