SDA logo

TMNT 4, originally a 1991 arcade game, was ported to SNES in August 1992 and closely follows the formula of TMNT 3 for NES. The twist comes when Shredder sends the turtles back in time, where the foot clan is already there to fight them. Battle Bebop and Rocksteady in the past, Krang in the future, and finally return to the present, 1992, to fight Shredder and recover the stolen Statue of Liberty.


Note: the game's timer at the end isn't used because it is buggy.

Best time as Raphael: Hard mode 0:20:20 by Mike Uyama on 2012-07-17.

Get Flash to see this player.

Author's comments:


It was my destiny to run this game. Not sure why Don is on that shirt though.

For anyone who wants to see the live stream of the run

Part 1:

Part 2:

"I will get around to improving this run someday, just not anytime soon."

-The last sentence from my 2006 TMNT4 run comments

Well I certainly didn't lie with that sentence because I didn't run this game again until 6 years later.

I didn't start to run this game because Hanzo started to run it and found some new strategies, nor did I start to run it because I practiced the game for marathons. Both of those reasons would be a perfectly logical transition into running this game, but the reason why I really got back into running the game lies in two streams.

It would also make sense if I started running the game because Sinister1 was streaming it and planning on improving my 2006 run. I did start playing TMNT4 a bit when Sinister1 was streaming it.

I really started running TMNT4 because I was a featured guest on The Sunday Sequence Break, a speedrunning talk show that features a game and the two best runners or most knowledgeable people about running said game. Despite TMNT4 being sloppy seconds and an emergency measure for the show :P, I decided to show some footage of new boss strats that have been discovered since my 2006 run. In doing so, I actually became motivated to run this game again because I realized there have been so many time savers and discoveries since 2006.

It took me three months of streaming 1-3 times a week, a marathon, and 525 attempts to get this run, but it was definitely worth it. Unlike my old 22:31, I am quite satisfied with this run. That isn't to say the run is perfect, because the game is hard to optimize, but I don't think I'll be trying to improve this game for a while.


Everyone who watched my stream: Streaming this game is a lot more fun than doing recorded attempts in solitude. Maybe my concentration wasn't 100% because I streamed, but the same applies to my frustration. :P It was always fun chatting to my stream in-between and even during attempts.

Hattori Hanzo: For finding some strats, most importantly, the one-round Slash kill.

Nitsuja: For making the TMNT4 TAS. It might be old (2006) and outdated in some parts, but I still learned things from the TAS that I didn't use in my old 2006 run. Also, he gave me enemy life and turtle damage values, both of which are incredibly useful and helped me plan out this run and my previous run.

Sinister1: For finding some better ways to deal with certain areas such as the second wave in the technodrome elevator, which I screwed up. :P He was also a good guy to discuss strategies and just general facts about running the game. I'm not so sure about his abilities as a coach, but he wasn't too bad at that. ;)

Oh yeah, since I'm such a nice guy, I'll give you 2% credit for this run, Sinister! Hey, that's twice as much credit as I'm giving The Sunday Sequence Break! :P

The Sunday Sequence Break: For having me as a guest on their show, which remotivated me to run this game.

Since I'm being a nice guy for Sinister1's sake, I can only credit 1% of this run to these guys. You claim I signed a waiver giving you guys full credit for my run, but I use my dictator powers to declare that waiver null and void.

Oh yeah Flicky, if you're so tired of TMNT4 runs appearing on the show, then you shouldn't have had me on the show in the first place! Consider it revenge for forcing an awkward filler segment during the TMNT4 episode.

Splinterface: For always keeping the hype strong and pissing off flicky, both good things!

Duckfist: For making the most moving tribute to the Splinterface possible. You are a champion among men.


The obligatory InFrequently Asked Questions (IFAQ). Or maybe it should be Never Asked Questions (NAQ).

Q: Hey, it's been so many years and you STILL haven't run the game with any other turtles. What's up with that?

A: Having practiced other turtles for marathons, I realized one thing, THEY ARE SO BORING TO PLAY! Seriously, the fact that Raphael is so fast and powerful makes the run more entertaining. Raphael's speed and power lets him skip enemies no other turtle can, and the other turtles are forced to slam throw a lot more than Raph because their other attack options are less powerful.

In other words, a Raphael run is the best because it showcases one of the best things about the game, the large and varied moveset. Any run as a different turtle would devolve into a slam throw fest.

That being said, the only other turtle I'm considering running the game with is Donatello, and that's only because he is faster against a couple of bosses (namely Baxter and Super Krang). Don't expect a Don run anytime soon though.

Q: So you say other turtles would have to resort to slam throws more. Wouldn't slam throws be faster with Raphael too because slam throws are an instant kill?

A: Not really. There's two factors that limit the effectiveness of slam throws.

1. Time - The slam throw animation takes a long time to complete, and after you're done slamming the foot soldier, it bounces up a bit before falling and exploding. You can reduce the number of times you slam an enemy by hitting them more, but that also takes time.

2. Luck and enemy AI - Your success in slam throwing depends whether the enemies run into your slams. On the same screen, I'll sometimes be able to hit every single other enemy, and other times, the slam throw will only hit one or even no enemies.

Speaking of enemy AI, if a blue sword foot soldier is outside of slam range, then it will almost never run into your slam throw. They will just patiently wait outside of slam range.

Q: Okay, so what makes Raphael so much more effective than the other turtles when it comes to attack options?

A: The biggest factor is that Raph is the only turtle who can kill purple foot soldiers and other enemies with equivalent health (like Roadkill Rodneys) in one special or dash special, which gives him a lot of flexibility and mobility over other turtles, because he can kill foot soldiers while scrolling the screen with a dash special.

All of the other turtles have to use a flip shoulder tackle (or cartwheel tackle) to kill a foot soldier in one hit. The flip shoulder tackle not only requires a lot of time and space to set up, but when you hit a foot soldier with it, you stop in place, making it difficult to kill multiple foot soldiers with the tackle. And yes, this is why the other turtles have to resort to slam throws so often.

Q: Why didn't you reset when you had to collect the pizza in stage 3? You said in your last run that it saves time not to get any points in the bonus stage.

A: I lost maybe 2 or 3 frames at most. I'm not resetting over something that insignificant in a game where you can gain or lose whole seconds from luck.

Basic facts about running the game

As I mentioned in my IFAQ, Raph's special and dash special are used a lot in this run because they are very strong and kill purple foot soldiers and equivalent health (16 hp) enemies in one hit. It also helps that Raph's special has an enormous hitbox, the real range on Raph's special is about 50% longer than it looks.

Optimizing this game is very difficult because of how many options you have, and how random enemies are. Any strategy that involves a slam throw involves at least a small degree of luck. I think I did a good job of optimizing this game, thanks in no small part to people who have contributed in the past such as Hanzo, Sinister1, and Nitsuja. However, I'm sure there's at least a couple of spots where I could be significantly faster because there's a better strategy I didn't think of using.

How difficult is this game to optimize? I think there's actually one or two strats that are slower in this run than in my 22:31. I just didn't realize the strats were slower until I watched my 22:31 a couple of days ago. No, the new strats are not much slower, I think I lost two seconds max from the slower strategies (if they really are slower).

It's sometimes very difficult to determine where time is gained or lost in this game. Anyone who watched my stream can attest to how many times I was confident I lost or gained time, and then the complete opposite happened. Sometimes it's random luck factors that determine how quickly you completed a stage.

Another thing to be aware of when running this game is that a small mistake that would only cost 1 second or less can snowball into a much larger mistake that costs 4-5 seconds because of how spawns work. A good example is what happens near the end of stage 5 with the bomb throwers and pink shield foot soldiers. I lost a lot of time because the bomb throwers didn't end up killing both pink soldiers that appear from the bottom.

As a general rule, if there's only one purple foot soldier on the screen, the next set of enemies will spawn, but if there's a foot soldier of any other color that's still alive, then the next set of enemies won't spawn.

Stage-by-Stage Analysis

Stage 1:

Split this run: 01:16.21
Personal best: 01:14.12 (My actual personal best is around 01:13.5x, but I didn't save it in my wsplit)
Time difference: 2.09 seconds

So funny thing about this stage. When I re-watched my 22:31, I think the way I handle the first 18 or so seconds before the barrel might be faster in that run than in this run. If my old method is faster, then it's faster by a half a second or a second, which is pretty significant at the level the run is at right now.

This stage 1 split is around the border of what I would accept for a submittable run. My execution was great this stage, but there was one thing that cost me a lot of time near the end. At the very end of the stage, a blue sai foot soldier dodges my dash special, and I'm not able to kill him until 2 or 3 seconds later. That blue sai foot soldier is significant because like I said earlier, new enemies can't spawn unless if all non-purple foot soldiers are dead in the final area.

My Baxter fight went without a hitch. The strategy I use in this run is faster than my 22:31 because I kill Baxter when he's on the ground, which makes him die and explode faster than if he's in the air.

Stage 2:

Split this run: 01:34.33
Personal best: 01:31.80
Time difference: 00:02.53
Total time: 02:50.54

Despite being 2 and a half seconds away from my personal best, my stage 2 time is quite good. There's a lot of places where I'm faster than my 22:31 run. One of the biggest spots is near the beginning, where I get a lucky slam throw and am able to skip a bunch of orange foot soldiers who normally spawn on the left. This good luck compounds and I'm able to spawn the first set of Rodkill Rodneys instantly. I wouldn't be shocked if I saved 10 seconds over my 22:31 in this area alone.

Honestly, the rest of this stage was quite smooth. The only two mistakes in this stage were hitting a Roadkill Rodney slightly late during a second spawn, and hitting the last nunchaku foot soldier twice before specialing. I'm pretty sure both of these mistakes combined add up to less than a second, so it's pretty insignificant. I'm honestly not sure how I was 2.5 seconds faster on my best split. Time loss in this game is so mysterious. :-/

The Metalhead fight is exactly the same as it was in the old run. There's really no way to improve the fight in a significant way.

Stage 3:

Split this run: 02:03.22
Personal best: 02:02.77
Time difference: 00:00.45
Total time: 04:53.76

Honestly, I'm not sure why I listed those splits, it's not like you can do much to speed up this stage. I have managed to lose a second from a bad Rat King fight though. :P

What's funny is that out of my 525 attempts, I've only been hit in this stage during 3 or 4 runs, which is less than 1%. I think it's pretty funny that my successful run is in that less than 1%. You could just say the hits add entertainment value. :D

The Rat King battle is exactly the same as it is in the 22:31. I've tried the TAS strat before and it's too difficult to line up the jumping chop.

Technodrome first half:

I like to split up stage 4 into two halves because it's a long stage with two very distinct areas, both of which can be fickle time-wise.

Split this run: 01:21.51
Personal best: 01:20.36
Time difference: 00:01.15
Total time: 06:15.27

So a big improvement over the last run is how I handle the first area with the Roadkill Rodneys. Not only did I execute well and not choke, but I also managed to minimize the dash specials needed to take out everything.

My luck here is insanely good. I managed to skip a lot of enemies and had good slam throw luck for all of my slam throws. It would have been perfect slam throw luck if the blue sword guy slam throw hit the sword guy on the left. I could have tried to skip the remaining sword guy, but if he followed me I would have lost a lot of time, so screen throwing him was a better idea. The remaining blue sword guy didn't really affect my time much, though that's probably why this split was a little over a second slower than my best.

Honestly, there's not much to say here because this was a great first half of the technodrome.

Technodrome second half:

Split this run: 01:31.98
Personal best: 01:26.56
Time difference: 00:05.42
Total time: 07:47.25

The biggest factor determining your time here is the Shredder fight. You can execute perfectly on the elevator, and then have it all wiped out and then some from a crappy Shredder battle. Actually, I think my best split here has a pretty crappy elevator, but my Shredder fight had TAS levels of luck.

That being said, I definitely lost some time on first two phases of the elevator. I guess my positioning was off on my flip tackles, but honestly, I've never managed to figure out what determines successfully hitting all of the foot soldiers with a flip tackle. I did manage to get all of the blue sai soldiers in one dash special, which made up a bit of time, but not enough to offset how badly the first two phases went.

My Shredder fight was quite good. My execution was mostly good, and for the most part, I was lucky with the foot soldier spawns.

Stage 5:

Split this run: 02:41.91
Personal best: 02:34.57
Time difference: 00:07.34
Total time: 10:29.16

This stage and starbase are my worst stages. My biggest time loss here isn't really my fault, it's more of a luck factor screwing me over and costing me at least 3 seconds.

The first mistake I make is how I deal with the pterodactyls. It's almost always faster to let the last pterodactyl fly to the right before moving to the right because if you scroll the screen to the right too early, then the pterodactyl follows you and you end up wasting time waiting for the pterodactyl to move off-screen. BTW, the only time I've seen anyone outrunning the final pterodactyl so you don't have to wait for it is in Nitsuja's TAS.

This stage is the beginning of what is possibly the biggest improvement over the 22:31 - the way I handle stone warriors (or worriors :P). In my old run, I fail to group pretty much every stone warrior encounter, and horribly screw up some of the fights. In this run, I'm successful at grouping all of the stone warrior spawns so I can take care of two at a time, which makes a huge difference. I wouldn't be shocked if I gained 30 or 40 seconds from how I handled stone warriors in this run.

So the second big mistake is more bad luck than anything else. After I collect the bomb pizza (the one that makes me spin), I get some bad luck with the bomb throwing soldiers, and only one of the pink foot soldiers is hit by the bombs. Usually at least 2 out of 3 are hit. This bad luck meant I killed fewer enemies than usual by the time the stampeding dinosaurs spawned. Why would that matter? Well if you kill enough enemies before the dinosaurs spawn, then two dinosaurs will stampede at once, which reduces the amount of stampedes you have to wait for from 3 to 2, and each stampede costs about 3 seconds. So yeah, I lost time from some dumb bs luck. :-/

As fast as this new Slash strat looks compared to the old one, it saves a whopping one second. The big kicker is that even though Slash doesn't get a chance to attack, I have to hit him many more times than the older strategy, which eats up a lot of the time gain. Still, a second gained is a second gained.

Stage 6:

Split this run: 01:59.33
Personal best: 01:56.06
Time difference: 00:03.27
Total time: 12:28.49

This was a very solid run of this stage. Most of the time lost was from bad slam throw luck and wasn't my fault.

The two big spots where I had bad slam throw luck were my very first slam throw, and the slam throw to take out the purple and pink foot soldiers. I lost more time from the second slam throw, because it delayed the screen scrolling long enough so I couldn't skip the three purple foot soldiers on the left. Fortunately I took out all three purple foot soldiers with one special, so I didn't really lose much time.

Again, my stone warrior fights were way better than my old run. I also knew how to walk all the way to the right to spawn them this time. :P

The big difference in the Rocksteady and Bebop fight over the old run is that I use a lot more special to not only make the fight faster, but using specials also reduces my life for the score countup at the end of each stage.

Stage 7

Split this run: 01:47.03
Personal best: 01:47.03
Time difference: None! This is my personal best.
Total time: 14:15.52

Yes, this stage is the one personal best I got during the run. Getting a personal best during a run is actually quite lucky, literally, because you do need good luck in a stage to get a personal best.

Honestly, my execution was on point, I had great slam throw luck, and I tore through the stage. The only thing that is somewhat debatable is how I handle the screen right before the final set of stone warriors because I dash special so much, but the way I handled it makes sure a foot soldier doesn't follow me, which would cost a lot of time.

What's funny is that the one spot where I lost time is Leatherhead because I let him tail whip, which almost never happens. The tail whip costs somewhere in between a half a second and a second, and who cares? This stage was great! The only tragedy is that beating the stage fast means less time spent listening to the awesome music.

Stage 8

Split this run: 02:23.15
Personal best: 02:22.83
Time difference: 00:00.32
Total time: 16:38.67

Again, just posting the splits for posterity. There is more finesse to the Super Krang battle than Rat King, but times still shouldn't vary by more than a second here.

This new Super Krang strategy saves about 2 or 3 seconds over the old one because it reduces the number of times Super Krang travels across the screen. Sometimes you'll need to hit Super Krang one extra time after burning through all of your specials, but I didn't have to do that this time. I have no idea what determines whether you have to hit Super Krang an extra time.

Stage 9

Split this run: 02:52.88
Personal best: 02:48.55
Time difference: 00:04.33
Total time: 19:31.55

As much as I hate it, this stage is a HUGE factor in determining your time. There are three reasons for why it's such huge factor.

1. The sheer length of the stage - It's the longest stage in the game with a ton of enemies, which means there's a lot of things that can go wrong.

2. A lot of luck factors in the stage - Get bad luck on robot walkers and you can waste 5+ seconds trying to kill them so the next set of enemies will spawn. A bad slam throw can delay you 2 or 3 seconds, and a good one can save you just as much time, or more. However, there's one luck factor that is so ridiculous and stupid it needs mentioning.

3. KRANG! - This is such a huge factor in determining a good time, and it's stupid because it's mostly luck. This is the worst boss in terms of randomness and luck. AFAIK, there's no way to manipulate where Krang spawns, and you just have to pray he spawns at a good height. Bad Krang luck can cost 10 seconds or more.

I won't lie, I was nervous going into this stage because usually at this point I was in the red on my splits when I arrived here, but this time I was ahead by 4.2 seconds, which is huge in this game. I definitely made some mistakes I normally wouldn't make in this stage, but at the same time, I did a decent job in this stage and didn't make any huge mistakes.

My execution before the first pizza was pretty dodgy. I screwed up some dashes and some slam throws. Fortunately, I had the presence of mind to make up for failed slam throws by going special crazy, so I didn't lose too much time from screwing up the slam throws. I even choked up a slide and did a really high hop kick instead.

Fortunately, by the time I collected the first pizza, I collected myself and executed really well for the rest of the stage. I managed to pull off every strategy successfully for the most part, and I got a key bit of good luck. Near the end of the stage where the lasers are there's an enemy spawn has a bunch of blue and orange foot soldiers. You can sometimes slam throw one and it will kill the rest, which saves a time of time on the next screen because it means the yellow bomb throwers instantly spawn. I didn't quite kill all of the foot soldiers in one slam throw, but I was close enough because I managed to skip the last one, and the bomb throwers spawned right away! That bit of good luck probably saved 5+ seconds, because taking out the bomb throwers early means the stone warriors at the end will spawn right away.

You might notice there are a lot of mistakes in my starbase and wonder how my personal best is only 4.33 seconds faster. One reason is Krang luck obviously, but that's far from the only luck factor in the stage. There's also factors such as the four purple soldiers spawning instantly (which happened this run) or if a robot walker runs in and prevents that from happening. To be honest, there are so many luck and execution factors in starbase that getting a perfect run of starbase is very difficult, if not nearly impossible.

My Krang luck was pretty good! Four passes is definitely a good battle, and most of the places where I lost time were from misjudgment on my part or poor execution. Fortunately, on the third pass when I missed my jump kick, I was actually pretty fortunate because I was able to hit three out of four robot walkers in one attack. When three out of four robot walkers are down, Krang spawns again. Krang's fourth spawn was low enough for me to special, which definitely worked out well. The irony is that if I actually hit Krang during his third spawn, my fight would have been slower.

Super Shredder

Split this run: 00:48.85
Personal best: 00:45.27
Time difference: 00:03.58
Total time: 20:20.40

This was a very safe Super Shredder fight. I could have actually been close to my personal best, or even beaten it had I taken advantage of my good luck, but I just wanted to finish the run and not choke it up on Super Shredder, which I fortunately managed to do.

What was the good luck in the fight? It was how far to the left Shredder went after his first attack. I had an easy chance to hit Shredder because he was right next to me, but I didn't want to choke it up, so I played it stupidly safe and didn't take advantage of my good fortune. Yes, I was that nervous.

Anyway, my Super Shredder strategy baits him to the left side of the screen so I can go nuts with dash specials, which is a strategy improvement over the 22:31. I also make sure to time my specials so they go through whichever attack Super Shredder chooses, even the flames, which linger for a while and normally make it difficult to hit Shredder.

One last thing to note is that where you kill Super Shredder on the Z-axis is also important. If you kill him at the bottom, it's about 3 or 4 seconds slower than if you killed him near the top because he has to walk all the way to the top before dying. I killed Shredder in the middle, so I didn't really lose or gain much time from his death animation.

Seeing my time I was very happy to be done with the run. :) 20:20 is definitely a solid time that isn't easy to beat, and I beat my 2006 run by over 2 minutes. In theory sub-20 is possible, but with my current strats you would need some insane luck for that to happen, and I had some great luck this run. I'm not saying sub-20 isn't possible. I'm sure there are better strategies for handling some areas than what I used, but this is a great run and I'm not going to bother looking into new strategies anytime soon, if at all.

Anyway, if you enjoyed this run (and these long comments), then please check out my stream.

P.S. I'm actually faster than the TAS in some spots now, but I can't get the slam throw luck the TAS gets. :P If any TASer is interested on working on this game, I'll gladly help. Just PM me on the SDA forums.

Return to the Game List, the FAQ, or the Home Page.