Booting to BASIC excludes mouse pointer

I’ve not searched the forum, so sorry if this is a well known issue, but if I configure RPCEmu to boot directly into BASIC (with *configure lang.) I cannot move the mouse into the window unless the *POINTER command has first been issued. I am reliably informed that this is not how a real RISC PC behaves, with that you can move the mouse even if it is invisible (i.e. MOUSE X,Y,B still works).

My system is Windows 11, RPCEmu 0.9.4, RISC OS 5.28.

Is the Settings→Mouse→Host Mouse Enabled option enabled?

Is the Settings→Mouse→Host Mouse Enabled option enabled?

‘Follow Host Mouse’ you mean? Yes, it is (without that setting I wouldn’t expect to be able to move the RISC OS mouse at all).

The only way I can get it to behave properly is to do:

*POINTER
MOUSE OFF

I would expect that to be the initial state, but it isn’t.

‘Follow Host Mouse’ you mean? Yes, it is (without that setting I wouldn’t expect to be able to move the RISC OS mouse at all).

I don’t think that’s a correct assumption… from the manual:

Settings → Follow Host Mouse – RPCEmu uses a hack to make the RISC OS mouse pointer follow the native pointer. In many games this does not work, so disable this to turn it off. Disabling this enables ‘Capture’ mouse mode where the pointer will be constrained within the emulators display, click in the window to capture the mouse, and press CTRl+END to release the mouse.

With the option off you have to manually “move” the mouse pointer “into” and “out of” the RPCEmu window, which is more clunky. However, it’s also more compatible.

‘Follow Host Mouse’ you mean? Yes, it is (without that setting I wouldn’t expect to be able to move the RISC OS mouse at all).

Sorry, yes. I got ahead of myself there. I run RPCEmu with it switched off for improved low-level compatibility.

I don’t think that’s a correct assumption

Either way, it’s peripheral to the apparent bug. I want to be able to move the RISC OS pointer around using my native mouse in Windows, and it’s not letting me do that until after issuing a *POINTER command (even though I don’t necessarily want to see the pointer). There’s no logical reason for that (and apparently it doesn’t happen in RPCEmu on Linux), it should initially behave the same as it does after:

*POINTER
MOUSE OFF