(Link to video)
Goals:
  • Aims for fastest possible time
  • Uses hardest difficulty
  • Contains speed/entertainment tradeoffs
    • Brightness is set to maximum for better viewing
IMPORTANT: This TAS does not take advantage of lag clipping, a major glitch that would save several minutes off of this time, because there is no emulator that can faithfully recreate this glitch. While this time is several minutes ahead of the fastest real time speedrun, there are a few points where I uncontrollably lose time to RTA (examples being the levels Final Outpost, Watch Your Step, and Breakdown.) Ultimately, it is up to a judge to decide whether this invalidates the run from being publishable.
General Information:
Doom 64 is a spiritual successor to Doom 2 on the Nintendo 64, made by Midway Games and published by id Software. It features 32 new maps, as well as new graphics and sounds on a modified version of the Doom engine (which is built around the Atari Jaguar port's codebase.) It is also the only Doom game to feature the Unmaker weapon. While monsters such as the archvile and the chaingunner have been removed due to low memory capacity, there are also two new monsters: The Motherdemon and the Nightmare Imp.
This run plays on the hardest difficulty: Watch Me Die, the equivalent of original Doom's Ultra-Violence. A seperate category exists for the lowest difficulty, Be Gentle, but I believe that this category is not different enough, nor entertaining enough in a TAS setting to warrant its own TAS. The only notable differences in Be Gentle are less monsters, slightly different routing in certain maps, and double ammo.
Notable differences between Doom 64 and other Doom games:
This will document everything there is to know that makes Doom 64 different between other Doom games in a speedrun/TAS setting.
Usage of speed modifiers (SR40/SR50)
Doom 64 is not designed to be played on a mouse and keyboard, but the weird octopus-styled Nintendo 64 controller we all love to hate. Unlike most games, however, the further you're pushing up on the control stick, the faster you move forward. The same can be said for backwards movement as well. However, the N64 analog stick has ranges far greater than the physical stick can allow (the normal limit is approximately 90, -90 for both the X and Y axes, where the real limit is 127, -128). As a result the console allows inputs greater than what is supposed to be possible on the analog stick. This allows us to get to speeds far greater than expected. In RTA, this involves abusing the calibration of the controller on startup, or manually calibrating using a set of inputs. Setting the neutral position to 90 forward allows access to inputs beyond the standard range while still using a standard N64 controller. In a TAS, however, this is irrelevant as we have access to the full analog range at all times. Therefore, it is important to categorize this trick differently for TAS.
Essentially, we get the full benefits of Neutral Reset but even greater. To make things easier to understand, I will be defining strafe running the same way you would with PC Doom. Holding fully up on the control stick and strafing gives a speed boost equivalent to SR40 speed. Additionally, holding down the button bound to "strafe on" (default: down C) locks your turn speeds, allowing the equivalent of SR50 speeds. As far as I know, there is no way to maintain this while turning, so I frequently have to default to SR40 again to make corners.
Player/Enemy size difference
All enemies are bigger than their PC counterparts, including Doomguy himself. Because of the latter, there are no places where the player can glide (exactly fit through an area the width of Doomguy's size) in any maps of the game.
MACRO scripting
MACROs are a unique, albeit basic and Doom-centric scripting language designed specifically for Doom 64. They offer far less flexibility than other Doom engine based languages such as ACS, but more flexibility than any other original Doom title. Only one MACRO script can be run at a time, which is important because this means that we may be temporarily unable to progress through a level until the MACRO runs out (example: end of MAP05). MACROs are used to call basic and advanced actions, sometimes in a designated sequence in order to change the layout of a level, spawn enemies, operate bridges and more. Sometimes the level will offer triggers that will terminate the MACRO that is currently functioning (example: MAP15), but this is often not the case.
Other notable differences in map design:
Fake bridges
Bridges work based on MACRO scripts that are often based on line triggers. These triggers raise and lower respective floors and ceilings out of our sight so as to give a fake-3D effect. The triggers can be skipped, but this isn't useful as it just ends up blocking our path.
Other/miscellaneous key differences between PC Doom and Doom 64:
  • Rocket jumping no longer works
  • Delay after hitting exit switch (for nearly all maps this delay is 15 frames long)
  • Linedef trigger skipping is no longer directional
  • Wallrunning to get extra speed no longer works
  • Archviles, chaingunners, revenants and spider masterminds have been removed due to the limited cartridge space
  • One of two projectiles can be spawned by the game at will (revenant rockets or darts)
  • Lost souls have reduced health but are extremely aggressive
  • Pain elementals can now spawn two lost souls at a time, and attempting to block them will result in you taking explosive damage that can kill you almost instantly
  • The SSG and rocket launcher have a small amount of knockback, which increases your speed when going backwards, but not enough to make a huge difference over the course of any map
  • The plasma rifle, while having virtually no cool down time, is highly nerfed, making it virtually useless in a run
  • The Unmaker weapon, which can be powered up by grabbing three demon artifacts across three of the secret levels. Currently useless in an any% speedrun
  • Additionally, grabbing the aforementioned demon artifacts allows you to exit the final map more quickly, but this is slower overall
  • Lag clips allow you to voidwalk where the level design permits, and you can enter the level easily from the void by lag clipping in
Tricks/Skips used in this run:
Speed modifiers
SR50 speed is used extensively throughout the entire game. Running at this speed gradually saves minutes throughout the course of the run.
Keybumping/keygrabbing
Throughout the game, we find keys on pedestals 64 units or thinner in length. Just like in PC Doom, we can thrust into these pedestals and grab the key early. This skips several puzzles and saves minutes of time, causing the run to be much faster paced.
Trigger skipping
There are two types of triggers that can activate events: LINEDEF based triggers and THING based triggers. The former kind can be skipped when passing through the line at the proper speed and angle. Not only can horizontal and vertical lines have skippable triggers, but using SR50, I can also skip triggers on diagonal lines, which is impossible with SR40 speed. This comes into play on MAP05, and saves some time. Most line-based triggers however you will want to activate, such as those for lifts, doors and other important events to progress the map. THING based triggers cannot be skipped as they possess a length and width bigger than possibly skippable, however there are no instances where skipping those triggers would help anyway.
Door jumping
There are a few places where doors are designed to open from the middle vertically and are divided into two halves. If we close the door on a precise frame, we can take advantage of the fact that we can get increased height from the bottom half. This is used in The Bleeding and Terror Core.
RNG
Doom, no matter what version you play, is a game that's heavy on RNG. RNG decides enemy movement and damage ranges, along with shotgun and SSG spread patterns. Thankfully, upon loading every level, the RNG seed is reset to zero, meaning that if improvements were to be made to the run, it will only affect RNG in that individual level.
RNG Manipulation without losing time
Shooting at walls allows us to manipulate RNG without changing our movement and delaying inputs by a specific number of frames. Therefore, we want to be able to manipulate RNG without losing time as much as the game allows us to. Also, in non-time sensitive moments such as autoscrollers, moving differently allows us to manipulate the RNG without using ammo.
Tricks/Skips not used in this run:
Lag clipping
Emulation for the Nintendo 64 is not cycle accurate, and therefore, no functional emulator in this present time emulates lag properly. While in some cases this isn't a major problem, in this game lag can be used in order to clip through impassable linedefs (and given the map design, even walls, making voidwalking possible!) This TAS loses several minutes to not being able to abuse any sort of lag.
Here is how lag clipping would work if it was used in this TAS: If we run into anything that allows us to build speed on it, such as a fence or a bar, the SSG can be used to generate lag if we're close enough to the wall. This lag somehow increases our speed beyond normal limits about a half second later. Having this increased speed can allow you to bypass nearly anything as long as you have enough. This currently allows us to bypass any impassable linedef, any 16-unit bar, as well as any 32-unit bar. There are also a huge number of theoretical uses that a TAS might be able to take advantage of, so the full scope of what's possible with this glitch is still not fully known.
Other notable differences between console and emulator:
Loading times
On BizHawk, loading times are doubled. This increases the length of the TAS by about 3 more minutes than if emulation was more accurate. This is unique because in many cases, the exact opposite is true from an emulation standpoint.
30-60FPS
The version of BizHawk I used runs all N64 games at 60FPS even if they run below that. Doom 64 runs at 30FPS. Interestingly, this has the small side effect of dithering between odd and even framecounts per level (Example: The Terraformer advances on even numbered frames, where Main Engineering advances on odd numbered frames).
Stage by Stage Comments
MAP01: Staging Area
Staging Area is one of the most technical levels in the entire game from an RTA standpoint. The level consists of many small hallways and orifices that sometimes have 8 unit wide supports sticking out, along with very many enemies blocking you on Watch Me Die, making optimal movement difficult. The IL record is 0:28.
MAP02: The Terraformer
This is where new additions to the Doom engine start becoming evident. This level is so-called because we activate a terraformer in the middle of it. This is the only instance in the run where we have more than a few seconds of time to waste as the level changes. This is also the first level where sequence breaks come into play in the speedrun. We keygrab the blue and yellow keys to get them far more quickly than the intended route. Grabbing the blue key skips a puzzle that takes more than 20 seconds to set up, while the yellow key skips having to press a switch that isn't too far away, but wastes a few seconds to go towards. This is also the first instance of trigger skipping, in which I skip the activation of ceiling crushers just before the red key.
MAP03: Main Engineering
This was the first major roadblock in making this TAS. The main culprit was the yellow key room, which is deceptively complex and requires a lot of effort to make optimized. Incidentally, I abuse a previously unknown bug in the mapping that displays a missing texture (missing textures in this game display the message "I suck at making maps"). The yellow key room is one of the most complex rooms in the entire game. It consists of me having to spawn prevent two cacodemons as well as two nightmare imps. I have to manipulate the RNG so that three of the cacodemons already present in the room were able to spawn prevent them. This leaves six lost souls to kill, which takes two SSG shots. Additionally, I had to redo this multiple times due to having bad luck with the way the cacodemons in the room moved around.
The rest of the level is otherwise straightforward and not too difficult. A lot of it is an autoscroller when taken down to such an extreme. I lose a small amount of time by intentionally spawning a rocket launcher (which in turn spawns a few nightmare imps) but I need every single rocket I can get as rockets are sparsely placed in this portion of the game.
MAP04: Holding Area
The level starts with five enemies that need to be killed. I kill three spectres and two shotgun guys in three SSG shots.
The rest of the level is mostly just movement. I activate a lift as I'm getting the blue key and just barely make it back on. Activating this lift early saves 4 seconds.
MAP05: Tech Center
This level is mostly about perfect movement with some moments of RNG manipulation throughout. Early on we encounter two sets of barrels. I boost through the second set very dangerously, leaving me with 1 health.
After getting the yellow key, I head into a very tight set of rooms that are connected by thin hallways with a lot of doors. These rooms have monsters in them, and at the health I was at, manipulating monsters was very important. After this, I head into another thin hallway with a bunch of crushers inside, but I skip the trigger that activates them, saving about a second. Shortly after this, I turn around and press a switch, then proceed to use the lift on my way to the red key through the wall. Using the lift early saves 2 seconds. After getting the red key, I open the red door and skip a very hard to skip trigger that activates a crusher that surrounds the blue key, saving another 2 seconds. Skipping this trigger requires SR50 speed, which makes it essentially impossible to skip in RTA runs.
After obtaining the blue key, a MACRO activates that takes some time to end. While casually, and even in RTA runs, it ends quick enough to activate the next switch seamlessly, I end up speeding to the switch very quickly, which forces me to wait a short bit before going ahead and activating it, as the switch also triggers a MACRO. From here, it's just a short walk to the exit.
MAP06: Alpha Quadrant
This is a pretty unforgiving level that starts off with a hexagonal hallway that was extremely difficult to optimize. After that, we flip a switch to spawn two hell knights, which we must kill to get the yellow key. After that, we head over to the yellow key switch and position a lost soul under a portion of the ceiling that is lowered, so we can open cause the door to open early. We don't need to enter the door, but doing so causes the MACRO to end early, and the platform directly after the jump uses a MACRO to function. We cross over a few large platforms to hit another switch. Going backwards once you're on the other side saves about 10 frames over taking the teleporter. Once we get the blue key, we can exit the level.
MAP07: Research Lab
Very simple level with a lot of sequence breaks. I make a jump to the blue key that you aren't supposed to make, and then a few seconds after, immediately jump to the exit early. I intentionally lose 3 frames here to pick up two rockets that I must pick up; if I avoided them, I would be one frame away from getting a 0:10.
MAP08: Final Outpost
Due to the fact that lag clipping cannot be done on emulator, we cannot skip this level. There are still a few sequence breaks present that you wouldn't otherwise see, however.
Just before I get the blue key, I shoot the lost souls in such a way that one of the lost souls activate the lift trigger. This saves about 1.5 seconds over having to activate it myself. After that, we keybump the red key immediately, skipping an entire section filled with zombies, imps, and a baron. Shortly after, we proceed to do another keybump for the yellow key, which skips a small puzzle and saves a small amount of time.
The last part of the map is the most complicated, and is a massive highlight for those watching. Normally, alongside the four barons that spawn, there are two hell knights, but we can have the nightmare imps spawn prevent them, leaving only the barons. TASing this part of the game was insanely fun once I got that step done, because of the actual fight between the barons. I tried making this look as flashy as possible while picking up the health and ammo I need. In my opinion, this is an awesome tradeoff for unfortunately not being able to skip the level.
MAP09: Even Simpler
This level is a reference to MAP07 of Doom 2's "Dead Simple" and plays similarly, though we cannot skip it and the improved engine provides more freedom towards what enemies can be placed in the level.
This level is highly complicated, and is possibly the hardest level to TAS in the game. To put it simply: I must first kill the first set of mancubi in 3 rockets each, then kill 8 hell knights in 2 rockets each while manipulating RNG to spawn prevent 4 of 6 barons using the cacodemons that spawned earlier. After this, I kill the second set of mancubi that spawn by using two rockets and the darts. Because I spawn prevented most of the barons, I just barely have enough ammo to continue the run.
MAP10: The Bleeding
We start this map by immediately door jumping onto the edge of a balcony, which is possible because not all the lines that make up the edge are impassible. This causes the sequence to be broken a little bit. While we could go immediately towards the direction of the yellow key at this point, there's a problem: we have to get back in the spawn area to exit, and since we can't use lag clipping, we have no way of finishing the map. Therefore, we first have to go to the opposite side of the level to activate a switch that lowers a set of pillars, allowing us to enter once again.
Once we activate the switch, we head towards the yellow key. Then we navigate our way through a room filled with spectres and activate a couple of switches. Once I get the BFG, I use it to clear my path once I open the outside door.
MAP11: Terror Core
This level begins with a large amount of monsters that I all have to kill to progress. My weapon of choice is the rocket launcher, because it clears out the area the most efficiently. After getting the blue key, I head my way toward the arena this level has.
It's worth noting that the hardest part of this level in a TAS is not the arena, but this hallway that's filled with spectres just before it. Going back to the hallway and moving through it is very difficult without being blocked by the spectres. Intense RNG manipulation was used to get the best possible movement in this specific area.
Anyway, once we get to the arena, I use my BFG to clear out the majority of the monsters, then finish the rest off using rockets. After this, I head up and use a door jump to skip the yellow key. This leads me to the red key, after which I use my BFG again on the lost souls and open the red door through the wall, which allows me to get a faster entry into the exit.
MAP12: Altar of Pain
Yet another simple, yet broken, level. Up until the lift, the level stays relatively normal. But once I make it up the lift onto the balcony, I make a huge jump that effectively skips the rest of the level.
MAP13: Dark Citadel
By contrast, this level is pretty complicated. What makes this level so difficult is that you can have a demon activate a lift for you much earlier than you can get to it, which saves seconds. While I have to use this lift twice, I wasn't able to get this to happen a second time due to the monster movement. In the library, I'm able to hit a switch through the wall, saving roughly 4 seconds. Otherwise, this level is focused on movement.
MAP14: Eye of the Storm
We start this level by killing four barons in order to lower the pedestal that the blue skullkey is resting on. We kill two of the barons using the BFG, and the other two using rockets. It's important to note that I lay 3 rockets on one of the barons, then let the arachnotron finish it off as I go to the other side. After that, it's mostly a rush to the exit, avoiding some lost souls along the way, as well as hitting the switch that lowers the exit through the wall, saving a second.
MAP15: Dark Entries
One of the simplest levels in the TAS, mostly focused on optimization. I keybump the blue key to skip a massive puzzle that would otherwise take a huge amount of time. Otherwise, what's going on in this level is self-explanitory.
MAP16: Blood Keep
This is a very movement-intensive level with a ton of enemies blocking our way throughout. The room that is locked by the red door traps us in once we're inside, but if we're fast enough, we can avoid getting trapped entirely. I grab a backpack from its pedestal so I can get the 20 cells it gives so that I can use my BFG later in the level to kill some spectres that block me. Once I get the yellow key, I make a jump onto a railing to get to the yellow door earlier, saving some seconds. The zombies open the door for me after that, saving me from having to open it myself. Once I open the blue door, I ready my BFG and use it against the spectres. Normally, the exit room traps you in, but since I'm fast enough, I'm able to activate the MACRO before the door has a chance to close, giving me the ability to walk in and out of the room freely. I use my BFG as early as possible to kill all the enemies that spawn, which opens the exit.
MAP17: Watch Your Step
Similar to Even Simpler, this map has more of an emphasis on killing monsters in an arena. But this time, far more spawn prevention is involved. First, I need to get the blue key so that I can press a switch. I have a pain elemental spawn a lost soul that I lure with some RNG manipulation. I head back into the arena to start the fight sequence. It's very important to note that I only kill the monsters that the game flags as being necessary to kill, and spawn prevent some of those when possible. While fighting each set of monsters, I lead the lost soul through the arena, making sure it doesn't die. The reason I need to do this is because I have it spawn prevent one of the two cyberdemons, which saves two seconds and a BFG shot. Of course, this level can be skipped with lag clips, but we can't use them here.
MAP18: Spawned Fear
This map is very long, with multiple sections and lots of backtracking throughout. The first part consists of getting the blue key; the trick is to do so optimally while getting enough monsters to open the doors for me ahead of time. I have to pass this area of the level 3 times, so I need to do this 3 times as well. After getting the blue key, I head to the blue key door, and perform the window jump that allows me to get the red key early. This saves about 40 seconds. I then head to the red door and use my only BFG shot in the level; this is to make it easier for me going out the room with all the hell knights in it, as well as make it so it's less crowded going back to that area. I'm actually able to avoid using a lift simply by moving fast enough, saving a couple seconds, which is really nice. Moving in the caverns at this point was a chore, not just because of monsters blocking you, but because there's a nightmare imp that's stuck in the map geometry, which really changes the jump to get the yellow key. Fortunately, it saves more time to just leave it there than to kill it.
MAP19: The Spiral
This level is short and incredibly simple: go up the spiral once, hit a switch that lowers a wall, go up halfway, teleport twice to the red key, go up once again and make your way to the exit. The red key door can be opened from the opposite side through a wall, saving a second. The IL record is 0:18.
MAP20: Breakdown
We start by immediately hitting a switch from below using the SSG. By abusing the shotgun spread, we can hit the switch to lower the blue key earlier than intended. After moving through the first set of hallways, I make it to the red key room, where I have to use 2 BFG shots to kill 18 demons. After this, I head back and go to the maze portion of the level. There are a couple bars that block our way, but there's enough space to get past them by running along the side rail.
MAP21: Pitfalls
After shooting the first three rockets, I grab the yellow key and head to the switch. Pressing it gives me access to the maze, where most of this level takes place in. It's harder to get past monsters in this area, because the maze is built diagonally. This means I have to shoot most of them to progress. I can make a jump to get to a switch that lowers a wall early, while avoiding a cacodemon below to save a few seconds. I also keygrab the blue key, saving a few seconds. After getting the red key, I had to manipulate RNG so that I could get the best possible distribution of the zombies after teleporting, making it so that I'm not blocked for too long.
MAP22: Burnt Offerings
This level has some of the strangest movement in the entire TAS, not only having to deal with very constrained spaces that are filled with monsters, but also requiring a lot of RNG manipulation on top of that. When I press the first switch, I go out of my way to rocket a cacodemon from below. I have to do this again, because both cacodemons completely block your path. I can get the blue key while the MACRO I just activated is still running, saving a few seconds. I am also able to activate the rocket launcher trap just before the MACRO is about to end, enabling me to not only run over the trap entirely, but to stand on top of the walls that surround it. Shortly after teleporting, I use a BFG shot against the demons, just so I can clear my path ahead. After I get the yellow key, I make my only real detour to get some ammo in this TAS, which wasted around 1.5 seconds, but will save me much more later on. I am also able to skip the trap that activates when you grab the red key, saving me more than 10 seconds. Luckily, the monsters didn't block me at all on the way to the exit.
MAP23: Unholy Temple
This is the longest level in the entire game. It starts with some arachnotrons and spectres that we need to kill to proceed. After that, we gain access to the lift that allows us to reach the barons, who we have to kill next. After that, we take a shortcut by standing on the switch that raises from the ground, and going over the border after we press the switch. We then go to the blue key area, skipping a trigger that spawns several sets of monsters, losing about 20 seconds. We'll have to go here again later. After getting the red key, we enter the room where we get the yellow key. Normally, we have to trigger a couple of MACROs and follow the key around in a convoluted fashion, but we can skip that. We need to trigger these MACROs later, but we need the yellow key immediately. We now go up so that we can do the puzzle, which gives us access to three switches that we need to press to exit the level. After the first switch, we have to go back to the lowest level to press the second one. We need to kill 3 arachnotrons with a BFG shot, and after that, we have to press two switches on the top of the room. Then, I press the lift so that I can make a quick exit from this room. Now, we have to trigger the aforementioned MACROs to get to the exit. For some reason, doing this as quick as I did causes one of the yellow keys that spawn to get in this weird state in between being spawned and being despawned (you can actually see it briefly in the TAS; you can tell because the key stays translucent the whole time.)
MAP24: No Escape
Appropriately named, No Escape is the hardest level in the Watch Me Die speedrun. The exit trigger is activated only once you kill the three cyberdemons. I have to go around part of the level twice, as I don't have lag clips, so I have to damage the first cyber with a BFG shot and 3 rockets. Once I get to the other two cyberdemons, I made it so that I kill two of them at the same time for entertainment purposes, although the level ended much earlier than normal, so I think doing this saved time as well.
MAP28: The Absolution
For some reason, this level is MAP28 despite being the 25th required level. I can only assume they wanted the player to wonder about MAP25-27, the "fun" levels. Anyway, this level has three gates that monsters come out of. Getting the three demon artifacts would close all the gates, but it takes less time to deal with the level normally. I need to manipulate RNG so that I don't need to wait between the time that any of the monsters spawn, which is easier said than done, because the monsters have to physically teleport into the arena, and the monsters' movement patterns vary a lot, especially when dealing with a lot of them.
The Motherdemon (known internally as the Resurrector) is the final thing we have to kill in order to finish the run! It has only 5,000 health, which means that with proper manipulation, it can be killed in two BFG shots. I shoot the first shot just after it becomes vulnerable, then set up the second shot in such a way where I get a height boost from the Motherdemon's attack. The result is that I die before the Motherdemon does, but the BFG tracers are still close enough so that I can kill it as early as possible.
TAS vs. RTA
LevelTASRTA WRDiff
Staging Area0:230:30-0:07
The Terraformer1:041:39-0:35
Main Engineering0:521:31-0:39
Holding Area0:260:45-0:19
Tech Center1:001:27-0:27
Alpha Quadrant0:401:27-0:47
Research Lab0:110:20-0:09
Final Outpost0:450:09+0:36
Even Simpler0:411:27-0:46
The Bleeding0:291:281-0:59
Terror Core0:432:03-1:20
Altar of Pain0:260:41-0:15
Dark Citadel1:051:391-0:34
Eye of the Storm0:320:591-0:27
Dark Entries0:290:55-0:26
Blood Keep1:022:16-1:14
Watch Your Step0:310:17+0:14
Spawned Fear1:101:501-0:40
The Spiral0:150:27-0:12
Breakdown0:550:30+0:15
Pitfalls0:431:231-0:40
Burnt Offerings0:522:17-1:15
Unholy Temple1:343:25-1:51
No Escape0:211:17-0:56
The Absolution1:071:50-0:43
TOTAL TIME18:1632:32-14:16
1Utilizes pausing for lag clips/OoB strats
Possible improvements:
There are a huge amount of places where lag clips can be incorporated to save time, as established earlier. The difference might be so staggering as to create a whole different category entirely, but that's debateable, and depends on how much we still have yet to discover using the glitch.
Other than that, there are a few places where further optimization could save some time. The beginning of MAP04 is probably where I could improve the most, as I get through that area in three SSG shots as opposed to two, which would save over a second in that map.
Conclusion:
This is my first attempt at TASing a full game. I have always appreciated TASes from a distance, but never thought I would end up making one until I realized that if I wanted a TAS of this game to exist, I had to make it myself. It took over two years, but I'm extremely happy with the resulting TAS.
Unfortunately, it simply can't be ignored that this TAS is unable to utilize the most broken and notable glitch in the entire game due to emulation issues that are out of my control. Some in-game times are significantly slower than what they could possibly be, occasionally being slower than even what real time runs have done. If this TAS isn't publishable for this reason, I fully understand.
Special thanks to Grav, for being one of the only speedrunners of this game and for providing insight that helped make this TAS possible.
Suggested Screenshots:
Frame 10167

adelikat: Claiming for judging

adelikat: Accepting for publication to Alternative

TheCoreyBurton: Publication is now under way.

TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 14776
Location: 127.0.0.1
This topic is for the purpose of discussing #6051: Headshot's N64 Doom 64 in 23:54.90
Joined: 11/12/2013
Posts: 17
I cant speak highly enough of this TAS. Headshot has done an amazing job implementing basically every possible time saver I knew about and dozens more that I wasn't aware of. On top of that, hes added tons of flair and swag in almost every map to make it an even more entertaining spectacle. I could only dream about Doom 64 having a well represented real time run and tool assisted run... now it has both! It is unfortunate about the lag clips not being possible on emulator, but such is life with the poor state of N64 emulation. I strongly believe this TAS should be published despite its inability to utilize that trick, if only because it could be ages before emulation catches up to console in this respect. Congrats again to Headshot on his hard work!
Editor, Experienced player (548)
Joined: 10/22/2016
Posts: 581
Location: Argentina
I can't believe that this day finally comming to see a doom 64 TAS. Here is my yes. And here is my suggestion for stars.
You can see more TASes on my youtube channel
ViGadeomes
He/Him
Active player, Judge (302)
Joined: 10/16/2017
Posts: 455
Location: France
Amazing ! Yes vote !
Dimon12321
He/Him
Active player (478)
Joined: 4/5/2014
Posts: 1120
Location: Ukraine
The moment of my life! I couldn't imagine someone will TAS the whole game. First steps were done by MTA, who TASed first 4 maps in 2011-2012. It impressed me and I decided to TAS the whole game drawing on current speedrun technics. In 2014, I submitted this: #4267: Dimon12321's N64 Doom 64 in 41:22.45 It was tearred down, but I had impressed many people anyway. Later, I decided to reTAS the game on Be Gentle! with Graviton together, but I dropped it. In 2015, MTA and Graviton attracted me to TAS the whole game again. Sadly, our team lasted for only 2 weeks if I'm not mistaken. The result is this: User movie #22952445841349675. A pair of maps were TASed by IcySpeedruns in 2016, but it seems to be a fun work. And now, in 2018, Headshot is submitting a great work that we all haven't managed to do! It is colourful headshot to my old submission! I'll delete my encode when your TAS is published. I think, your TAS automatically worth an award! I'm still impressed of how fast you are beating the game though I know that Doomguy moves faster than in any other port and I'm really shocked how long you've been running with 1 HP kinda I forgot it's TAS. The game is pretty dark and I don't think in-game brightness even work. So, for those who hasn't watched the video yet, I'd recommend to download the Bizhawk, the ROM and watch it with increased emulator brightness. Also I vote for publishing a brighter TAS since some people may complain about the game is so dark. I have only one question. Did my TAS inspired you to make this run? What inspired? BTW, Doom 64 was my first TAS too. Yes vote from me! Let me hug you guys!!!
TASing is like making a film: only the best takes are shown in the final movie.
Site Admin, Skilled player (1234)
Joined: 4/17/2010
Posts: 11251
Location: RU
I was also waiting for this ever since I had to reject Dimon12321's submission. Will watch/vote/post later.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Banned User
Joined: 3/10/2004
Posts: 7698
Location: Finland
Is it really necessary to spend several seconds setting up the brightness?
Memory
She/Her
Site Admin, Skilled player (1515)
Joined: 3/20/2014
Posts: 1762
Location: Dumpster
Warp wrote:
Is it really necessary to spend several seconds setting up the brightness?
Yes. This game can be incredibly dark, so anything to remedy that is appreciated.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
Banned User
Joined: 3/10/2004
Posts: 7698
Location: Finland
Maybe the encode itself could be made brighter so the run doesn't need to be made slower?
Memory
She/Her
Site Admin, Skilled player (1515)
Joined: 3/20/2014
Posts: 1762
Location: Dumpster
You can literally just fast forward on the emulator or skip ahead in the video. It's only done once at the start of the run and the rest of the run is constant action. This is not a problem.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
Banned User
Joined: 3/10/2004
Posts: 7698
Location: Finland
Memory wrote:
You can literally just fast forward on the emulator or skip ahead in the video. It's only done once at the start of the run and the rest of the run is constant action. This is not a problem.
That's not the point.
Memory
She/Her
Site Admin, Skilled player (1515)
Joined: 3/20/2014
Posts: 1762
Location: Dumpster
There has been plenty of precedence for spending time in settings in order to maximize entertainment. Having encoders do extra work for things that can easily be done within the TAS itself is unnecessary.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
Active player (405)
Joined: 2/5/2012
Posts: 1677
Location: Brasil
lol i actually think there's not enough brightness in the run and would appreciate some encode that changes that, but i appreciate the effort the runner made of inserting that for viewers
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general
Headshot
He/Him
Player (118)
Joined: 2/25/2017
Posts: 13
Dimon12321 wrote:
I have only one question. Did my TAS inspired you to make this run? What inspired? BTW, Doom 64 was my first TAS too.
Doom 64 was one of the games from my childhood that has stuck with me over the years. Ever since I discovered speedrunning I increasingly became obsessed with seeing this game pushed to its limits in a TAS, to the point where I had to try it myself. Anyway, yeah, this game is hard to get to look right. If you turn up the brightness outside of the game, it might look a bit washed out, but if you don't, it might look too dark for some people. For the encode, I'd personally advise against turning the brightness up externally, because Doom 64's lighting is much more specifically composed in comparison to other games. I'd personally recommend that anyone interested in this pick up Doom 64 EX on PC to get an idea of how this game should really look like. While not a perfect port of the game, it looks amazing with the in-game brightness slider turned up alone. In my opinion, that's the type of look the encode should try and emulate, if possible.
Banned User
Joined: 3/10/2004
Posts: 7698
Location: Finland
Memory wrote:
There has been plenty of precedence for spending time in settings in order to maximize entertainment. Having encoders do extra work for things that can easily be done within the TAS itself is unnecessary.
If I made a run that's otherwise identical, except it doesn't set that brightness setting, would it be rejected? It would be faster.
Memory
She/Her
Site Admin, Skilled player (1515)
Joined: 3/20/2014
Posts: 1762
Location: Dumpster
Warp wrote:
Memory wrote:
There has been plenty of precedence for spending time in settings in order to maximize entertainment. Having encoders do extra work for things that can easily be done within the TAS itself is unnecessary.
If I made a run that's otherwise identical, except it doesn't set that brightness setting, would it be rejected? It would be faster.
That would be plagiarism so yes it would. If the original author did it, I'd recommend still recommend turning up the brightness since as the author states, you'd lose a lot of colors if you did it artificially. The Guidelines recommend to keep movies watchable. If a movie that is objectively less watchable is faster than the movie that is watchable but have no differences otherwise, we will take the latter based on grounds of entertainment. Speed/Entertainment tradeoffs will not count against a movie when aiming for Moons.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
Banned User
Joined: 3/10/2004
Posts: 7698
Location: Finland
Memory wrote:
Warp wrote:
If I made a run that's otherwise identical, except it doesn't set that brightness setting, would it be rejected? It would be faster.
That would be plagiarism so yes it would.
Creative Commons Attribution 2.0 wrote:
You are free to: Adapt — remix, transform, and build upon the material for any purpose, even commercially. Under the following terms: Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. The licensor cannot revoke these freedoms as long as you follow the license terms.
Is it too much to ask to abide by your own rules and licenses? You cannot publish something under a certain usage license and then arbitrarily revoke people's rights provided by that license. (Well, I suppose technically speaking you can. Nobody is stopping you. But for consistency and fairness you shouldn't. If you promise something, and use a legal license to give people some rights, you should stand behind your words. If different rules are being imposed than granted by that license, use a different license. (And I'm using the plural "you" here, for the record.))
Editor, Experienced player (548)
Joined: 10/22/2016
Posts: 581
Location: Argentina
I think she refers to "plagiarism" because you only want to made a new movie only without brightness using Headshot inputs (and doing nothing new/relevant or saving frames) that is stupid for me, I'm not sure how much you know D64 but it can be incredibly dark 90% of the time. Of course after pasting the inputs you need to have a lot of luck to maintain the RNG of the game same as this movie (wich is a 3D game based on full RNG)
You can see more TASes on my youtube channel
Memory
She/Her
Site Admin, Skilled player (1515)
Joined: 3/20/2014
Posts: 1762
Location: Dumpster
Warp wrote:
Memory wrote:
Warp wrote:
If I made a run that's otherwise identical, except it doesn't set that brightness setting, would it be rejected? It would be faster.
That would be plagiarism so yes it would.
Creative Commons Attribution 2.0 wrote:
You are free to: Adapt — remix, transform, and build upon the material for any purpose, even commercially. Under the following terms: Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. The licensor cannot revoke these freedoms as long as you follow the license terms.
Is it too much to ask to abide by your own rules and licenses? You cannot publish something under a certain usage license and then arbitrarily revoke people's rights provided by that license. (Well, I suppose technically speaking you can. Nobody is stopping you. But for consistency and fairness you shouldn't. If you promise something, and use a legal license to give people some rights, you should stand behind your words. If different rules are being imposed than granted by that license, use a different license. (And I'm using the plural "you" here, for the record.))
...Creative Commons means that you can do whatever you want with it... But it absolutely does not mean that others have to accept what you do with it. If you uploaded a version to youtube that doesn't have the brightness screen, that is fine. However, TASVideos has no obligation to accept TASes from people that do minimal amounts of changes to the work in question. If you tried copying off of a Creative Commons work in real life and turned it into a professor, you'd still get failed for plagiarism. See the Movie Rules for more information.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
Site Admin, Skilled player (1234)
Joined: 4/17/2010
Posts: 11251
Location: RU
Warp wrote:
Memory wrote:
That would be plagiarism so yes it would.
Creative Commons Attribution 2.0 wrote:
You are free to: Adapt — remix, transform, and build upon the material for any purpose, even commercially. Under the following terms: Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. The licensor cannot revoke these freedoms as long as you follow the license terms.
Which part of Creative Commons Attribution 2.0 is talking about plagiarism and rejections? Because I can't find how it regulates any of that, nor can I see the relevance.
Warp wrote:
Is it too much to ask to abide by your own rules and licenses? You cannot publish something under a certain usage license and then arbitrarily revoke people's rights provided by that license. (Well, I suppose technically speaking you can. Nobody is stopping you. But for consistency and fairness you shouldn't. If you promise something, and use a legal license to give people some rights, you should stand behind your words. If different rules are being imposed than granted by that license, use a different license. (And I'm using the plural "you" here, for the record.))
Which part of whose actions appears as not abiding by our own rules and licenses to you?
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Banned User
Joined: 3/10/2004
Posts: 7698
Location: Finland
Plagiarism: "an act or instance of using or closely imitating the language and thoughts of another author without authorization and the representation of that author's work as one's own, as by not crediting the original author" By definition it cannot be plagiarism if the usage license gives me permission to copy the content, as long as I credit the original author. The claim was made that the submission would be rejected because it would be "plagiarism". It cannot be plagiarism if it's published under Creative Commons Attribution (and you give proper credit), no matter how much of it is copied verbatim. So, again, why would it be rejected? It literally cannot be because of "plagiarism".
Editor, Skilled player (1402)
Joined: 3/31/2010
Posts: 2081
Warp wrote:
So, again, why would it be rejected? It literally cannot be because of "plagiarism".
It would be rejected because the old run would be the same but 10 times more watchable.
fsvgm777
She/Her
Player, Senior Publisher (221)
Joined: 5/28/2009
Posts: 1180
Location: Luxembourg
Notes for the publisher/encoder: The TAS desyncs around 30k frames in with GlideN64 (roughly the second map after the intermission text). However, it's also the only plugin that emulates the "melt" effect when you reach the exit of a stage or when you die. I currently don't know which setting in Glide64mk2 (which was used here) restores that effect, but I do know that the "Framebuffer read every frame" setting in Glide64mk2 restores the fade-outs (after selecting the difficulty and after pressing any button on the results screen after each map).
Steam Community page - Cohost profile Oh, I'm just a concerned observer.
Noxxa
They/Them
Expert player, Moderator (4131)
Joined: 8/14/2009
Posts: 4083
Location: The Netherlands
Warp wrote:
Plagiarism: "an act or instance of using or closely imitating the language and thoughts of another author without authorization and the representation of that author's work as one's own, as by not crediting the original author" By definition it cannot be plagiarism if the usage license gives me permission to copy the content, as long as I credit the original author. The claim was made that the submission would be rejected because it would be "plagiarism". It cannot be plagiarism if it's published under Creative Commons Attribution (and you give proper credit), no matter how much of it is copied verbatim. So, again, why would it be rejected? It literally cannot be because of "plagiarism".
You are conflating copyright law/infringement with plagiarism, which are two different concepts. The Creative Commons attribution license allows you to legally redistribute the work, provided credit is provided, and all that. It defines what can be done with the work without violating copyright law. Plagiarism is not a legal concept, but an academic one. Performing it is considered academically dishonest, or a violation of ethics/integrity, but it is not legally considered a crime. Obviously there can be overlap, but that is not necessarily the case. It's possible to infringe copyright without plagiarism, by e.g. giving credit to a work that is licensed not to allow redistribution of that sort. And it is possible to commit plagiarism without infringing copyright, by taking a work that's licensed to allow redistribution, but using it as personal credit. Now, while TASVideos' usage of Creative Commons allows redistributing published work (among other things) as long as credit is given, that still doesn't mean you can take such a publication, copy the entirety of gameplay TAS work of it, (co)-author yourself on it, and expect it to be published. Because we find that to be (academically) dishonest behavior, and we don't support such a thing. You are legally allowed to do it, but we are in no way obligated to accept and publish it. Other posters have already addressed this, which you conveniently ignored entirely. Oh, and even if the whole authorship issue is put aside, it would still be rejected as the movie rules declare that an improvement movie must contain improvements in gameplay to be publishable.
http://www.youtube.com/Noxxa <dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects. <Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits <adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
Site Admin, Skilled player (1234)
Joined: 4/17/2010
Posts: 11251
Location: RU
Creative Commons Attribution 2.0 only talks about what you can do with a work published under its terms. Of course you can take someone's submission, tweak it as you like, and even submit here. This is perfectly fine according to the license we use. Accusing the site in not abiding by it, despite of explicitly using it for all our contributions, is a very bold move that has to be backed up, otherwise one is going to lose a bit of respect. So please quote which parts of Creative Commons Attribution 2.0 we violate please. As far as plagiarism goes, it is covered by our existing rules:
MovieRules wrote:
Do not claim authorship for something you haven't made Taking another user's movie and submitting it under your name is strictly forbidden. Such offense is potential grounds for a ban. This doesn't apply to situations where a similar or identical solution is discovered and implemented independently.
This is exactly what Warp quotes as a definition of plagiarism, and it is indeed not allowed. However, I should clarify, that plagiarism wouldn't be a reason to reject his theoretical movie. Once such a submission is made, it's trivial to properly credit the actual author, and I don't have doubts Warp wold do that according to our license. The actual reason to reject such a movie is two-fold. The first part was addressed by Moth. The second part doesn't seem to be covered in the Movie Rules. The thing is, it would have the same gameplay length as the old movie, but even then obsoletions can happen. For Moons, we can accept an improvement that has the same input length as the old movie, or beats the game within the same time, if it's significantly more entertaining. This would have happened to the latest warpless SMB run if it had no time-wise improvements to MrWint's movie. There was also an accepted submission or Magician Lord that had the same gameplay overall and only tweaked the ending a bit. It was accepted because it was significantly more entertaining, despite of having no gameplay improvements. In cases when there's no time-wise improvement, nor entertainment improvements, just cosmetic changes that don't actually make the run more entertaining (the contrary in fact), the movie gets rejected.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.