45.7s for the first mission -- even possible?

Started by Ceraus, October 11, 2013, 03:47:27 AM

Previous topic - Next topic

Ceraus

The current best time for the first mission is 45.7 (and is tied with another 45.7, though both seem to be from the same user). I'm fairly sure it's impossible to attain that time, so I'm wondering if I'm missing something.

The Command Node takes 3.0s to land. The Nullifier (as close as possible) takes 26.7s to build, 14.0s to charge and 2.0s to fire. Adding up those times gives us the best possible mission time*: 45.7s.

Let's break down the timing:
0.0 - Mission starts
3.0 - Command Node lands, begins building Nullifier
29.7 - Nullifier finishes building, begins charging
43.7 - Nullifier finishes charging, begins firing
45.7 - Creeper Emitter is blown up, mission ends

Problem is, both building and charging a Nullifier that fast require more energy than the first mission's single Command Node can output. Timings for finishing building the Nullifier with only one Command Node:
- Nullifier only: 36.8s
- 1 Collector, then Nullifier: 34.7s
- 2 Collectors, then Nullifier: 36.4s

Building even a single Collector makes us miss the target time by a large margin. But where else could the extra energy come from? 2 Collectors are needed for full-speed Nullifier building; a whopping 6 are needed for full-speed charging!

My current best time is 57.4; I'm making this topic instead of trying to improve it. I get that times as low as 55.5s are possible, but 51.1s stretches my credulity and I believe 45.7s to be entirely impossible.


* Disclaimer: the best possible time for any mission is actually 45.1s, achieved by hovering a Command Node over the Nullifier when it begins and ends charging. Interesting fact: two Command Nodes are needed to perform that trick!

Fisherck

You have this down to a science ;) I agree with the implausibility of not just those two times, but a few of the ones after it. My current best on that map is 57.0, and the fastest time by a player that I know is definitely a real time is ea3401's 56.8. I'd have to see the strategy involved in beating it faster than that to believe it. But I doubt such a tactic exists.
My CW2 Maps
My CW1 Maps
Quote from: Sqaz on August 28, 2011, 02:49:35 PM
The comments are here to comment, dare to use them.

Grauniad

Almost on the hour of release, someone joined the chat and posted hacks of the game. :(

So, as with prior releases, there are and will be hacked scores. When Virgil has time, and the violations are egregious, he cleans it up and blacklists those keys from submitting further scores.

In the interim, it's only electrons in a game. Don't sweat the small stuff.
A goodnight to all and to all a good night - Goodnight Moon

NFITC1

And to think I'm tied in 57th place with 1:01.2 :( I thought that was a good time.

If you could generate 4.0 energy from the get-go you would be able to service both the nullifier and a collector at capacity.

Nullifiers want 100 energy packets for construction and ammo. It should take 25 seconds to build and 12.5 seconds to charge (ammo packets go out twice as fast), but it looks like it takes more time than that. The data here isn't quite clear. The smallest delay is the moving of packets from CN to nullifier is the distance from the center of the CN to the midpoint of its side. By my best calculation that's 6 units. I don't see any actual data on packets' initial speed, but it looked like it took .6 seconds to travel that far. That delay will happen twice, once after receiving the command to build and receiving the command to supply. There's also the delay of CN landing to beginning to distribute packets. It looked like to me it would send the first packet at 3.5 seconds.

0.0 - mission start
3.0 - CN lands
3.5 - first build packet sent (rate of -2.0 per sec)
4.1 - first build packet received
{during this time, a collector is built to obtain the shield key}
29.1 - last build packet out
29.7 - last build packet received; first supply packet out (rate of -4.0 per sec)
30.3 - first supply packet received
42.2 - last supply packet out
42.8 - last supply packet received
44.8 - emitter destroyed, end of mission.

That's my hasty estimation and could be a bit off.

Harkler

JonneeG and I speedran the hell out of this map and shared techniques. We both ended up not being able to pass 55.5. I can safely say that anyone getting a time of under 55 is either hacking or knows something that we don't. (but probably hacking)

Ceraus

NFITC1, I agree with your theoretical times, but all my tests lead to 45.7 (45.1 with hovering). Even without the packet travel time, building the Nullifier takes about 1.1s more than should and charging it, 0.7s.

... let me check something.

...

Building a Bertha, from the arrival of the first packet to that of the last, takes 132.8s instead of the expected 125.0s; that's 7.8s extra. So every packet is delayed by... pretty much exactly one frame. That... could explain a lot!


Grauniad, that confirms my suspicions. Well, at least we have a good idea which scores are legit, at least for this mission... namely Harkler and JonneeG's. Too bad I'll always have to doubt the high scores I see from now on.


I still own the last mission, though.  ;D

Cavemaniac

Quote from: Ceraus on October 11, 2013, 06:51:32 PM
Grauniad, that confirms my suspicions. Well, at least we have a good idea which scores are legit, at least for this mission... namely Harkler and JonneeG's. Too bad I'll always have to doubt the high scores I see from now on.
I still own the last mission, though.  ;D

It's a fine line - be careful...

On Farbor I currently have the second or third highest score - and I just blasted through it on my way to the end so I could unlock the Alpha sector, I wasn't trying for a good time.

I've played that map quite a number of times during beta testing and I've got pretty good at it. I wonder what kind of time I could turn in if I really tried?

When CW2 came out I spent a whole morning playing and replaying one great map, dropping my time by a few seconds each time until I had a really tight score...

...only to be accused of cheating/hacking by one of the mapmods!

I sent him an 'umm, no actually I worked really hard to get that time/score' message and he apologized and we became good (forum) friends.

It was that experience that prompted me to become a mapmod and from there a beta tester.

But I agree - it's disheartening to see someone move the goalposts...



Be yourself. Everyone else is already taken.

4xC

This is the same exact suspicion I have always had about several best times I found in the last 2 games. They were so low that I tried to do the levels down to a science in my head and nearly always deduced they were impossible to finish so quickly without hacking.
C,C,C,C

vhl

How the hell you finish in 0:56.5 on "trololo map". Thats is the same question :)

Alter Old

My scores are reached honestly though even in CW1 I repeatedly happened under suspicion. On this map is unreal to receive the score better 9848 and time better 55,4-55,5s.
At once I will make a reservation - my English is in poor shape.
As usual, forgive for my English.

Amram

Quote from: vhl on October 12, 2013, 03:19:35 AM
How the hell you finish in 0:56.5 on "trololo map". Thats is the same question :)
I wonder that myself seeing as I couldn't figure out how to get quite that low, though my failure is no indication of anything other than, im not that quick, lol.  Still im happy with what i did get.




Alter Old

Best time with infinity energy is 45.7 without fly, as proved Slovak (Czech) player "Tusim, ze som prvy" ("I think I'm first"). With fly best time 44.7, easy.
I consider steps between the events, each step is equal 1/30 seconds:

90 - CN lands
1 - first build packet sent
16 - second build packet sent
16 - third build packet sent
...
16 - last build packet sent
17 - last build packet received (or 2 for fly)
1 - first supply packet sent
16 - second supply packet sent (!!! exactly so !!!)
8 - third supply packet sent
...
8 - last supply packet sent
17 - last supply packet received (or 2 for fly)
1 - Nulifier on
60 - Nulifier destroyed; end.

Overall 90+1+(16*49)+17+1+16+(8*48)+17+1+60=1371 steps or 1371/30=45.7 sec. With flight 30 steps (1 sec) is saved.
As usual, forgive for my English.

Ceraus

Meh, I tried flying again and got 45.0, so I guess your timings are right. Frame by frame is more precise, too!