fnf@mcdsun.UUCP (03/23/87)
At various times over the last few months, I have posted details of my problems in trying to get the products from different vendors of expansion hardware products for the Amiga 1000 to work simultaneously on my system. So far, I have tested the following pieces of hardware in various combinations: o Two different Amiga 1000's of late '85 early '86 vintage o CLtd's SCSI controller with Rodime 20Mb hard disk drive o ASDG Inc's Minirack-C with one/two 2Mb memory boards o ASDG Inc's Minirack-D with one/two 2Mb memory boards o Access Associate's Alegra 512K memory board First of all, remember that we are talking about one, or sometimes two, specific units from a given vendor. This is hardly a representative sample to start drawing conclusions about *ALL* hardware from any of the four vendors involved, C-A, CLtd, ASDG, or Access Associates. For all I know, replacing one of the three subsystems with a new one would completely cure all of my problems. Nevertheless, I have been working this problem long enough now, and talked to enough different hardware designers and other users with similar problems, that some reasonable trends and conclusions seem self evident. To recap some of my previous opinions and impressions about the products involved, as they relate to problems with the expansion architecture: o Amiga 1000 --- Expansion product designers have reported that the Amiga 1000 is very noisy internally and some parts used on early production models are run so close to specs that any variation due to more than a single expansion product is enough to turn some working Amiga 1000's into a non-working ones. Indications are that the Amiga 2000 is much better in this respect. o CLtd --- Marginally working hardware (my controller card is full of cuts and jumpers) and the quality of the metal case is about that of an internal company prototype (thin sheet metal, cable runs through the crack between top and bottom of the case). Buffers? What are those? o ASDG --- Highly professionally designed and manufactured hardware. Their boards buffer the signals in and out of the board, their boxes look virtually indestructible, and their new boxes buffer the signals in and out of the box to cut any loading effects to the absolute minimum possible. o Access Associates --- Quality comparable to ASDG. I only got to use the board briefly but I was impressed with its quality, as with ASDG's. Like ASDG, they have taken steps to reduce the loading and noise problems. Note the following: o All of the expansion products worked correctly when used by themselves on either Amiga 1000. One of the Amiga 1000's is totally "stock", the other has had the recommended PAL changes, has had the additional PAL ground modification, and has had the plastic 68000 replaced with a ceramic version. o Since only the CLtd controller passes the bus (a mixed blessing apparently), the only combination of CLtd, ASDG, or AA products that could be tried was CLtd+ASDG and CLtd+AA. With the unmodified Amiga 1000, neither of these configurations would even boot. With the modified Amiga 1000 the system would boot but not run reliably enough to be useful. o The controller + memory configuration was tried with two different ASDG units and one AA unit with identical results, so I don't believe the problem to be in the design of either the ASDG or AA product, or the result of a faulty unit. From these experiences I would recommend that: o If you plan to use a mix-and-match approach to expansion on your Amiga 1000 that you insist on trying the products with *your* machine *before* you lay out any cash. Or at least have an understanding that you can return any product that does not perform to your satisfaction (even if not "defective" by the vendor's standards). o Do not accept "solutions" that require you to hack at the guts of your machine unless you really enjoy this sort of puzzle solving. o If you need both expansion memory and a disk controller for the Amiga 1000, try to get everything from a single vendor. The PAL series from Byte-By-Byte looks like a possibility. ASDG should have their controller out soon also, and the design specs that I have read in their newsletter look impressive. o If you absolutely *must* have a disk controller *now* and at the lowest possible cost, the CLtd solution may be acceptable for you, but remember point numbers 1 and 2 above. The saving grace here is that you can toss the controller and just use the non-CLtd designed SCSI disk box with someone else's controller card (ASDG or C-A's A2000) when they become available. o If you plan to expand your Amiga but do not already have a large investment in existing products, wait for the A2000. On this system, every vendor will be competing on more "even ground", and those that choose to produce high quality products at a fair price will be rewarded rather than penalized by the market prices and finger pointing confusion. Disclaimer: Any flames or opinions expressed within are mine and mine alone. Neither my employer nor my Amiga knows anything about this text, generated by a random number generator hooked to my serial port... :-) Feel free to cross post this message to information services that I may not have access to. -Fred -- = Drug tests; just say *NO*! (Moto just announced new drug testing program) = = Fred Fish Motorola Computer Division, 3013 S 52nd St, Tempe, Az 85282 USA = = seismo!noao!mcdsun!fnf (602) 438-5976 =
page@ulowell.UUCP (03/25/87)
fnf@mcdsun.UUCP (Fred Fish) wrote in article <269@mcdsun.UUCP>: that CLtd's hard disk doesn't work with other vendor's memory boards; or does so in such a poor fashion that you should stay away from it. This is just a note to say I have had _no_ problems with CLtd's disk along with their RAM product. They perform reliably together. If you post Fred's article to other information services, you might want to post this one too. ..Bob -- Bob Page, U of Lowell CS Dept. ulowell!page, page@ulowell.CSNET
perry@sfsup.UUCP (03/26/87)
In article <1152@ulowell.cs.ulowell.edu>, page@ulowell.cs.ulowell.edu (Bob Page) writes: > fnf@mcdsun.UUCP (Fred Fish) wrote in article <269@mcdsun.UUCP>: > that CLtd's hard disk doesn't work with other vendor's memory boards; > or does so in such a poor fashion that you should stay away from it. > > This is just a note to say I have had _no_ problems with CLtd's disk > along with their RAM product. They perform reliably together. Bob, I think you've missed a very important point. One cannot look at this situation and ask: Is the glass half empty or half full. The fact is, the glass *isn't* full. Specifically, saying that *your* C Ltd products work well together is simply the EXPECTED case. Fred reporting that the EXPECTED case is not always acheived carries significantly more weight than your two boards working together. To put it another way, if you had just paid for two expansion products and found they didn't work for (no matter what you or the manu- facturer could do) would it make you feel any better if your next door neighbor had the same hardware and it worked fine? I know that ASDG customers didn't pay their money to take a chance that their hardware would work together. Other full spec Zorro manufactur- ers similarly designed their products to work in *every* case, not just on certain Amiga, but every Amiga. Perry
root@sbcs.UUCP (03/27/87)
> In article <1152@ulowell.cs.ulowell.edu>, page@ulowell.cs.ulowell.edu (Bob Page) writes: > > fnf@mcdsun.UUCP (Fred Fish) wrote in article <269@mcdsun.UUCP>: > > that CLtd's hard disk doesn't work with other vendor's memory boards; > > or does so in such a poor fashion that you should stay away from it. > > > > This is just a note to say I have had _no_ problems with CLtd's disk > > along with their RAM product. They perform reliably together. > > Specifically, saying that *your* C Ltd products work well together > is simply the EXPECTED case. Fred reporting that the EXPECTED case is not > always acheived carries significantly more weight than your two boards > working together. > > Perry What Perry says with respect to board interoperability is, of course, correct; however let's place the blame where it belongs - on Commodore. Face it, the Zorro standard is not a standard in the way the VME Bus, MultiBus II, NuBus, etc are spec'ed. As far as bus specs go, the Zorro "standard" is more of a edge connector pinout and a bracket mechanical than a formal "here is what ya gotta do to be 99.99% sure the widget will work". (Yup, I've designed Amiga boards). As far as I am concerned, the only sane way to buy peripherals for the 1000 is to buy them all from ONE manufacturer. This doesn't guarantee (in all cases) that said products will work with ALL rev Amigas - just that the manufacturers involved can't play "no it's THEIR problem" with you as the helpless consumer! Sure, we've all heard lots and lots of claims that manufacturer X, Y, or Z's stuff never fails, but reality is that no third party manufacturer operating in the Amiga market has sold into a statistically significant portion of the installed base. Rather than listening to all sorts of claims, the average Amiga consumer should concern himself with: 1. Is the company responsive to customer input? 2. Does the company provide easy accessible technical assistance if so required? 3. What guarantees on are placed on the equipment? 4. Is there local support for the equipment - i.e. at this point it is really better to purchase equipment from your local dealer so that you have someone to beat on.. 5. Unless you're an engineering type, don't bother with all the smoke concerning issues such as whether the pass though bus is buffered, board trace lengths, F vs LS loads on the bus, rev A/B pals in the Amiga, motherboard noise problems, current consumption, etc are all engineering posturing. The average consumer is simply unable to evaluate ANY of the above in the context of the real system that is sitting on their desk. All you really want to know is whether it WORKS! In closing, my great hope is that Commodore will get their collective act together (between Germany/Westchester/East Jahunga) and put out a full engineering spec for both the A-500/A-2000 as a SERVICE to their customers. Until then, it is a roll of the dice whether your new widget will work.. Rick Spanbauer SUNY/Stony Brook
hah@omepd.UUCP (03/29/87)
I have been very quiet here so far but I feel I must correct a few erroneous assumptions made by some of the hardware designers. Assumption 1 : I can have as many TTL gate loads as I want on the output lines of the ZORRO BUS. A) The Zorro bus std allows for only one LS gate load per line. Assumption 2 : Passing the buss thru a board is a good idea. A) All boards that pass the buss thru may only have one LS load per line and therefore will work fine by themselves. Adding one or more boards with pass thrus adds additional LS loads because the first board does not buffer the signals before they were passed thru and neither do the remainder of the boards. Assumption 3 : As a designer I dont need to read all the fine details of the ZORRO BUS specs. A) The Zorro spec is also quite clear in stating that the maximum length any output trace to buffer can have. (I'm sorry but I do not have the specs with me now.) My point in this is quite simply to point out that while a single board by itself will work perfectly, as soon as a second board is attached to the first boards pass-thru buss they both violate the Zorro bus specs. If you as a user need to have more than one expansion board on your system then you must use one of the CORRECTLY buffered expansion boxes. NO EXCEPTIONS!! Hans
bakken@tahoma.UUCP (03/30/87)
In article <269@mcdsun.UUCP>, fnf@mcdsun.UUCP (Fred Fish) writes: > o All of the expansion products worked correctly when used > by themselves on either Amiga 1000. One of the Amiga 1000's > is totally "stock", the other has had the recommended PAL > changes, has had the additional PAL ground modification, > and has had the plastic 68000 replaced with a ceramic version. How much does the ceramic 68000 help (Fred or anyone who knows)? > o If you plan to expand your Amiga but do not already have > a large investment in existing products, wait for the A2000. > On this system, every vendor will be competing on more > "even ground", and those that choose to produce high quality > products at a fair price will be rewarded rather than > penalized by the market prices and finger pointing confusion. > Fred, are you saying that the coming of the A2000 will give the A1000 owners a better selection of quality peripherals, or, as it seems to me, suggesting that the A1000 owners should consider getting rid of their A1000s and buy A2000s? The posting was very helpful - thanks (and for your PD efforts). Dave Bakken Boeing Commercial Airplane Company Flight Simulation Lab uw-beaver!ssc-vax!shuksan!tahoma!bakken (206) 237-5890 My views are my own, not my employer's. Don't let them deter you from buying the 747 you've been saving hard for.
daveh@cbmvax.UUCP (03/31/87)
in article <371@sbcs.UUCP>, root@sbcs.UUCP (Root) says: > What Perry says with respect to board interoperability is, of course, > correct; however let's place the blame where it belongs - on Commodore. Face > it, the Zorro standard is not a standard in the way the VME Bus, MultiBus II, > NuBus, etc are spec'ed. As far as bus specs go, the Zorro "standard" is more > of a edge connector pinout and a bracket mechanical than a formal "here is > what ya gotta do to be 99.99% sure the widget will work". (Yup, I've > designed Amiga boards). Whaddaya want! You've got ONE document to follow, the Expansion Specification booklet. It gives you power, loading, and timing data. Much of the timing data does come from the 8MHz 68000 spec, but its still there and it is saying, DO THIS OR ELSE. I've heard lots of stories about boards that don't work, and I've yet to see one that's not in the OR ELSE category. A standard expansion box from CBM (or the built-in slots of an A2000) would help force manufacturers into closer accord with the specs, and it does give them a bit more slop factor (2 "F" series loads per most pins, versus 1). I agree that this book could have been written better, as time permits I'm sure it'll change more of a "here's everything you need, all in one place" sort of book. But this is a well defined bus, not just "build it to plug in here and hope for the best". -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Dave Haynie Commodore Technology // /| ___ __ __ __ {ihnp4|caip|rutgers}!cbmvax!daveh |\ // /_| | / \ / \ / \ Commodore rarely admits to knowing me, \\// / | +--+ | | | | | | much less sharing my personal opinions. \/ / | |___ \__/ \__/ \__/ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
fnf@mcdsun.UUCP (04/01/87)
In article <142@tahoma.ARPA> bakken@tahoma.ARPA (Dave Bakken) writes: >In article <269@mcdsun.UUCP>, fnf@mcdsun.UUCP (Fred Fish) writes: >> changes, has had the additional PAL ground modification, >> and has had the plastic 68000 replaced with a ceramic version. > >How much does the ceramic 68000 help (Fred or anyone who knows)? I just replaced it on the hunch that it might have slightly better electrical characteristics. If it costs more it must be better, right :-) It did seem to help some though. I've heard that if you can find a CMOS version of the 68000 it helps even more. >> o If you plan to expand your Amiga but do not already have >> a large investment in existing products, wait for the A2000. >> On this system, every vendor will be competing on more >> "even ground", and those that choose to produce high quality >> products at a fair price will be rewarded rather than >> penalized by the market prices and finger pointing confusion. > >Fred, are you saying that the coming of the A2000 will give the A1000 >owners a better selection of quality peripherals, or, as it seems to >me, suggesting that the A1000 owners should consider getting rid of their >A1000s and buy A2000s? The difference in price between a cheapo slap-on-the-side expansion product and a quality board in a quality expansion box is significantly greater than the difference between a cheapo board and a quality board. Thus the incentive to by cheapo boards for the A2000 will be greatly reduced. This may help the A1000 owners also, if venders like ASDG can produce their expansion boxes, that allow use of A2000 cards on an A1000, in a timely manner. -Fred -- = Drug tests; just say *NO*! (Moto just announced new drug testing program) = = Fred Fish Motorola Computer Division, 3013 S 52nd St, Tempe, Az 85282 USA = = seismo!noao!mcdsun!fnf (602) 438-5976 =
root@sbcs.UUCP (04/02/87)
> sort of book. But this is a well defined bus, not just "build it to plug > in here and hope for the best". > -- > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > Dave Haynie Commodore Technology // /| ___ __ __ __ Sorry Dave, get a copy of the VME bus spec (for example) if you would like to see a _reasonable_ bus specification. Even with the VME bus defined concisely as it is, some users _still_ experience problems getting boards from different manufacturers to run.. Rick
papa@bacall.UUCP (04/05/87)
Fred Fish writes: > The difference in price between a cheapo slap-on-the-side expansion product > and a quality board in a quality expansion box is significantly greater > than the difference between a cheapo board and a quality board. Thus > the incentive to by cheapo boards for the A2000 will be greatly reduced. > This may help the A1000 owners also, if venders like ASDG can produce their > expansion boxes, that allow use of A2000 cards on an A1000, in a timely > manner. > I just got the Lates AmigaWorld which finally has some interesting articles! Among them one by Eric Grapham, the author of the "juggler". Anyway, in the what's new section new products from CSA are listed. Among them, is an new expansion chassis for the Amiga 1000, called the Turbo Amiga Tower, which attaches to the 86-pin expansion giving the A1000 the capabilities of the A2000. It provides a CPU slot, SEVEN (!!) 100-pin slots, four IBM PC AT slots and a 200 watt power supply. Basically all the features of the A2000 - the features of the A1000. The PC-AT slots are supposedly arranged in the same fashion as on the A2000 so one can use the A2088 card. 5 of the seven slots are autoconfig The other two hold extra 32-bit memory (optional). These slots are 32-bit since CSA is priducing a board with a 68020 and 68881. This board supposedly plugs into both the Turbo AMiga Tower and the A2000. The 68020/68881 board runs at 14 MHz. According to CSA this board outperforms the PC AT by a factor of 40 in the Savage benchmark. A whole list of periperals follows, from Static RAM boards to SCSI adapters, disks of various sizes. Prices. The Turbo Amiga Tower lists for $1095. This could help people evaluate whether to trade in their machines for an A2000 or get the CSA Tower. The 68020/68881 board, $1480. A SCSI controller, $695. A 20M hard disk, $910. Look in the magazine, for the prices of the rest of the peripherals. My question is: is this still vaporware, or is the CSA tower already available? Did anybody see it at any of the recent meeetings and shows. It's the weekend, so I had no time to call CSA and find out. I'll try monday. With the CSA tower and ASDS's 2000-and-1 box there should be good competition, and for the moment I am not yet ready to give away my A1000. As usual, I have no connection with CSA, just interest in their products. -- Marco -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Marco Papa 3175 S. Hoover St., Ste. 275 (213)747-8498 Los Angeles, CA 90007 USC: (213)743-3752 F E L S I N A Now working for ::::::: BIX: papa But in no way :: :: Officially representing ::::::: ...!usc-oberon!bacall!papa S O F T W A R E papa@usc-cse.usc.edu -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
lachac@topaz.UUCP (04/07/87)
In article <2359@bacall.UUCP> papa@bacall.UUCP (Marco Papa) writes: >My question is: is this still vaporware, or is the CSA tower already available? >Did anybody see it at any of the recent meeetings and shows. It's the weekend, >so I had no time to call CSA and find out. I'll try monday. With the CSA >tower and ASDS's 2000-and-1 box there should be good competition, and for >the moment I am not yet ready to give away my A1000. Not from the rumors I have been hearing. ------------------------------------------------------------------------------ Disclaimer: (not to put words in people's mouths, but an attempt to solicit a response...) ------------------------------------------------------------------------------ From what I was told recently, some hardware companies aren't too happy that CBM is "undercutting the third party hardware developer" by offering the current A1000 owners A2000 upgrades. It seems they feel that there may not be a reasonable market share of people willing to buy A1000 to A2000 card cages to warrant going ahead and building them. I would like to know if any of the hardware peoples on the net would care to comment/warn we consumers if this is true. Anyone?? Perry?????? -- "Truth is false and logic lost..." - Neil Peart (who at the time didn't realize he was talking about RU) lachac@topaz.rutgers.edu <--------OR--------> {seismo|ames}!rutgers!topaz!lachac
kim@amdahl.UUCP (04/08/87)
In article <10791@topaz.RUTGERS.EDU>, lachac@topaz.RUTGERS.EDU (Gerard Lachac) writes: > > From what I was told recently, some hardware companies aren't too happy that > CBM is "undercutting the third party hardware developer" by offering the > current A1000 owners A2000 upgrades. It seems they feel that there may not > be a reasonable market share of people willing to buy A1000 to A2000 card > cages to warrant going ahead and building them. On the other hand, this could be very good incentive for 3rd party developers to provide some "value added" features that are not available in the A2000 itself ... like MORE SLOTS. With an expansion box offering more fundamental capability than the A2000, the only reasons I can see to upgrade to the 2000 would be for the VIDEO slot capability it has, and for a reduced overall footprint. /kim -- UUCP: kim@amdahl.amdahl.com or: {sun,decwrl,hplabs,pyramid,ihnp4,seismo,oliveb,cbosgd}!amdahl!kim DDD: 408-746-8462 USPS: Amdahl Corp. M/S 249, 1250 E. Arques Av, Sunnyvale, CA 94086 CIS: 76535,25 [ Any thoughts or opinions which may or may not have been expressed ] [ herein are my own. They are not necessarily those of my employer. ]
lachac@topaz.UUCP (04/09/87)
In article <6133@amdahl.UUCP> kim@amdahl.UUCP (Kim DeVaughn) writes: >With an expansion box offering more fundamental capability than the >A2000, the only reasons I can see to upgrade to the 2000 would be >for the VIDEO slot capability it has, and for a reduced overall footprint. ------------------------- ^ | Not really if you think about the HUGE keyboard and no garage. (I think!) -- "Truth is false and logic lost..." - Neil Peart (who at the time didn't realize he was talking about RU) lachac@topaz.rutgers.edu <--------OR--------> {seismo|ames}!rutgers!topaz!lachac