Tournament: Vector War VIII (10th Nov - 17th Nov 2018)

[fruktodlaren]

Player: Fruktodlaren


Blitz!: 23

[Chainsaw]

Scramble: 126440 (reachted Level 5)




Star Hawk: 138300 (Too tired for this game today...)




Spike: 69900

[fmml]

Hate to post my scores after Chainsaw
I lost my highscore for Spike by pressing reset while using my Vecfever. This appearently frustrated me enough I couldnt get anywhere near again. Oh well. 
Release is such a fun game!


Starhawk 43900

picload.org/view/dcolcrrr/img_6977.jpg.html

Web Wars 14

picload.org/view/dcolcrra/img_6973.jpg.html

Release 12777

picload.org/view/dcolcrrd/img_6969.jpg.html

[Malban]

Gz Chainsaw.
Looks like I have to squeeze some small points out of those games again :-).

[VectorX]

^Chainsaw can be a tough guy to beat.

[Chainsaw]

Malban, yeah, it was already hard to beat your Scramble score. But I did some mistakes, so more points are possible :-) Finally you will run out of fuel in level 5 of Scramble, I think in the second section of the level... Or is more possible?
But your scores (Malban, VectorX and of course of the other players) are also great... I think, finally you will beat me...

[hcmffm]

Highscore list has been updated.

fmml has qualified.
Ranking: 1. Malban (99,61%) 2. Chainsaw (99,40%) 3. Fruktodlaren (82,75%)

___

fmml: Congratulations, Roland, pretty good scores!

fruktodlaren: Both fun and a bit strange experience once you realize that your children can do better than yourself. Perhaps Len give you some lessons in race driving.

@Malban, Chainsaw: The German and the Swiss discuss about optimizing Scramble scoring.

Fourth day of the Vector War. Many scores updates and an interesting phase of the Vector War:

More and more games have a high score which is hard to beat. I guess for Vector Patrol the "Star Castle effect" will reappear: The game won't be played anymore because its current highscore is so high that you get very little reward (percentages) for playing it.

I'm nosey whether more players will participate or even veterans like coleco1981 and wazzal will join in. And sure enough I'm nosey which player will win.

[jbrodack]

Tried to do blitz but once I figured out how to play it kept crashing on me while using my vecmulti. Maybe I need to patch it or something.


Vector Patrol 95180 (I reset my scores before playing for Vector war)

http://instagram.com/p/BqG7JcnlTxN

Release! 32107

http://instagram.com/p/BqG9twDlMlF

Bedlam 95000

http://instagram.com/p/BqG_ytelOnT

Star Hawk 77800

http://instagram.com/p/BqHBgeQFin6

I'm wondering how long it took to get the top score for Vector Patrol

[nexus6]

Phew, these scores are hard to beat. So little time, so much scores … Need more practice.

photos.app.goo.gl/Yu2C86kFvukSCn8o9

Asteroid Cowboy: 246

photos.app.goo.gl/MNfkpkwmQrijwT6N8

Floor is Lava: 38 (It was more, but i was too slow to take a picture)

photos.app.goo.gl/oFXJPa82xW6hF3GY7

Scramble: 97090 (Update) There must be more … too many stupid crushes in the corridors

[fmml]

To be fair the score is for Starhawk, you mistakenly put it down for Spike. But, thank you for trying to help me out Helmut. Appreciate it.

[Malban]

Hi,

I asked if I might look at the sources of Asteroid Cowboy in order to find the cause of the crashes.

I located one code section, that could overwrite the STACK and thus cause a crash.

The fix was an additional "if clause". I don't know if this was the only place. But if you play Asteroids Cowboys - you might as well use the "fixed" version - and please report if you experience another problem.

vectrex.malban.de/tmp/AsteroidsCowboy.bin

Regards

Malban

[Peer]

Greetings everyone,

first of all, a big thank you to Helmut for selecting two of our games (Asteroid Cowboy and Floor is Lava) for this year’s Vector War event. This is a huge honor for us and a big motivation for all the students. It is really great to see you all playing the games. Also many thanks to those of you who sent personal feedback. I will make sure that your words are relayed to the students.

Volcanus, who created Floor is Lava, was kind enough to provide a special Vector War Edition binary. It is 100% identical to the previous binary as far as gameplay is concerned, but it features a slight redesign of the scoreboard with more time for the screen capture of the final score of a game. Some of you had reported difficulties with taking pictures of their scores. You can find the new version on the project page of Floor is Lava. The binary also features a new melody called "Serenade" to say thank you to all the players.

Have fun and enjoy the games!

Cheers,
Peer

[nexus6]

Thanks for the fixed versions, Malban and Peer!

[andyi]

Hello everyone,

@malban many thanks for your efforts! It would be awesome if you could tell me what parts of the source you suspect to crash the game. I hope to find some spare time in the next weeks/months to fix it and release a crash-free version, but can't promise anything because atm I'm preparing for my bachelor's thesis. If you have any questions about the source, please contact me :-)

@all:
I observed most crashes happen when an asteroid breaks the lasso, so if this is the issue, you could prevent it by resetting your lasso by pressing button 4.

Thank you very much for choosing the game and the great feedback, I am really happy about it! :-)


Regards,
Andy

[Malban]

I played quite a while in Vide till it crashed, than I tracked down, that the stack was corrupted.

In the case I found, in the function within file: "Lasso.h":

FORCE_INLINE void Lasso_addVector(Vector direction)
{
     // if lower array is full, copy it to upper array
    if (Lasso_content.firstElementIndex == 0) // check if array boundary is reached and if yes, copy values to the top
    {
        Lasso_deleteLastElement();
        // get head array (copy destination)
        Vector *upper_array = Lasso_content.elements + LASSO_BUFFER_SIZE - Lasso_content.currentLength + 1;
        Cell_Compressed *cell_upper_array = Lasso_content.cells + LASSO_BUFFER_SIZE - Lasso_content.currentLength + 1;
        // don't copy the tail of lasso, it'll be replaced anyway -> -2

        // BUG:
        // lasso LEN can be 0 here
        // than index = -2 = $fe (253)
        // max (allowed) index of array is LASSO_MAX_LENGTH (times two  == 110)
        // with the larger index ->
        // the stack is overwritten and a crash happens
        uint8_t index_of_byte_to_copy = Lasso_content.currentLength - 2;

            while (index_of_byte_to_copy != 255) // 255 means underflow, so everything copied
            {
                upper_array[index_of_byte_to_copy].yx = Lasso_content.elements[index_of_byte_to_copy].yx;
                cell_upper_array[index_of_byte_to_copy] = Lasso_content.cells[index_of_byte_to_copy];
                --index_of_byte_to_copy;

            }

Nice game Andy, played it a lot the last couple of days.

Regards

Malban