This topic is for the purpose of discussing #3893: MESHUGGAH's NES Driar in 05:14.45

Link to video I tried to play the game and couldn't beat the second level :D The movie represents quite some action and requires planning and testing, it's pretty fast-paced. I'm voting Yes. People seem to know nothing about the game and so I'm posting the encode.

Good enough for me, it was short and fast. Yes vote.

MESHUGGAH LEARN TO ENCODE!!!! FCEUX ISN'T THAT HARD TO ENCODE!! <3 Loved the run, Yes vote.

The game looks pretty funny, and the play has a fairly good pace. Yes vote for entertainment.

MESHUGGAH LEARN TO ENCODE!!!! FCEUX ISN'T THAT HARD TO ENCODE!! <3.

If I would make a list of how bad I'm at different things, video encoding would be on the top of the list (okay, maybe english would be the first and video encoding the second). Deblinking is a full of mistery for me... I'm currently trying to figure out the proper resolution for the tutorial video, cause cropping this has 2.05 ratio which I don't think is good, I'm thinking about repositioning the elements and somehow fit them to 2.0 ratio. And yes, I don't even know what ratios are good... I hate videos...

He probably means uploading straight emulator dumps at least.

Thanks for the encode, feos! I would've tried to comment more on this movie earlier, but I've been on vacation. The game is very nice. It feels like a quicker version of Nuts 'n Milk. I had fun with this, though I didn't particularly like the death abuse in level 31. If you prefer to use it, though, I'm fine with it.

I had fun with this, though I didn't particularly like the death abuse in level 31. If you prefer to use it, though, I'm fine with it.

It saves time. There was 3 other levels where it was so close to save time but it was 5-10 frames longer. Btw, a quick comparison of different levels shows that actually it's faster to load level 18 in FCEUX 2.2.1 than other levels for example level 17. The other thing I observed that in Bizhawk (and both in puNES which according to the emulator tests, it's the etalon of NES emulation) there's a 1 line of artifact after a level ends. So I have no idea which version should be preferred. Maybe the MasterjuBizhawk one.

I have no idea which version should be preferred. Maybe the MasterjuBizhawk one.

Whichever one allows you to "improve" short runs without having to do most of the work and watch as the world marvels due to the previous TAS being produced on an obviously horribly broken emulator. Wahoo!

Whichever one allows you to "improve" short runs without having to do most of the work and watch as the world marvels due to the previous TAS being produced on an obviously horribly broken emulator. Wahoo!

I can't decide that it's supposed to be funny (I was laughing on this at ~08:00 AM) or you are the first person who found out what I did in those runs. Which is both a good (for you) and a bad thing (for others) =(

I'm kidding.

I'm kidding.

Good! Now I think I should detail what I was talking about. When I was selecting vault submissions, I found that some TASer was used the same file while the "author" tag of the previous TASer remained. While I made these vault runs, I thought I want to check how the judges do their jobs. So I intentionally used their file and just deleted the input part for 2 tases (this way I can have nearly the same text they wrote). The reason for this is I always wanted to help TASvideos, and I thought that judges are like "super improving masters" and both "judges like a boss" who make deep analysis of these submissions to investigate it's authority (I don't know what word should I use). So before I wanted to apply to be a judge (however I don't think I will ever do it because of time), I thought I should do this test. This isn't a conspiracy or anything like a "secret plan to dominate the world". I'm pretty sure that judges saw that my movie files greatly differs from the obsoleted runs (as I sad, I deleted the input parts). There was no fail on the judges side and I'm happy with that. Also notice that these 2 tases are really unpopular. I wanted to make a conclusion when a submission gets wrongly accepted (everyone knows what I'm referring to), but since I already hinted it in my pervious post, I thought I should clarify everyone before thinking something worse. The 2 TASes I'm talking about: #3756: MESHUGGAH's NES Archon: The Light and the Dark in 01:10.00 #3754: MESHUGGAH's NES North & South in 01:10.93 I hope nobody makes a negative conclusion mistakenly regarding the excellent work of TASVideos users (TASers, judges, admins etc). I apologize if this broke any rules and I will understand if this leads to a negative consequence on my side.

LOL, well, but your little experiment didn't prove anything (except the obvious notion that nobody looks at the author field), because it is common technique in the traditional TASing method to start improving old movies by replaying the existing file and switching to Read+Write near the frame where a potential timesaver is noticed. Naturally, the previous author name is often left in the new run (until you deliberately change it later), as well as some initial Input (e.g. the titlescreen and everything until the Level 2), which is not important, because you still have to put a lot of effort into making your new improvement and then rerecording the remaining parts of the movie (since the old movie does not sync after you've made the improvement, and there's no easy way to resync the old Input using traditional TASing method, in fact it was often easier to just reproduce the old strategy by recording totally new Input). So, it does not matter if some parts were copied literally (using copypaste) or indirectly (using human memory and a reference movie), the main question is: was the new contribution significant enough. That's what influences judging decision the most. And analysing the author field doesn't give any useful information about the "authority" of the submission (I think you meant "authencity").

Some submissions really need a good amount of investigation. Some are just clear improvements. No one would check how the movie file header looks unless it causes a bug or something.