I seriously think you should fill the whole site with huge flashing gif banners and even huger flash ads as well as unrequested popups which refuse to close and which obstruct your view until you click on them...
Just kidding, of course. The ads seem ok because they don't bother.
DW4 is perhaps the best nes game I have ever played and which I have the warmest memories about (it feels incredible how much empathy you can feel towards a 3-color 16x16 pixels sprite on a computer screen).
While I'm a bit too lazy to actually get the emulator, the rom and watch the movie (and thus to vote), I'm really looking forward for the avi...
You don't specify your system, but if you are using Windows, try a codec pack. This one is superb: http://www.free-codecs.com/download/K_Lite_Codec_Pack.htm
(When installing, make sure that ffdshow/h264 is checked. At least some version had it unchecked for some reason.)
Isn't Mario Paint just a painting/composing program? Plus the bug squatting game, for which there already is a TAS, I think. What would a TAS of Mario Paint aim for?
Sheesh, people don't have a sense of humor.
It was a joke. A joke. :P
So, anyone willing to do a TAS of this wonderful game?
Another one which I would really like to see is the old NES classic color-a-dinosaur, even though mario paint is technically slightly better.
I have to say that very old NES games suck big time, at least technically. :P
Just compare this elevator thingie with eg. the bucky o'hare game. The difference is huge.
I see this as redundant; people know that there are more movies on the site than the starred movies. If they are a fan of the game, they'll go looking anyway, regardless of whether there is a link in a starred description.
I think you didn't quite understand what I meant.
Some people felt the need to give a start to all those smw movies because they thought they were all good. The problem with this is that there are too many of them, and less would be enough. Now, what I meant was that if someone watches one of the smw movies and gets interested, the text&links will tell him that there are other "almost starred movies" of the same game (series) too. This is almost like giving all the movies the stars, but the actual list gets less cluttered.
Besides, it might not occur to a visitor viewing the starred list that there might be more movies for the same game so there's a high probability that he won't even try to search.
The same thing could be done to other starred movies with alternative runs for the same game (or series) such as rygar and metroid.
One would think that RPGs would be one of the least suitable genres for a TAS, but this one proves that concept wrong.
(Of course it doesn't mean *all* RPGs are suitable, but at least some are, and the runs can actually be pretty interesting, such as in this case.)
Meh...I think excellent games are excellent games, no matter how many of them there are.
Games may be excellent, but watching a TAS is a differen thing... :P
The problem I see with the current super mario/demo world stars is that, counting all their playing time together, it's a whole lot. I haven't actually counted, but it's probably several hours in total. Less would be more than enough for a first-time viewer.
I have an idea to alleviate the dilemma of removing the stars from some of them: In the description of the starred one put links to the other movies with a text like "for other similar and equally excellent movies see also: link link link". This way the list of starred movies gets less cluttered but those interested in the marioworld runs in particular (after they see the one starred) can still find all the runs.
Another thing which just occurred to me is that the "least downloaded movies" gets actually filled mostly by the newest movies. Perhaps there could be a category like "least downloaded movies older than 1 month" or something to truely show the least popular ones
Actually, thinking more about it, that problem is very small, almost nonexistent. New movies are published rather rarely and if they are at all popular, they get off the list very fast (probably in a few days at most). In other words, I retract my statement: I don't think the list gets filled with the most recent movies, so there's not really a need for an "older than 1 month" version of the list.
Other statistics using the date of publication could be interesting though.
I'm not sure about giving it a star, though... don't we have too many stars right now?
I think one possibility is to remove stars from some currently starred movies. Do we really need two starred movies of the same supermarioworld game? Overall, I think there are too many "super mario/demo world" stars. I propose that we choose one or two of the best of them (ie. the most entertaining ones).
The same could perhaps be said of the two starred sonic movies. Even though they are movies of different games, they are neverhteless movies of the same game series which are extremely similar to each other. I propose we choose the best one (ie. most entertaining) of the two.
This same principle is being used with other games too, such as Rygar and Metroid (more than one run for those exist, but only one, the most entertaining, is starred), so why the same principle cannot be used with the supermarioworlds and genesis sonics?
IMO the stars to both SMB2 and SMB3 are justified because they are quite different as games (even though they, technically, belong to the same game series). I don't understand why SMB1 is not starred, though.
Megaman 1 and 2 are rather similar as games, but I think in this case it's also justified to give stars to both because the TASes are unusually stunning.
One thing that's been strange to me about downloading statistics is that obsoleted movies are not accounted for. Someday when SMW-11 is obsoleted, its download statistics are discarded, and as far as the goal of statistics, it doesn't seem accurate to me.
I'll have to consult Bisqwit about whether bittorrent statistics are preserved for obsoleted movies or not. It would indeed be interesting to have "most downloaded movies of all time" statistics.
Another thing which just occurred to me is that the "least downloaded movies" gets actually filled mostly by the newest movies. Perhaps there could be a category like "least downloaded movies older than 1 month" or something to truely show the least popular ones (assuming it's possible to get publishing dates; I'll have to consult Bisqwit about that too).
(Btw, if your movie gets into the least popular movies list, don't get too depressed. That list might actually make the movie more popular, helping it get out of the list... :P )
Suggestion: Add the ability to select a different number than 10 to show of each category - I'd be interested in seeing what the top 20 or 30 of something are, what the shortest run is that's longer than marble madness, etc.
The problem with that is that calculating the contents of the page is a more or less heavy process and that's why it's done only each 10 minutes (what you see is just a cached version of the page). If it was done each time someone loads the page, that would be quite a security risk (someone could cause extremely heavy loads to the server by requesting the page repeatedly very fast). This may especially be true in the future if parsing of the movie files is implemented, which is a very heavy operation.
We started creating a new statistics page as a joint effort with Bisqwit:
http://tasvideos.org/MovieStatistics.html
Note that this is a heavy Work In Progress (TM). Our plan is to make it versatile so that statistic tables could be much more easily created anywhere in the site.
Any ideas and suggestions are welcome.
(Note: One idea we had was to create statistics based on the amount of keypresses in movies. However, making code to parse all the different movie file formats is rather laborious, and the process itself is rather heavy, so for now we left those statistics out.)
If you have problems watching videos in Windows, this should be the ultimate solution:
Download and install the K-Lite Codec Pack: http://www.free-codecs.com/download/K_Lite_Codec_Pack.htm
Make sure that during the installation the H.264 support of ffdshow is turned on (I think that in at least some versions of the pack it was turned off by default). During the installation it will offer you to uninstall unneeded codecs/players already installed in your system (to keep it clean); afaik it's safe and a good idea to let it do that.
If you still have problems viewing the videos, play them using media player classic or bsplayer, both of which come with the pack. Winamp5 also is able to play most videos (except multi-soundtrack ones). Sometimes just having the codecs installed is not enough for a player (such as Windows Media Player) to be able to correctly play a video; media player classic and bsplayer should support basically everything.
People have a (reasonable) misconception that doubling the image size will double the size of the file where the image is stored, and doubling the number of frames in a movie will double the size of the movie file. Testing different-sized images saved as JPEG seems to about corroborate this (even though you probably don't need to double the size of the file in order to get the same visible quality, but the size of the file increases notably nevertheless).
Although that is intuitive, MPEG-4 doesn't really work that way (it's probably the same with MPEG-2; I'm not sure about MPEG-1 - there you probably will have to increase file size noticeably).
According to my own tests (which I made many years ago), increasing the resolution of the video while keeping the same filesize (of the resulting MPEG-4 file) will actually *increase* the image quality instead of decreasing it. Not much, but some. In the same way increasing the framerate while keeping the same file size will most probably only have a positive effect on the image quality.
You can probably think about it like "increasing resolution or framerate will give the encoder more data to work with and create a higher-quality result while keeping the same file size".
Of course there are most probably some limits. You probably can't expect to keep the same image quality if you compare a 512x384 video and a 16000x12000 version of the same video with the same file size. However, with practical resolutions it indeed seems to be so that bigger resolution gives better image quality without needing more space.
Umm... Sanskrit is a dead language because there are no native speakers.
One should make a distinction between dead languages and artificial languages.
Neither of them are spoken natively (as their "mother tongue") by anyone (or at least not by any significant group of people). The difference is that a dead language has been a native language in the past but isn't anymore, while an artificial language has never been a native language and isn't the result of natural linguistic evolution, but it just has been artificially created by someone.
For example latin is a dead language while klingon is an artificial language. I think esperanto is also an artificial language (probably by far the most widespread one).
How about a tool-assisted competition? Take a game and set a goal (other than finishing the game) and set a time period for submissions. The one who gets the goal done in less frames wins.
My first idea was: Get 99 lives in SMB3. The last frame of the movie must be exactly when mario gets the 99th life.
For extra excitement, during the submission period there could be a public list of accepted submitters with their framecount (but not the video itself in order to not to spoil anything).
The time period? I don't know. It should not be too short. I would say at least one month, if not more.
The price? Well, it would be all for the glory and babes, I suppose. And the winner video would probably be fun to watch. There could also be an aftermath where people could submit their improvements to the winner video if they can.
Any takers?
In Japanese, there are only ~5 different patterns in how to conjugate a verb into past tense, but in Finnish, the number is more like >30.
There are over 130 ways to conjugate Finnish verbs altogether (compare that to the at most 5 ways to conjugate English verbs). Also nouns have about 14 cases (English nouns don't have them at all).
Curiously, though, Finnish is one of the easiest languages for a child to learn as his native language. An extensive study about this was performed recently. It measured how fast children of different nationalities learn to speak their native language fluently. Finnish was among the easiest ones. Among the hardest ones was German and, perhaps a bit surprisingly, English.
By the way, I understand and speak Finnish very well, as it is my native language (I'm actually rather confident that I can more easily write grammatically correct Finnish than the average Finn... :P ).
I also understand spoken and written Spanish very well. I was completely fluent in speaking and writing it 10 years ago (again, mostly better than natives themselves, especially the writing part), but due to lack of usage it's very difficult now.
I understand spoken and written English quite well. I can write English somewhat fluently (in some cases I need to stop to think, and sometimes even consult the dictionary), but my spoken English is very untrained.
I had a couple of Japanese courses, so I knew some rudimentary things about that language, but I have forgotten much of it due to lack of usage.
