it seems to me like you are worried it will take a long time to jump over your subs at the start of the program, the only time lbl-goto have a speed issue is if you are repeatedly doing it in a loop while looking for something really far away. Splelers method combines the strengths of both skipping structures and jumping structures so his way is defenitely the fastest(while in the program). That, correct me if im wrong, is better than having a half second or even a second shorter start up time. also you mention jumping over the 1000.. byte loader? the only time this would be a concern would be if you had a lbl after this loader but if you include all the subs before the loader it shoodnt be a problem, splelers method uses while loops to get past the loader, this is optimal because the start of the while is random access, so the calc will immediately go to it when you hit the end, as opposed to sequential access like in lbl-goto where it has to look over everything to find the lbl. I hope i answered everything...

just remember that lbl/goto have to start jumping at the start of the program where if-then's skip straight from the if-then. it seems like it would depend how much stuff you have before the subroutines. if you have alot then i'd use if-then otherwise i'd use lblgoto. If you are unsure what "alot" is you can just force lbl-goto be the best by rewriting your code so that the subs are up top.

going halfway in is odd... the most plausible thing would be some mixup between bit numbers, like your scrolling an 8*8 char using a 16x16 tile map  or something. code would defenitely be useful though

Edit: just thought of some stuff... one thing that may be wrong is the fact that x and y on the matrix corrospond to the top left spot on the screen. so when you are calculating what spot you are on you have to subtract one from x and y when you add screen location coordinates with regular position coordinates. another thing that might be a problem is that if you are using smooth movent instead of jump movement (moving a pixel at a time instead a tile at a time) you have to check for hit detection on every corner of the sprite.

i belive the manual says the sprites must have width of multiples of 8, so 16 is definitely fine, you only have to write 2 for sprite width instead of 1 (along with 16 in height not

it took under 24 hours (setup yesterday, got it today) but the voting is closed i think so its no use

the links to all tutorials after arrays don't work.  

Do you need a ticalc.org acc to vote? if not could someone please point me to the place i can enter my vote, all i can find are pages that say VOTING HAS OPENED! i can't find any VOTE HERE

i'm not sure if you meant to do this but sometimes it is easier to move and at others it is harder. like sometimes i will run and jump and it will carry me to where i thought it would, and other times i fall well short of my projected destination. Im not sure if this was intended but its prevented me from beating screen 2 a bunch.