Christophe de Dinechin
Christophe de Dinechin
Caused by an incorrect use of `-` operator on `uint` value, which causes the epsilon to be very large instead of very small.
It is a bit difficult to adjust the sides while preserving the aspect ratio (which is necessary to accurately identify the keys when touching).  The best I could do...
Indeed, preserving aspect ratio and respecting the area that Apple marks as "safe" are what guided the implementation. For a few releases, I tried to enlarge things to the whole...
So @marsell on the real device, does this feel cramped? See also #1570.
I had actually tried to improve the distribution in some earlier release. It is surprisingly difficult to do right for integers, and the range -9 .. 9 is particularly annoying...
This is really a duplicate of #1378
Oops. Good catch, @BadtoysWEF. Fixed on `dev`, will be fixed in the next release.
> I haven't looked at the code for derivative (could not find it, yet). The core of that code would be [here](https://github.com/c3d/db48x/blob/02409d2f698e28fd62b3d9f49b06c8fb1f2fd04e/src/expression.cc#L3906) > So I am speculating. Probably it stumbles...
Hi @evgaster > I am very new to this db48 stuff, coming from a HP28S and HP50g. A, perhaps stupid, question: which key sequence activates the substitution operator | in...
It's really weird, because variations of ``` rpl '4/3·Ⓒπ·x³' 'x' ∂ ``` work better: ``` rpl '4/3·x³' 'x' ∂ @ Works: 4*x^2 '4·Ⓒπ·x³' 'x' ∂ @ Works: 12*pi*x^2 ``` In...