T&F -time to vote

greenspun.com : LUSENET : MAME Action Replay : One Thread

Yes Chris, there wasn't such a cheat in original arcade. Ofcourse I understand your program...not. But anyway it works(wow!). If only that autofire cheat could be spotted without playing back inps. I suggest we remove all T&F and hypersport scores and have fresh start without "rlh". Let's see after few days (how about tuesday cet) the results. My votes: ----remove all T&F scores =yes ---remove all hypersport scores=yes ----"run like hell" cheat not allowed=yes Ofcourse you can upload your removed scores again if they are not cheats.

-- Tommi Tiihonen (Tiihoto@hotmail.com), August 06, 1999

Answers

I'll agree to that. Some of us can't affors autofire controllers. :)

-- Q.T.Quazar (qan@home.com), August 06, 1999.

I`ll buy that for a Dollar....( yes on all counts , that is ) ALEX

-- Alex Weir (alexweir@indigo.ie), August 06, 1999.

Hey if you can afford a Sidewinder or what ever cooky joystick you mentioned you had then you CAN afford an autofire :)

I vote to remove the T&F and clone scores, but notify all those players who have scores and give them a week to download their non cheating score to be able to upload it again, or maybe just send it to them via mail attachment when they are removed.

However, i think you should first vote to specify what will be allowed and what will not be allowed before cleaning the slate and have the specification of the uploads in the Description of the game, so you can avoid any confusion in filling the slate back up.

-- Chad (churritz@cts.com), August 06, 1999.


Once again, I'm going to be complicated and annoying - just what I do.

I didn't play T&F on MAME until the cheats were only availible if you used the -cheat switch at startup, but I have the feeling part of JoustGod's reaction is because it was there without a special switch at one point. Using RLH instead of other methods was at least an effort to keep some sort of fairness, so each player was dependent on jump button timing & accuracy. Without other detection - any hope of barring other autofire methods, deciding to allow it was probably the best move availible then.

To go to button tapping only we need an autofire panel - some people to look over questioned recordings and give a best guess ruling about whether autofire was used. Since T&F seems to draw the fastest legit button tappers, I think its expert players will have the best eye for the job. It's not always as much of a competition killer on other games (allowing impossible scores), but not having to tap the buttons is almost always a clear advantage. Even though someone might manage to fool them, judging would help more than it would hurt. If we don't want autofire judges, the sanest move (aside from deleting the game) is to go RLH ONLY. Since it's use is detectable by Zwaxy's program, requiring all recordings to be made with it should be reasonable easy to enforce. No, it wouldn't allow the fastest button tappers in the west to show off, but it would keep the field level so they could at least hit 45 degrees on the long jump in peace.

Aqua

-- Aquatarkus (aquatarkus@digicron.com), August 06, 1999.


I don't think you can eyeball an autofire recording, the best people have done was to say "shit that's too fast, so it must be autofire." which isn't very scientific.

I'm still working on something that should give a tangable result as to wether or not autofire is used. Timing differences between average button presses and depresses is a more logical way to determine if autofire is being used.

And about the RLH cheat, it is a cheat regaurdless if it is part of the MAME distribution or not. Using cheats are entertaining and make getting big scores easier, but usually competitions don't measure how well someone cheater better than someone else. I enjoy using autofire and timing the 43/5 jumps as well as watching others do it with better cheating techniques. But does that make it justified to upload a score to compete with someone else who isn't cheating?

I think cheating and honest recordings deserve upload/download time. But i doubt zwaxy would split all games and have a cheating leader board and a honest leader board. (see this time i didn't assume you were going to change anything :) So i'd say we stick to uploading recordings that don't have any external persuation on playing the games.

-- Chad (churritz@cts.com), August 06, 1999.



Bullshit.

I think at one point Krool brought it up, and I haven't really said much about autofire detection before, but some of us can give a better answer than "shit, that's too fast".

Quick example of a recording that looks like autofire but isn't is Tommi's Tron. The firing "speed limit" in Tron is high, but reachable with a keyboard or keyboard/mouse combo. The trick is you can only do two things at once - you can't move, spin, and shoot max speed. When max shooting speed is needed, he consistently chooses to stops moving to spin & fire in one place. This technique isn't hard to duplicate - I can pull it off and I only get 900cm/s on the 100m dash legit. Tommi's recording also demonstrates a clear understanding of how the game works. I can't do his light cycle patterns, and the tank waves are hard to pull off with my inept spinning, but I can use his methods of dealing with the grid bugs & MCP cone. Why is this important? I never passed level 3 in the arcade. After watching his recording a few times and practicing for several hours in MAME I got to the recognizer stage.

The bullfightj leader is an autofire recording. The leader has mesmerized the bull into very long series of spins. You have to tap cape at the precise times - the bull breaks off if you're slow, and too fast will run the bull into you. Without autofire this would eat all his concentration. But at the same time he is also hammering stab like a madman to gain more points during the spins. It is not possible to do both at once. Aside from this, his fight with the bull shows no greater knowledge of the game than the 20k to 30k recordings - he doesn't know how to get away from the bull's final direct charge or how to kill the bull.

If the recording looks too fast, but isn't autofire, there will be some sort of special firing technique that can be explained to a skilled person well enough to be duplicated. Mastering the T&F "six fingers" should be sufficient qualification to tell if someone's defense email is garbage.

Aqua

-- Aquatarkus (aquatarkus@digicron.com), August 07, 1999.


I didn't mean to say there shouldn't be judges determing what is or what isn't autofire. It is BULLSHIT that all i heard in the track and field discusion was that the meter is too high so it must be autofire. That didn't seem very logical to go by just that, there proly is a way for a person to make the meter go higher than Tommi's and not use autofire. But If autofire is used, a big clue is that the presses are going to be much more equally spaced than a human can equally space button presses. And you can't eyeball that.

-- Chad (churritz@cts.com), August 07, 1999.

And I didn't mean a program to analyze the shot pattern wouldn't help. I love the idea behind what you're building, even if I don't think it's the only way.

Aqua

-- Aquatarkus (aquatarkus@digicron.com), August 07, 1999.


Moderation questions? read the FAQ