Fixing Pole Position (and enhanced source code)

[YANDMAN]

This is awesome, thanks guys for all your efforts, would love to see the 7800 version altered in this way with kevins sprite hack included , that would be one sweet version.

 

Cannot wait to try this out later.

[Lauren Tyler]

I never thought Pole Position could be made better, but it is!

[Thomas Jentzsch]

Late to the party again...

Now if the game can be hacked to use a paddle or steering controller instead of the joystick...

Paddles are most likely impossible, but a driving controller might work. It is still better to have only 2 or 3 different steer angles than only 1.

[Keatah]

Not sure why I see paddles as being impossible. Shouldn't it be as simple as reading the value and then using that value to either turn the car's steering wheels? Or pulse-modulate them. And I think all that'd be needed is a couple of steps in either direction. Maybe it's more complex than that. Idk..

[Tempest]

Did we ever verify if there was a mysterious 'extra car' or not? I could have sworn we were having an argument about that not too long ago.

[Thomas Jentzsch]

Not sure why I see paddles as being impossible.

Depending on the resolution required, paddles have to be polled very frequently during the display kernel (usually every scan line). I strongly doubt there is enough CPU time for this.

 

But now that I think about it... If we are lucky, we maybe find the time required every n scanlines. That might be sufficient for e.g 10-20 different paddle positions. But I have to look into the code and debugger first. And then we have to utilize this input in the code if you want more than just left or right.

Edited August 18, 2017 by Thomas Jentzsch

[Thomas Jentzsch]

Did we ever verify if there was a mysterious 'extra car' or not? I could have sworn we were having an argument about that not too long ago.

I remember that, but IMO that's a myth.

[Keatah]

I think that 5 positions left and 5 positions right would work for paddles. Should provide enough granularity/resolution.

 

And, somehow, driving controllers for this game don't interest me.

[+Mitch]

The arcade version used a spinner not a pot...

 

Mitch

[Keatah]

That's what I don't like. Spinners spin "forever" unlike a real steering wheee that goes from lock to lock.

[Swami]

This is all cool! Can't wait to see the next developments!

[+Mitch]

That's what I don't like. Spinners spin "forever" unlike a real steering wheee that goes from lock to lock.

 

So I assume you don't like the arcade version of Pole Position then?

 

Mitch

[Keatah]

Oh I liked the game. But I always felt the controls could've been more to my liking. I really like the 400/800 version even though it's a joystick. Ahh well..

[bigdcaldavis2k]

Now if the game can be hacked to use a paddle or steering controller instead of the joystick...

The thing is, it's impossible to shift gears with the paddle or driving controllers. You would have to design a new 2-button paddle/driving controller and used the left button for your brakes and the 2nd button to shift gears, and you would have to re-code the game to support the new controller.

[Swami]

The thing is, it's impossible to shift gears with the paddle or driving controllers. You would have to design a new 2-button paddle/driving controller and used the left button for your brakes and the 2nd button to shift gears, and you would have to re-code the game to support the new controller.

Or, way out there, use the paddle as the wheel and accelerator and the joystick in port 2 for shifting and the joystick button for breaking, like on the Colecovision EM2. Not sure if any 2600 games did this two joystick play except Stargate/Defender2.

 

There is also the Star Raiders pad.

[Arenafoot]

The thing is, it's impossible to shift gears with the paddle or driving controllers. You would have to design a new 2-button paddle/driving controller and used the left button for your brakes and the 2nd button to shift gears, and you would have to re-code the game to support the new controller.

or use the CBS Booster Grip (from Omega Race)

[Asaki]

I'll have to try this out on the Portable. I know the game already has a slight bit of graphical glitching going on there.

 

In my tests, it was certainly possible to go through the laps without ever slowing down (which would be impossible in reality! Keep in mind that the track is based on the then-current version of Fuji Speedway, but greatly simplified to fit into the cart).

The original arcade game is the same way, though

[Arenafoot]

I'll have to try this out on the Portable. I know the game already has a slight bit of graphical glitching going on there.

 

The original arcade game is the same way, though

I tried the 3 files [PolePosition(updated).zip] on post #23, they work but are flawed. The enemy cars (just like the original ROM on the AFP) show up off the track in the distance.

[Asaki]

The enemy cars (just like the original ROM on the AFP) show up off the track in the distance.

Yes. Some of the glitching at the bottom of the screen is less distracting now, though.

 

There's also a few blocks of image on the very left side of the screen visible that weren't before.

[+Random Terrain]

In the last days, I've been reading Lou's Pseudo 3D page after reading up on how Sega's sprite-scaling and drawing hardware works.

Wow, that is a good page:

 

www.extentofthejam.com/pseudo/

 

I wonder if that article would be any help for making roads on the Atari 2600 using batari Basic since bB has limitations that assembly language users don't have?

[MayDay]

I've messed around with this a few times over the years, but I always wondered if something like this could be possible for a hack?  The idea is that the playfield used for the main car could match the quad-width pixels of the competitor cars, so that all the cars look similar, and better.  If possible, I wouldn't think it would be too much harder to add variant colors to the other cars as well.  Thoughts?

[MayDay]

On 8/18/2017 at 12:48 PM, Tempest said:

Did we ever verify if there was a mysterious 'extra car' or not? I could have sworn we were having an argument about that not too long ago.

I know this was a couple of years ago, but I have been looking into this a bit.  I'm not sure about the grand prix, but I do know that sometimes you pass nine cars on the qualifying lap and sometimes ten.  Not sure if the grand prix has one less or not if you get the pattern with ten.  I do have screen captures of both nine and ten cars in the qualifying lap to compare against.

[Tempest]

23 minutes ago, MayDay said:

I know this was a couple of years ago, but I have been looking into this a bit.  I'm not sure about the grand prix, but I do know that sometimes you pass nine cars on the qualifying lap and sometimes ten.  Not sure if the grand prix has one less or not if you get the pattern with ten.  I do have screen captures of both nine and ten cars in the qualifying lap to compare against.

Weird.  I wonder what causes that?

[MayDay]

Not sure.  However, looking through the source code, there is a counter for passed cars.  I guess it might be needed if you dont finish the grand prix, but if it was always the same I would think it would know based on how far you made it in the race.  I wish I understood the code better so I could figure out what actually cause lane (and more in particular double) changes deep in the grand prix.

[MayDay]

So, I've found another extra car anomaly.  In the grand prix, there are almost always 14 cars that you pass on the first lap.  I have a recording where I pass 15, and it's not even a clean lap.  It seems on occasion, the cars just generate a bit earlier than other times.  Unfortunately I crashed out, so I couldn't see how it affected it at the end.

 

There is definitely things going on with the code that I simply don't understand.  I've maintained for years that the emulation version of the game is MUCH harder than the hardware original based on the fact that there are way more late lane changes, as well as double lane changes.  I also discovered something else I find very interesting- normally 56.73 is the fastest you can do the qualifying lap, and it depends on the pattern you are given.  This is listed as the fastest Atari2600 lap on speedrun.com, matched by several others, and I have about 20 recordings of running a 56.73 lap.  Even driving a "perfect" lap hugging the rails and not even squeaking the tires does not improve this time, yet it doesn't have to be perfect to the get the 56.73.  There are at least two patterns that give this time for a near perfect lap, while many others give a 56.74 or some a 56.75.  By simply disabling the collision detection (and still taking the normal route around enemy cars), that same route becomes a 55.22. 

 

There are other effects to this collision detection change- the music and sound effects become an octave higher.  There are rarely lane changes at all for enemy cars.  If and when they do change lanes you often see patterns of 4 double lane changes in a row.  The time goes down, and the highest score goes up by quite a bit (62660 on first try) even though I don't believe you are passing any more vehicles than normal.  Finally, the grand prix time does not alternate with the score, it simply only shows the score each time it flashes back.

 

I'm most interested in the lane changes, as understanding when/how they do that is key to playing the game to as close to perfect as possible.