panon@cheddar.cc.ubc.ca (Paul-Andre Panon) (12/07/89)
Ho, I could be wrong but I don't see how they could fit PC or Nubus slots in
an A3000 in a way similar to the way it's done on the A2000 and still have
it fit on a desktop (i.e. have a reasonable footprint). I would expect the
Zorro III slots in the machine have AT LEAST 24 more pins/lines to support, so
that 32 data and 32 address lines can be accessible by all peripherals. That
doesn't include any additional lines Commodore engineers may have decided would
be useful since the Zorro II specs. were hashed out for the A2000. If they add
the lines on (at one end or another) a la AT bus, form factor problems develop
using existing bridgeboard designs.
I guess there's also the question of whether Commodore feels the PC
compatability will be an asset in the A3000's preferred market. The A2000
needed PC compatability because of a small software base. With the A3000 aimed
at the high-end-Amiga/low-end-UNIX market, PC Compatability wouldn't be as high
an issue/priority as when C= was trying to break the business market with the
A2000.
Finally, there's board space. I'd much rather see Commodore incorporate the
A2091 drive controller and more 32-bit RAM on the motherboard of the A3000
for a lower overall system cost (I assume that would be the case since that
was the argument presented for the built-in keyboard on the A500). The machine
could also probably be reduced to a MacIIcx-like footprint or have more room
for internal hard-disk expansion if that approach were used. Of course, having
a2091-equivalent on the motherboard would mean there would be one more slot
available for additional uses.
So, yeah, IBM compatability is nice - but I think the A3000 would gain a lot
more by NOT having it. People who want/need IBM compatability could probably
still get an A2500/30 or some similar beast and slap a bridgeboard in it.
--
Paul-Andre_Panon@cheddar.cc.ubc.ca or USERPAP1@UBCMTSG
or Paul-Andre_Panon@undergrad.cs.ubc.ca or USERPAP1@mtsg.ubc.ca
"What should the role of the University be? It should be to enlighten Society."
-Luis Sobrinoswarren@eugene.uucp (Steve Warren) (12/08/89)
In article <5906@ubc-cs.UUCP> panon@cheddar.cc.ubc.ca (Paul-Andre Panon) writes: > > Ho, I could be wrong but I don't see how they could fit PC or Nubus slots in >an A3000 in a way similar to the way it's done on the A2000 and still have >it fit on a desktop (i.e. have a reasonable footprint). I would expect the >Zorro III slots in the machine have AT LEAST 24 more pins/lines to support, so >that 32 data and 32 address lines can be accessible by all peripherals. That >doesn't include any additional lines Commodore engineers may have decided would >be useful since the Zorro II specs. were hashed out for the A2000. If they add >the lines on (at one end or another) a la AT bus, form factor problems develop >using existing bridgeboard designs. There is an easy way to design zorro III so that the same space is used as zorro II, while allowing zorro II to plug into the same connectors: #I I I I I I I I I I I I I I I I I I I I I I I I I I I# <-zorro II fingers # # <- ridge to keep zorro II boards from III fingers #I I I I I I I I I I I I # I I I I I I I I I I I I I I# <-zorro III fingers ####################################################### <-connector case The bottom row, "zorro III fingers" only has the additional data, address, and control lines that are unique to the zorro III standard, and uses the same lines as zorro II for the rest of each bus. Older boards can plug into the connector and only make contact with the zorro II fingers. Newer boards have a slice cut out of them so they can seat down into the ridge, and they have two rows of contacts to connect to both the zorro II and the zorro III fingers. This technique is already being used in "blue" land. Yes, they do manage to do some things right out there, and I think this is a good example. --Steve ------------------------------------------------------------------------- {uunet,sun}!convex!swarren; swarren@convex.COM
panon@cheddar.cc.ubc.ca (Paul-Andre Panon) (12/09/89)
In article <3751@convex.UUCP& swarren@convex.COM (Steve Warren) writes:
%There is an easy way to design zorro III so that the same space is used
%as zorro II, while allowing zorro II to plug into the same connectors:
%#I I I I I I I I I I I I I I I I I I I I I I I I I I I# <-zorro II fingers
%# # <- ridge to keep zorro II boards from III fingers
%#I I I I I I I I I I I I # I I I I I I I I I I I I I I# <-zorro III fingers
%Older boards can plug into the connector and only make contact with the
%zorro II fingers.
%Newer boards have a slice cut out of them so they can seat down into
%the ridge, and they have two rows of contacts to connect to both
%the zorro II and the zorro III fingers.
%This technique is already being used in "blue" land. Yes, they do
%manage to do some things right out there, and I think this is a
%good example.
%--Steve
%-------------------------------------------------------------------------
% {uunet,sun}!convex!swarren; swarren@convex.COM
I guess you mean the EISA kludge. Sorry, I don't consider it "doing it right".
--
Paul-Andre_Panon@cheddar.cc.ubc.ca or USERPAP1@UBCMTSG
or Paul-Andre_Panon@undergrad.cs.ubc.ca or USERPAP1@mtsg.ubc.ca
"What should the role of the University be? It should be to enlighten Society."swarren@eugene.uucp (Steve Warren) (12/12/89)
In article <5936@ubc-cs.UUCP> panon@cheddar.cc.ubc.ca (Paul-Andre Panon) writes: >In article <3751@convex.UUCP& swarren@convex.COM (Steve Warren) writes: [discussion of how to get 32-bit connections while allowing zorro II boards to plug into the same slots] >%This technique is already being used in "blue" land. Yes, they do >%manage to do some things right out there, and I think this is a >%good example. >I guess you mean the EISA kludge. Sorry, I don't consider it "doing it right". > ^^ >-- > Paul-Andre_Panon@cheddar.cc.ubc.ca or USERPAP1@UBCMTSG >or Paul-Andre_Panon@undergrad.cs.ubc.ca or USERPAP1@mtsg.ubc.ca >"What should the role of the University be? It should be to enlighten Society." Depends on what you mean by "it". Of course I have no idea what you mean by "it", since all you stated was your opinion with no reasons or justifications. That must be why you prefaced it with an apology. Apology accepted. --Steve ------------------------------------------------------------------------- {uunet,sun}!convex!swarren; swarren@convex.COM
panon@cheddar.cc.ubc.ca (Paul-Andre Panon) (12/12/89)
In article <3854@convex.UUCP> swarren@convex.COM (Steve Warren) writes: >In article <5936@ubc-cs.UUCP> panon@cheddar.cc.ubc.ca (Paul-Andre Panon) writes: >>In article <3751@convex.UUCP& swarren@convex.COM (Steve Warren) writes: > > [discussion of how to get 32-bit connections while allowing zorro II > boards to plug into the same slots] > >>%This technique is already being used in "blue" land. Yes, they do >>%manage to do some things right out there, and I think this is a >>%good example. > >>I guess you mean the EISA kludge. Sorry, I don't consider it "doing it right". > >Depends on what you mean by "it". > >--Steve Sorry (again :-). You effectively said that you thought the two level connection EISA uses is one of the things "they" managed to do correctly in Blue land. Most of the reviews I'd heard about EISA weren't exactly-shall I say-raving. I can't remember the exact points. One that occurs to me is that you would need extra circuitry so that-should somebody not push in a card far enough so that, if the lower level fingers are in contact with the higher level contacts, the board and machine don't get fried from juice going the wrong way. If something can go wrong, it will. I find the AT or MCA type extension more esthetic and less mechanically error-prone, and the mechanics of the extension is, after all, what's involved isn't it? You might have greater crosstalk at high frequencies with the EISA design too, no? I seem to remember that was part of the reason why so many of the lines in MCA are have to be ground. Hopefully I haven't put my foot too far in my mouth and won't wind up saying sorry in this thread again. -- Paul-Andre_Panon@cheddar.cc.ubc.ca or USERPAP1@UBCMTSG or Paul-Andre_Panon@undergrad.cs.ubc.ca or USERPAP1@mtsg.ubc.ca "What should the role of the University be? It should be to enlighten Society." -Luis Sobrino
steve@ntmtka.mn.org (Steve Wahl) (12/13/89)
As I recall, this whole discussion has to do with whether or not there will be IBM-PC type slots int the 3000, and the argument has been "Probably not, because of the extra room taken by the 32 bit connections on the slots." How about this solution: (NOT to scale) |----------------------------------------------------------------------| B o M | 16-bit Zorro 32-bit Zorro extension | a f a | ============================ ========================== | c c | | k h | ============================ ========================== | i | (Repeat N times) | n | ============================ ============ ====================== | e | 16-bit Zorro (Bridgecard Slots) AT & XT bus | | ============================ ============ ====================== | | | | ============ ====================== | | (AT/XT Bus in whatever size) | | ====================== | | | . . . | | |----------------------------------------------------------------------| Of course, there would have to be some kind of logic so that the 16-bit zorro slots are mapped only into one 16-meg portion of the 32-bit address space. This would have to be possible to use older Zorro boards, wouldn't it? Due to the opinions that I've heard in this group regarding the utility of a bridgeboard (which I agree with - I don't see a reason to get one for my 2000, and I don't know what I'd do with my 5.25" hard drive), I don't think anyone is going to argue that the bridgeboard has to be connected to the full 32 bits of the Zorro bus; I'm sure it can get by on a 16-bit port. Does anyone see anything wrong with this scheme? Personaly, I'm not sure whether or not the 3000 will have a PC Bus or not, but I do think it is possible. As to whether it would be a desired feature, I must point out that for *some* reason Sun Microsystems has both a software and hardware option for (some of?) their Sparc Stations (and maybe other systems) to run MS-DOS based software. My only guess is that there are some executives who like to buy the most expensive computers that can run 1-2-3 and dBase :-). These are only my opinions, of course! --> Steve -- Steve Wahl Northern Telecom, Inc. (612) 932-8079 S-100, 9701 Data Park steve@ntmtka.mn.org Minnetonka, MN 55343 {rosevax,bungia}!ntmtka!steve
navas@cory.Berkeley.EDU (David C. Navas) (12/14/89)
In article <1935@ntmtka.mn.org> steve@ntmtka.mn.org (Steve Wahl) writes: >As I recall, this whole discussion has to do with whether or not there will > >|----------------------------------------------------------------------| B o M >| 16-bit Zorro 32-bit Zorro extension | a f a >| ============================ ========================== | c c >| | k h >| ============================ ========================== | i >| (Repeat N times) | n >| ============================ ============ ====================== | e >| 16-bit Zorro (Bridgecard Slots) AT & XT bus | >| ============================ ============ ====================== | >| | >| ============ ====================== | >| (AT/XT Bus in whatever size) | >| ====================== | >| | > . > . > . > >| | >|----------------------------------------------------------------------| > >Due to the opinions that I've heard in this group regarding the utility >of a bridgeboard (which I agree with - I don't see a reason to get one >for my 2000, and I don't know what I'd do with my 5.25" hard drive), >I don't think anyone is going to argue that the bridgeboard has to be >connected to the full 32 bits of the Zorro bus; I'm sure it can get by >on a 16-bit port. Unless you remember that all graphics memory is transferred via a 16 bit bus rather than a 32 bit bus if you keep the old architecture... And we have a '386 version on the way -- and I'd like EGA capability on of these days.. Of course, the BridgeBoard would have to be souped up a litle. > >Does anyone see anything wrong with this scheme? Yeah -- see above. The answer is easy, really. ---------16-bit Zorro----------------32 ext--------------- | =============================== ========== || | | || | | =============================== ========== || | | || 32 bit ext for bridgbrd | =============================== ========== || | | || | | =============================== ========== | | || | | C P U C O - P R O C E S S O R S L O T || | | || | | || | | || | | || | | || Zorro II for BrdgBrd | || | | || | | || | | || | | || | | || | | | | | | =32 MSDOS= =AT ext.= =normal PC====== || | | || | | ========== ========= ================ | | || | | ========== ========= ================ || | | || | | ========== ========= ================ || | | || | | | | || | | || | ---------------------------------------------------------- I don't know if this makes any more sense, however... You might want to move the Bridgeboard card to the front of the machine, or something like that. Main point is to keep the expansion possibilities open, right? > >Personaly, I'm not sure whether or not the 3000 will have a PC Bus or >not, but I do think it is possible. As to whether it would be a desired >feature, I must point out that for *some* reason Sun Microsystems has both >a software and hardware option for (some of?) their Sparc Stations (and >maybe other systems) to run MS-DOS based software. My only guess is that >there are some executives who like to buy the most expensive computers >that can run 1-2-3 and dBase :-). Of course, isn't that why I want a '486 machine? >These are only my opinions, of course! Me too. > >--> Steve >-- >Steve Wahl >Northern Telecom, Inc. (612) 932-8079 >S-100, 9701 Data Park steve@ntmtka.mn.org >Minnetonka, MN 55343 {rosevax,bungia}!ntmtka!steve David Navas navas@cory.berkeley.edu
farren@well.UUCP (Mike Farren) (12/14/89)
In article <3751@convex.UUCP& swarren@convex.COM (Steve Warren) writes: > <discussion of EISA technique of allowing two different cards to interface differently, but using the same slot by having two levels of connector pins deleted > >This technique is already being used in "blue" land. Yes, they do >manage to do some things right out there, and I think this is a >good example. I can't consider this technique "right" in any real way. It has struck me, from the beginning, as some of the shakiest mechanical and electronic engineering work I've ever seen. Not only does it require special card-edge connectors which are going to be hell to fabricate, and therefore expensive, but it takes too little account of such things as PCB tolerances, random glorp dropping into the connectors, and the biggest problem of all, the near-total lack of any knowledge on the part of the people sticking the damn cards into the slots. It's easy enough to misalign a card in an existing slot, but at least the existing slots don't carry the problem of dealing with signal and power lines shorting to one another... -- Mike Farren farren@well.sf.ca.usa
panon@cheddar.cc.ubc.ca (Paul-Andre Panon) (12/14/89)
In article <20643@pasteur.Berkeley.EDU> navas@cory.Berkeley.EDU.UUCP, (David C. Navas) lays out a proposed method for having 32-bit Amiga and PC busses (sp?) in an Amiga 3000 > >David Navas >navas@cory.berkeley.edu and it looks very nice but it would have to be in a tower or else your desk will disappear. I'd rather see something lean and mean with a Mac II cx sized footprint with C= keeping the 2000/2500 machine around for the people who need PC compatability. Since, from what I've heard, the 2630 runs asynchronously at 25 MHz, the A3000 probably won't run that much faster for CPU-bound stuff. Customers would have to make a decision as to which is more important to them, PC slots or 32-bit Amiga slots. A more expensive tower could come out later but I would rather see a lower-cost '030-based amiga first. -- Paul-Andre_Panon@cheddar.cc.ubc.ca or USERPAP1@UBCMTSG or Paul-Andre_Panon@undergrad.cs.ubc.ca or USERPAP1@mtsg.ubc.ca "What should the role of the University be? It should be to enlighten Society." -Luis Sobrino
swarren@eugene.uucp (Steve Warren) (12/15/89)
In article <5971@ubc-cs.UUCP> panon@cheddar.cc.ubc.ca (Paul-Andre Panon) writes: >In article <20643@pasteur.Berkeley.EDU> navas@cory.Berkeley.EDU.UUCP, >(David C. Navas) lays out a proposed method for having 32-bit Amiga and PC >busses (sp?) in an Amiga 3000 [...description of why Paul feels this would be bad...] >Customers would have to make a decision as to which is more important to them, >PC slots or 32-bit Amiga slots. A more expensive tower could come out later >but I would rather see a lower-cost '030-based amiga first. You are probably right, although sometimes I wonder if the designers watch us discussing things about which we know nothing substantive (that is, what will their design decisions be) and smile. Oh well, such is the lot of the ignorant peasants ;^). If Commodore wants to really beat the Mac price point and sell a lot of A3000s they will probably need to come out with a box such as you describe. Maybe they could come out with a neo-sidecar connected with flex-cable to an expansion port ;^). Use a standard baby-AT case so there would be some expansion lots; yeah, that's the ticket... --Steve ------------------------------------------------------------------------- {uunet,sun}!convex!swarren; swarren@convex.COM
filbo@gorn.santa-cruz.ca.us (Bela Lubkin) (12/15/89)
What kinds of bus might we expect in there? Zorro II -- 100-pin -- Amiga 2000 compatible. Zorro III -- more pins -- Amiga 3000 specific. IBM PC -- 62 (?) pins -- Amiga 2000 compatible. IBM AT -- 30 (?) pin extension -- Amiga 2000 compatible. 386/486 -- many "standards" or make one up -- mainly for 32-bit memory. What would be really nice is if the same form factors could be made to work for all. From the Amiga perspective there might be a Zorro II slot with an extension that makes it into Zorro III. From the IBM perspective there might be an AT slot with a 32-bit extension for 386/486 memory. It would look something like this: +---------------------------------------------------------------------+ | ----- PC ----- -- AT -- == 32-bit == ========== Zorro II ========== | 1 | | -------------- -------- ============ ============================== | 2 | | -------------- -------- ============ ============================== | 3 | | -------------- -------- ============ ============================== | 4 | | -------------- -------- ============ ============================== | 5 | ~~~~~~~~~~~~ | ============== ======== ============ ============================== | 6 | | ============== ======== ============ ------------------------------ | 7 | | ============== ======== ============ ------------------------------ | 8 +---------------------------------------------------------------------+ All "==" and "--" are physical slots. "==" are active, "--" are inactive. "~~" represents electrical isolation between adjacent slots. In the absense of a bridge board, all slots are 16/32-bit Zorro II/III. In the presence of a bridge board, slots above the bridge board are Zorro II/III; the bridge board itself uses the Zorro II part of the bus to interface with the Amiga side, and activates the other three sections as an AT bus with 32-bit extension. This allows all the slots to be used for Amiga (Zorro II or III) boards; or, with a bridgeboard in the first slot, allows all the rest to be used for PC boards. You can reallocate slots by simply moving the bridgeboard. In the setup above, slots 1-5 are allocated for Amiga boards, 6 has the bridgeboard, and 7-8 are available for PC, AT or 32-bit PC boards. Unfortunately, there's a big detail I'm leaving out. HOW does the bridgeboard electrically isolate the Amiga 32-bit extension from the PC 32-bit extension? I don't know. I'm sure there's a way. "A way": I would not buy it if this was how it was done, but as an example to prove that it's possible, you could put hinges in the 32-bit extension bus traces between each two slots and design the bridge card to physically swing open those hinges. Now, is there a >reasonable< way to do it? Bela Lubkin * * // filbo@gorn.santa-cruz.ca.us CI$: 73047,1112 (slow) @ * * // belal@sco.com ..ucbvax!ucscc!{gorn!filbo,sco!belal} R Pentomino * \X/ Filbo @ Pyrzqxgl +408-476-4633 and XBBS +408-476-4945
sparks@corpane.UUCP (John Sparks) (12/15/89)
<3854@convex.UUCP> <1935@ntmtka.mn.org> Sender: Reply-To: sparks@corpane.UUCP (John Sparks) Followup-To: Distribution: na Organization: Corpane Industries, Inc. Keywords: 3000 slots In article <1935@ntmtka.mn.org> steve@ntmtka.mn.org (Steve Wahl) writes: >I don't think anyone is going to argue that the bridgeboard has to be >connected to the full 32 bits of the Zorro bus; I'm sure it can get by >on a 16-bit port. Ok I'll argue for it :-) I think it would be nice to have a way for the bridgeboard slot (if they put on in) to be able to use 16 or 32 bits of the Zorro bus. Why? Well the 16 bit so you can plug in the XT and AT bridgeboard, and the 32 bits so commodore can come out with a new 386 bridgeboard (utilizing the full 32 bit bus so it doesn't end up crippled in a 16 bit slot) I think IBM compatibility will be important in the 3000. Remember, the reason it was put in the 2000 in the first place was to help the amiga get it's foot in the door of the business market. Well the 3000 is more of a business machine than the 2000 is. It needs the same advantage, if not even more so. -- John Sparks | {rutgers|uunet}!ukma!corpane!sparks | D.I.S.K. 24hrs 1200bps ||||||||||||||| sparks@corpane.UUCP | 502/968-5401 thru -5406 : I'm not as dumb as you look.
monty@sagpd1.UUCP (Monty Saine) (12/16/89)
In article <14997@well.UUCP> farren@well.UUCP (Mike Farren) writes: >In article <3751@convex.UUCP& swarren@convex.COM (Steve Warren) writes: >> <discussion of EISA technique of allowing two different cards to interface > differently, but using the same slot by having two levels of connector > pins deleted > >of the people sticking the damn cards into the slots. It's easy enough >to misalign a card in an existing slot, but at least the existing slots >don't carry the problem of dealing with signal and power lines shorting From what I have seen of the EISA design used by big blue there are alignment ridges that prevent the old style boards from contacting the extra contacts down deeper. These samw ridges and the edges of the overall connector should prevent the kind of problem you fear. The killer is going to be the cost of these kind of "custom" connectors. Remember trying to finds a DB-23? Only time will tell what design they have choosen. Monty Saine
panon@cheddar.cc.ubc.ca (Paul-Andre Panon) (12/18/89)
In article <557@sagpd1.UUCP> monty@sagpd1.UUCP (Monty Saine) writes: > From what I have seen of the EISA design used by big blue there are >alignment ridges that prevent the old style boards from contacting the extra >contacts down deeper. These samw ridges and the edges of the overall connector >should prevent the kind of problem you fear. The killer is going to be the >cost of these kind of "custom" connectors. Remember trying to finds a DB-23? > Monty Saine I don't think big blue wants credit for EISA. The problem I had in mind was more along the lines of: somebody does not push one of the newer cards in deep enough so that the card's bottom row of fingers is in contact with the slot connector's upper row of contacts. Either way, there is too much potential for something to go wrong that has to be kludged around one way or another; I can't feel comfortable with the design. -- Paul-Andre_Panon@cheddar.cc.ubc.ca or USERPAP1@UBCMTSG or Paul-Andre_Panon@undergrad.cs.ubc.ca or USERPAP1@mtsg.ubc.ca "What should the role of the University be? It should be to enlighten Society." -Luis Sobrino
farren@well.UUCP (Mike Farren) (12/20/89)
In article <557@sagpd1.UUCP> monty@sagpd1.UUCP (Monty Saine) writes: > From what I have seen of the EISA design used by big blue there are >alignment ridges that prevent the old style boards from contacting the extra >contacts down deeper. These samw ridges and the edges of the overall connector >should prevent the kind of problem you fear. The killer is going to be the >cost of these kind of "custom" connectors. Remember trying to finds a DB-23? The last time I saw the EISA spec (admittedly, only a quick look), I saw one, and only one, alignment device - a key which fit into a slot approximately halfway down the card-edge on an EISA card, and which would, supposedly, block a non-EISA card from being inserted fully. By the way - ascribing this to "big blue" is one of the funniest bits of total mistake I've seen in this group for months. Big Blue (IBM to you), of course, promotes the MicroChannel bus, not EISA - a bus design brought out solely to stick it to "big blue". (Fat chance, IMHO, but we'll see...) -- Mike Farren farren@well.sf.ca.usa
monty@sagpd1.UUCP (Monty Saine) (12/22/89)
In article <15126@well.UUCP> farren@well.UUCP (Mike Farren) writes: >In article <557@sagpd1.UUCP> monty@sagpd1.UUCP (Monty Saine) writes: > >The last time I saw the EISA spec (admittedly, only a quick look), I saw >one, and only one, alignment device - a key which fit into a slot approximately >halfway down the card-edge on an EISA card, and which would, supposedly, >block a non-EISA card from being inserted fully. ^^^^^^^^^^ Why supposedly? Have you ever tried to plug a circuit card in that had a key slot in it that didn'ty match the key it was being pluged into? T'ain't easy my friend. Also remember that connectors have ends that also guide a card in otherwise you would run the risk of doing what you fear plugging in any pc card. > >By the way - ascribing this to "big blue" is one of the funniest bits >of total mistake I've seen in this group for months. Big Blue (IBM to you), of course, promotes the MicroChannel bus, not EISA - a bus design brought >out solely to stick it to "big blue". (Fat chance, IMHO, but we'll see...) On this you are totally correct. I "blue"d it by mistake and appoligize to the net for spreading this error. Monty Saine
farren@well.UUCP (Mike Farren) (12/23/89)
In article <569@sagpd1.UUCP> monty@sagpd1.UUCP (Monty Saine) writes: > Have you ever tried to plug a circuit card in that had > a key slot in it that didn'ty match the key it was being pluged into? > T'ain't easy my friend. Also remember that connectors have ends that also > guide a card in otherwise you would run the risk of doing what you fear > plugging in any pc card. Yes, as a matter of fact, I have done exactly that. It may not be easy, but there are these things called tolerances (you know about tolerances?), and all it takes is an out-of-tolerance slot, or out-of-tolerance key, or (as in my case) a key which had been broken, allowing the card to "slip". Also remember that the EISA key is NOT a full-height key - it only extends half-way up the connector body. Just a little tolerance slippage, and you'll be able to play see-saw with your card to your heart's content. Have you ever had to design a connector system? I have, and you might try it sometime - the constraints are much more subtle than you might suspect... -- Mike Farren farren@well.sf.ca.usa
tron1@tronsbox.UUCP (HIM) (12/24/89)
>>In article <557@sagpd1.UUCP> monty@sagpd1.UUCP (Monty Saine) writes: >> >>The last time I saw the EISA spec (admittedly, only a quick look), I saw >>one, and only one, alignment device - a key which fit into a slot >>approximately >>halfway down the card-edge on an EISA card, and which would, supposedly, >>block a non-EISA card from being inserted fully. ^^^^^^^^^^ I can tell you definately that this is NOT the case. There is at least ONE ket at the boundry where IBM XT 8BIT slots ended, possibly one where the old AT 16 bit slot ends as well. Any others are purely EISA. BUT , be aware that the EISA specs main claim to fame is that the old XT/AT cards will plug in and run without modification. **************************************************************************** "Perfume and leather baby , you and me together baby, what good is living in paradise, if you don't let yourself once or twice." -Tiffany Kenneth J. Jamieson ---- THE BOSS at Xanadu Enterprises Inc. UUCP: tron1@tronsbox.UUCP BEST PATH ---> uunet!tronsbox!tron1 Sysop, Romantic Encounters BBS - (201)759-8450 / (201)759-8568 ****************************************************************************