D0430@PUCC.BITNET.UUCP (06/23/86)
Newsgroups: mod.computers.vax
Subject: dma question
Date: Monday, 23-Jun-86 11:09:57 EDT
Userid: D0430
I received only one reply to my last query so let me try a different tack:
I am interested in any experience people have had with
1) DEC DRU11-C (double buffer DMA card)
2) ABLE QNIVERTER II
3) any generic dma cards other than DR11-W or DRV11-W
specifically interested in q22-bus.
thanks in advance. paul lansky (princeton!winnie!paul or
d0430@pucc on bitnet)
609-452-4241
(seismo, ucbvax, etc, all reach princeton)D0430@PUCC.BITNET.UUCP (06/26/86)
Newsgroups: mod.computers.vax Subject: Re: VME-bus or Multibus-II interfaces for Uni/Q-bus? Date: Wednesday, 25-Jun-86 23:49:53 EDT Userid: D0430 In article <[G.BBN.COM]24-Jun-86.16:28:11.WOLF>, WOLF@BBNG (Jerry Wolf) writes: >We're looking to develop some signal processing software for an >array processor such as the Sky Warrior or Sky Vortex, and it would >be convenient to use one of our VAXen or MicroVAXen as the development >system. Such AP's ......... Sky makes a version of the Warrior for the Microvax. There are also others such as Mercury's Zip which will run on the Q-bus. Sky also has an inexpensive board that runs signal processing on the Microvax using TMS32010 chip ($2-3k) I think the Warrior is in the $15k ballpark.
D0430@PUCC.BITNET (Paul Lansky) (06/27/86)
Newsgroups: mod.computers.vax net.unix-wizards net.arch Subject: Q-bus/uvaxII question Date: Thursday, 26-Jun-86 19:51:58 EDT Userid: D0430 This a tricky question: on the MicrovaxII (or Qbus), under Ultrix, if you have a card which has Q-bus memory, but *not* PMI memory (e.g. it doesnt connect to the cpu by the PMI, but has its memory on the Q-bus) ---- how do you copy from PMI memory to Q-bus memory? I understand how ubasetup() works vis-a-vis PTE's and mapping DMA transfers etc, but I don't understand this issue at all. please reply here or to d0430@pucc , or ...princeton!winnie!paul. thanx Paul Lansky
D0430@PUCC.BITNET.UUCP (07/05/86)
Newsgroups: mod.computers.vax net.unix-wizards net.unix Subject: MDB Unibus-QBus convertor Date: Friday, 4-Jul-86 23:39:55 EDT Userid: D0430 I'm interested in any experience people have with the MDB Unibus<->Qbus converter, perhaps as opposed to the Able Qniverter II. thanks Paul Lansky Music Department Princeton University ......princeton!winnie!paul or .....princeton!winnie!paul or D0430@PUCC
D0430@PUCC.BITNET (Paul Lansky) (08/09/86)
Newsgroups: mod.computers.vax Subject: tk50 Date: Saturday, 9-Aug-86 12:11:14 EDT Userid: D0430 To people who have been reporting problems with tk50 tapes, and drives shredding each other, I believe(?!?) a small preventative measure will help out. Before you insert a cartridge, simply swing the flap of the cartridge opened by depressing the tab at the bottom of the back side, and check to see if the little triangular hook is showing there at the end of the tape. According to one experienced serviceman, the problems arise when the hook is not positioned properly to start with. I'm not utterly sure about this but it seems reasonable, and I would appreciate hearing other views and experiences.
D0430@PUCC.BITNET (Paul Lansky) (09/14/86)
Newsgroups: mod.computers.vax net.unix-wizards
Subject: Mercury Array Processor
Date: Sunday, 14-Sep-86 10:45:35 EDT
Userid: D0430
Does anyone out there have any experience with the Mercury Zip Array Processor
on a uvaxII? Thanks.
Paul Lansky
seismo .. etc. ...... princeton!winnie!paul
d0430@puccted@blia.UUCP (Ted Marshall) (11/26/86)
Subject: Re: Re: Ethernet promiscuious mode (VAX/VMS) Newsgroups: mod.computers.vax Summary: small correction References: <8611241538.AA10222@ucbvax.Berkeley.EDU> In article <8611241538.AA10222@ucbvax.Berkeley.EDU>, LEICHTER-JERRY@YALE.ARPA writes: > WARNING: Be aware of one restriction, which may not be obvious when you first > read the documentation: An Ethernet interface in promiscuous mode cannot be > shared. In particular, you will not be able to run DECnet through the device > while you are using it in this mode. This is not quite true in my experience. There can only be one port on a DxxxA set to promiscuious mode at any time. "Normal" ports can co-exist with the promiscuious port. However now, instead of the hardware doing incoming address filtering, the hardware must give EVERY message on the ethernet to the driver which then does the address filtering for the normal ports. The upshot of this is that if you have a slow VAX or ethernet interface (i.e. the DEC DEQNA) and a busy ethernet, the driver will miss so many messages that most protocols will fail. I wrote an ethernet traffic monitor using promiscuious mode. While running it on our 11/750, DECnet worked ok most of the time. If I tried it on our u-VAX II, all of the DECnet links would break every time.