Scott Stilphen

Latest version.
Cylon_Attack-v2.bin

Latest version.
Cylon_Attack-v2.bin

Latest version.
Cylon_Attack-v2.bin

Latest version.
Cylon_Attack-v2.bin

Found this on Instagram.  Sent this to a good friend of mine who had this memory to share:
"I paid 40 bucks for Asteroids, I'll never forget it.  Went to 4 or 5 stores when it came out and they were all charging 45-50 bucks and already sold out.  Ending up standing in a line in JC Penney's that went out of the store all the way out into the mail itself to get a copy, it seemed like they had hundreds of copies for sale.  Same thing with Missile Command; that was also an expensive cart when it was new."
 

All the issues of this are on my site and have been for years:
https://www.ataricompendium.com/archives/newsletters/atari_owners_club/owners_club.html

I've never seen this one, either.  What's the source for it?  Some issues with it that make me think it's either not from Atari, or not shown in the U.S., or not real:
As was pointed out, the screenshot on the TV is of Pitfall, which would rule it out from coming directly from Atari.  Atari never hyphenated Atari Club, and the address shown (1265 Borregas Ave) was Atari's headquarters in Sunnyvale, CA.  The ad mentions filling out the form to join the club, but doesn't mentioned the city, state, or zip code.  The early Atari Age newsletters mention the Atari Game Club and the club and the newsletter were based in Half Moon Bay, CA.  The Atari Age magazine talks about the Atari Club (no hyphen) and was based in Philadelphia, PA.  Galaxian was released in March 1983 and was the newest game of all the others mentioned (Asteroids, Vanguard, Ms. Pac-Man, Centipede), so if this ad is real and was published somewhere, it likely would have been sometime in early 1983.
 

Here's an ad that was recently posted in the Atari 2600 Facebook group.  I've asked where it game from but as of yet nobody has offered a source for it.  The text posted over it obviously wasn't originally part of it, but note the different box art for Gravitar!

Here's an ad that was recently posted in the Atari 2600 Facebook group.  I've asked where it game from but as of yet nobody has offered a source for it.  The text posted over it obviously wasn't originally part of it, but note the different box art for Gravitar!

I've never seen this one, either.  What's the source for it?  Some issues with it that make me think it's either not from Atari, or not shown in the U.S., or not real:
As was pointed out, the screenshot on the TV is of Pitfall, which would rule it out from coming directly from Atari.  Atari never hyphenated Atari Club, and the address shown (1265 Borregas Ave) was Atari's headquarters in Sunnyvale, CA.  The ad mentions filling out the form to join the club, but doesn't mentioned the city, state, or zip code.  The early Atari Age newsletters mention the Atari Game Club and the club and the newsletter were based in Half Moon Bay, CA.  The Atari Age magazine talks about the Atari Club (no hyphen) and was based in Philadelphia, PA.  Galaxian was released in March 1983 and was the newest game of all the others mentioned (Asteroids, Vanguard, Ms. Pac-Man, Centipede), so if this ad is real and was published somewhere, it likely would have been sometime in early 1983.
 

All the issues of this are on my site and have been for years:
https://www.ataricompendium.com/archives/newsletters/atari_owners_club/owners_club.html

Vendel ended up with "everything that survived".... how, exactly?  What's your source of this information?  Did Vendel personally tell you that?  Or did he publicly state that somewhere?
You say all materials were trashed or auctioned off in 1996 and that the hard drives "and everything" just disappeared, and Vendel didn't seem to have it (so I guess if Vendel doesn't have it, nobody does?).  "None of the BI/WN team kept anything"...?  Then explain how BJ West has 2 different revisions of the game (with the later being the very last version of the game burned to disc before Atari closed up shop), plus the soundtrack?
http://bjwest.com/BIWN/
In this interview with BJ West, he claims the game was some 50%-70% complete:
https://www.arcadeattack.co.uk/bj-west/
 

Prior to the Harmony cart (and subsequent Melody boards), Yarusso was getting boards from Joe Grand (AKA Pixels Past).  But when the option to use flash carts opposed to programming EPROMs and soldering boards, that's when he changed over.  Unfortunately, Yarusso kept a monopoly on those Melody pcbs, to the point that nobody can even purchase the boards for their own projects.  Likewise, the information on how to flash them has never been shared, either.

The pcbs always came from Fred - both he and Chris Walton created the Harmony, and the Melody pcb (and variants of it) are a Harmony board w/o the USB + SD interface board.  But is Fred programming them, or is John C?
Yarusso always printed his own labels and manuals, and possibly the boxes? (though I think he might have out-sourced those).

Who's making the carts?

Exactly.  Atari never even publicly announced it had licensed Star Castle AFAIK, so only his contact(s) within Atari would have told him that.  Plus, if Atari was that concerned about not using it for a VCS version, they could have released one for the 400/800 computers, but they didn't.  Apparently they didn't care about not having a direct port, since it led to Yars, and Yars was a major success for them.  Another thing worth noting is that the Vectrex had a version of Star Castle for it, so either Atari's license for the game was only effective for a specific amount of time (possibly depending on whether or not they released a home version of it), or Atari worked a deal with GCE for them to have it (who released their version in 1983).
 
I clearly remember seeing the ad in Atari Age for Yars and Defender, and immediately wanting Yars.  It sounded amazing from the description, and the artistic rendering of the screen only fueled my interest in having a copy.  As for Defender, I eventually got a copy, but the game had some of the worst flickering of any VCS game, and the fact the ship disappeared completely every time you fired made the game look cheap.  And yet EG's review of the game was so over-the-top in its unbridled enthusiasm for it, I started to seriously question their judgement from that point on.  I've been slowly going through all their VCS reviews in-depth and putting together an article about them, and it's truly astounding how downright awful some of them are.  EG might have been the first US mag dedicated to video games, but it was hardly the best.

Just to reiterate some of Frye's contradictions:
 
Development - has claimed:
6 weeks - as quoted in the book, Racing The Beam (pg. 67)
5 months - as quoted in the April 1998 Next Generation article
6 months - as quoted in the documentary Stella At 20 - 
(12 min in)
4K vs 8K - in the same Next Gen article, Frye mentioned 8K ROMS weren't available when he started programming it.  VCS Asteroids came out in July/August 1981 and was the first 8K VCS game released.  The bank-switching technique was developed (but not put into production) 2 years before, for Video Chess.  In this thread (http://atariage.com/forums/topic/232660-pac-man-review-from-1982/page-8), Goldberg claims to quote Tod from a Facebook conversation they had regarding the story about him being offered use of an 8K ROM for Pac-Man, which is something Rob Zdybel said happened in Once Upon Atari:
  Goldberg: Were you offered to move to 8K for Pac-Man towards the end of coding for it?   Frye: Nope. 8k wasn't even an option until after Pacman coding was complete.  I did have a meeting after Pacman came out, to assess the possibility of a quick revision with less flicker, if we used 8k as an option.   Goldberg: So where did the claim that you originally asked for 8K come from?   Frye: It came from thin air. I never considered 8k. rom was not really an issue. ram was.   Rob Zdybel, Frye's co-worker and office mate at the time, claimed as far back as the late 1990s in Howard Scott Warshaw's Once Upon Atari that Frye absolutely asked for 8K during the project.  Rob was interviewed twice by Randy Kindig of ANTIC: The Atari 8-bit Podcast.  The first one was in June 2016:  https://ataripodcast.libsyn.com/antic-interview-184-rob-zdybel-star-trek-missile-command-bug-hunt
 
At 1 hour and 5 minutes in, he briefly talks about Tod Frye and mentions, "Tod's changed a lot over the years.  Tod's now denying stories that, I don't know, man... he's saying he never asked for more than 4K for Pac-Man.  I was there when you did", to which Frye apparently replied, "That's okay, maybe I don't remember that one."   Popularity - Frye also stated, "Pac-Man wasn't a particularly big game.  'Pac-Man fever' hit between the start and the finish of the project."   PuckMan was released in Japan in May 1980 and the Midway Pac-Man version in October 1980.  Working backwards, VCS Pac-Man came out late March 1982, and production took a good 10 weeks (2.5 months, so Tod likely finished it no later than December 1981 (since the game's copyright date is 1981 which reflects when programming was completed).  Go back 5 months at most for programming, so let's say he started no later than July 1981.  So since October 1980, Pac-Man wasn't a huge hit by the following summer?  A friend of mine remembers the Tomy and Entex handhelds coming out almost immediately together and Coleco's arriving quite a bit later (all in 1981).  The earliest ad for Odyssey2 K.C. Munchkin I could find from my area was December 21st, 1981 (attached); it was reviewed in the March 1982 issue of Electronic Games.  Also, Buckner and Garcia's song was released December 1981 (and the album in January 1982). 
 
Colored background - In a keynote from the 2015 Portland Retro Gaming Expo, Frye states he wish he had made a black background with a blue maze, but claims Atari had a rule against black backgrounds because it would have burned the maze into the CRT (apparently this rule didn't apply to space games...).  This makes no sense since Atari touted the anti-burn-in effects of the VCS from day one, plus Tod included the color cycling code routine in his Pac-Man game!  And I've never heard any other Atari VCS programmer state such a requirement, either.  The story I heard back then was that Pac-Man had a colored background and muted colors, to help make the flickering monsters less noticeable (and they were relabeled as ghosts because of their flickering, which was more logical to accept than having flickering monsters).
 

Another keynote showing more contradictions from Tod Frye.  At this year's PRGE, Frye admits what we already knew - that he didn't care much for arcade Pac-Man:
https://youtu.be/Olagk6ZI_Sc?si=xAXGhbm9hXSe9v0l&t=1260
Frye talks about the development of Pac-Man, but offers a different excuse as to why he didn't use the earlier kernel he developed:
"We were mastering the kernels that reused the players vertically down the screen.  I didn't do that in Pac-Man.  I had a kernel that did it, but I had to throw it away because I ran out of time to finish it.  No, I had the kernel but I didn't have the flicker manager which is a completely different thing.":
https://youtu.be/Olagk6ZI_Sc?si=3vPtIN9p0DoPbl6j&t=1468
The explanation he offered in Once Upon Atari was that he was under probation at the time, and after someone remarked that nobody had ever designed a kernel like that before, he opted not to use it (out of spite for being under probation, I guess).
Later on, Frye talks about when Atari sent GCC a copy of all their VCS game source code listings.  Frye claims that when GCC later sent Atari a copy of all of their VCS listings, that, "I will say with no false humility whatsoever a lot of their tech looked a lot like what I'd done in Pac-Man with horizontal columns.":
https://youtu.be/Olagk6ZI_Sc?si=zoy1Q7GyMO3tkOer&t=2417
So, he not only contradicted himself with previous keynotes and interviews he's given about Pac-Man's kernel, he's done it in the same keynote!  Now, how could GCC had a copy of his earlier Pac-Man kernel if he'd thrown it away?  If he's mistaken about that bit and GCC actually did end up with a copy of his earlier work, and used that for the basis of their Ms. Pac-Man and other VCS games, then he has no reason to complain about it.  He had the chance to use it - and be the first to use it - and didn't.  But to claim he had a better kernel, but threw it out because he didn't have time to finish it... but yet had time to start over and create a new kernel? That makes zero sense.  Rob Zdybel is right - there's something off about Frye's recollections, and it seems the more time that passes, the more 'fried' (pun intended) his memory seems to get.
Also, Rob Z. claimed in a previous keynote that GCC never shared any of their VCS listings with Atari (or if Atari had copies of them, they never shared them with the VCS group).  I again will side with Zdybel over Frye's account.  We certainly know GCC didn't reuse anything from the released version of Pac-Man, because there was nothing special about how it was programmed as far as any innovative tricks.
Frye also criticizes Warshaw for not asking for help with E.T.  Well, he did have help with it.  Jerome Domurat created 1.5K of graphics code for it.  If anyone should have been asking for help, it was Frye with Pac-Man.  He also mentions someone at the show told him he should have chosen to do Defender, since that was one of his favorite games.  Well, Frye claimed in Once Upon Atari that he was offered first choice to do either Defender or Pac-Man, and left it up to Bob Polaro to chose; Polaro couldn't see how Pac-Man could be done (apparently without it sucking) and picked Defender.  Both were poor conversions, but if given the 'stuck on an island' decision of picking one or the other, I'd pick Defender.  Much like Pac-Man, Defender's kernel didn't rely on vertical separation, which is why it flickers as badly as Pac-Man.  Chopper Command was clearly Activision's take on Defender, and shows what a kernel that uses vertical separation looks like.

Just to again refute one of Tod Frye's comments about Pac-Man not being very popular by the time he was assigned to do the VCS version.  In Steve Golson's GDC talk in 2017 about Ms. Pac-Man and the history of GCC, he mentions why and when they decided to make an enhancement kit for Pac-Man, in June 1981:
 
https://youtu.be/rhM8NAMW_VQ?t=18m00s
 
He mentions 2 Pac-Man pattern books, one of which - Mastering Pac-Man - originally came out in 1981 (I don't know which month it was released, but a revised edition came out in January 1982). 

Another keynote showing more contradictions from Tod Frye.  At this year's PRGE, Frye admits what we already knew - that he didn't care much for arcade Pac-Man:
https://youtu.be/Olagk6ZI_Sc?si=xAXGhbm9hXSe9v0l&t=1260
Frye talks about the development of Pac-Man, but offers a different excuse as to why he didn't use the earlier kernel he developed:
"We were mastering the kernels that reused the players vertically down the screen.  I didn't do that in Pac-Man.  I had a kernel that did it, but I had to throw it away because I ran out of time to finish it.  No, I had the kernel but I didn't have the flicker manager which is a completely different thing.":
https://youtu.be/Olagk6ZI_Sc?si=3vPtIN9p0DoPbl6j&t=1468
The explanation he offered in Once Upon Atari was that he was under probation at the time, and after someone remarked that nobody had ever designed a kernel like that before, he opted not to use it (out of spite for being under probation, I guess).
Later on, Frye talks about when Atari sent GCC a copy of all their VCS game source code listings.  Frye claims that when GCC later sent Atari a copy of all of their VCS listings, that, "I will say with no false humility whatsoever a lot of their tech looked a lot like what I'd done in Pac-Man with horizontal columns.":
https://youtu.be/Olagk6ZI_Sc?si=zoy1Q7GyMO3tkOer&t=2417
So, he not only contradicted himself with previous keynotes and interviews he's given about Pac-Man's kernel, he's done it in the same keynote!  Now, how could GCC had a copy of his earlier Pac-Man kernel if he'd thrown it away?  If he's mistaken about that bit and GCC actually did end up with a copy of his earlier work, and used that for the basis of their Ms. Pac-Man and other VCS games, then he has no reason to complain about it.  He had the chance to use it - and be the first to use it - and didn't.  But to claim he had a better kernel, but threw it out because he didn't have time to finish it... but yet had time to start over and create a new kernel? That makes zero sense.  Rob Zdybel is right - there's something off about Frye's recollections, and it seems the more time that passes, the more 'fried' (pun intended) his memory seems to get.
Also, Rob Z. claimed in a previous keynote that GCC never shared any of their VCS listings with Atari (or if Atari had copies of them, they never shared them with the VCS group).  I again will side with Zdybel over Frye's account.  We certainly know GCC didn't reuse anything from the released version of Pac-Man, because there was nothing special about how it was programmed as far as any innovative tricks.
Frye also criticizes Warshaw for not asking for help with E.T.  Well, he did have help with it.  Jerome Domurat created 1.5K of graphics code for it.  If anyone should have been asking for help, it was Frye with Pac-Man.  He also mentions someone at the show told him he should have chosen to do Defender, since that was one of his favorite games.  Well, Frye claimed in Once Upon Atari that he was offered first choice to do either Defender or Pac-Man, and left it up to Bob Polaro to chose; Polaro couldn't see how Pac-Man could be done (apparently without it sucking) and picked Defender.  Both were poor conversions, but if given the 'stuck on an island' decision of picking one or the other, I'd pick Defender.  Much like Pac-Man, Defender's kernel didn't rely on vertical separation, which is why it flickers as badly as Pac-Man.  Chopper Command was clearly Activision's take on Defender, and shows what a kernel that uses vertical separation looks like.

Another keynote showing more contradictions from Tod Frye.  At this year's PRGE, Frye admits what we already knew - that he didn't care much for arcade Pac-Man:
https://youtu.be/Olagk6ZI_Sc?si=xAXGhbm9hXSe9v0l&t=1260
Frye talks about the development of Pac-Man, but offers a different excuse as to why he didn't use the earlier kernel he developed:
"We were mastering the kernels that reused the players vertically down the screen.  I didn't do that in Pac-Man.  I had a kernel that did it, but I had to throw it away because I ran out of time to finish it.  No, I had the kernel but I didn't have the flicker manager which is a completely different thing.":
https://youtu.be/Olagk6ZI_Sc?si=3vPtIN9p0DoPbl6j&t=1468
The explanation he offered in Once Upon Atari was that he was under probation at the time, and after someone remarked that nobody had ever designed a kernel like that before, he opted not to use it (out of spite for being under probation, I guess).
Later on, Frye talks about when Atari sent GCC a copy of all their VCS game source code listings.  Frye claims that when GCC later sent Atari a copy of all of their VCS listings, that, "I will say with no false humility whatsoever a lot of their tech looked a lot like what I'd done in Pac-Man with horizontal columns.":
https://youtu.be/Olagk6ZI_Sc?si=zoy1Q7GyMO3tkOer&t=2417
So, he not only contradicted himself with previous keynotes and interviews he's given about Pac-Man's kernel, he's done it in the same keynote!  Now, how could GCC had a copy of his earlier Pac-Man kernel if he'd thrown it away?  If he's mistaken about that bit and GCC actually did end up with a copy of his earlier work, and used that for the basis of their Ms. Pac-Man and other VCS games, then he has no reason to complain about it.  He had the chance to use it - and be the first to use it - and didn't.  But to claim he had a better kernel, but threw it out because he didn't have time to finish it... but yet had time to start over and create a new kernel? That makes zero sense.  Rob Zdybel is right - there's something off about Frye's recollections, and it seems the more time that passes, the more 'fried' (pun intended) his memory seems to get.
Also, Rob Z. claimed in a previous keynote that GCC never shared any of their VCS listings with Atari (or if Atari had copies of them, they never shared them with the VCS group).  I again will side with Zdybel over Frye's account.  We certainly know GCC didn't reuse anything from the released version of Pac-Man, because there was nothing special about how it was programmed as far as any innovative tricks.
Frye also criticizes Warshaw for not asking for help with E.T.  Well, he did have help with it.  Jerome Domurat created 1.5K of graphics code for it.  If anyone should have been asking for help, it was Frye with Pac-Man.  He also mentions someone at the show told him he should have chosen to do Defender, since that was one of his favorite games.  Well, Frye claimed in Once Upon Atari that he was offered first choice to do either Defender or Pac-Man, and left it up to Bob Polaro to chose; Polaro couldn't see how Pac-Man could be done (apparently without it sucking) and picked Defender.  Both were poor conversions, but if given the 'stuck on an island' decision of picking one or the other, I'd pick Defender.  Much like Pac-Man, Defender's kernel didn't rely on vertical separation, which is why it flickers as badly as Pac-Man.  Chopper Command was clearly Activision's take on Defender, and shows what a kernel that uses vertical separation looks like.

Another keynote showing more contradictions from Tod Frye.  At this year's PRGE, Frye admits what we already knew - that he didn't care much for arcade Pac-Man:
https://youtu.be/Olagk6ZI_Sc?si=xAXGhbm9hXSe9v0l&t=1260
Frye talks about the development of Pac-Man, but offers a different excuse as to why he didn't use the earlier kernel he developed:
"We were mastering the kernels that reused the players vertically down the screen.  I didn't do that in Pac-Man.  I had a kernel that did it, but I had to throw it away because I ran out of time to finish it.  No, I had the kernel but I didn't have the flicker manager which is a completely different thing.":
https://youtu.be/Olagk6ZI_Sc?si=3vPtIN9p0DoPbl6j&t=1468
The explanation he offered in Once Upon Atari was that he was under probation at the time, and after someone remarked that nobody had ever designed a kernel like that before, he opted not to use it (out of spite for being under probation, I guess).
Later on, Frye talks about when Atari sent GCC a copy of all their VCS game source code listings.  Frye claims that when GCC later sent Atari a copy of all of their VCS listings, that, "I will say with no false humility whatsoever a lot of their tech looked a lot like what I'd done in Pac-Man with horizontal columns.":
https://youtu.be/Olagk6ZI_Sc?si=zoy1Q7GyMO3tkOer&t=2417
So, he not only contradicted himself with previous keynotes and interviews he's given about Pac-Man's kernel, he's done it in the same keynote!  Now, how could GCC had a copy of his earlier Pac-Man kernel if he'd thrown it away?  If he's mistaken about that bit and GCC actually did end up with a copy of his earlier work, and used that for the basis of their Ms. Pac-Man and other VCS games, then he has no reason to complain about it.  He had the chance to use it - and be the first to use it - and didn't.  But to claim he had a better kernel, but threw it out because he didn't have time to finish it... but yet had time to start over and create a new kernel? That makes zero sense.  Rob Zdybel is right - there's something off about Frye's recollections, and it seems the more time that passes, the more 'fried' (pun intended) his memory seems to get.
Also, Rob Z. claimed in a previous keynote that GCC never shared any of their VCS listings with Atari (or if Atari had copies of them, they never shared them with the VCS group).  I again will side with Zdybel over Frye's account.  We certainly know GCC didn't reuse anything from the released version of Pac-Man, because there was nothing special about how it was programmed as far as any innovative tricks.
Frye also criticizes Warshaw for not asking for help with E.T.  Well, he did have help with it.  Jerome Domurat created 1.5K of graphics code for it.  If anyone should have been asking for help, it was Frye with Pac-Man.  He also mentions someone at the show told him he should have chosen to do Defender, since that was one of his favorite games.  Well, Frye claimed in Once Upon Atari that he was offered first choice to do either Defender or Pac-Man, and left it up to Bob Polaro to chose; Polaro couldn't see how Pac-Man could be done (apparently without it sucking) and picked Defender.  Both were poor conversions, but if given the 'stuck on an island' decision of picking one or the other, I'd pick Defender.  Much like Pac-Man, Defender's kernel didn't rely on vertical separation, which is why it flickers as badly as Pac-Man.  Chopper Command was clearly Activision's take on Defender, and shows what a kernel that uses vertical separation looks like.

Another keynote showing more contradictions from Tod Frye.  At this year's PRGE, Frye admits what we already knew - that he didn't care much for arcade Pac-Man:
https://youtu.be/Olagk6ZI_Sc?si=xAXGhbm9hXSe9v0l&t=1260
Frye talks about the development of Pac-Man, but offers a different excuse as to why he didn't use the earlier kernel he developed:
"We were mastering the kernels that reused the players vertically down the screen.  I didn't do that in Pac-Man.  I had a kernel that did it, but I had to throw it away because I ran out of time to finish it.  No, I had the kernel but I didn't have the flicker manager which is a completely different thing.":
https://youtu.be/Olagk6ZI_Sc?si=3vPtIN9p0DoPbl6j&t=1468
The explanation he offered in Once Upon Atari was that he was under probation at the time, and after someone remarked that nobody had ever designed a kernel like that before, he opted not to use it (out of spite for being under probation, I guess).
Later on, Frye talks about when Atari sent GCC a copy of all their VCS game source code listings.  Frye claims that when GCC later sent Atari a copy of all of their VCS listings, that, "I will say with no false humility whatsoever a lot of their tech looked a lot like what I'd done in Pac-Man with horizontal columns.":
https://youtu.be/Olagk6ZI_Sc?si=zoy1Q7GyMO3tkOer&t=2417
So, he not only contradicted himself with previous keynotes and interviews he's given about Pac-Man's kernel, he's done it in the same keynote!  Now, how could GCC had a copy of his earlier Pac-Man kernel if he'd thrown it away?  If he's mistaken about that bit and GCC actually did end up with a copy of his earlier work, and used that for the basis of their Ms. Pac-Man and other VCS games, then he has no reason to complain about it.  He had the chance to use it - and be the first to use it - and didn't.  But to claim he had a better kernel, but threw it out because he didn't have time to finish it... but yet had time to start over and create a new kernel? That makes zero sense.  Rob Zdybel is right - there's something off about Frye's recollections, and it seems the more time that passes, the more 'fried' (pun intended) his memory seems to get.
Also, Rob Z. claimed in a previous keynote that GCC never shared any of their VCS listings with Atari (or if Atari had copies of them, they never shared them with the VCS group).  I again will side with Zdybel over Frye's account.  We certainly know GCC didn't reuse anything from the released version of Pac-Man, because there was nothing special about how it was programmed as far as any innovative tricks.
Frye also criticizes Warshaw for not asking for help with E.T.  Well, he did have help with it.  Jerome Domurat created 1.5K of graphics code for it.  If anyone should have been asking for help, it was Frye with Pac-Man.  He also mentions someone at the show told him he should have chosen to do Defender, since that was one of his favorite games.  Well, Frye claimed in Once Upon Atari that he was offered first choice to do either Defender or Pac-Man, and left it up to Bob Polaro to chose; Polaro couldn't see how Pac-Man could be done (apparently without it sucking) and picked Defender.  Both were poor conversions, but if given the 'stuck on an island' decision of picking one or the other, I'd pick Defender.  Much like Pac-Man, Defender's kernel didn't rely on vertical separation, which is why it flickers as badly as Pac-Man.  Chopper Command was clearly Activision's take on Defender, and shows what a kernel that uses vertical separation looks like.

It looked great compared to the XE version.  There's some screenshots on my site: https://www.ataricompendium.com/game_library/easter_eggs/7800/78gato.html

The voice module was first mentioned at the Summer 1982 CES show in Chicago.  RS Baseball was first mentioned in the March/April 1983 issue of Atari Age magazine, but makes no mention of it having voice. Matthew Hubbard did the initial programming for the game, before leaving for Activision.  Jim Andreasen took it over.  Keith Hayena mentioned at CGE2K7 that consultants Larry Car (sp?) and Andy Barber (sp?) devised a method to insert digitized sounds into a program, and Hayenga used it to add voice to the game.  Programmer Frank Hausman was able to adapt a voice compression algorithm developed by Electronic Speech Systems into the game.  According to the V1N6 March/April 1983 issue of Atari Age, this was to be available in April for $31.95, but it didn't appear for sale until the Sept/Oct issue for $28.95.  Michael Current says it was released November 1983: https://mcurrent.name/atarihistory/wci_games.html
Whether or not an actual voice synthesizer module was being developed or not, we don't know.  Atari had the habit of announcing a lot of things back then that ended up being vaporware (5200 keyboard, expansion module, etc).  I'm sure the voice module was talked about in-house, especially since the Intellivision had one in 1982, as did the Odyssey2.  Remember, Atari's Marketing was looking to one-up Mattel's Intellivision with the 5200, which is why the controllers were designed the way they are, with having a keypad and a "360 degree" joystick, which was deemed far superior to the Intellivision's 16-position pad controller.  But it took Atari a year from announcing a VCS Adaptor to finally releasing it, and it's possible that if a voice module was being worked on, the development of RS Baseball being delayed to add voice into the game probably convinced someone within Atari that the module wasn't needed if voice could be done in software.  Besides, Mattel's Voice Module wasn't the runaway success it had hoped for:
https://history.blueskyrangers.com/hardware/intellivoice.html