Screenshot of "The Core", the lost homebrew from 2001

[+Propane13]

This game seems to have been long forgotten, so I thought it was an appropriate time to share the story.

 

Around 2001, a new game called "The Core" was advertised and near completion.  Unfortunately, due to an unexpected and unfortunate basement flooding, the hard drive that contained the game was completely lost, even though pre-orders were already taken.  The author had talked about attempting a redo, but it sadly never happened.  Eventually, I believe all money was refunded (even if it took a while), so that was at least a positive note to end on.

 

Here is the only screenshot (as far as I'm aware):

 

 

Some highlights of described gameplay:

• The game used the Driving Controllers (a la Tempest).
• Was NTSC only (no PAL)
• Two players only (No AI in the current version)
• The outer wall had 20 segments; one hit turned it grey, two hits turned it red, and 3 hits lost the game.  Via Select, you could choose a "2 hit" option instead.
• The inner player had to defend the entire center from a number of hits defined by the Select switch.  The remaining number of hits was represented by the horizontal black lines inside the center (in the above screen shot, 3)
• Using Select, you could choose how many rounds were needed to win.  Players' wins are on top of the screen.
• The difficulty switches could "lock" your defender such that it rotates along the circle.  Alternatively, an unlocked defender is rotated using the fire button. 

 

This is one game I think about every few years, just because I thought it would be a really interesting spin on Pong and Jedi Arena.  I have always been curious if someone could duplicate a mock-up of the screenshot above with modern tools, just to see how it would look if the image was more crisp.  The circle is really well defined and looks great, I think.

 

Well, that's all for now.  A vaporware title, but an interesting piece of history.

[GrizzLee]

Awe yes, I remember this well. I was one who got a refund. 
 

I truly believe had good intentions. Life events got in the way. 
 
Thanks for the memory lane trip.

[alex_79]

That screenshot looks quite suspicious to me.

 

You need to use two players or two missiles to smooth the outer circle like that, and there aren't enough objects left to draw the two paddles, the ball, and to smooth the inner circle (which needs other two objects) without flicker.

But since, according to the original website, the screenshot was taken from a real console using an hardware device ("Snappy"), some objects would either be missing or displayed as horizontal stripes (if the device captured two fields at once) in case flicker was really being used.

Some "on-the-fly" color updates would also be required, as missiles and relative players share the same color.

 

Quite a lot of advanced techniques for someone's first attempt at assembly programming:

 

Quote

[...]

I wrote "The Core" on my PC as a text file in assembly for the 6507 (the microprocessor for the 2600.) I found "The Stella Programmer's Guide" by Steve Wright, and the book "6502 Assembly Language Programming" by Lance Leventhal invaluable while writing the game. I also used the Biglist Stella site for a lot of my information, though I am ashamed to admit I haven't joined the Stella Programmer's list which the Biglist site is a archive site for, but that was partially because I wasn't sure I could actually write a working game. (This is actually my first assembly language program I have ever written.) I didn't want to announce I was working on this program, and then not be able to do it. So I was pretty quiet about the whole thing. Now that I know I can, I will be joining the Stella list.

[...]

 

EDIT:

BTW, the game was supposedly finished in 1998. This is from "rec.games.video.classic" usenet newsgroup, and is from November 1998:

From: "Lee R. Krueger" <xxxxxx>
Subject: Re: Where is "The Core " ?
Date: 1998/11/12
 [...]
Newsgroups: rec.games.video.classic

MGedeon216 wrote:

> I have E-mailed the responsible party twice over the last 10 days and have not
> received a response yet !  I am starting to get worried as I purchased 5 of
> them.  I dont like wasting money ,: (
> If anybody has heard any news on this very delayed project please fill me in.
> Mike Gedeon : )
>
> (www.videogameconnection.com )
>
> below is the link to the Core site:
>
> http://www.plethora.net/~paulo/thecore.html

Paul just emailed me the other day. I assumed he emailed everyone else. Here is
his message:

Take care,
Lee

***********************************************************

Since I e-mailed you last time a couple of months ago, here is what has
happened:

I have soldered almost half of the eeproms to the boards, but it has
been fairly slow work.  I have not had as much free time as I expected,
because the transmission on my car went out in mid-September, so I've
had to take the bus to work, and it takes an extra hour each way.
(Repairing the transmission would cost more than the car is worth.)  So
I haven't been able to stay up as late to work on the soldering.

I have also had a lot of weird things happen over the last month.  For
example, our cat peed in our dryer earlier this week, and I've spent the
first part of this week taking apart the dryer and washing it out with
bleach.  Also, in the last two weeks, both of my children fell down and
cut their head open.  The second time, on Halloween, when we went to the
hospital they used Dermabond, which is essentially sterile Superglue, to
glue her cut shut.  Well, the doctor was completely incompetent and
glued our daughter's eye shut, and then tried using nail polish remover
on her eye.  Then the doctor left us, and told us to get her eye open,
and then the hospital sent us home without fixing the problem.  Her eye
opened later on in the week, but I had to spend my time on that instead
of getting other stuff done.  (Our cat peed in the dryer the night of
the day our daughter's eye finally opened.)

This kind of stuff has been happening for the last month and a half, and
so I have had trouble finding time to work on the game.  However, I have
still been working on it, and making slow progress. I'm really sorry
that I don't have it out yet, but it will probably be two more months.
If it isn't ready by the end of the year I will mail what I do have
done, and mail the rest when it is finished.  The instructions are back,
but I still haven't done the box.  (If someone else wants to do the box,
please let me know.)

I know some of you are probably tired of waiting.  I still will refund
anyone's money who doesn't want to wait any longer.  I didn't realize I
would have so little spare time, or have so many weird things happening
to me.  (I was promoted to head of my dept. for this year, and it is
taking a lot more of my time than I expected, too.)  Please e-mail me if
you have any questions.

One thing that I never mentioned because I wasn't planning on working on
it is the fact that as I wrote The Core I also wrote a Yahtzee game for
the 2600 just to try to see how certain things worked for it, such as
the graphics and random numbers.  I never added a computer opponent, and
didn't concentrate on making it polished.  However, I feel really bad
about how long The Core is taking to get out.  I'm going to try to
polish Yahtzee and add a computer opponent after The Core is shipped.
I'm going to have someone else who has time make a copy for each of you
when I get done.  (No instructions or box, just the cart.)  This is one
per person, not one per order of The Core.  (I can't afford to do that.
But yes, I am going to pay for it myself.)  You will get this by the end
of next year.  That's probably how long it will take for me to get the
spare money to have the carts for Yahtzee made.

Again, I'm very sorry about the delay in shipping The Core, but I have
been making progress, though very slowly.

Thanks,
Paul


--

__________________________________________________________________

 

Edited December 28, 2022 by alex_79

[Thomas Jentzsch]

1 hour ago, alex_79 said:

That screenshot looks quite suspicious to me.

/signed

[Rom Hunter]

That screenshot rings a distant bell.

Is that even possible on a VCS?

The story around this game sounds to complicated for a hoax IMO.

[Thomas Jentzsch]

BTW: Driving controllers seem like a bad idea for such a game. The resolution (16 directions per 360°) is just too low. 

[Thomas Jentzsch]

Last snapshot while the website was still up:

http://web.archive.org/web/20010419194407/http://www.plethora.net/~paulo/thecore.html

 

At least the pixels of the screenshot seem correct. In the vertical middle, there are 28 pixel outer circle and 40 black pixel gap left and right each, plus 24  pixel green inner circle. Since everything is hires, there must have been quite a lot of flickering. E.g the inner circle would flicker the left and right 8 pixel using one player. The middle 8 pixel could have been the missile, but that is obviously used as paddle too. For the screenshot it could be faked. Else the code would have to use playfield. Which means there are 3 playfield color changes per scanline. With two changes being at varying positions! I suppose the code had several kernels just to get he playfield color changes timed correctly.

 

For smoothing the outer circle's border, one could use the ball. But only either the left or the right half of the circle. So more flicker here.

 

The green paddle is either the missile (where the inner circle is), or the player (to allow a better locking horizontal line). Since there are multiple kernels for the playfield anyway, implementing this should be no major problem.

 

I think one could create the kernel for the screenshot display quite easily.

Edited December 28, 2022 by Thomas Jentzsch

[alex_79]

After digging a bit more in the old rgvc newsgroup archives, I found this post where the author gives some technical details:

Quote

The 3 colors were done just by writing to the COLU** registers at the
appropriate times.  I place the ball and paddles after the score and
before the circles are drawn so I don't have to waste time doing it
during the drawing of the circles.  The large circle is drawn with the
playfield, plus smoothed with the ball sprite and one of the missile
sprites, which are moved using a lookup table every two lines.  (The
reason for those black lines down the left side.)  The paddles are the
player sprites, and the ball is the other missile sprite.  During
overscan the location of everything is determined, so the kernel just
has to know when to turn on the paddles and ball, since everything else
is drawn the same every time.  But I can place the ball and paddles
horizontally right after the score, so all I have to do while the circle
is drawn is to write the graphics at the correct time for the players
and enable the ball which is really a missile at the correct time.
Which leaves enough time for hitting COLU** at the right time.

 

The only hard thing was that damn center circle.  Well, except for a top
and bottom line (only playfield graphics), while it is being drawn, the
larger circle is flat and drawn using only the playfield graphics.  I
reposition the ball and missile for the larger circle and draw the
middle circle (also using playfield graphics) then when that's done move
them back to the large circle.

 

The inner player does have an easier time, so switching off is
recommended for the game setting.  However, the outer player can get hit
much more often if it's not the same spot.  While playing the outer
player, when I only need to hit the inner player one more time, I can
sacrifice one of my own spots if it isn't red to hit to deflect it to a
spot the inner play can't reach.  It really balances out more than you'd
think.  There is a LOT of strategy in this game.

 

Paul

 

The part highlighted in bold doesn't match what we see on the screenshot, which makes me think it is photoshopped (maybe merging different screenshots, each one from a program drawing only part of the screen. The author seems to actually have some knowledge about 2600 programming, or at least he studied the docs before posting.

 

Here are the relevant messages about "The Core" that I could find in the archives after a quick search. The entire story seems fishy, to say the least.😄

thecore.zip

 

EDIT:

Finally, the "2600 Connection" newsletter, issue 66 (May/Jun 2001), page 6, under "News & Notes" says:

Quote

A long time ago a game called The Core was announced by Paul Oswood, who took preorders and then disappeared. It never seemed quite like a scam despite the lack of a game in hand by many purchasers... It turns out it wasn’t a scam, as Paul has recently come forward, giving refunds for preorders and hoping to redo The Core (the original was lost in a flood that killed his computer) at a future date. If you preordered the game and haven’t heard from Paul yet, let me know.

 

Edited December 29, 2022 by alex_79
added "2600 Connection" article

[Thomas Jentzsch]

3 hours ago, alex_79 said:

The part highlighted in bold doesn't match what we see on the screenshot,...

I think it is just described not very well. After rereading multiple times, I think I got him. The outer circle is (mostly) flat while the inner circle is drawn. That allows him to use the smoothing missile and ball for the inner circle there. Pretty clever.

 

Nevertheless, there still is something missing. If one missile is used for smoothing the outer circle, this has to be either black or the current color of the outer circle. That doesn't match the green paddle color. So I suppose he is also updating one player color per line.

 

This all adds up to quite a lot of cycles. I am tempted to find out if it is feasible by doing my own kernel.  

3 hours ago, alex_79 said:

The author seems to actually have some knowledge about 2600 programming, or at least he studied the docs before posting.

Definitely.

Edited December 29, 2022 by Thomas Jentzsch

[alex_79]

What I meant is that if you look closely, the smoothing of the inner circle extends beyond the area where the outer circle is flat:

[Thomas Jentzsch]

8 minutes ago, alex_79 said:

What I meant is that if you look closely, the smoothing of the inner circle extends beyond the area where the outer circle is flat:

Got you. He writes about one top and bottom line which is playfield only. That one could be done. But there are two lines at top and bottom. So how is the 2nd line done?

 

Also, the repositioning would hide (the first copy of) the missile. And IIRC the ball too. So both cannot be used for smoothing the circles then.

[TrekMD]

I had never seen this before.  Interesting.

[Shawn]

If he had already soldered chips to boards there would be a very good chance those boards would still work perfectly, even after a flood. Perhaps a copy or three are still out there?

[glurk]

2 hours ago, Shawn said:

If he had already soldered chips to boards there would be a very good chance those boards would still work perfectly, even after a flood. Perhaps a copy or three are still out there?

He'd also still have the entire SOURCE CODE really, just a disassembly away.  Would lose all labels and comments, etc., but it should be enough for the original programmer.  But this whole thing smells like a big fish tale to me...

[Thomas Jentzsch]

On 12/29/2022 at 2:25 PM, Thomas Jentzsch said:

Got you. He writes about one top and bottom line which is playfield only. That one could be done. But there are two lines at top and bottom. So how is the 2nd line done?

I experimented a little bit and it seems that the overlapping wasn't even necessary. 

To overcome the limitations of the driving controller (16 steps/full turn), we need definitely would need (much) larger paddles. For the inner player at least ~11 pixel, for the outer player at least ~32 pixel. That way the paddles could cover the whole circle within one 360° turn. They would move in steps, with no overlapping then. With overlapping, the paddles would have to become even larger.

 

Also I think different paddle sizes would make the game more fair and balanced.

[+x=usr(1536)]

9 hours ago, Thomas Jentzsch said:

To overcome the limitations of the driving controller (16 steps/full turn), we need definitely would need (much) larger paddles. For the inner player at least ~11 pixel, for the outer player at least ~32 pixel. That way the paddles could cover the whole circle within one 360° turn.

Just thinking out loud: there's a way that this could possibly be compensated for.

 

360° in a circle divided by 16 positions in the driving controller results in 22.5° of sector for each controller position.

 

Have each pulse of the controller move the player only one half of a sector, giving a resolution of 11.25° segments for 32 positions within 360° of movement.

 

The downside to this is that the player would move at one-half the speed compared to having 32 dedicated positions on the controller itself.  While it may be possible to work around that to some extent by spinning the controller faster than usual, precision of movement would likely be adversely affected.

 

The Trak-Ball might be a better alternative.  Hell, even joystick control à la Gyruss could potentially be an improvement over half-speed movement.

[Thomas Jentzsch]

@x=usr(1536)Not sure if I get you, but I think your suggestion would mean, that for a full turn of the paddle, the player would have to rotate the controller twice.

[+x=usr(1536)]

7 hours ago, Thomas Jentzsch said:

@x=usr(1536)Not sure if I get you, but I think your suggestion would mean, that for a full turn of the paddle, the player would have to rotate the controller twice.

Effectively, yes.

 

What I was proposing would double the resolution of the player's movement, but at the cost of halving the speed at which the player could move.

[Thomas Jentzsch]

After I invested some more time, I think the game is feasible. Just not exactly as shown on the screenshot.

 

I wanted to see if a driving controller is good enough. So I implemented the inner paddle, and there it seems to work quite well. If I increase the size of the outer paddle (unlike in the screenshot) to reflect the size of its large circle (~3x), then it would work well too. Else the outer player would need three full 360° turns to move around once. And both paddles in the screenshot are even a bit smaller!

 

Now I wonder how the paddles would have to look like. Straight like in the screenshot or rounded like the circle they are moving on. I suppose the former would convert into better gameplay. E.g. imagine a rounded outer paddle, angled bounces would feel quite unnatural. 

 

Anyway, I think the concept has a major problem due to very small gap between inner and outer paddle. The ball has to move quite slow, else the reaction time is too low. I don't think this game would provide a lot of fun. In the end it is still more or less just Pong.

 

The screenshot shows my code's screen and kernel (the shading) layout. I have also attached the ROM, just in case someone wants to test the driving controller responsibility on real hardware. Within an emulator you have to tell it to use the DC. Note: The outer circle is just there as a placeholder.

The_Core_004.bin

[CrazyChris]

I always thought this was an excellent game idea.

[diddlybopshubop]

On 12/29/2022 at 2:57 PM, glurk said:

He'd also still have the entire SOURCE CODE really, just a disassembly away.  Would lose all labels and comments, etc., but it should be enough for the original programmer.  But this whole thing smells like a big fish tale to me...

It looks like he's still a mathematics instructor in Minnesota. I won't link to it here, but it's pretty easy to track down his contact info online if anyone wants to reach out to him and see if he still has any of the original materials available.

[N0mi]

Yeah, also saw this yesterday, but I doubt he will be eager to revisit The Core considering how the whole thing turned out for him in the end. 

Edited January 3, 2023 by N0mi

[Thomas Jentzsch]

Now that I have a working kernel, does anyone have a good idea for some interesting gameplay?

[+x=usr(1536)]

2 hours ago, Thomas Jentzsch said:

Now that I have a working kernel, does anyone have a good idea for some interesting gameplay?

2-player Tempest, with both players - two humans or one human and one AI - shooting at each other from the inner and outer circles while trying to avoid and/or destroy enemies that appear?  Add in some sort of co-op aspect that makes it less appealing to just nuke the other player and it might be workable.

[diddlybopshubop]

4 hours ago, N0mi said:

Yeah, also saw this yesterday, but I doubt he will be eager to revisit The Core considering how the whole thing turned out for him in the end. 

Fair enough! Thinking about it a bit further, he has more avenues to release what he had - assuming he still has anything - to the community today than he had 20 years ago.

Maybe things have settled down for him a bit since then.