All this SCSI stuff makes me realize how fortunate I've been with my system. It has all worked perfectly from the very beginning. I guess It was "beginner's luck" as reading all the problems some of you are having makes me realize how fortunate I really have been. Of course, I don't have a lot of peripherals, (scanner, printer, Zip drive & such) but they are all "daisy chained" together with a "terminator" & have worked perfectly. I was very scrupulous to make sure that they all had the right "identification number", etc. & that the terminator was hooked up correctly to the right device. Ginny
> From: Jeff Walther <[EMAIL PROTECTED]> > Reply-To: "PCI PowerMacs" <[email protected]> > Date: Thu, 19 May 2005 13:48:52 -0500 > To: "PCI PowerMacs" <[email protected]> > Subject: Re: [PCI] SCSI > >> Date: Thu, 19 May 2005 10:24:29 +1100 >> From: David Elmo <[EMAIL PROTECTED]> >> >>> From: [EMAIL PROTECTED] >> >>> SCSI is easy. >> >> If it was so easy how come it gives us all such headaches? It is not so much >> the basics that is in sensible dispute (each device must have a unique ID, >> the ends of the chain must be terminated) but in the implementation on >> particular chains and devices and the peculiarities of the different types >> of SCSI and the different types of termination. BTW, let me renounce any >> faith in the possibility of a program to answer how to go about anything to >> do with SCSI. After the basics, all we can do is collect strategic ploys and >> facts and muddle along... like most chess players, even quite good ones... >> >> Anyway, I noticed that on one of my CD SCSIs there was a pin pair marked >> "term power" and another was "termination" Most of my other CD SCSIs had >> only "Term power" marked on one pin pair, no mention of "termination" on any >> other pin pair. What gives here? > > My experience with SCSI is that if one configures it properly (i.e., > according to the "rules" for SCSI) it works properly. To give that > some context, I've been using SCSI systems since about '92 starting > with a Mac Plus. At one point I was administering about 50 machines > (PCs) at work, all with simple SCSI cards so that they could > interface with external tape and CDROM drives (back when CDROM drives > were several hundred dollars and sharing them made economical sense). > I have not gone beyond LVD SCSI (80 MB/s) but I've configured RAIDs > and the like. At one point I had a four drive RAID in an 8100 clone > spread across the two built-in SCSI busses and a JackHammer card (it > really was faster that way). > > All of the "SCSI Voodoo" that I have seen has been from folks who had > an improperly configured chain, which, by some miracle, worked, and > then stopped working at some point. The fact that it worked at first > convinces them that it was configured properly, so they then conclude > that their "properly" configured chain stopped working for no good > reason. And they declare "SCSI Voodoo". > > However, I would not agree that SCSI is easy. It is a bit complex. > By complex, I mean that it has several rules and the applications of > some of the basic rules are not always as obvious as one might wish. > > For example, the term power that you mention. In order for SCSI > termination to work properly, termination power must be present. In > general, the SCSI host (controller) supplies proper termination > power. But, in some cases inferior cables fail to pass that > termination power on to the next device and so it never reaches the > end of the chain where it is needed. Also, if more than one device > is supplying termination power to the chain, it can cause problems, > perhaps contention resulting from differences in the level settings > of the voltage regulators involved. > > And IIRC, the Mac Plus is not terminated internally and does not > supply term power. So the first SCSI equipped Mac from Apple already > required screwy configuration, needing the first SCSI device attached > to have a short cable and a pass-through terminator and term power > enabled. > > And then there's the fact that folks often enable termination power > thinking that it is termination, and so have too much of the one and > not enough of the other. Your CDROM drives are likely Apple models > which have no provision to supply termination. On machines shipped > with those drives there was a termination module on the end of the > SCSI cable itself. I did once get an actual telephone help person > on the line who was familiar with those drives (4X or 8X IIRC) but it > was more than six years ago and I can't remember if he told me that > there was no termination available on those drives, or if there was > an unlabeled jumper on the jumper bank that was actually termination. > > Another tangle of yarn is adapted drives. Adapting between 50 pin, > 68 pin and 80 pin drives causes termination issues because it is > possible to leave the top 18 wires swinging in the breeze with no > termination, or terminate them twice where only once was expected. > Choosing the correct adapter and making the proper termination choice > is not a simple matter with adapted drives and the requirements > change depending where in the SCSI chain you place the adapted drive. > My rule of thumb with adapted drives is that if you have to ask, then > you don't understand SCSI well enough to be happy with adapted > drives. They're a crap shoot (leading to SCSI voodoo) for the less > experienced. > > Finally, SCSI chain length is confusing. The length of the chain > supported varies with the SCSI implementation, e.g. original SCSI, > SCSI-2, Ultra SCSI (AKA SCSI-3), LVD, etc. The first problem is > identifying which flavor of SCSI you're actually using. The second > problem is remembering which chain length goes with the flavor that > you are using. The third problem is having cables of sufficient > quality that that maximum length will actually work. > > So, while my experience has shown me that SCSI works as expected and > according to its rules, I would not say that SCSI is easy. None of > the rules are difficult to understand, but there are a number of them > beyond the basics and some sublety to the ones that are well known. > > Jeff Walther > > -- > PCI-PowerMacs is sponsored by <http://lowendmac.com/> and... > > Small Dog Electronics http://www.smalldog.com | Refurbished Drives | > -- Sonnet & PowerLogix Upgrades - start at $169 | & CDRWs on Sale! | > > Support Low End Mac <http://lowendmac.com/lists/support.html> > > PCI-PowerMacs list info: <http://lowendmac.com/lists/pci-powermacs.shtml> > --> AOL users, remove "mailto:"; > Send list messages to: <mailto:[email protected]> > To unsubscribe, email: <mailto:[EMAIL PROTECTED]> > For digest mode, email: <mailto:[EMAIL PROTECTED]> > Subscription questions: <mailto:[EMAIL PROTECTED]> > Archive:<http://www.mail-archive.com/pci-powermacs%40mail.maclaunch.com/> > > iPod Accessories for Less > at 1-800-iPOD.COM > Fast Delivery, Low Price, Good Deal > www.1800ipod.com -- PCI-PowerMacs is sponsored by <http://lowendmac.com/> and... Small Dog Electronics http://www.smalldog.com | Refurbished Drives | -- Sonnet & PowerLogix Upgrades - start at $169 | & CDRWs on Sale! | Support Low End Mac <http://lowendmac.com/lists/support.html> PCI-PowerMacs list info: <http://lowendmac.com/lists/pci-powermacs.shtml> --> AOL users, remove "mailto:"; Send list messages to: <mailto:[email protected]> To unsubscribe, email: <mailto:[EMAIL PROTECTED]> For digest mode, email: <mailto:[EMAIL PROTECTED]> Subscription questions: <mailto:[EMAIL PROTECTED]> Archive:<http://www.mail-archive.com/pci-powermacs%40mail.maclaunch.com/> iPod Accessories for Less at 1-800-iPOD.COM Fast Delivery, Low Price, Good Deal www.1800ipod.com
