Congrats on the new pb and 2nd place run in Arcade, Nate! Seeing the run and how arcade runs have been timed, I wanted to ask you, Polly, anyone who knows, as I'm the newcomer; is there a reason that you guys have your rule set as time ends when player control is lost, as opposed to when the final hit is delivered on the boss? I should clarify that I've been timing my Original runs with this final hit as the end point. To me it would make plenty of sense to use this rule across both modes, but since there are a multitude of runners with established times in Arcade I wanted to discuss first, and I would certainly be comfortable tweaking my time if that makes things more cohesive.

Just had a run end cos of the tanks at the end of L7, very irritating because I was destroying it before that without farming for medals and much better technique on bosses. I was 13 seconds ahead of myself, so it seems sub 17:00 is eminently possible with more tuneups. Most of the improvement came solely from more efficient handling of the jeeps boss in 5, and the pipes boss in 6 - if you avoid crippling your health bar at these points and save enough M's for the battleship, there's no real need to save medals for 8, and in fact I may be able to cut a few out after more runs; saving a few more seconds. I seem to have immense trouble with the turrets at the end of L2, so it might work better to skip the M bomb and 5medal there since an M bomb on the boss there is utterly wasted.

Hey nate, no it's just still uploading now cos I have the upload power of half a potato xD Should be finished imminently.

Upgauges add 4 max health to any character, Rifle's max is 48 from a start of 32; Laser starts at 36. When hit by a bullet or grunt, you have 90 frames invincibility which includes the 20 you are frozen in place. The level 4 dinghy's speed is not great, will check (then edit this) whether it's actually better to just walk through the water if constantly vertical, but the jeep in level 5 is essential; its top speed is 12% quicker than the character. And to clarify from earlier, moving h/v in water gives 2/5 null frames, but moving diagonally gives 3 in 6, so do not do that xD

Okay! I submitted my first good run of the game having culled a lot of info on stuff. Horizontal or vertical movement are optimal, purely diagonal (45 degree) movement adds one null frame every 5. Water movement adds 2 null frames in 5 and is to be avoided. Turns out Laser doesn't really need beefing up with Power units - his base firepower is 11 damage every other frame, this can only be increased to 14 which only tends to make a difference if you use him on later bosses where you'd use M bombs anyways. Being hit by a bullet or grunt costs ~20 frames so it's always worth jinking to avoid this unless you deliberately want the invincibility, which is a more advanced part of route plans I am yet to look at. In time I'm going to map the game based on position of essential pickups, and medals related to that - basically the rule is, if medals are within the rectangle drawn from where you are, towards where an essential pickup or progress point is, it's ok to grab them so long as you don't sacrifice vertical speed. The only other consideration is that if you are going to move diagonally, it's best to do it all in one motion, rather than constant little jinks where no enemies are involved - the game pauses your diagonal movement on the 2nd frame of 5 so if you're constantly fiddling about, you'll lose more time than if you just go straight across and up. I'm sure there are areas in my run where I flout this rule but I'll make plenty of improvements as I go. If I multiply ntsc by 1.2 as a rough guide, compared to the run I posted from someone else earlier in the thread, I'm now faster than he was in levels 4/5, latter losses are mostly down to being rusty on bosses and losing too much health, which accounts for ending up a minute and a half down on him comparatively in level 8. So already my run has some points of being superior to anything done before, which I'm proud of. You can watch the run at

I made a good start on lots of health and speed values. I can post some values here or create a document with stuff in as you prefer. I remember I had a question though: is there a specific reason why rapid fire disabled is part of the rules? It strikes me as strange since it's part of the options menu by default ie not a cheat nor something that could only be achieved with a turbo button.

Lol no I mean why he doesn't use Laser at other points as well - it might be that Rifle's ultimate power is close enough that his extra speed works out more beneficially. I'll ask in the general forum in case anyone knows of specific issues that might affect this game re emulators. Once I got the hang of ram searches/watches in emuhawk I'll post all the info I glean here ^^

I've already seen a pretty blistering unofficial speedrun at - but using rapid fire so not countable, lucky for his fingers. I will have to investigate why he disavows using Rifle for segments - it's certainly way quicker than my old mark on vhs. I asked Adelikat who made a TAS of Arcade mode but he lacked the info. Are there preferred emulators which are known as more accurate, or conversely, any to avoid using?

Noticed no submissions for Original, wondering have any of you ever tinkered with it or tried speedrunning it, and might have info about weapon attack values and such - it'd be easy enough to use ram watches and work it out myself but if you have anything already known about original mode it'd be a great help. Is there a preferred emulator for this?

My gosh, you people are hostile xD I did talk to him, it wasn't something intentional he did; and to reiterate, a large part of the reason I posted the thread was in case there are any people who played the game before casually who might be able to shed light on the subject.

Hiya, I have a submission from a runner playing on emulator, but which features a glitch unknown to me since I didn't grow up with that console's version - on 2 occasions during run, rather than as usual having to kill a specific enemy for a required scroll, the scroll appears at random, and in a position incongruous to where he has been shooting at.

My inclination as the moderator for that game is to disallow the submission until the cause is established; whether it's a bad disk image, or something that happened to other players on Amiga consoles in real life too. While I have no reason to distrust the player, and he is the only player for the category currently, it feels way too important to let by until researched.

Any advice on this conclusion would be great, and more importantly, if any of you ever played Amiga Flimbo in your youth, did you ever have this happen to you? If you know anyone outside the forum who had this version I would really appreciate this being passed on, since I'm not the most technical guy and wouldn't know where to start.

EDIT: Have requested the game be tracked as a separate entity, happy to mod/verify for you :)

@btrim, still interested? Didn't hear from you, no problem if you're busy as I've paused development while strackel has a play but would love your feedback too ^.^

Great, email me at daniel.a.odowd@gmail.com and I can send you some stuff!

Hi, I was advised to post here - I'm making an Alex Kidd in Miracle World ROM hack with the help of the kidded tool http://emulicious.net/kidded/ and I was wondering since speedrunners are typically very able gamers, would anyone be interested in playtesting a few level designs to help me calibrate the difficulty? The main difference in the game compared to the original is more fluid physics; squats are no longer slow and drawn out, most importantly I beefed up the jump a lot to enable long jumps, and even something approximating wall kicks, like in the mario series. I would like at the very least to be able to release something that will challenge excellent gamers, whether or not I also make something for the casual fan. Let me know if you're interested!

Heh, didn't think of that, but I did contact many of the site's akimw runners through youtube so I should get some input from them and others :)

Hi, I'm making an Alex Kidd in Miracle World ROM hack with the help of the kidded tool http://emulicious.net/kidded/ and I was wondering since speedrunners are typically very able gamers, would anyone be interested in playtesting a few level designs to help me calibrate the difficulty? The main difference in the game compared to the original is more fluid physics; squats are no longer slow and drawn out, most importantly I beefed up the jump a lot to enable long jumps, and even something approximating wall kicks, like in the mario series. I would like at the very least to be able to release something that will challenge excellent gamers, whether or not I also make something for the casual fan. Let me know if you're interested! If you've speedrun AKIMW or similar platformers before you would be perfect to try this out.

Lol, wow, didn't realise speedrunning was so much in the dark ages even after fifteen years! Guess my wr is safe then nvm xD

Hi all, am new here!

I speedrun an old C64 game Flimbo's Quest and there are only two of us I am aware of who currently do this. Recently I discovered through memory watching and such tools that the game has item drops similar to Legend of Zelda [on a counter in each level] but more interestingly, that the rng for target enemies is very arcane. Since that discovery I've halted my speedruns because it's going to turn the whole game upside down.

-Game has a memory byte whose value runs between 00 and FF. -At the start of each level there is a fixed uneven distribution so that each value corresponds to a set enemy, and some have higher frequency. There are hot spots for particular enemies or groups. -Each time an enemy is pulled, all slots in the memory with it in, are supplanted by n-1 in the order.

The byte's incrementation is not fixed, I am working on mapping this because it means that there are 'dead' zones on each level where standing in a particular location can freeze or slow it. Because of the distribution this also means that one can sort of manipulate the byte to give higher probability of better enemies.

This manipulation can be achieved by normal means but much, much less accurately because it is not directly correlated with in-game timer. You would know the enemy targeted but have no clue where in the distribution you were, and so further manipulation would be sheer guesswork.

I am curious whether any other games have a feature like this where being able to view the memory [either a. only the byte's value between enemies, or b. the changes during gameplay] makes such a big difference to runs, and further, what those games' positions are currently on the rules regarding memory viewing of any sort, during speedrun attempts.

I am currently inclined to think that it would be reasonable to permit runners to view the static value between enemies, but not to see the changes during play. This would enable you to know what pocket of the distribution you landed in, but would mean you still rely on your skill to manipulate it. Thus the game remains outside of TAS bounds to my understanding. That would be as opposed to having the ability to mess around on screen until the hot spot approached in view.

Opinions, info would be greatly appreciated, the more info I have before consulting the game's other runners would be very helpful.