Couldn't resist... 
[Roddie's Normal]
01:09:651 (1,2) - could you allign them to be parallel? Somehow the different orientation makes me cringe.
01:57:938 (4,5) - probably better to turn these into a slider. This is the easiest diff of the set and something like this might throw beginners off too easily.
[Hard]
00:05:541 - end the break here since you don't resume the mapping on the downbeat (drag break ending to the downbeat)
00:21:979 (2) - maybe use new slider algorithm to make a perfect blanket, the current one looks really bad (also re-adjust 00:22:801 (4) - afterwards)
01:33:281 (1,2) - this feels very strange. The slowdown with the lowered SV works perfectly but the low spaced circle feels out of place. I'd rather see a repeat on (1), (2) needs to be removed before. Also don't use a NC here as it is the same phrase and a consistent combo works better.
01:35:130 (1,2) - is the different orientation intended?
02:00:199 (3) - maybe use the new slider algorithm here to make the blanket more fitting...it kinda stands out as not as perfect as the rest
03:03:075 (8) - consider adding a NC here to indicate the new phrase
03:13:760 (5,6,7) - messed up spacing. Surrounding pattern is symmetric, why don't you center the stack in the middle of 03:12:938 (1,2,3,4) - @268/200? Having it centered in the middle of the playfield doesn't make any sense here unless the previous pattern is centered as well.
dubstep kitty says:

Anyway the set is really fun to play alrady and all you have to do is level out the last bumps. Have a star for that.
[Roddie's Normal]
01:09:651 (1,2) - could you allign them to be parallel? Somehow the different orientation makes me cringe.
01:57:938 (4,5) - probably better to turn these into a slider. This is the easiest diff of the set and something like this might throw beginners off too easily.
[Hard]
00:05:541 - end the break here since you don't resume the mapping on the downbeat (drag break ending to the downbeat)
00:21:979 (2) - maybe use new slider algorithm to make a perfect blanket, the current one looks really bad (also re-adjust 00:22:801 (4) - afterwards)
01:33:281 (1,2) - this feels very strange. The slowdown with the lowered SV works perfectly but the low spaced circle feels out of place. I'd rather see a repeat on (1), (2) needs to be removed before. Also don't use a NC here as it is the same phrase and a consistent combo works better.
01:35:130 (1,2) - is the different orientation intended?
02:00:199 (3) - maybe use the new slider algorithm here to make the blanket more fitting...it kinda stands out as not as perfect as the rest
02:23:623 (1,2,3,1,2,3,4,...) - wrong timing in the dubstep part! Can't blame you - dubstep usually can't be described by normal musical rules. This is probably too complicated to do "right" because the arranged 146BPM layer is still used in the background together with something else for the noises. As you follow the other layer(s) (with whatever BPMs are used) with some notes you should use the correct timing for the respective notes OR don't follow them at all:
02:23:623 (1) - this slider is too fast because the wobble is gradually getting slower. Snapping to 1/16 would fit much better but you should never use 1/16 snap. So a new timing section hat just affects the slider with the following circle is needed. You have to reset the timing @02:24:445 - and @02:25:267 - to be in the correct measure again.
02:25:062 (1,2,3,4) - (1,2) seem disconnected and not very intuitive as they don't follow anything dominant. I can see three possibilities here:
1) you delete the two circles and start the combo on the downbeat with 02:25:267 (3) -
2) you add a circle on 02:24:959 - to match the (very feint) whatever-that-sound-is with a complete stream (probably needs a different shape for the previous slider)
3) you follow the "lighter wobble". You'll need a new timing section beginning with 02:24:445 (3) - and end the slider on the first wobble timing the following circles . Reset timing on 02:25:267 - for correct measures. The jump back to the regular timing will be hard to read so I suggest using a kickslider for this instead of the circles.
02:23:623 (1) - this slider is too fast because the wobble is gradually getting slower. Snapping to 1/16 would fit much better but you should never use 1/16 snap. So a new timing section hat just affects the slider with the following circle is needed. You have to reset the timing @02:24:445 - and @02:25:267 - to be in the correct measure again.
02:25:062 (1,2,3,4) - (1,2) seem disconnected and not very intuitive as they don't follow anything dominant. I can see three possibilities here:
1) you delete the two circles and start the combo on the downbeat with 02:25:267 (3) -
2) you add a circle on 02:24:959 - to match the (very feint) whatever-that-sound-is with a complete stream (probably needs a different shape for the previous slider)
3) you follow the "lighter wobble". You'll need a new timing section beginning with 02:24:445 (3) - and end the slider on the first wobble timing the following circles . Reset timing on 02:25:267 - for correct measures. The jump back to the regular timing will be hard to read so I suggest using a kickslider for this instead of the circles.
03:03:075 (8) - consider adding a NC here to indicate the new phrase
03:13:760 (5,6,7) - messed up spacing. Surrounding pattern is symmetric, why don't you center the stack in the middle of 03:12:938 (1,2,3,4) - @268/200? Having it centered in the middle of the playfield doesn't make any sense here unless the previous pattern is centered as well.
dubstep kitty says:
Anyway the set is really fun to play alrady and all you have to do is level out the last bumps. Have a star for that.