Faucon2001 wrote:I have also noticed that if I don't boost Videl resolutions (standard 640x480x4b instead of 800x592x8b with VI2), SCSI transfers under TOS + CT60 seems to be stable. I didn't have time to do many tests nor to test this with Magic to confirm this observation, so I will stay with "seems".
If I go for a clock patch, should I consider Exxos one or just the simple resistor one explained by Rodolphe?
Faucon2001 wrote:Yes, I am using HDDRIVER 10.3, that's the first thing I did this weekend.
I did some more tests tonight.
In 030 mode, under TOS or Magic, with or without VI2 up to 800x592x8b, SCSI transfer to or from IDE is working OK.
In 060 mode :
TOS without VI2 : SCSI transfer is OK
TOS with VI2 in 800x592x1 : SCSI transfer is OK
TOS with VI2 in 800x592x8 : SCSI transfer crashs
MAGIC with or without VI2, 1 bit to 8 bits video mode : SCSI transfer crashes all the time
So on one hand under TOS in 060 mode it really looks like a sdma clock issue (bus saturation?), on the other end I wonder why Magic is crashing in every case. Is Magic more demanding on the bus or tighter on bus timing, or is there something funky with Magic 060 patch?
Fortunately I have bought an EXXOS clock patch just in case, so I will install it this weekend. I am not very comfortable with scrapping track on my Falcon.
Faucon2001 wrote:- HD Driver 10.3 may still have an issue. I need to ask to Uwe.
checksum wrote:As i said in my previous post, if you don't have any issue in 030 mode, then i think it's a software issue.
checksum wrote:Yes you should.Make a try with HDDRIVER and ST-RAM (configure hddriver) and disable 060 caches (use general6.cpx from ct60 package). As i said in my previous post, if you don't have any issue in 030 mode, then i think it's a software issue. Uwe can't test all configurations and this tests (alloc only STRAM, caches off) will help a lot if this is a driver problem.
Users browsing this forum: No registered users and 4 guests