CompuServe Thread

At least I'LL admit when I'm wrong. ,<|:^) Since you have the literature, and I don't, I will assume you are correct (in this rare instance) Erik Flom P.S. When are you going to get an actual machine (it might be nice to make sure you can live up to all the claims you are boasting about)? P.P.S. (Way back in the days of the A3000, they called it the Processor Direct slot (kinda' like the Mac), now I guess they are calling it a Local Bus (maybe trying to woo those unwary IBMers out there.))

I stand corrected, I'm willing to admit when I was wrong. (Though they did just change the name recently.) Yeah, I thought it WAS pretty funny when I heard about IBMers touting the 'Local Bus'. (Oooohhhhh, wow man!) A few years ago (circa: A3000), they were calling it a Processor Direct slot (same name as the Mac PDS slot), now I guess they are trying to woo all those IBMers by pointing out that, YES, the Amiga does have a 'Local Bus'. Thanks, Erik Flom

Dave, The 68EC020 _is_ a surface mount part. It is a 100 pin PQFP (Plastic Quad Flat Pack) with a package designator of 'FN'. I would suggest _you_ get some information on the processor you intend to design for, as there are considerable differences between the EC020 and an 030. And for your information. A PLCC package is considered to be a surface mount package, it just so happens to have thru hole sockets available for it. Dean DKB Software

The 68EC020 in the A1200 is DEFINITELY surface-mounted. This is darned close to the actual size of the chip: _____ | || |_____|

Betty, CBM told my dealer flatly that CBM would not allow mail-order of any of the AGA machines, so that dealers would be supported. They implied that there were some early trickles of these machines that may have gotten through before the policy, but after that, nada. And if you buy it mail-order, or from any source other than a dealer, the dealer cannot honor the warranty. I smirked and said, we'll see. This isn't the first time CBM has paid lip service to keeping mail-order at arm's length to support dealers. I heard of someone who actually tracked a truck from the WestChester plant to Montgomery Ward and unloaded machines. Hearsay, of course. I hope it's true this time. Amiga dealers have a rough time of it; CBM changes their policies daily at times. Karl

Betty, Maybe I should get out more. <g> Your explaination sounds quite reasonable. I think I might go back and ask him anyhow. I'll tell you if it's anything interesting (or different). -Sean

Mark, Betty seemed to have a good explaination (that it's being mass marketed and is therefore leaving dealers out in the cold). I think I'm going to go back and ask anyhow though. Thanks. -Sean

Mark, Sounds nice. My dealer know nothing about it. Do you know how they are going to distribute a 77MB anim? Is it in HAM8? Karl

This is a Quad Flat Pack CPU? FN suffix designates a PLCC CPU….I have the Motorla specs right here. RC = PGA @ .100" centers FE = Surface mount @ .025" centers FN = PLCC @ .050" centers in this text they make no mention of anything in Quad Flat Pack package Besides, there are devices that can be placed over one, even at it's .025 centers….they are pretty expensive though BHAH-HUMBUG! Dave Elite MicroComputers – via Whap!

It would all come down to a socket….it would have to fit the chip just right….and I have a few makes that just might do that…as long as all the leads make a soild contact it is a very good way of doing it. Dave Elite MicroComputers – via Whap!

All I would need is about 1/4" of clearance above the chip, to make the socket fit. I don't want my product to be a 'this or that' item. Dave Elite MicroComputers – via Whap!

Dave, The point isn't that the FCC will or won't test the device, the point is that it is illegal to sell the device before it is FCC certified. It doesn't even matter if when the FCC tests it…it passes with flying colors, what matters is the selling of an uncertified device. Don

We have some superb developers on Amiga, and I for one am grateful for them, such as Softwood, GVP, DKB, etc. Some aren't quite so hot, some just fly away … Ron

Dave, You didn't answer my question. Are points A, B and C in my message # 82599 correct? I'm not asking your logic, I'm asking if those points were correct. Don

You have just made it perfectly clear what separates law abiding, honest folk from yourself. Your logic would suggest that it would be just fine for a factory to continue to dump pollutants in to a water supply, in violation of federal law, simply because the cost of cleaning up the process would be too high. And your logic would also dictate that broadcasting a pirate radio station is 'justififed' because the licensing fees are too high. Regardless of your technical expertise, or the quality of your products, you have, IMO, no credibility as a dealer. You might want to consider selling your products to a larger company that is willling to have the products tested, thereby allowing you to concentrate on development, and leaving the legal issues to a responsible third party.

Yep, well, we know exactly what happened in that case. Here's the poop: When we got our first little communication from the FCC, among the other things in the envlope was a photostatic copy (scan) of the bottom of a HAM-E unit, complete with FCC ID tag, and the serial number of the unit. We know who had that unit – we gave it to them free of charge at their request – and it's as simple as that. There is a great deal of competition, and sometimes (NOT always) that turns into backstabbing. Sometimes there is legitimate defense from a company that has been attacked, and I'm a bit unwilling to characterize that as backstabbing, although if cornered I might go for it. There is always the problem of what to do when attacked; Me, I don't expect people (companies) to act like mother Theresa. It would seem that many users DO expect that, though. Competition can exist side by side with community; I ran a martial arts school for some time and that is *exactly* the attitude that you want to foster in that environment. My view is that the Amiga marketplace can be considered extremely similar – similes and parallels abound. Nice is better – if people want to play, I'll play. If they want to play rough, I'll play rough. And if they want to fight, fine. I'll fight. I'd always prefer soft to hard, though. –Ben …via AutoPilot

You're getting a lot of good advice for free.

Don't bet against the king of dirty pool….I can be VERY difficult if I want to be. I didn't get this far with nothing. Dave Elite MicroComputers – via Whap!

That's good to hear. :^) –Ben …via AutoPilot

Jonathan, This isn't a simple answer, so you'd probably better save it. A PC clone is a bit of a different case than an Amiga-based approval. Some of our costs were incurred because of distance (no labs in Montana), and a large amount due to the Amiga 500's inherent non-compliance. Labs tend to know more about PC clones, and that will "grease" the process, and amplify any help they offer. Cost is one of those highly variable things. Probably what you were being told ($5k) is the lab fee; which can vary based on a number of things. Let's talk about the process. First, you take your "thing" into the lab proper, in a building somewhere. They put it on a bench and point some test gear at it. They measure all over the RF spectrum and graph it. That usually takes about a day. A good lab accounts for that day in the quoted price. A good lab will also point out to you exactly where in the RF spectrum your device appears to fail, and then will use an RF sniffer to help you localize the problem unless it's totally obvious (and those problems are NOT usually obvious, btw). Perhaps a small cap can help, or a choke. A good lab might have those around for you. If not, and you forgot, you're coming back, and that will cost you more – a lot more, usually. Now, let's say you get thru the "lab day". A good lab will also have counted a day in the field. And it's really "in the field", too… they have what some (me) like to call a "calibrated cow pasture" where they are outdoors, away from all RF generating devices, or where RF interference is known *exactly* (such as a nearby FM broadcast station) and accounted for in the report(s) to the FCC. You go thru the same tests, and again, if there are NO problems, you're looking at about a day. The testing is more stringent, so you are very likely to find new problems. Again, if problems are found, and you have the *fast* means to fix them in your "pocket" either by the virtue of the lab's preparedness or your own, you can still <continued>

get out of there in a day. If not, you're coming back, and those costs go up fast, again. Now, the above seems to indicate that a well prepared lab (or testee) can avoid extra tries – well, yes, but not usually. Since most people don't know poo-poo about RF interference, often the problems are more basic than just whacking (a) cap(s) on (a) signal(s). Sometimes it's how long a PCB trace is, or *where* a PCB trace is ajoining another. Sometimes it's the physical size of the case, or the ventilation of that case. Holes for feet, plugs, wires, outlets, you name it. Sometimes it's the attachment of YOUR device that makes the HOST device's radiation exceed the limits, and you always test as a system, not as a "thing". That's related to why the Amiga 500 is such a nasty thorn waiting to be stepped on; it just barely passed under the old FCC rules, and it CANNOT pass under the recent FCC rules, even with NOTHING attched to it. So if you go in there with a 500 (we did) to test with, you cannot pass, period, and you're coming back, period, and you're spending more money, period. Now, in addition to the lab fees, there is a $750 FCC registration fee (which most labs do NOT include in their price, because after all, they aren't the ones charging you, eh?). There is a *ton* of paperwork to be processed, photos to be taken of the device, the device in use, top of the board, bottom of the board, etc. Does the device use cables? Oh boy, more fun. :^) Anyway, it's just not an easy, or simple process. A good lab makes a lot of difference, but it does NOT make it "easy". Then there's travel; lost work; will you need a consulting RF engineer to solve your problems? Unless you're lucky, or you're like us and have lots of RF types available in-house, you probably will. Maybe the lab can do that for you, but guess what – it'll cost more money, almost certainly. You'd have to factor a number of things in about that clone; and also, class A (not too stringent) or class B (very stringent, a real pain in the <continued>

a$$) ratings make the test completely different. The bottom line will vary wildly as a result – and the more so for smaller organizations that don't have RF testing equipment on site (we do – but do most Amiga developers have $10k to $50k for a spectrum analyzer? If they rent or lease one, will they know how to use it? Will they use the right antennas? Could they set up a test site? You get the idea…), plus the relevant FCC regs for finding out just what it is you're testing for. Remember that one of the costs of FCC approval is *designing* for that approval; doing it just right takes a great deal of extra time, usually in the mechanical areas such as the case, but sometimes right down to where and how PCB traces are run. If you don't have to worry about RFI, your design process (or fix-it-up-afterwards-process) will be shortened considerably. You should count that as part of the costs, because that's what they are. Ready for the clincher? You're not, but here it is anyway. These labs, see, they are "approved" by the FCC, you bet. They have to be to get results that are meaningful. But the FCC, if it wants to, can decide that they don't "like" the results from the lab, and then require you to travel to an actual FCC site (read, Washington D.C., or at the very least, "somewhere else") and repeat the ENTIRE series of tests under the eye of a bona-fried FCC engineer person. Why would they do this? You've got me there, but in fact they DO do it, and you'd better account for that as well in your budget – remember that you MUST complete this process once you start it, or you're well and truly fried – you have NO product and you've lost EVERYTHING you spent to get it to whatever point it got to, pre-final-approval. The FCC is very like the IRS in one way – they aren't particularly accountable to anyone. They can hang you out to dry without benefit of "due process" and you have little or no recourse when and if they do. That's one of the reasons I cautioned Dave as strongly as I did – it's not <continued>

the same as a court situation at all, and those, Darwin knows, are bad enough. I can offer one hint: If you're going to test, use an Amiga 3000 or 4000 if at all possible. These machines were produced in the time frame where they had to pass under the more stringent rules, and that means a lower noise floor than you'd have with a 2000-class machine, which gives YOU more headroom to make noise. We only brought an Amiga 500 with us to the test the first time. Baaad mistake – and we bought it on the way TO the test FOR the test, so we hadn't done any RFI analysis with it, only on sevreal of our 2000's.. We thought the 500 would be quieter than the 2000; another baaaaad mistake. And we're really not too dumb when it comes to RF work… Just goes to show you, you can't assume anything about this particular process. In regards to this, when we came BACK with a 2000, the HAM-E passed in that frame with considerable margins right across the spectrum; in all but two locations in the *entire* RF spectrum, the HAM-E adds no discernable noise on a 2000 frame – partly because the 2000 is so noisy itself, partly because it's just a quiet device. In those two "spikes", the HAM-E clears the limits by plenty anyway. About the lab; We would use the same lab again; they were very helpful. I will not outright name the lab, for one reason that we may indeed use them again, and I don't think it would be prudent to let the info out – most new hardware we think about isn't anything we want made public in any way, shape or form, and for another reason, The research and time that went into choosing the lab was a use of assets which included considerable ca$h money. We have several hardware designs on the shelf, where they will in all likelyhood remain forever unless the Amiga market changes completely. I think others need to do their own research here, as required. I might $ell you the name, though. :^). –Ben …via AutoPilot

The $10,000 figure I used was what I was quoted (roughly) to get a PC Clone approved. I was also told that this would the EXACT system I sent. Thus if I were to change one board, being brand or type, the system would then be illegal to sell. In theory, every advertiser in Computer Shopper is breaking the law. Even more so then I am because they are claiming certification when they don't have it. As I've said it's not what they're doing, but who's doing it. (Kind of like C=) Dave Elite MicroComputers – via Whap!