Page 22 of 57 FirstFirst ... 12171819202122232425262732 ... LastLast
Results 421 to 440 of 1128
Like Tree12Likes
  1. Spyral's Avatar
    Posts
    128 Posts
    Global Posts
    182 Global Posts
    #421  
    I might as well just uninstall every other game I've bought/homebrewed for my Pre. VBA makes everything else obsolete. I've donated, and I hope others do too.
  2. #422  
    Quote Originally Posted by Josephc1991 View Post
    I've edited nights0223 custom skin, all I did was fix the sharpness around the A and B buttons.

    All credits belong to nights0223, and me for the edit.
    Hey I appreciate you pointing that out, I don't know how I let that get by me. I fixed it so is looks better, no offense to your edit, but it was a little too blurry. It should look much better now.

    Also of note, the L and R buttons still work at the top, but since this skin is really supposed to be for regular GB/GBC games, they aren't visible.
    Attached Images Attached Images
    Attached Files Attached Files
  3.    #423  
    Quote Originally Posted by alex.dobeck View Post
    Well maybe I just can't seem to get the configuration right with the current implementation...

    But I can't seem to get the deadzone right. The area between left and right always feels too big, or I can't drag enough to the left/right before it either doing nothing or suddenly going up or down because I'm no longer perfectly straight.

    I'd like to be able to define a polygon in the shape of one of the attached (badly drawn) images, either something where it would feel natural to drag (and you could include a skin with a simply dot rather than the whole d-pad graphic), basically extending the area recognizable of right/left/up/down but have them close together in the center. Obviously this would make it quite a bit more difficult to do diagonal, but for instance the area completing the square in the first drawing would fit.

    Currently it almost feels like the third image, as the deadzone shrinks the top and bottom come closer together, but if the deadzone value is increased the distance between the two opposing directions begin to feel more unnatural.

    Maybe I don't quite understand this yet, it is the first time I've tried to deal with a virtual joystick, so it would be great if someone could point me in the right direction.
    Hey, I'm super glad you're thinking about what a good joystick would be.

    The current joystick implementation is as follows:

    Imagine a full circle, then imagine a chord cutting across it horizontally, just above the center (so it almost makes two semicircles, but does not). All the area *above* this chord is 'up'. The perpendicular distance from the center of the circle to the chord is the 'deadzone' size. Very similar for left/right/down-- so four overlapping almost-semicircles. No I'm not cracked out on geometry, it was on the easy side to code and gave reasonable playability. I did however not particularly like it, but it was decently simple and so was hoping would work out (simple solutions usually work best ).

    However, I also find it not the best, and would like to improve it. Or at least try to improve it .

    I rather like your second idea--it's simple and effective, and a marked improvement over what there is now. Or at least I think it would be, I'd have to go implement it to verify--but it encodes the idea that if you're really far to the right, for example, you should have to go farther 'up' to indicate you want to go up (farther than if you're close to the center)--the fact that you're so far right should mean something. Something intuitive that my current scheme gets wrong and has plagued me from the start. For some reason this simple idea didn't occur to me . I'm actually a little that my original idea was as convoluted. This is why I don't design interfaces hehe.

    However it does not have diagonals marked, and since your thumb can only hit exactly 1 1-dimensional point, this means there's no way to move diagonally with that layout.

    I'm imagining (warning another description coming, I'm no good with pictures ) a similar idea, the basis being a circle still, but
    in each quadrant of this circle, there are two lines going out radially from the center--essentially what you have (1 line for each quarter) but have two--at about 30/60 degrees or whichever makes sense. This gives the 'cone' for each direction, but also has explicit diagonal directions. To be clear these lines are separating markers between regions, not the regions themselves.

    Perhaps I'll embarrass myself later and draw it up, but hopefully that makes some sense.

    This new implementation would, as I said, encode the concept that how far away you are should absolutely matter, yet near the center also gives a decent ability to 'turn' quickly since these radially lines obviously converge at the center .

    I still think perhaps a deadzone in the center would be good, just so you can rest your finger and not have it super sensitive--but that could probably be noticeably smaller than what we have now since the cone idea would help stabilize as you move your finger away from center.

    Anyway I have to get going, but let me know (and everyone else feel free to chime in) what you think of that idea, here's hoping we can make the controls that much better.

    I suppose if anyone's super fond of the current implementation speak up--but I think you're all okay with it, happy to have touchscreen controls, and letting me hide behind the fact that it's a touchscreen when it comes to dpad usability . Or letting the app hide anyway.

    Oh and I don't think I'll be implementing arbitrary polygon hit detection in the near future. Absolutely sounds like overkill and just makes things that much more complicated. I'm a bit stubborn in that simple and clean will win over something super complicated. If you do have a compelling idea, and perhaps want a toggle or to try it, I'm open to being convinced, but .. it'll have to be convincing .

    If we can figure out something that sounds compelling I can probably push it in the next few days.

    Thanks for your thoughts and feedback in helping making this excellent .
    Quake
    thread
    VisualBoyAdvance
    thread App Cat
    X11/XServer/Xterm
    thread
    SuperNES
    thread App Cat


    If you like my work, please consider donating or buying VBA!
  4. #424  
    Hey dtzwill....2 questions. Can the latest update log be found on The wikisite? including all the changes that was made? and have you developed a way to go back to the rom select screen straight from the game???

    update...found the changelog... have you developed a way to get back tot he Rom selector straight from the gamescreen or do I still have to exit out of the whole program first???
  5. #425  
    Still have to exit. I don't think dtzWill has any easy way of returning to the ROM selector right now, that is, it would be hard to implement, if not impossible.
  6. navinag's Avatar
    Posts
    455 Posts
    Global Posts
    489 Global Posts
    #426  
    Quote Originally Posted by Vincent Law View Post
    Still have to exit. I don't think dtzWill has any easy way of returning to the ROM selector right now, that is, it would be hard to implement, if not impossible.
    Suggestion....Can't a bind key(or a key on the virtual keypad) be created to exist and restart the application....thus going back to the rom selection screen?
  7. #427  
    Quote Originally Posted by dtzWill View Post
    Hey, I'm super glad you're thinking about what a good joystick would be.

    The current joystick implementation is as follows:

    Imagine a full circle, then imagine a chord cutting across it horizontally, just above the center (so it almost makes two semicircles, but does not). All the area *above* this chord is 'up'. The perpendicular distance from the center of the circle to the chord is the 'deadzone' size. Very similar for left/right/down-- so four overlapping almost-semicircles. No I'm not cracked out on geometry, it was on the easy side to code and gave reasonable playability. I did however not particularly like it, but it was decently simple and so was hoping would work out (simple solutions usually work best ).

    However, I also find it not the best, and would like to improve it. Or at least try to improve it .

    I rather like your second idea--it's simple and effective, and a marked improvement over what there is now. Or at least I think it would be, I'd have to go implement it to verify--but it encodes the idea that if you're really far to the right, for example, you should have to go farther 'up' to indicate you want to go up (farther than if you're close to the center)--the fact that you're so far right should mean something. Something intuitive that my current scheme gets wrong and has plagued me from the start. For some reason this simple idea didn't occur to me . I'm actually a little that my original idea was as convoluted. This is why I don't design interfaces hehe.

    However it does not have diagonals marked, and since your thumb can only hit exactly 1 1-dimensional point, this means there's no way to move diagonally with that layout.

    I'm imagining (warning another description coming, I'm no good with pictures ) a similar idea, the basis being a circle still, but
    in each quadrant of this circle, there are two lines going out radially from the center--essentially what you have (1 line for each quarter) but have two--at about 30/60 degrees or whichever makes sense. This gives the 'cone' for each direction, but also has explicit diagonal directions. To be clear these lines are separating markers between regions, not the regions themselves.

    Perhaps I'll embarrass myself later and draw it up, but hopefully that makes some sense.

    This new implementation would, as I said, encode the concept that how far away you are should absolutely matter, yet near the center also gives a decent ability to 'turn' quickly since these radially lines obviously converge at the center .

    I still think perhaps a deadzone in the center would be good, just so you can rest your finger and not have it super sensitive--but that could probably be noticeably smaller than what we have now since the cone idea would help stabilize as you move your finger away from center.

    Anyway I have to get going, but let me know (and everyone else feel free to chime in) what you think of that idea, here's hoping we can make the controls that much better.

    I suppose if anyone's super fond of the current implementation speak up--but I think you're all okay with it, happy to have touchscreen controls, and letting me hide behind the fact that it's a touchscreen when it comes to dpad usability . Or letting the app hide anyway.

    Oh and I don't think I'll be implementing arbitrary polygon hit detection in the near future. Absolutely sounds like overkill and just makes things that much more complicated. I'm a bit stubborn in that simple and clean will win over something super complicated. If you do have a compelling idea, and perhaps want a toggle or to try it, I'm open to being convinced, but .. it'll have to be convincing .

    If we can figure out something that sounds compelling I can probably push it in the next few days.

    Thanks for your thoughts and feedback in helping making this excellent .
    I wouldn't mind seeing a change to the dpad. I think the touchscreen controls are awesome, but for hardcore play, I definitely play in Portrait view. The hard keys quicker response time can't be passed up when moving a character around the screen.
  8. #428  
    I am so impressed with the Homebrew community. With the frequent updates to WebOS, new Preware and app catalog games coming out rapidly, and now VBA, I feel like the Pre is the gift that keeps on giving.
  9. #429  
    Sorry for the picture again but I wanted to double check I understood your idea.

    The picture should hopefully illustrate the four quarters and the eight quadrants for each direction and diagonal.

    I do like this idea, it seems easier to figure mathematically by only needing the location, radius, and degrees of either each individual control (equaling 360), or possibly just the diagonal degree (less than 90) so than you could provide the location and radius of 60, followed by possibly the degree of 20 for the diagonals, so then each direction would be a degree of 70, in this case.

    I think that would work very well as it would allow in many degrees a larger area of forgiveness as you move away from the center, but by moving up or down you should logically hit the diagonal, or if your close enough to the center, eventually up. This should allow skinners to provide a simple dot for the center and a circle for diameter of the circle or not and still be able to provide enough accuracy for the individuals.

    And your explanation of the current method helped me alot, it also reviled why I couldn't find the right solution, the lower I made the dead-zone the closer right and left would get to each other, but the more they would be overlapped by the up/down meaning the need for more precises left/right movements.

    Once again please excuse my horrible ability to draw on the computer.

    Also, for coolness sake you could then use the umbrella corp symbol for your d-pad
    Attached Images Attached Images
    Last edited by alex.dobeck; 01/26/2010 at 01:19 PM.
  10. #430  
    Are there any tutorials out there on how I could make a skin for this? I want to try to create the skin I requested earlier
  11. #431  
    Quote Originally Posted by dirtygreek View Post
    Is it possible to have skins in portrait mode or a skin that otherwise doesn't shrink the screen so much?
    I agree we need skins for Portrait mode
  12. #432  
    Yeah portrait mode skins please, I submitted an idea earlier. I know that due to limitations we can't have both the dpad and buttons on screen, but could we have at least maybe the L/R buttons on screen? I prefer portrait using the keyboard controls but no matter how I map them I always find L/R to be uncomfortable or unusable. Specially with the faster action games like metroid fusion.
  13. #433  
    I don't think the portrait skins are implemented yet. I mean if you press orange+/ you won't get a skin in portrait, only in the keyboard-right view, so you'll have to wait.

    I'm interested to know how close presses can be to register as other people have said. I don't see why it would be a problem from the Pre's side, it seems to detect pinch's which are close just fine. Though I mean it's possible the limitation of the sdl libraries are there or something, but I'm interested to hear from Will on this, though he might be planning to surprise us coming up here anyway.
  14. #434  
    Anyone else having issues when trying to play GBA games from the "Classic NES Series"? I get a "game pak error Turn the power off" message in the game window.
  15.    #435  
    Quote Originally Posted by fastamx79 View Post
    Anyone else having issues when trying to play GBA games from the "Classic NES Series"? I get a "game pak error Turn the power off" message in the game window.
    This is a known issue with vba on all platforms and most emulators I believe.

    I won't link it, but there are 'patches' available to get these roms to play.

    I've only looked into this briefly; I'm not sure I understand what the issue is, why the patches are necessary, or what fixing it would entail.

    But there you have it
    Quake
    thread
    VisualBoyAdvance
    thread App Cat
    X11/XServer/Xterm
    thread
    SuperNES
    thread App Cat


    If you like my work, please consider donating or buying VBA!
  16. #436  
    dtzWill, I think it's time you get a subforum for your work on this.
    There's well over 400 posts in this thread and no one will volunteer to read every single page when they come with a concern or such. Having a subforum should make the work and questioning much more streamlined.

    For example, there should be a thread for announcements and changelog, thread for skins, a thread for controls, a thread for suggestions, thread for issues, thread for coding(such as the complex on screen control system discussion).

    Of course this could all be just a wiki but some people desire the ability to discuss freely.

    This would make it easier on users who want to check updates on a specific section for vba.
  17. #437  
    Quote Originally Posted by dtzWill View Post
    This is a known issue with vba on all platforms and most emulators I believe.

    I won't link it, but there are 'patches' available to get these roms to play.

    I've only looked into this briefly; I'm not sure I understand what the issue is, why the patches are necessary, or what fixing it would entail.

    But there you have it
    Thanks for the info!!
  18. #438  
    Quote Originally Posted by sjsamphex View Post
    dtzWill, I think it's time you get a subforum for your work on this.
    There's well over 400 posts in this thread and no one will volunteer to read every single page when they come with a concern or such. Having a subforum should make the work and questioning much more streamlined.

    For example, there should be a thread for announcements and changelog, thread for skins, a thread for controls, a thread for suggestions, thread for issues, thread for coding(such as the complex on screen control system discussion).

    Of course this could all be just a wiki but some people desire the ability to discuss freely.

    This would make it easier on users who want to check updates on a specific section for vba.

    yeah i agree, though i have read everyword of this thread, and continue to
  19. #439  
    Quote Originally Posted by dtzWill View Post
    This is a known issue with vba on all platforms and most emulators I believe.

    I won't link it, but there are 'patches' available to get these roms to play.

    I've only looked into this briefly; I'm not sure I understand what the issue is, why the patches are necessary, or what fixing it would entail.

    But there you have it
    which patch is it?
  20. #440  
    Quote Originally Posted by Irocdaspot513 View Post
    which patch is it?
    I think he means there is a ROM patch, not a patch you can install on the Pre.

Tags for this Thread

Posting Permissions