Wanted: SIMH VMS quick test

Rich Alderson RichA at LivingComputerMuseum.org
Tue Dec 16 13:13:02 CST 2014


From: Al Kossow
Sent: Tuesday, December 16, 2014 9:59 AM

> On 12/16/14 9:39 AM, Chuck Guzis wrote:

>> On 12/16/2014 06:29 AM, Al Kossow wrote:

>>> On 12/15/14 9:04 PM, Chuck Guzis wrote:

>>>> Okay, I've been working with my SCSI tape stuff and I'd like someone
>>>> to check a .TAP image on a VAX VMS SIMH setup.

>>> I looked at it with my tools. There are THREE extra bytes at the end of
>>> the first record, before the 50 00 00 00.

>>> Bob may pad to a short, but I don't think he pads to a long. The images
>>> I create don't pad a record at all, what you see is what you get.

>> http://simh.trailing-edge.com/docs/simh_magtape.pdf is the document
>> that I'm using for this.  I clearly have some extra padding in the
>> records--that's a bug in my own padding routine.

>> The document does say that records will be padded to an even byte
>> boundary--is this incorrect?

> No, if that's the format you want to support. I personally think it is
> incorrect since that wasn't what was written on the tape.

It's a little more complex than that.  Bob was using PDP-11 tape drives as
his model, and they *do* insert an extra null(?) byte at the end of odd-
length records.  I think, but have not verified since my KS systems are
not running right now, that the RH11 on the KS's Unibus may do the same
thing, but for now that's just a guess.

Ken and Mark's tape format does not add a padding byte for KL simulation
by default, but distinguishes several different formats, including one
that is the same as Bob's.

My utilities which I wrote for TOPS-20 interactively inquire whether odd
length records should be padded or not, depending on which emulator I'm
aiming at.

                                                                Rich


Rich Alderson
Vintage Computing Sr. Systems Engineer
Living Computer Museum
2245 1st Avenue S
Seattle, WA 98134

mailto:RichA at LivingComputerMuseum.org

http://www.LivingComputerMuseum.org/


More information about the cctalk mailing list