Tom7 was pretty clear on this matter in the rules of the game:
"The bizarro world and regular world both share the same destination.
that is, if one contains a panel and the other contains a transporter,
both will necessarily reference the same spot on the playing field."
Although it would be a handy feature, I find it a bit scary as its abuse could lead to some ridiculous and near impossible levels. Just imagine a double-layered 'Dimensional Trap'.
And the next silly block idea:
HOMING SPHERES.
Function like a normal sphere except that a soon as they encounter a panel they can activate they will stop on it. |
| |
I see. But it would be a really nice feature! |
| |
If you place any two tiles that require specified destinations in the same place (one in bizarro and one in normal) they will have the same destination (this includes both colored and non-colored panels, transporters, and remotes). I believe that this is because destinations of tiles are associated with the grid space, and not the tile itself. Making the two layers have different destinations would be complicated because they would have to flip between layers themselves. D-maps would also be a mess and very confusing.
I agree it would be nice if we could change that, but I just don't see it happening. |
| |
This might be a bug report?
-----> Two colored "dots" at the same place in normal as well as in bizarro world will ALWAYS lead to the same point / aim / place. Unfortunately it is not possible to define separate aims for each of them.
I hope this is understandable......?! |
| |
I was just thinking about the idea of ice tiles for escape. I know blocks moving across it would be hard to implement, but would seem obvious. So, as a variant that would make sense and add a bit more to the idea, an alternative tile could be thin ice. Thin ice would make the player and bots slide across it, but the ice would be too thin to support blocks, and would look like if there was just blackness under it.
The property that would be special about it being thin would be that an otherwise generic type block could be made to interact with the thin ice like grey blocks do to holes, in this case crushing the ice and falling into it, turning it to floor. I'm not sure whether an existing block should do so or if it should be a new block.
Since this might present problems with gold and spheres, I offer these interaction possibilites:
1. Stopping just before it (definitely should be used for gold blocks at least) or
2. Sliding onto the first thin ice tile encountered and falling through it, leaving a hole (definitely not for gold blocks)
Let me know what you think. |
| |
(semi-spoilers) I kept on thinking you had to get everything somehow inside either Legal Pad or Minsweeper since those were the things you could directly close, and that having the games in a won state maybe had something to do with it. (The key name pun made me think you had to Dragon Drop them somewhere, heh.) Also I wasn't sure whether the excess of 7s was relevant to the puzzle or just a namely flourish :)
I commented on the LD page re: performance issues btw, it seems mostly based on the browser. |
| |
|
|
| |
Hmm, what red herrings? Almost everything is there for a reason! |
| |
Haha this is awesome. I feel like there's a few red herrings though, or maybe I just think of weird possibilities that make no sense. |
| |
Oh drat, that is a bug! Should be easy to fix that (and some others) in the jam version. |
| |
BUG: Attempting to finish by closing out Legal Pad doesn't end the game. |
| |
|
|
| |
11803. Tom 7 (67.186.7.254) –
26 Apr 2013 20:56:12
[ Quake 3 Done! ]
Ssssssuppp?? |
| |
On a scale from "I want my time back now" to "ZOMG! Reading this changed my life, I'm quitting my job ASAP," your paper was a solid 8. I am happy to see that you're still working on interesting things and look forward to more. |
| |
11800. fetjuel (70.199.134.0) –
23 Apr 2013 13:16:01
[ SIGBOVIK 2013! ]
I must say I was not expecting to wake up to Steven Pinker tweeting a link to a New Yorker article about Tom 7!! Congratulations man. :D
http://m.newyorker.com/online/blogs/elements/2013/04/can-super-mario-save-artificial-intelligence.html |
| |
In the case of Square Root of 2, the two levels are different (one is 14x14, the other 14x13) but this is not immediately obvious. I did this to change the checksum but I can't remember why, there was a reason at the time. |
| |
The comments/ratings/solutions will not be carried over to the copy in triage if you delete your level from triage and then re-upload it, from what I've seen. I'm not sure why that is the case if the level is still exactly the same, but I wouldn't be too suprised if a few minor flags and other bits of data included in the file serve behind-the scenes purposes. For example although your copy of "Square Root of 2" in triage has a speedrun associated with it and is rated difficulty 6, the copy in the graveyard still remains unrated and has no speedruns uploaded, although it is the same level. (although the original solution is two moves longer, but I can't remember whether that actually gets incorporated into the .esx file; my guess is no.) |
| |
It was a silly question really. Obviously, when you download a level you have uploaded yourself your existing run is applied to the triage version because the solution is referenced to the MD5 in the esp file.
So, to clarify what you just said, because I'm not going to test it, if I delete my own level and then re-upload it, the server based comments and solutions will be applied to the new version?
If that is the case, it's a bit of a shame, when you consider recent discussions, that the title/author/etc. header part of the esx is included in the checksum calculation at all. |
| |
Tom, I like the approach you took to making your program. How do you think it would work playing the stock market? |
| |
It can't be in the .esx itself, otherwise the md5 su would change, causing the level to lose all its comments and ratings, as those are associated with the level by its md5 sum. Tom has mentioned that himself when re-uploads have had to be made and people asked if the ratings and comments could be transferred from the original levels. |
| |
I assume it is a dummy folder for debugging the download process. Even if available, I doubt you'd find much there.
Either way, there is coding involved in the alternate graveyard because somehow you, I, or any other deleter (is that a word?) needs to be able to choose which graveyard the levels goes to.
All the files are in one collection, I have no idea what determines which folder any particular esx lands in when it arrives at destination. noname or radiant (or Tom himself of course) may be able to enlighten.
I assume it's not in the esx itself because I thought the downloaded version has the same MD5 as the uploaded version... must check that.
|
| |
11789. meemoo.org (62.78.227.100) –
16 Apr 2013 20:21:11
[ SIGBOVIK 2013! ]
I have been thinking about making a Tetris bot... where would you recommend one to start with that? |
| |
From my experience when downloading levels (I might be wrong), Escape will list the oldest version that is fully compatible with a given collection by the folder's name. For all of the available folders, that release is 200704130. However the routine that gets the folder info from the internet lists another collection/folder that is not available and cannot be found on this site, the "impossible" folder, which has its compatible version number set at 299900000. Even if you were to hack escape to retrieve this folder and have the version number 299900000 it would likely return an error. I do believe that some of the folders have not been around as long as others. The move to alternate graveyard would probably involve slightly more work. |
| |
I see three main reasons for deletion:
1. A cook fix has been uploaded. In this case the deleted level can be used as a spoiler (should anybody wish to).
2. Second thoughts. Maybe it was just a bad idea in the first place.
3. It has been improved on (as opposed to cooked).
90% of the 'mods' levels would be in the third category. If any category should be kept it is this one because there should be a level playing field. Knowledge of the previous levels, and their various solutions, was used in the creation of the next one, this knowledge should be publicly available otherwise the creator has an unfair advantage over the solver. Permanently deleting them would be hiding information from future participants or anybody trying to catch up and I feel this is contrary to the principle of the teamwork concept.
A separate folder for these types of deleted levels makes sense because:
1. They should not be considered spoilers, rather aids to design, so there is a legitimate reason for playing them.
2. Under other circumstances, most of them would easily deserve a place in triage. The only reason that is not acceptable here is because of the sheer quantity. I don't think any of us would be happy to see triage clogged up with them.
Unfortunately, this would require either a change to the source code and, therefore, a new version, or they could be flagged with a keyword in the deletion comment which would be detected by a modified server program and cause them to be automatically redirected to the alternative graveyard. Either way it's more work for Tom. |
| |
Tom think your your playfun/learnfun video and your paper research are really great.
This thing you are creating maybe used for something usefull if you think about it.
Let me congratulate you and congratulate you again |
| |