MAMEWorld Forums - EmuChat - Re: Some videos of my frontend in progress

> I think this is a step in a good direction, and cool to see some efforts being put
> forth with it.
>
> If its possible, additional un-used processor cores could be used for running a few
> virtual cabs at the same time... even if they are sort of buffered and streamed.
>
> I thinks it would be nice to have a row of cabinets put on a circular path. That
> way, you can see the sides of a cabinet as you are scrolling thru them.
>
> It would also be nice to be able to adjust the level of depth that you stop in at.
> For example, you might not want to zoom completely into the game. Rather, you would
> stop just about over 2ft away, so you could still see some of the cabinet
> model/artwork while playing.
>
> It also looked like there were too few frames when the games zoomed in and out.
>
> The biggest problem of course, are the models themselves. It appears for example,
> that there is no 3d bezel surrounding the normally curved monitor. The monitor should
> be a little deeper set into the game, and not right up against the plexiglass face.
> Typically, about 2 to 3 inches.
>
> If there was a curved invisible surface, it would be nice to wrap the games display
> around it.
>
> At very least, a squared bezel could be created as a temporary way to set the
> monitor image a little deeper into the game.

Thank you for your suggestions!

In terms of using extra cores - I just added a feature (that I had intended to use all along but hadn't implemented yet) where the game that the highlight is on starts up after 1/3 of a second of hovering on that game. This makes games start up pretty much instantaneously as you scroll through the list - the lag time between stopping on a game and having it running is very small.

The purpose of this is because one of my goals is for game starting and running to be as fast as possible, so I use the extra CPU of the system (i.e. extra cores and extra cycles available even on single core systems) to start games fast. The previous game hasn't even stopped running yet, it's just 'moved out of the way' in the background and shuts down while the hovered game starts up. The result is that you can pretty much start playing any game within a fraction of a second of highlighting it. By the time you press the button to zoom in it's usually already going.

That's one way that I'm soaking up extra cycles. I had also thought about maybe running more than one game at once, although my goal is less to offer a '3d arcade' style interface, and more just to showcase games one at a time and allow fast selection and play of them.

Modern graphics cards don't even begin to break a sweat with the amount of polygons and textures I'm throwing at them so multiple cabinets at once is definitely a possibility. Multiple games going all at once could be done too with the infrastructure I have created but like I said, it's kind of not really aligned with my focus which is to choose and play individual games rapidly.

I also thought about adjusting the zoom depth, with a stop like you said that stopped close to the game but not all the way there, either to view the cabinet (with game playing) closer up, or to allow the artwork to be viewed while playing the game. Although my personal preference is not to see artwork while I am playing, for some reason I just find it distracting.

In terms of more correct simulation of the screen, with an inset and rounded surface, that could be done but it would be something of a headache, mostly because it would involve work on the model side, and I am just trying to leverage free models, not make my own, because I don't really have the time or skill to do that; and also because projecting a texture onto a curved surface is not the easiest thing to do unless it's been authored that way when the model was created, which would, for various reasons, not work well with my system (I can dynamically position flat screens in space, but wrapping them around curved surfaces is quite a bit harder).