Jump to content
IGNORED

ECS Program Expander was planned... in brown!


Lathe26

Recommended Posts

Mattel marketing would have known about colecovision from at least february 1982, when it was introduced at new york toy fair.

Not that I'm sure it matters much, but supposedly Coleco unveiled the ColecoVision already at Winter CES in January 1982, together with their new games for the Atari 2600 and Intellivision. Also at the Winter CES, the Vectrex, the Bally [Astrovision] Professional Arcade, the Commodore Ultimax (*) and I believe also the C64 were presented. Even Mattel Electronics were there, so in case they sent out someone to check on their competitors, most likely the August release of the ColecoVision couldn't take them by surprise.

 

http://vidgame.info/vid1982.htm

 

(*) Which is referenced by three sources to have been released in June, though recent discoveries have found that the first specimens didn't get assembled until December the same year... So much for trusting what the magazines write.

Link to comment
Share on other sites

Obvious to people that knew about the ecs but not everyone knew.

 

I think work stopped on the KC earlier that summer, maybe spring. Hardware that is, not sure about software.

 

Actually, dzjay is probably right about the sales meeting. Even if there was no KC announcement, the defining moment is when the ecs becomes an official product.

 

Edit:

The "history and philisophy" document on papaintellivision.com has the following.

 

"August, 1982 - terminated the program"

 

Alright, so we do have a very likely candidate, if not in fact an actual, official date: August, 1982. That means, they cancelled the KC hardware project, and a month or two later made the official announcement to the rest of the company via the marketing and sales meeting.

 

-dZ.

Link to comment
Share on other sites

I see on all the pics that the notches of gue are sometime small, and it appears to be of that glue that dry and no longer stick well over age.

It's not impossible that at some point on the assembly line, they ran out of glue and they let the ECS go anyway because really, the cover is a tight fit and wouln't come out by mistake.

Link to comment
Share on other sites

 

  • P3 089756
  • "172"
  • U52 date code 8335
  • U42 date code 8315 (?)
  • AY-3-8917 date code 8307
  • Glued exp port
  • Connector.
  • Daughter board has extra bodge wires and is a lighter shade.

attachicon.gifIMG_20190310_214918.jpgattachicon.gifIMG_20190310_215017.jpgattachicon.gifScreen Shot 2019-03-10 at 11.25.53 PM.pngattachicon.gifIMG_20190310_215231.jpgattachicon.gifIMG_20190310_215146.jpgattachicon.gifIMG_20190310_215252.jpgattachicon.gifIMG_20190310_215353.jpg

 

 

Interestingly, this one has 4182-0033 for the MMI chip rather than 4182-0050. I should go re-open that one and get a better pic of the MMI chip. I don't think the date code on U42 is 8315. Through the glare in the photo it looks more like 8301. I should have paid more attention.

 

I think this unit is similar to Lathe26's "prototype" unit in many ways. It doesn't have the L000... serial number and the daughter card is a distinctly different color with lots of bodgework on it.

Edited by intvnut
Link to comment
Share on other sites

The "history and philisophy" document on papaintellivision.com has the following.

 

"August, 1982 - terminated the program"

 

I assume you're referring to this slide:

 

post-14113-0-20868800-1552347649_thumb.png

 

That would be a pretty clear indication of when the Keyboard was terminated.

 

In memory of the KC, here's a pic of some EPROMs with the PicSee code on a KC board that I took in the museum at CGE2K14. RIP KC and CGE.

 

post-14113-0-31339800-1552347830_thumb.jpg

  • Like 2
Link to comment
Share on other sites

 

  • P3 089756
  • "172"
  • U52 date code 8335
  • U42 date code 8315 (?)
  • AY-3-8917 date code 8307
  • Glued exp port
  • Connector.
  • Daughter board has extra bodge wires and is a lighter shade.

This particular board looks A LOT like my Mattel-internal ECS 00018. The biggest difference between them is that my bodge parts are far more "air-borne" and yours are compact and closer in arrangement to later ECSs.

 

When comparing both ECSs to later ECSs, pin 1 of the Expansion Connecter is not cut (ours are still tied to 5V) and the MMI PAL chip's part number ends in 0050 instead of 0033. I suspect that the 0050 PAL chips are not programmed to handle the CTRL_IN_DIS signal that comes from pin 1 of the connector, but I have not confirmed that yet.

Link to comment
Share on other sites

This particular board looks A LOT like my Mattel-internal ECS 00018. The biggest difference between them is that my bodge parts are far more "air-borne" and yours are compact and closer in arrangement to later ECSs.

 

When comparing both ECSs to later ECSs, pin 1 of the Expansion Connecter is not cut (ours are still tied to 5V) and the MMI PAL chip's part number ends in 0050 instead of 0033. I suspect that the 0050 PAL chips are not programmed to handle the CTRL_IN_DIS signal that comes from pin 1 of the connector, but I have not confirmed that yet.

 

I think you swapped part numbers again. 0033 is the "proto" w/out CTRL_IN_DIS, while 0050 is the "production" version which supports that pin.

 

I'll dig that unit back out and take some more detailed photos before I take the box back to storage.

Link to comment
Share on other sites

I think you swapped part numbers again. 0033 is the "proto" w/out CTRL_IN_DIS, while 0050 is the "production" version which supports that pin.

 

I'll dig that unit back out and take some more detailed photos before I take the box back to storage.

 

Actually, not in this case.

  • Only the very earliest ECSs have 0050 and with lots of bodge parts on the pale main daughterboard (4182-4239, no rev mark). It is suspected it doesn't support CTRL_IN_DIS due to lack of cut pin 1 trace to the Expansion Connector.
  • The majority of ECSs have 0033 on a sometimes darker daughterboard (4182-4239 Rev B) regardless of whether the Expansion Connector is present. The 0033 does support CTRL_IN_DIS, or at least on the very few I tested; I assume there are not chip variants with the same markings. Note that there are 2 variants of the 4182-4239 Rev B board, one dark and one pale, but traces got cut on the pale Rev B to make the two variants functionally identical.

Another interesting difference I saw between your early ECS and my early ECS is that yours has the dual ribbon cable constructed of wires and rubber. My early ECS's ribbon cables are constructed of paper and thin traces, which makes me rather nervous about messing around with (worried it will get damaged).

Link to comment
Share on other sites

 

Actually, not in this case.

  • Only the very earliest ECSs have 0050 and with lots of bodge parts on the pale main daughterboard (4182-4239, no rev mark). It is suspected it doesn't support CTRL_IN_DIS due to lack of cut pin 1 trace to the Expansion Connector.
  • The majority of ECSs have 0033 on a sometimes darker daughterboard (4182-4239 Rev B) regardless of whether the Expansion Connector is present. The 0033 does support CTRL_IN_DIS, or at least on the very few I tested; I assume there are not chip variants with the same markings. Note that there are 2 variants of the 4182-4239 Rev B board, one dark and one pale, but traces got cut on the pale Rev B to make the two variants functionally identical.

Another interesting difference I saw between your early ECS and my early ECS is that yours has the dual ribbon cable constructed of wires and rubber. My early ECS's ribbon cables are constructed of paper and thin traces, which makes me rather nervous about messing around with (worried it will get damaged).

 

*d'oh*

 

You are correct, of course, and it was *I* that got the part numbers swapped this time. :-P

 

 

Note that there are 2 variants of the 4182-4239 Rev B board, one dark and one pale, but traces got cut on the pale Rev B to make the two variants functionally identical.

 

Does it appear they were cut in the photomask, or cut manually? As in, the darker Rev B really should be called Rev C?

Link to comment
Share on other sites

...

 

 

Does it appear they were cut in the photomask, or cut manually? As in, the darker Rev B really should be called Rev C?

 

My thinking as to the evolution of the 4182-4239 daughterboards:

 

Pale 4182-4239 (no Rev marking) daughterboard came first since it is attached only to the lowest ECSs internal serial numbers that are manually written and the daughterboards are missing the ink layer (no labels like U42). Here is the underside of this board where U42 pin 11 is grounded. U42 is the 4182-0050 here.

post-37124-0-82044400-1552458523_thumb.jpg

 

Pale 4182-4239 Rev B daughterboard came next. I haven't checked the internal serial numbers yet but these numbers are printed on manufactured stickers. However, these have ink layer on them (more professionally manufactured). On this particular board, here is the underside where U42 pin 11's traces are manually cut (and bodged to pin 1 of the Expansion Connector). U42 is the 4182-0033 here.

post-37124-0-30291000-1552458687_thumb.jpg

 

Dark 4182-4239 Rev B (I would have called it Rev C as well ;)) daughterboard came last. Again, I haven't checked the internal serial numbers yet but these numbers are printed on manufactured stickers. Same as before, these have ink layer on them (more professionally manufactured). On the underside under U42, its pin 11's traces are designed as disconnected from ground. Some dark boards are missing the Expansion Connector that was removed in the last group of ECSs. U42 is the 4182-0033 here.

post-37124-0-67263600-1552459357_thumb.jpg

 

Note, there may be more variants out there. Perhaps there are some already in the photos already provided and I haven't looked close enough... yet.

  • Like 1
Link to comment
Share on other sites

Wow, that middle one looks rather impressive—quite the gouge there!

 

For the final Rev B boards, it looks like they did just edit the photomask. That's also probably why they didn't change the revision in the metal layer. I didn't look at my units (and would need to dig them out and tear them back open again), but I do seem to recall that some boards had different document-control numbers and revisions on the top side and bottom side. That is, the top-side mask and bottom-side mask were separately numbered and controlled.

 

I noticed this particularly with the KC BASIC cartridge.

Link to comment
Share on other sites

I made a few new photos of my first ECS (P3 00115).

Now I know why I didn't find the U52 on it, it's not written on the daughterboard.

On the underside of the mainboard: Rev. PR...does that mean Prototype?

Connector cable = paper

 

 

Interesting that your ROM chip says 4182-0127 instead of the typical 4182-0128. I wonder if the contents are different.

 

I originally was thinking that Rev PR meant "PRototype" but I vaguely recall seeing "Rev PR" on a game cartridge so maybe it means "Production Run". Not sure about it.

  • Like 1
Link to comment
Share on other sites

 

Interesting that your ROM chips says 4182-0127 instead of the typical 4182-0128. I wonder if the contents are different.

 

I originally was thinking that Rev PR meant "PRototype" but I vaguely recall seeing "Rev PR" on a game cartridge so maybe it means "Production Run". Not sure about it.

 

I'm not sure about the Rev PR either. Several components in this Keyboard Component BOM rollup are listed as "adding Rev PR to xxx." Another data point, but not much additional insight I'm afraid.

Link to comment
Share on other sites

I made a few new photos of my first ECS (P3 00115).

Now I know why I didn't find the U52 on it, it's not written on the daughterboard.

On the underside of the mainboard: Rev. PR...does that mean Prototype?

Connector cable = paper

 

 

I just noticed that your main board has the appropriate cut traces and bodge wires for pin 1 of the Expansion Connector to be setup as CTRL_IN_DIS (and not hardwired as +5V). Can you show a photo of the underside of the large daughterboard? I'd like to see if the traces to pin 11 of U42 have have been cut as well.

Link to comment
Share on other sites

Ok, with all the photos everyone has been sending, a pattern is starting to form. This is great work, folks.

 

If the internal serial number is low, then the Expansion Connector is installed. The cut-off is somewhere between L00018629 and L00025223. Intvnut has the one oddball Expansion Connector with internal serial number of L00052302.

 

Also, the 3 hand-written internal serial number units all have the MMI PAL chip of 0050 (instead of the common 0033) and a early date code of 8301.

 

post-37124-0-65436300-1552496852_thumb.png

Link to comment
Share on other sites

  • 2 weeks later...

8K

Have any prototypes of the expansion module ever surfaced?

 

None that I know of. Most folks think the images in the catalogs were just carved blocks of wood (similar to the Intellivoice II).

 

All it would have been is more RAM and ROM. Values vary but IntellivisionLive.com (before the pages were removed but archived here), said that it added 16K of RAM and 12K of ROM (catalogs listed 8K of ROM with the same RAM).

Link to comment
Share on other sites

  • 1 month later...

That's 8K rom and 16kB ram, would all those chips fit on that skinny expansion card.

 

The white ecs and brown ecs should be the same on the inside, although some might be missing the expansion slot.

 

Does the ecs case have a removable panel to access the expansion slot?

 

When I opened my brown ECS I noticed that the edge connector (tongue of the ECS that plugs into the cart slot of the Intv) has a black wired soldered to a pair of traces jumping them together

 

The reason I noticed this is because when I used my white intv2 with white ECS and system changer together the Rf output was very snowy. When I used my brown ecs with the intv2 and system changer the image was clear. When I used the white ecs with my brown intv 1 the image was now perfect? ; not snowy like on the intv2. Not sure why so I opened both units intending to swap the boards and keep the matching colour cases. That's when I noticed the difference on the underside of the bus connector

Link to comment
Share on other sites

  • 3 years later...

I was thinking about the Program Expander today. Do we know anything about its technical specs? Amount of RAM, any ROM, memory ranges, that sort of thing? It would be pretty simple to adapt part of my Raspberry Pi Pico code from the ACC and have a Pi Pico become a Program Expander.

Link to comment
Share on other sites

  • 1 month later...
On 3/7/2019 at 11:13 PM, intvnut said:

"PAGE 7" was reserved for Extended BASIC, but doesn't say what address range. So, at best we can surmise it's at least 4K, probably at least 8K, and maybe bloating toward 12K by the time it was canceled.

Why would you think 12K bloated? Depends on the functionality, right? This was a CP1600 cartridge that included an EXEC-like run time system and a lot of built-in graphics, and was on track to being delivered in 1983, when ROM prices had dropped a bit.

 

On 3/9/2019 at 11:20 AM, intvnut said:

After the June 1983 CES in Chicago, Josh Denham and Stav Prodomou, Mattel Electronics' President and Senior Operations VP, resigned. Josh and Stav had been blamed for pushing the company too far into hardware production; hundreds of millions of dollars had been tied up in the development, beyond the original Intellivision, of the Keyboard Component, Intellivoice, Intellivision II, the System Changer, ECS, Aquarius (and peripherals), Intellivision III and the top-secret Intellivision IV.

Prodromou and Denham inherited the Keyboard Component, and Intellivoice was given high priority at the strong "suggestion" of Bob Anderson. There was no way that the development cost of the other hardware listed was "in the hundreds of millions of dollars." The Intellivision II project was a cost-reduction effort that repackaged the original in a less-expensive housing and initiated the development of the STIC 1A and CP1610A, with GI funding the latter. The ECS was a trivial piece of hardware by 1982 standards. The Aquarius was a pre-existing design that had been fully developed by Radofin. The development cost of new ROM types was almost immediately recouped by lower cartridge manufacturing costs. Total expenditures on the two biggest hardware projects, Intellivision III and IV, couldn't have exceeded $20 million before they were cancelled. The "hundreds of millions of dollars" number thrown around in support of this narrative just doesn't add up.

That's not to say poorly considered hardware projects didn't contribute to Mattel's problems, but excess inventory, excessive promotional commitments and uncontrolled software development expenses played a far greater role.

No, blaming Mattel's problems on hardware development is a narrative pushed by proponents of turning Mattel into an Activision-like software house that only made software for other manufacturer's systems, a camel that got its nose under the tent with the M-Network cartridges and survived Denham and Prodromous' departure. That idea didn't work out so well either, eh?

 

On 3/9/2019 at 11:44 AM, intvnut said:

Supposedly, the internal pitch was that the Keyboard Component and LUCKY (the ECS) were supposed to coexist in the marketplace. That internal story was supposedly to keep Dave "Papa Intellivision" Chandler from using his political clout from getting Mattel to cancel the ECS. You can see in some of the later memory maps (CCF_10232011_00026.pdf) that both the Keyboard and ECS were shown in the same memory map plan, alongside the Intellivision III planned EXEC II.

 

Yup. LUCKY was at a much lower price point, one that Denham was more comfortable with.

 

On 3/9/2019 at 11:44 AM, intvnut said:

That internal story was supposedly to keep Dave "Papa Intellivision" Chandler from using his political clout from getting Mattel to cancel the ECS.

 

Nope. (1) Chandler wasn't that stupid, (2) knew about the ECS from the beginning and (3) never had that kind of clout.
On 3/9/2019 at 11:44 AM, intvnut said:

You can see in some of the later memory maps (CCF_10232011_00026.pdf) that both the Keyboard and ECS were shown in the same memory map plan, alongside the Intellivision III planned EXEC II.

 

post-14113-0-20386100-1552160948_thumb.png

I've mentioned elsewhere how Mattel took compartmentalization to extremes. This memo is a case in point: look at the tiny distribution list. Only ten people, including the author, were privy to this vital information. It's also interesting to note that although Chandler knew at the time of this memo that the Keyboard had been killed, either word hadn't yet filtered down to Rudd or Rudd wasn't going to be the one to let the cat out of the bag to the others on this distribution list.

 

On 3/9/2019 at 12:57 PM, Lathe26 said:

Do we have an idea when the Keyboard Component was finally killed off?

End of August 1982.

On 3/9/2019 at 12:57 PM, Lathe26 said:

Do we know when work started on the Intellivision III, even with the limited known history of it?

1977. Really. But in earnest the week after the Keyboard Component was killed.

 

On 3/9/2019 at 1:35 PM, Lathe26 said:

U52 (the 24-pin chip on the 4239 daughterboard) is confirmed to be the ROM. ... U42 is strongly suspected to be an MMI PAL chip.

Yup.

 

On 3/10/2019 at 12:07 AM, mr_me said:

Mattel marketing would have known about colecovision from at least february 1982, when it was introduced at new york toy fair.

Earlier; it certainly knew by Winter CES, the month before. Even knew it used a Z80 and the TMS9918.

 

On 3/10/2019 at 5:38 AM, DZ-Jay said:
On 3/9/2019 at 12:57 PM, Lathe26 said:

Do we have an idea when the Keyboard Component was finally killed off?

 

According to the Intellivision Lives site, in Fall 1982, during a marketing/sales meeting where the ECS was introduced:

Nope. The Keyboard Component was killed in a carefully planned assassination. The ambush took place before Denham and Prodromou, who informed those who needed to know within 24 hours of witnessing the deed. Although the feeding of the corpse ceased immediately, those who didn't need to know were kept in the dark pending the fabrication of, er, preparation of a cover story as to how the ECS was a credible replacement.

 

On 3/10/2019 at 1:52 PM, intvnut said:

It was effectively dead. When was the project actually cancelled though? I've worked on projects that were effectively dead and didn't actually get cancelled for months.

Not here. The Keyboard Component had the full support of management immediately before the ambush and was completely dead immediately thereafter.

On 3/11/2019 at 4:44 PM, intvnut said:

I assume you're referring to this slide:

 

post-14113-0-20868800-1552347649_thumb.png

As you can see from the milestones listed on the slide, the Keyboard Component was fully alive in June and fully dead in August.

Here’s an interesting perspective: Based on the milestones listed in that slide, it took only 20 months for the Keyboard Component to get from the start of design work to first units on the market. The product was being sold for almost a year before being canceled, albeit at a low level. Design of the ECS began in mid-1981 and first units arrived in stores in December of 1983, about thirty months later. The ECS was only for sale for a month before Mattel quit the market.

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...