Username:
B
I
U
S
"
url
img
#
code
sup
sub
font
size
color
smiley
embarassed
thumbsup
happy
Huh?
Angry
Roll Eyes
Undecided
Lips Sealed
Kiss
Cry
Grin
Wink
Tongue
Shocked
Cheesy
Smiley
Sad
1 page
--
--
List results:
Search options:
Use \ before commas in usernames
Edit history:
mikwuyma: 2010-06-05 07:13:39 pm
My feelings on The Demon Rush
Game Page: http://speeddemosarchive.com/SplinterCellChaosTheory.html

Michael 'Tigger77' Welle's PC Expert ILs

Verifier Responses

Quote:
I've watched the run now and before i send you my notes, i have a question for the Runner.

I've noticed a difference between the new and the old run. In the last level of the new run, almost the entire conversations are different from the old run from the point where he enteres the vent above the war room.
In the new run, the conversations (which are unskippable) are shorter and therefore he can finish faster.

Just a little example:
Old Run (after Otomo committed suicide): "Otomo just comitted Seppuku"
New Run (after Otomo committed suicide): "Now what?"

So what is up with that? Is he running a different version or why is there such a big difference?


The runner's response to the verifier's question

Quote:
Yeah its different, that's because I have alarm level 4. It’s different in many levels because Lambert blames you for being so unstealthfly. It’s part of the game as a story element also all bonus objectives are cancelled if you reach alarm level 4 that’s not a version difference.
Btw thanks for verifying.
Mfg Tigger77


Verifier's final response:

Quote:
I have watched the new run several times now and made my notes before reading the runner's comments. These are my notes I wrote down immediately after i watched each level.

When i write "better", "faster" or "different route" i mean compared to the previous record holding run.

Quote:
1. Level (Lighthouse)
-Better inventory management (saves time)
-same route, different weapons (saves time)
-doesn't wait for boat description (saves time)
-less stealth (saves time)
-takes out enemies, instead of avoiding them (saves time)

2. Level (Cargo Ship)
-less stealth and less waiting (saves time)
-2 small mistakes in the engine room which cost about 1 second each (failure to climb over certain objects at the first try)
-faster route in the engine room (even with the mistakes, it still saves time)
-at about two thirds into the level, he takes a different route so there is less backtracking (saves time)

3. Level (Bank)
-different route, uses front door instead of roof access (saves a lot of time)
-takes out the guard and hacks the retinal scanner instead of using him (saves time)
-less stealth (saves time)
-takes out enemies, instead of avoiding them (saves time)

4.Level (Penthouse)
-bypasses guards at the beginning of the level a little faster
-slightly different route in the penthouse (saves time)

5. Level (Displace)
-takes out a lot of enemies instead of going around them (saves time)
-triggers alarms to be faster(saves time)
-no stealth at all (saves time)
-same route
-ignores cameras (saves time)

6. Level (Hokkaido)
-no stealth (saves time)
-takes out enemies instead of bypassing them (saves time)
-triggers alarm to be faster
-ignores cameras (saves time)


7. Level (Battery)
-uses different way to take out the guards by the bomb (saves time)
-uses computer to finish objective instead of listening to conversation (saves a lot of time)
-ignores cameras (saves time)

8. Level (Seoul)
-much faster handling of the guards in the elevator (saves a lot of time)
-kills enemies instead of sneaking by (saves time)
-aiming is sometimes sloppy (wastes time, about 2-3 seconds max)
-different route at the first tank (saves time)
-takes fall damage (saves time)

9. Level (Bathhouse)
-ignores cameras and alarms (saves time)
-takes out the guards fast and precise (saves time)
-makes good use of the given tools (airfoil rounds and sticky shockers) to take down enemies
-same route
-bypasses a wallmine instead of taking it (saves time)
-shoots shetland sooner (saves time)
-small mistake (falls off the plank into the water) (cost about 1 second)

10. Level (Kokubo Sosho)
-takes out enemies instead of sneaking by (saves a lot of time)
-ignores cameras and alarms (saves a lot of time)


Now for the overall performance:

-I haven't noticed any kind of cheating
-The video quality was very bad, sometimes you couldn't see anything at all because it was so dark (but as it says in the verification guidelines "The verification copies of the runs do not usually represent the quality of the final products", so i hope the final videos will be of better video quality)
-I haven't noticed any audio problems
-I have noticed  a few minor mistakes which add up to about 5-10 seconds

Since the new time is a big improvement to the old one, i recommend to accept the run.


Quote:
No cheating, deaths, or major technical problems.

I could break the run down piece by piece and give a really detailed essay on it like I sometimes do, but it's significantly faster than the existing run--which I already thought was pretty good--so there isn't much point. I didn't see anything that made me think "hmmm that could be done differently", and everything was done fairly smoothly. Even the random delays didn't seem too bad.

My only concern is that I can't find my game disc to double check the guard patrols as mentioned in the comments, to verify that is in fact played on expert, so I'm hoping one of the other verifiers can check that.

Accepted.


Quote:
Verifier: general player skill / execution was decent throughout
Verifier: but
Verifier: there were quite a few places where i thought minor or major planning errors had been made, and while i can't confirm them all without being able to load up the game here at the moment i can confirm a few by comparing with mrguytodd's unfinished 100% run from a while back
Verifier: i think pretty much every level has several seconds of planning mistakes or points where performing actions in a subtly different sequence would've saved time
Verifier: but the one that really stands out as appalling
Verifier: is battery
Verifier: the runner does it in 7:23
Verifier: there's a 100% run of the level in 6:39
Verifier: and this run isn't even 100%
Verifier: and unless i'm incredibly stupid it's got nothing to do with a secondary objective being carried over from a previous mission in the any% that isnt carried over in the 100%, im pretty sure the any% has no disadvantages over the 100% on this level
Verifier: mrguytodd's run is over 2 years old as well, so the runner should have looked it up
Verifier: so if this game didnt already have a run up, itd be a reject for battery alone and would probably barely scrape through if it weren't for battery being so bad
Verifier: unfortunately sda has an incredibly embarassingly shite run of the game up already, which this beats
Verifier: so i have to accept


Decision: Accept

Reason: It's a good improvement over the previous run, despite what the last verifier says. Tongue
Thread title:  
I survived MIKE-Fest 1
Thanx for vertifiing

Quote:
Verifier: the runner does it in 7:23
Verifier: there's a 100% run of the level in 6:39
Verifier: and this run isn't even 100%

Question:
do you have a video link or the route for this time so I can redo it because if sutch a time is possible 100% I definatly has to beat it.
Just reuploaded the Battery 100% run. I think this will work:

http://www.mediafire.com/?dlmzwmtwdzj
Tigger77, I have a question: did you do this run on a PC running Windows Vista? If so, how the fuck do you get the game to run on Vista? If you can point me towards the necessary downloads / cracks, I'd hugely appreciate it.

Unrelated, but didn't seem worth bumping the topic for since it's probably not relevant to anyone except Tigger77 right now:
Links to all mrguytodd's 100% runs:

http://dl.dropbox.com/u/4533383/Bank532.avi
http://dl.dropbox.com/u/4533383/Battery639.avi
http://dl.dropbox.com/u/4533383/CargoShip540.avi
http://dl.dropbox.com/u/4533383/displace527.avi
http://dl.dropbox.com/u/4533383/Hokkaido358.avi
http://dl.dropbox.com/u/4533383/Lighthouse238.avi
http://dl.dropbox.com/u/4533383/penthouse649.avi

I've also given Tigger77 a bunch of suggestions on possible improvements to his runs, which I'll happily forward to anyone who decides to try to beat them or post in a planning thread if someone makes one.
I survived MIKE-Fest 1
So I manage to get 4:35 in Bank thanx to Exploding Gabbage I send Mike a PM over submitting because i WILL NOT IMPROVE any other Level in the near future because of Exams in the next 3 month and other stupid not speedrunning relevanting stuff  Undecided

So If someone want to improve my current run GO FOR IT don't wait of another improvement from my.

thx to ExplodingGabbage for your great tips I testet a few of them and with those tips and a lot of pations this run is beatable to make a fair competition out of this I post your tips hier (hope you don't mind)
sorry that I only managed to improve only one Level  :-[

By the way, a few things I wondered about / ideas that might help you tweak your runs. I don't have a working version of the game right now so forgive me if any of these are stupid suggestions; I'm not in a position to test them myself.

* Did you try running straight across the natural bridge over the canyon at the beginning of lighthouse and spraying down the two guards who come round the corner, instead of jumping across and climbing up the ladder?
* I recall that it's possible to get the name of the Maria Narcissa in three ways: by overhearing the radio conversation, interrogating the operator, or checking the files. Clearly if overhearing the conversation requires delaying your progress up the stairs, then interrogation is faster. But is it possible to leave a sticky camera near the radio operator, then toggle back to the sticky camera for an instant at the moment that the Maria Narcissa gets mentioned to overhear the radio conversation by proxy? Would this be faster?
* Something I didn't understand: on Cargo Ship, how come the sticky camera you fire into the Engine Room goes off instantly? And why didn't the engine room explode?
* On Cargo Ship, did you try a more direct and violent approach to the very first section, taking out the two guards who are talking at the start (or maybe using a grenade on them) and running straight to the entrance to below decks, rather than taking the stealthy way between the cargo containers? You end up having to shock two people anyway, and the more direct approach would save a few seconds of movement.
* Did you consider going to the captain's quarters before the bridge, leaving a sticky camera to kill Lacerda with when he spawns, then
going to the bridge, interrogating the captain, and firing gas from the camera to kill Lacerda? Seems like it would save backtracking.


Bank:
* Just going up the stairs would be MUCH faster than climbing through the vents to get to the president's office. I just checked mrguytodd's 100% run, and he does this, and winds up getting a time of 5:32. Since you won't be wasting time on the opportunity objective, and can run straight through lasers, you should be able to crush this in an any% run. As an added bonus, he was able to leave the treasurer's office (I think... the one downstairs with the guy fixing the door, anyhow) via the door, instead of the crawlspace, which was probably somehow related to him not killing the mechanic before he went in (since he approached from the opposite direction).
* Also on bank, is turning off the lights at the beginning REALLY the fastest way to get in without setting any alarms off? If the threat is being seen by the guards outside, can't you create a distraction, or just knock them all out? If the threat is the guard at the desk, did you try picking the lock, bashing the door with the assault rifle out, and shockering him before he can react? Or just charging past him, if you can do that before the door closes and without getting killed. I'm sure there must be some faster way.
* Wouldn't getting the suspicious transaction records at the same time that you grab the telemetric lock picks be faster than doing it later? I was thinking perhaps you could do it after detonating the charges on the door, while waiting for the door to open, but after looking at some times on the video it looks like this would be slower.

Penthouse:
Not much to suggest here but...
* I can't help but think that you would be slightly better served by the sniper rifle than the launcher, here. Every guard you take out with shockers or airfoil rounds could've been safely eliminated with the sniper rifle instead, and it would save a few frames by eliminating the need to swap to airfoil rounds.
* Why not snipe / airfoil the guy at 2:20 instead of cutting his throat?
* Mightn't shooting the two windows at 3:03 be faster than punching them?
* Couldn't you run straight past the two guys you flashed, instead of taking them down?

Displace:
* The gas grenade at the start was unneccessary. mrguytodd lets the guards hear/glimpse him so that they'll start walking out of the door, thus turning away from the switch, then jumps over the fence while their backs are turned.


Anyway, the rest of my suggestions:
Hokkaido: Looks solid to me. My only two doubts are
* Whether it might be quicker, at 2:01, to run around the glass tank and use a shocker on the guard, instead of climbing up the pipe?
* At 2:53, in the 100% run mrguytodd just kept out of sight in the darkness to the right and ran to the door without taking down the guard. I think your way is quicker, but I'm not sure.

Battery:
You should definitely be able to beat the 100% run by more than just two seconds, here. Most of the improvements can be found just be comparing with the 100% run. However, I have another idea to improve even further, which is to use the Assault loadout and tackle this mission MUCH more violently. First of all I'll list the improvements that can be found just by comparing with the 100% run, then I'll suggest how things could be tackled with the Assault loadout at the end.
* It looks like the route the 100% run takes to the first door is a few seconds faster?
* It also looks like knocking out the two sleeping guards, as done in the 100% run, is a couple of seconds faster than sneaking past them.
* Trying to get the gas grenade takedown at 3:00 to go as smoothly as in the 100% run would save time here, obviously, but that's just a matter of execution.
* Maybe I'm missing something, but why do you even need to take out both guards before checking the computer at 6:10?

Now, if you were to take the assault loadout...
* You could take out the two guards at 1:20 with a frag grenade, instead of a sticky camera
* You could take out the guards at 3:00 with frag grenades and either the shotgun or sniper rifle, which would save a couple of seconds.
* Since punching someone in the face takes a second, using the shotgun to take out the two guards at 4:33 would probably be faster, as long as the second guy runs round the corner into your sights as soon as he hears the shot. The guard upstairs could be shotgunned when you go back up, and everyone you take out after that could be shotgunned just as well as shockered or airfoiled.
* The final two guards you take out would probably be faster with a shotgun.


So hopefully the run gets published soon otherwise only the vertifier would have something from this tips because with out the vids ... I think you got the point.
My feelings on The Demon Rush
Okay Tigger77 did indeed improve his bank run.

Quote:
Hey, just saw the Splinter Cell Bank run. Accept. over 2 minute improvement with an excellent path through the level.


Quote:
okay, SCCT bank verification:

The audio is slightly fuzzy for a lot of the video. I don't think it's just me because I don't get it with other videos. It's still watchable and doesn't prevent you hearing what's going on, but it's slightly irritating. My view is to let it slide.

The run itself is a huge improvement over the runner's last attempt, mainly thanks to him borrowing heavily from the strat of mrguytodd's 100% run. I think it's probably near-optimal now - the only thing I wonder about is whether there isn't possible to just take out the guards outside the front of the bank before entering through the front door, instead of turning off the lights, which would save a few seconds? Hopefully the runner will say if he tried this.

With this and battery improved, the main obvious flaws from the first set of ILs the runner submitted have been cleaned up and the time has come down a nice chunk. I'd probably even accept these ILs now on their own merits instead of just because they're an improvement to the ones already on the site (though I hope the runner will return to this game and try to whittle some more seconds off once he's done with his exams - there's still a second or two to be shaved here and there on several levels, I suspect).


I'M SORRY THIS HAS TAKEN SO LONG TIGGER77. I'll work on getting this run posted ASAP.

P.S. This really isn't a very good way of saying thanks for hosting me. Tongue
I survived MIKE-Fest 1
No problem Mike I had not mutch time the last month worring about Speedrunning in generly

To the improving mabey I will come back on this game some day but not planned in the near future
Tigger77, I know this is a bit late but I just noticed for the first time that on the game page you credited me for the route improvements on Bank and Battery (Mike didn't send us your changed comments when you improved those two levels so I hadn't seen this before). I'm not really comfortable with this since in both cases the improvements basically consisted of swapping to the route from mrguytodd's 100% run and then tweaking it for any%, rather than doing any original research. I feel as though I'm being given credit for mrguytodd's work.

Would you mind me asking you to change the comments to reflect the fact that the actual routes used on Bank and Battery are heavily based on mrguytodd's run? You're free to credit me for bringing the improvements and his videos to your attention, of course, and I'd still appreciate that.