jbreeden@netcom.UUCP (John Breeden) (01/10/91)
In article <NELSON.91Jan9224628@sun.soe.clarkson.edu> nelson@clutx.clarkson.edu (aka NELSON@CLUTX.BITNET) writes: > >Therefore, you need packet drivers. As far as I know, none of the free >TCP/IP software, nor Novell use NDIS drivers. > Russ is right, BUT FTP's dis_pkt (packet-driver-to-NDIS-driver adapter) will allow SOME packet driver software (KA9Q, NCSA, CUTCP, PC/TCP) to run on an NDIS driver (dis_pkt makes an NDIS MAC driver "look" like a packet driver). (It's available on ftp.vax.com and I think Russ' machine). Netware (BYU) will NOT run on dis_pkt, therefore there is no free software out there that will allow Netware to run on an NDIS driver. So if you want to run Lan Manager (ie: AT&T StarGroup, 3Com 3+ Open, DEC Lan Works etc) AND tcp-ip (using freeware), use a packet driver tcp-ip and FTP's dis_pkt. There are two commercial software packages out there (and could be more) that supply Novell drivers with NDIS interfaces. Hughes Lan Systems' Prolinc (which also includes a WORKING NDIS driver for Vines and a full copy of PC/TCP) and Walker Richer and Quinn's Reflections. If you need to run Lan Manager AND Novell (and/or Vines), these will do it. The down side is they cost $$s (and you don't get source). I thought ODI was a dog :-) (Garfy's friend). -- John Robert Breeden, netcom!jbreeden@apple.com, apple!netcom!jbreeden, ATTMAIL:!jbreeden ------------------------------------------------------------------- "The nice thing about standards is that you have so many to choose from. If you don't like any of them, you just wait for next year's model."
donp@na.excelan.com (don provan) (01/12/91)
(*sigh*) In article <NELSON.91Jan9224628@sun.soe.clarkson.edu> nelson@clutx.clarkson.edu (aka NELSON@CLUTX.BITNET) writes: >NDIS, packet drivers, and ODI all perform the same function. They >serve to hide the difference between network cards, and allow multiple >protocol stacks to access the same card. Technologically, there isn't >much difference between them. I've said this before, but ODI serves to hide the datalink protocol itself from the network protocol. In contrast, packet drivers and NDIS hide the details of the hardware from the network protocol, but the network protocol is still required to create datalink packets appropriate for the media being used. *Technologically*, this makes ODI about as much like packet drivers as it is like the BSD socket interface. On the other hand, i do agree that commercially there's not much difference between them, as the major advantage of all three of these interfaces is the freedom to choose the hardware independently of the network product. This doesn't make ODI better than packet drivers, just different. >ODI drivers are fairly scarce because manufacturers would rather >write dedicated Novell drivers since they're more efficient I'd have to see some references here before i'd believe this. From where i sit, it appears that manufacturers don't have ODI drivers yet because they already *have* dedicated Novell drivers so there's no rush. > (a dedicated driver will always be more >efficient than a general purpose driver). Well, i suppose this is true, but i don't believe the difference in efficiency is necessarily significant. The things that make a driver general purpose just aren't that costly compared to the other aspects of a datalink interface. don provan donp@novell.com
donp@na.excelan.com (don provan) (01/12/91)
In article <NELSON.91Jan9224628@sun.soe.clarkson.edu> nelson@clutx.clarkson.edu (aka NELSON@CLUTX.BITNET) writes: >The ONLY solution to running both Novell and TCP/IP at the same time >is to use the packet drivers. Of course Novell has their own solution >if you have more money than brains.... Imagine that i were a representative of a company which sells products that competed with packet drivers. Imagine that i started a paragraph by claiming that my product was the ONLY solution to the problem our competing products tried to solve. Even if i went on to retract that statement in my next sentence and said, "Packet drivers also provide a solution, if you have more spare time than brains", i'd still be dangerously close to commercialism. Since you give away your packet drivers for free, i guess about the best we can label you with is "name caller". But i do wish you'd keep your posts a little more technical and a little less political. don provan donp@novell.com
nelson@sun.soe.clarkson.edu (Russ Nelson) (01/14/91)
In article <2604@excelan.COM> donp@na.excelan.com (don provan) writes: In article <NELSON.91Jan9224628@sun.soe.clarkson.edu> nelson@clutx.clarkson.edu (aka NELSON@CLUTX.BITNET) writes: >The ONLY solution to running both Novell and TCP/IP at the same time >is to use the packet drivers. Of course Novell has their own solution >if you have more money than brains.... Imagine that i were a representative of a company which sells products that competed with packet drivers. ... Chiding accepted and apology granted. My experience with commercial software is that you pay an awful lot, and if you don't like what you've gotten, you're stuck with it. If you get what you paid for, you're lucky. Perhaps other people are more comfortable spending money on software, and I shouldn't be accusing them of lacking intelligence. -- --russ (nelson@clutx [.bitnet | .clarkson.edu]) FAX 315-268-7600 It's better to get mugged than to live a life of fear -- Freeman Dyson I joined the League for Programming Freedom, and I hope you'll join too.
nelson@sun.soe.clarkson.edu (Russ Nelson) (01/14/91)
In article <2603@excelan.COM> donp@na.excelan.com (don provan) writes: (*sigh*) Well, I knew when I stated such a bald opinion that people would disagree. I stand by that opinion. There *isn't* that much difference. >ODI drivers are fairly scarce because manufacturers would rather >write dedicated Novell drivers since they're more efficient I'd have to see some references here before i'd believe this. From where i sit, it appears that manufacturers don't have ODI drivers yet because they already *have* dedicated Novell drivers so there's no rush. > (a dedicated driver will always be more >efficient than a general purpose driver). Well, i suppose this is true, but i don't believe the difference in efficiency is necessarily significant. The things that make a driver general purpose just aren't that costly compared to the other aspects of a datalink interface. Well certainly if Novell sits down to make a driver for their product it's going to be efficient. I had more in mind the hypothetical case of Novell and Microsoft getting together and writing a driver that would serve both Netware and LAN Manager (if you *can* imagine such a thing happening :). Such a driver would likely please neither party completely, and would not be as efficient as NDIS is for LAN Manager or ODI is for Netware. Which is, of course, part of the reason why we have NDIS and ODI and packet drivers. The other part of the reason is political. -russ -- --russ (nelson@clutx [.bitnet | .clarkson.edu]) FAX 315-268-7600 It's better to get mugged than to live a life of fear -- Freeman Dyson I joined the League for Programming Freedom, and I hope you'll join too.
wittmann@erb1.engr.wisc.edu (art wittmann) (01/15/91)
In article <2604@excelan.COM> donp@novell.com (don provan) writes: >In article <NELSON.91Jan9224628@sun.soe.clarkson.edu> nelson@clutx.clarkson.edu (aka NELSON@CLUTX.BITNET) writes: >>The ONLY solution to running both Novell and TCP/IP at the same time >>is to use the packet drivers. Of course Novell has their own solution >>if you have more money than brains.... > >Imagine that i were a representative of a company which sells products >that competed with packet drivers. Imagine that i started a paragraph >by claiming that my product was the ONLY solution to the problem our >competing products tried to solve. Even if i went on to retract that >statement in my next sentence and said, "Packet drivers also provide a >solution, if you have more spare time than brains", i'd still be >dangerously close to commercialism. > >Since you give away your packet drivers for free, i guess about the >best we can label you with is "name caller". But i do wish you'd keep >your posts a little more technical and a little less political. > I've got to say, I've been trying to get info from novell on ODI for some time now. I've tried various channels, all without success. We're interested in writing software to various specs. We've done some for packet drivers, and 3com has been nice enough to supply us with alot of information on NDIS. I had no trouble at all getting info on either of these two specs. ODI is a completely different story. The only word I've heard is from a friend who works in the industry. According to him, I may as well not even try - unless I feel like barfing up $7500.00. When I hear numbers like that, I tend to agree with Russ' statement, it doesn't look like ODI is a reasonable interface to which to program. Of course I haven't even seen any technical literature, so I can't really comment on the technical merits of ODI. Please correct me if I'm wrong about pricing & let me know where I can get information on ODI for a "reasonable" price. Art =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Art Wittmann Phone: (608) 263-1748 Network Manager Email: wittmann@engr.wisc.edu Computer Aided Engineering Center or: wittmann@cae.wisc.edu University of Wisconsin, Madison