*This is a tool-assisted speedrun!*

For https://dsdarchive.com/

Author: RockyGaming4725
Date: August 19, 2026
IWAD: doom.wad
PWAD:
Map(s): E1M1
Skill: 4
Category: UV Speed
Sourceport (Tool): XDRE-R v2.24
Sourceport (Video): DSDA Doom 0.29.4
YouTube: https://youtu.be/kda0Gw2iUIk

Time: 7.66
Comments:

One tic improvement over almostmatt1's 7.69s.
I realized the movement after opening the first door could be slightly improved. By killing momentum while further back west, the curve towards the first .000000 could be improved, and the player would have more eastern momentum on the frame that they pass under the door. This happens on tic 81. My eastern mommentum is 17.9648, and Matt's is 17.6495. This alone saves units, but not any frames. The damage boosts were key here. I get one boost while in the tech room just before I wallrun. This allows for quite a significant time save due to higher momentum during a wallrun. This saved around a frame. To stay a frame ahead of Matt, damage boosts were obviously needed. Unfortunately, it was very difficult to get good RNG rolls after requiring one damage boost before the blue wall wallrun, since that in of itself is uncommon, it slims down how many RNG sets you can really work with. Eventually I got something that seemed workable. The major problem was that I couldn't get a damage boost from the shotgunner while curving into the green slime room. In Matt's demo, he does, and this is a very beneficial damage boost, due to the combination of speeding up the curve itself, and being midair for a decent amount of tics during the curve. To stay ahead, I had to settle for a damage boost when just entering the green slime room. To open the second door, I needed a *lot* of damage boosts. I thought it was going to be impossible, but once I added a better curve, and added the speed boost from jumping across the green slime (you need to land with a Z change of 24 units to get a significant boost, this is what propels NoMo demos), I finally got the second door to open a frame earlier than Matt. Then, I had to get the door boost, which was tricky, but doable. I couldn't afford to lose any speed on the door, so I had to slow down enough to leave more room for speeding up toward the door (slow down over 5 frames, door boost happens 2 frames later), get an extremely close position to the wall to allow for an optimal frame "achieving" the door boost, all while not slowing down too much to keep the door in range to open it. A very frustrating set of components.
The room after the second door was difficult. I didn't have much health, so whatever damage boost I got had to be weak but still powerful enough to send me to the door one frame earlier than running with no damage boost. I also didn't have a lot of room to play with RNG, since shooting the zombieman closest to the 2nd door was needed for getting enough damage boosts. I was going to give up, and it was getting late, but it bothered me enough to make me keep messing with the RNG, and eventually I got a damage boost that left me at 1hp. Unfortunately, it *still* wasn't enough. That's when I decided to optimize the door boost. After having done that, I was very, very close to opening the final door one frame ahead of Matt, but I was just barely not close enough. Then, I decided to change my curve in the final room. After some quick tweaking, it was done. I opened the final door on tic 233. I needed to build the rest just to see it done. So I did.
Unfortunately, the final door being only barely reachable means there's not really any time to slow down before opening it. This means instant .00000s on the wall and door are not doable (in this demo). Luckily it doesn't matter, since you can still gain enough speed in time to hit the switch just as fast as if you did have instant .00000s. It's just a cleanliness thing in this case, instant .00000s look much nicer, but killing speed and then getting them is still just as fast (at least here).

Now, I must say, this was originally going to be a collaborative demo. I invited Matt to join, and he decided to, but unfortunately between both our busy schedules, none of us made any real progress for days. When I decided to have a good sit-down and swing at it, I ended up building the rest of the demo in a few hours (after copying my NoMo demo for the inputs to the first door). Matt had messed around with different places for damage boosts, but only while experimenting. This means he did not place any inputs into this demo, so he is not an author. However, I must thank him for messing around with damage boosts and discovering that a damage boost before the blue wall wallrun was able to save a significant amount of units. I believe he got 15 units ahead, which made me choose to use a boost in the same room, and optimize it to get around a full frame (20 ish units) ahead. I'd also like to thank Matt for making the 7.69 demo, a still extremely optimal demo. This run is only a 1-frame difference over nearly 8 seconds of play - there really wasn't any room for error here. This demo was chalk-full of micro-optimizations, tiny things that added up to make a difference.

I'm quite proud of this demo. I honestly though 7.69 was the barrier, but it isn't anymore. Is 7.66 the barrier? Who knows. 7.63 won't be done by me, though, I know that hahaha :)


Rocky

rockygaming4725@gmail.com
