Why extremely heavy and expensive aluminum castings for RF shielding in the 400 and 800?

[+DrVenkman]

41 minutes ago, Winston2022 said:

Yes, I'm aware of all of that, but I'm exclusively and ONLY wondering about the actual necessity of thick aluminum castings in the 400/800 which caused higher retail prices than what would have been possible if the castings had not been used and the stamped sheet metal used by everyone else during the same time period had been used instead, those other machines passing FCC requirements then and for multiple years afterwards.

You keep missing the point that has been reiterated numerous times: 1) FCC regulations for home computing devices were not known and it was believed they would be much stricter than they turned out to be; 2) the design was done in the 1978/early '79 time frame, and tooling costs were essentially all spent and done by the time the FCC regs were finalized, so the only costs associated with the RF shielding left were unit production costs, which for a couple pieces of cast metal were not high compared to the parts in the rest of the machine. In 1983 when I paid $369 or whatever it was for my 48K 800, it's not like it would have cost $199 had the RF shield been replaced by the foil-covered cardboard garbage of the C64. Rather, the sale price for the 400/800 can be better and more reasonably attributed to the multi-board designs; the great number of discrete components; and the associated assembly, test, verification and warranty return/service costs that flow directly from the above. Single-board designs are inherently cheaper to manufacture, package, ship and warehouse, and a single-board 400/800 was not possible due to the high cost and types of DRAM chips available and the marketing-driven desire to make the 800 easily expandable. 

tl;dr: The high cost of the 400/800 compared to the C64 are not attributable to the RF shielding in any reasonable way. 

[_The Doctor__]

Computing Device vs Game system? VCS becomes the 2600? Rules were stiff in '75 not so much later on? These are the seeds that have been in the conversations, do a deep dive in the forums, you may find some answers. There was a focus on the TV modulator at certain times as well. Maybe a look at all of that would give you the answers you seek, the internal communications were not always on the Atari servers and the keepers of that information have all but a few passed away.

These machines were not just for your living room either, they had serious uses in industrial, nuclear, and military applications early on and not so much later.

rinse repeat.

Edited December 3, 2022 by _The Doctor__

[Faicuai]

3 hours ago, _The Doctor__ said:

There was a focus on the TV modulator at certain times as well.

Ding, ding, ding! Bingo, again!

 

It is the RF-modulator (plus electro-magnetic radiation and noise created by simply powering up the CPU, Antic, RAM, etc), all combined, which was eventually the tipping point.

 

Plus we never judge a government agency by what they write, say, etc. (just look at the current administration! 😵‍🤯)  We only judge them by what they DO, and back in the '77-80' time frame, it appears to have been "hunting-season" already for the FCC, when it comes to chasing down noise-offenders.

 

What is VERY interesting is that the Apple II/+/e have basically NO shielding other than some metal parts here and there... and the case-design remained almost untouched all the way to end-of-life... yet the key detail is that there was NO built-in RF modulator from factory that I know, so far...

 

I maybe wrong but... sounds as if Atari was quite afraid of what the FCC might do, considering what they were already doing...

Edited December 4, 2022 by Faicuai

[adam242]

The RF modulator is precisely what makes the difference. Without one, machines like the Apple II can claim they are intended for "office use" and are exempt from the more stringent regulations applied to "home use" computers. The 400/800, having an RF modulator, are clearly designed to use the good-ol' living room TV as their display.

[Pfaffa]

The Apple IIs had less shielding because they were designed to connect to monitors not TVs (pay no mind to Woz's friend selling a cheap and easy RF addon board down the street) and thus used much more relaxed rules.

[JamesD]

On 11/29/2022 at 3:02 PM, Faicuai said:

 

Because the FCC standards of design-time dictated so. On the other hand, my 800's internal shielding won't rot and as crap-looking as my 800-XLs tin-foil shielding...

 

Much better quality and better isolation on the 800. Simply put, built to a higher standard.

FCC standards required pretty strict levels, but it didn't require heavy aluminum. 
I'm guessing the aluminum was to serve as a heatsink.
Later designs that met the same standards used lightweight metal that had holes for ventilation.
I think it's just a matter of knowledge at the time. 
The standard was pretty new when Atari released the machines, RF testing is expensive, and nobody had to deal with it before that.
 

[Winston2022]

On 12/3/2022 at 2:25 PM, DrVenkman said:

You keep missing the point that has been reiterated numerous times: 1) FCC regulations for home computing devices were not known and it was believed they would be much stricter than they turned out to be

Current FCC RFI regulations for computers and video games are under the LOGICALLY combined category of "Digital Devices and Personal Computers." 

 

Since the sheet metal shielded game systems released at the same time as the Atari 8-bits had the same video-synced clock speed for their CPUs as the Atari computers, that frequency being what determines the harmonics which cause the RFI, WHY would Atari think that home computer standards could be justified to be any more strict than for the already FCC approved video game systems? 

 

Also note, as I've pointed out, the very early Atari VCS was cast aluminum shielded.  Why?  Were they worried about FCC RFI standards for home computers then?  Of course not.  Perhaps they were worried about pending FCC standards for video games, but if so that was later proved to be unwarranted by the sheet metal shielded video game systems later approved for release.

 

Thus, my theory, until I see actual engineering notes and FCC documents, is that this "uncertainty" about FCC standards excuse is exactly that, an excuse for a bad business choice.

Edited December 8, 2022 by Winston2022

[+DrVenkman]

You just keep beating that dead horse into (incorrect) glue.

 

One last time, and louder for those in the back: THE CAST METAL SHIELDING IS NOT WHY THINGS COST AS MUCH AS THEY DID.

 

 

[Winston2022]

2 minutes ago, DrVenkman said:

You just keep beating that dead horse into (incorrect) glue.

 

One last time, and louder for those in the back: THE CAST METAL SHIELDING IS NOT WHY THINGS COST AS MUCH AS THEY DID.

 

 

I agree that it wasn't the SOLE reason for the high cost, but it certainly didn't help.  Somewhere I've seen a document that was the component cost breakdown for SOMETHING made by Atari, but I don't recall what it was or where I saw that.

[_The Doctor__]

the rf modulator, cards, video cards et all being subject fcc rf and emi restrictions were (stringent) tough in 1978 and tightening into what was implemented in January of 1981. Look up the classes. There was the usual hijinx of business and government connected entities with the FCC having a hard on over things as directed by the rules and relationships. How about you do your diligence and look up what Texas Instruments was up to, look at what was dropped in 1981 (TRS models etc). Your answers not only lie within fcc documents but readily available interviews and documentaries. As we go into this again remember memories fade and people make mistakes or mess things up. Let's avoid poo flinging and pissyness as people chime in on the subject with different nuggets of wisdom, fact, and supposition. Otherwise this will coalesce into madness.

 

Seriously, Atari needed to put an end to the BS and get the machines out there. Consequentially Atari was an awesome amateur radio companion and could be used in just about any industry setting as a result.

Edited December 8, 2022 by _The Doctor__

[+x=usr(1536)]

On 12/3/2022 at 6:27 PM, Faicuai said:

It is the RF-modulator (plus electro-magnetic radiation and noise created by simply powering up the CPU, Antic, RAM, etc), all combined, which was eventually the tipping point.

This is basically correct.  To clarify a bit:

 

There are two broad device classes (amongst others) that the FCC has traditionally placed electronic devices into: intentional radiators, and unintentional radiators.  All this effectively means is whether or not one of the device's primary design goals is to be an RF transmitter.  If yes, it's intentional; if no, it's unintentional.

 

While it is absolutely correct to say that the electronics in a computer give off some degree of RF, those are considered to be unintentional radiators.  Their purpose is to form the computer itself, not act as a transmitter.  By association, this means that the rest of the machine is an unintentional radiator.

 

The minute that an RF modulator - a device specifically designed to be fed a signal and transmit it on a given frequency for external reception - is introduced into the system, the entire system becomes an intentional radiator because it now contains a component designed to do exactly that.  This leads to a different certification process, and a more stringent one at that regardless of how low-powered the RF modulator may be.

 

Also remember that the FCC was having to deal with certifying the first wave of consumer electronics that were designed to connect to a TV set.  This was new territory for them, nobody really knew what limits needed to be set, what testing would encompass, or if next year's testing requirements would be the same as last year's.  The chunky aluminium cases in the 2600 and 400/800 were probably (at least in part) a hedge against having to make mid-production run changes to meet RF emissions compliance.

 

6 hours ago, DrVenkman said:

THE CAST METAL SHIELDING IS NOT WHY THINGS COST AS MUCH AS THEY DID.

1000% correct.  They cost as much as they did because electronics were (and still can be) expensive.  The shield had very little to do with the overall cost - it's not like it was a $100 item in 1979 dollars.

[+x=usr(1536)]

6 hours ago, Winston2022 said:

Current FCC RFI regulations for computers and video games are under the LOGICALLY combined category of "Digital Devices and Personal Computers." 

Emphasis above mine.

 

The regulations in 2022 for RFI levels are completely different to what they were in 1979.  Drawing a direct comparison isn't possible; about as close as you can get is to say that there are a very limited set of similarities between the two.

 

I mention this because trying to understand a design decision made in the late 1970s based on today's RF regulations is not going to be accurate.

[zzip]

When the device is already expensive, then adding heavy shielding or case materials is not a huge deal.   Maybe it adds IDK $10-20 to the cost?   That's nothing when your MSRP is $1000 as the 800 was in the begging.

 

But when you are cost-reducing your design to ~$200 like for the XL line, that $10-20 is a much bigger deal, 

 

So no it doesn't add a lot to the cost by itself,  but it is a barrier to getting the cost down.

[Faicuai]

On 12/1/2022 at 6:30 PM, ClausB said:

The case design was done by 1978. Also the metal frame is structurally integral to the case.

And here's the second Bingo, here! Not sure how/why I missed it, in first place...

 

So the system had to be designed as "compact" as possible but also expandable (with cards standing up-right, like most designs of the time), plus providing excellent RF suppression.

 

Upon close examination of the mechanical design of the 800, it becomes visible that the internal metal-cage not only effectively addresses the RF suppression requirement, while also holding most cards up-right in the smallest possible space... but also provides  an "endo-skeleton" upon which the outer plastic shells actually ride and attach-to, while giving the entire frame adequate structural rigidity.

 

No wonder, then, why that RF-cage is hard and thick. The entire assembly depends on it, plus isolating the external shell from the internal forces coming from either pulling or pushing the L-cart, R-cart, RAM, ROM, 80-Col and other cards, vertically and toward the bottom-board.

 

That's why the 800 is built like the Abrams tank it is. 😜💪

Edited December 8, 2022 by Faicuai
Cleanup, clarity...

[Winston2022]

On 12/8/2022 at 8:56 AM, _The Doctor__ said:

How about you do your diligence and look up what Texas Instruments was up to, look at what was dropped in 1981 (TRS models etc). Your answers not only lie within fcc documents but readily available interviews and documentaries.

You'll see in this thread that I said that asked this question here in the hope that someone had already done that and could remember the location of the documentation that I believe I have seen which dealt with production costs and FCC issues.  Instead, some get argumentative with comments starting with, "How about you do your diligence and look up..." when I came here in the first place to see if anyone already had.

[Winston2022]

On 12/8/2022 at 2:31 PM, Faicuai said:

Upon close examination of the mechanical design of the 800, it becomes visible that the internal metal-cage not only effectively addresses the RF suppression requirement, while also holding most cards up-right in the smallest possible space... but also provides  an "endo-skeleton" upon which the outer plastic shells actually ride and attach-to, while giving the entire frame adequate structural rigidity.

This has already been covered earlier in the thread and it would not have been impossible to come up with a design based upon sheet metal.  I also showed that the tendency for Atari to use cast aluminum shielding may have been cultural because even the 2600 used it, unlike the other game systems which used sheet metal.

[_The Doctor__]

If you don't want to hear or read and you don't want to believe what the designers have said in interviews or speeches let alone the governmental papers it kind of makes the thread pointless. The Video Computer System (VCS aka 2600) has time frames around it with and without the thick shielding as well as it's name and classification. The reasons are not just 'that one thing' or 'this one thing', the reasons were many but the rf modulator was the main issue. Having migrated from the 400 and 800 to discuss the toys and gaming items which lead to the later tightening of rules because of interference complaints (antennae television/ham radio/general radio) shows that which is quite clear during the evolving process. Having pointed out that when the tightening occurred entire models were dropped by other computer makers of the time because they couldn't meet the requirements still hasn't sunk in. Atari did the correct move given what they were up against and all of the settings the machines were to be used in. Having performed a couple searches today, I have found everything people have stated online and skimmed through it. Seems pretty well documented and chronicled, heck there are even magazines with pertinent information. I implore you to combine all of the information as a whole an realize the complete and encompassing reasons why the machines practically all ended up in metal boxes and cages until the rules were relaxed in one way or another many many years later.

Edited December 15, 2022 by _The Doctor__

[Winston2022]

On 12/15/2022 at 8:01 AM, _The Doctor__ said:

If you don't want to hear or read and you don't want to believe what the designers have said in interviews or speeches let alone the governmental papers it kind of makes the thread pointless. The Video Computer System (VCS aka 2600) has time frames around it with and without the thick shielding as well as it's name and classification. The reasons are not just 'that one thing' or 'this one thing', the reasons were many but the rf modulator was the main issue.

I came here to ask for a link to any FCC documents.  Also, as I've clearly shown, game systems other than those from Atari during the 400/800 time period using CPUs with the same clock frequencies as the 400/800 due to the need to sync with the NTSC signal requirements used sheet metal shielding and sometimes minimal and very crude shielding at that.  They also had RF modulators.  As I've also said, the use of cast aluminum on the 2600, the only game system I can find that did, makes me think cast aluminum may have been a choice more about engineering culture than an FCC necessity and, in hindsight, perhaps a bad choice economically.  Basically, when I see things that don't seem to make sense, I don't blindly buy into the excuses, I want to see documents... which is why I came here, for leads to them.

[+x=usr(1536)]

3 hours ago, Winston2022 said:

I came here to ask for a link to any FCC documents.

Not a problem!  Here you go:

 

https://apps.fcc.gov/oetcf/eas/reports/GenericSearch.cfm

 

All you have to search for is applicant name 'Atari' and that'll give you all of Atari's FCC filings, complete with supporting documentation and correspondence.  If it's not there, it either hasn't been digitised or doesn't exist.  Enjoy!

[+CharlieChaplin]

On 12/8/2022 at 4:15 PM, DrVenkman said:

You just keep beating that dead horse into (incorrect) glue.

 

One last time, and louder for those in the back: THE CAST METAL SHIELDING IS NOT WHY THINGS COST AS MUCH AS THEY DID.

 

 

 

Ebay 2023: Atari 800 without metal shielding: 1$, with metal shielding: 1000$

Seller: Winston2022

 

 

[+x=usr(1536)]

13 minutes ago, CharlieChaplin said:

Ebay 2023: Atari 800 without metal shielding: 1$, with metal shielding: 1000$

Seller: Winston2022

Somewhere around here there's an 800 RF shield kicking around that doesn't have a machine to go into.  Winston2022 (and only Winston2022) may buy it from me for $500, thus allowing him to double his money on eBay once it's installed in an 800.  Everybody wins!

[+DrVenkman]

3 hours ago, Winston2022 said:

during the 400/800 time period using CPUs with the same clock frequencies as the 400/800 due to the need to sync with the NTSC signal requirements used sheet metal shielding and sometimes minimal and very crude shielding at that.  They also had RF modulators. 

Not generally in 1976 - 1978 when the VCS and 400/800 were designed, and AGAIN ... the FCC regulations for home computing devices didn't exist and the proposals were all over the place. Atari took the conservative, worst-case approach. Why you refuse to grasp that is the more interesting question at this point.

 

 

[+Stephen]

4 hours ago, Winston2022 said:

Also, as I've clearly shown, game systems other than those from Atari during the 400/800 time period using CPUs with the same clock frequencies as the 400/800 due to the need to sync with the NTSC signal requirements used sheet metal shielding and sometimes minimal and very crude shielding at that.  They also had RF modulators.

Please post a list of what other machines available for sale in January 1979 used a 6502 or Z80 CPU, had an RF modulator, and used thin sheet metal shielding.  I'm drawing a blank.

 

You do realize that the 2600 design started in 1976 correct?  The 400/800 design was started in 1977/1978 time frame.  Unless you are stuck fixating on the RCA Studio 2 or Fairchild Channel F, I am simply dying to know of these other systems using sheet metal shielding.

[Winston2022]

On 12/23/2022 at 11:08 AM, Stephen said:

Please post a list of what other machines available for sale in January 1979 used a 6502 or Z80 CPU, had an RF modulator, and used thin sheet metal shielding.  I'm drawing a blank.

 

You do realize that the 2600 design started in 1976 correct?  The 400/800 design was started in 1977/1978 time frame.  Unless you are stuck fixating on the RCA Studio 2 or Fairchild Channel F, I am simply dying to know of these other systems using sheet metal shielding.

See my post on page 1 of this thread.  It even has internal images of the game systems, their shielding, specs and dates of initial release and end of sales.

Edited December 29, 2022 by Winston2022

[Winston2022]

On 12/23/2022 at 10:02 AM, DrVenkman said:

Not generally in 1976 - 1978 when the VCS and 400/800 were designed, and AGAIN ... the FCC regulations for home computing devices didn't exist and the proposals were all over the place. Atari took the conservative, worst-case approach. Why you refuse to grasp that is the more interesting question at this point.

 

 

And I'll ask again:

 

Since the sheet metal shielded game systems released at the same time as the Atari 8-bits had the same video-synced clock speed for their CPUs as the Atari computers, that frequency being what determines the harmonics which cause the RFI, and all of them having video modulators, WHY would Atari think that home computer standards could be justified to be any more strict than for the already FCC approved and fielded video game systems?