NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/sys/amiga/hardware-part1/8138

Internet Message Format | 1996-08-05 | 5.1 KB

Path: news.mel.aone.net.au!usenet
From: [e-mail] (Mark P Allen)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: VLAB Motion/MovieShop "Data Transfer Phase Error"?
Date: 22 Mar 1996 04:47:57 GMT
Message-ID: <[e-mail]>
References: <[e-mail]> <[e-mail]>
NNTP-Posting-Host: cnsta02.iig.com.au
X-Newsreader: THOR 2.22 (Amiga;TCP/IP) *UNREGISTERED*


On 21-Mar-96 17:30:07, Phil Wright wrote:

>>1) Does anyone know what aspect of the data transfer phase produces the
>>"Data Transfer Phase Error" of MovieShop's vmotion.library?

 PW> Have you checked the jumpers on the WarpEngine to make sure it's
 PW> supplying Terminatio Power to its SCSI bus?  I seem to recall they come
 PW> without the jumper installed.

Yes, and left it at the default. As far as I can see, older drives can
only supply their own term power, and newer drives come configured to do
the same. All my drives are configured to supply their own as far as I
have been able to determine, though I need to confirm the ATLAS jumper
settings.

>>3) Does anyone know the orientation of the SCSI terminating resistors on the
>>A3000T motherboard?

 PW> My A3000 manual has a motherboard diagram and you can see on it where
 PW> pin one is for each term pack socket.  Perhaps the one for your tower
 PW> will have one as well.

Yes, it does, however the physical orientation is not shown in the
diagram. I was not able to determine which was the 5v power line on the
motherboard itself, though I will check next time by using a volt meter
(thanks Michael).

 PW> Not that this has anything to do with your problem since you only have
 PW> the CDROM drive on that bus.  Proper termination is always good but I
 PW> doubt it's a factor in your case.  (As far as the native A3000T bus
 PW> goes.)

That's one reason it's the only device on that bus (see below)

>>Due to the problems mentioned above, I wanted to fit resistors to see if it
>>helped in any way (I have 3 spare 221/331 8-pin resistors, while the
>>schematics ask for 220/330 resistors. Should these work?).

 PW> They should be fine.

Thanks

>>I also have 3 x SCSI-2 (2 are FAST) HDs and a SCSI-2 Exabyte 4200c DSS-1
>>(DAT) streamer attached internally to the warpdrive device, configured
>>thus:

 PW> I'd recommend putting every drive that isn't used for your VLAB video and
 PW> audio files on the native controller and leave the WarpEngine SCSI bus
 PW> for the performance critical stuff.  Have you tried disconnecting the
 PW> tape drive to see if there is a difference?

I tried to correct the A3000T SCSI termination by placing the Quantum
LPS525S system drive (terminated) on it's internal bus (ie.,
ext/CDROM/term -> motherboard/noterm -> int/LPS525/term). Immediate and
frequent lockups resulted. Frustrating to say the least. Needless to say I
immediately returned the LPS525 to the WARP Engine.

>>While I know the warpdrive.device will scan it's SCSI bus from ID#0 to
>>ID#6, I didn't think this arrangement would produce errors, as long as the
>>physical chain is correctly terminated.

 PW> The arrangement of the SCSI ID'S in relation to their position on the bus
 PW> is irrelevant.

Thanks

>>Of course, the parallel existence of the A3000T's SCSI bus, with one
>>external, and no internal devices attached, and with no termination on
>>the motherboard throws confusion into the equation (as far I am concerned
>>anyway).

 PW> Forget about it.  Unless you are actually accessing the native SCSI bus
 PW> at the same time as VLAB is using the WarpEngine SCSI bus, one is not
 PW> going to affect the other.

And why I tend to suspect problems may get worse rather than better if I
try to mix the devices and increase the general transaction levels between
the two SCSI busses. Will experiment further though.

>>Does this configuration offer any obvious explanation of problems being
>>experienced (occasional SCSI READ error with Exabyte 4200c,

 PW> Sounds like a possible termination problem, as in no Term Power perhaps,
 PW> or one of the drives in that chain isn't happy with one or more of the
 PW> others. Take off all the drives but the one you need to playback your
 PW> video and see what happens then.  If that works fine, add back in the
 PW> drives one at a time until you see a problem again.

OK. After confirming whether I do have the ATLAS correctly unterminated
and termpowered, I will try this.

 PW> In regards to your Exabyte, have you tried www.exabyte.com?

 PW> They have the SCSI id jumper settings and instructions for installing the
 PW> term packs online.

No, that's great. Thanks.

Thanks very much for your input.

BTW, while it doesn't affect the SCSI READ error on the Exabyte as far as
I know, changing the video and audio partitions to separate drives seems
to have markedly improved MovieShop's "data tansfer phase error". Ie., I
played back a 12 minute video several times last night, and though the
video glitch problem (caused by adding extra audio) is still there, I
didn't see one transfer error during playback. Testing continues...

Regards,

         Mark

---
   [e-mail]     -    Ph:[phone]     -         Cairns
   Wishful Vision     -    Fx:[phone]     -      Australia
---