• Odpowiedz

@ak47, post #1

• Odpowiedz

@Marcin74, post #2

• Odpowiedz

@ak47, post #3

• Odpowiedz

@ak47, post #3

• Odpowiedz

@ak47, post #1

• Odpowiedz

@ted, post #6

• Odpowiedz

@MariuszK, post #5

• Odpowiedz

@ak47, post #8

• Odpowiedz

@ak47, post #8

• Odpowiedz

@Adinfo, post #10

• Odpowiedz

@ak47, post #8

• Odpowiedz

@baderman, post #11

• Odpowiedz

@flops, post #13

• Odpowiedz

@ak47, post #8

• Odpowiedz

@Marcin74, post #15

• Odpowiedz

@alt_, post #17

• Odpowiedz

@ak47, post #18

• Odpowiedz

@MariuszK, post #19

• Odpowiedz

@ak47, post #20

• Odpowiedz

@MariuszK, post #21

• Odpowiedz

@ak47, post #22

• Odpowiedz

@ak47, post #22

Rozumiem że jak na głównej stronie nie ma, to oznacza że nie ma nigdzie?
Plik romstartup.txt w paczce z diagromem...

The DiagRom startup process explained: (Work in progress on this)

COLORCODES AT END OF TEXT
=========================

IF Serialoutput is strange. you MIGHT have to check paula and affected ICs disturbing serialoutput.



First Set powerled to HIGH. (so user should notice on the powerled that ROM works as all Amigas (except very early) start with led HALF bright)
(when telling something is printed at this strage, it is via the SERIALPORT, not the screen)

Prints Versiondata on Serialport.	

but FIRST Prints:
<number of binary chars>
"Garbage before this text was binary numbers: 1, 2, 4, 8, 16, 32, 64, 128, 240, 15, 170 and 85
To help you find biterrors to paula. Now starting normal startuptext etc"



Prints "- Parallel Code $ff - Start of ROM, CPU Seems somewhat alive" on Serialport, also set parallelport to $ff

Sets PRIMITIVE Exceptionhandles, will print data on serialport at an exception and try to flash screen with colorcodes.

Checks status of mouseports. (will be moved in startup later I guess)

Sets up IRQ and DMA registes (printing status on serialport, DONE is printed AFTER the mode is set)

Set Screencolor to Grey, do a loopbacktest, change greycolor while doing this.
(this is supposed to turn OFF serialport of a loopbackadapter is attached, this however doesn't work at this stage)

Prints "- Parallel Code $fe - Test UDS/LDS line" on serial and start test of tests to check if cpu is able to do proper even/odd writes.

Prints "- Parallel Code $fd - Start of chipmemdetection" on serial and set parallelport, start to detect chipmem  Screen flickers in GREEN/BLACK while doing this.

Also prints status on serialport while detecting, if failed data written data and read data is printed., Screen flashes in GREEN/RED if failure.
IF a working memoryblock is found, failure will be taken as end of chipmem.
While detecting current address checked is printed. also number of working 32KB blocks found. Powerled is also flashing.
After detection is complete, it shows start and end address of chipmem.

Prints "- Parallel Code $fc - Start of motherboard fastmemdetection" on serial and start to detect fastmem

Screen is set to different values of blue while this is happening. changing on every different block being tested
IF left mousebuttin is pressed at this stage, memorydetection will be canceled.

Prints "- Parallel Code $fb - Memorydetection done" on Serial

Prints "- Parallel Code $fa - Starting to use detected memory" on Serial.  Memory is now getting used, crash here means you have issues using memory.

	if no chipmem detected and if fastmem is found fastmemis used, unable to put data on screen,
	instead screen will flash when chars is printed on screen.  IF LMB is pressed during poweron, Fastmode will be forced.
	(if no fastmem avaible, chipmem is used but screen is not turned on)

	If no memory is found atall, a bright green/dark green screen will be flashing on screen and powerled flashing telling user no memory is avaible.
	also a message will be sent to serialport.

	END of allocated block will be used. if the RIGHT mousebutton was pressed during poweron, the beginning will be used instead.

Prints detected memory in serialport, and Basememoryaddress.  Basememory is the address where the ROM uses as workspace.

Prints "- Parallel Code $f9 - Detected memory in use, we now have a stack etc" on serial. if you come so far. your machne can handle memory.

Setting up screen, printing data on serialport.  "Done" is printed AFTER the value is set, so if your machine do not print "Done" after:
	"Set Start of copper (COP1LCH $dff080): Done"  then your machine crashed while setting that register.

Sets up more advanced exceptionhandlingcode, will be able to print data on screen etc.

Printing "- Parallel Code $f8 - Starting up screen, text echoed to serialport" to serialport. now stuff will be printed on screen AND serialport
	Meaning that you will have a duplicate of versionstring etc. 

(FROM This point, Prints will mean that data is printed ON Screen, and only on serialport if serialport is enabled)

Testing if we have a working raster (by simply reading beamposition, wait for a long time and read it again, if it is different the system guess we have a working raster. this will make timings better.

Prompt user to press a key via serialport or mouse or press RMB. if not, serialoutput will be disabled from this point.



Prints Menu and starts rom.


-----------------------------

COLORCODES:

NOTE: Sometimes hardwareproblems might do that colorcodes are not visible. text is also dumped on serialport.


Green/Black		Detecting Chipmem
Green/Red		Detected Chipmemerrors
Green/Dark Green	No Memory detected
Red/White		CPU Bus Error
Red/Purple		Address Error
Red/Cyan		Illegal Instruction
Red/Yellow		Division By Zero
Red/Blue		Chk Instruction
Black/White		TrapV Instruction
Black/Purple		Privilege violation
Black/Cyan		Trace
Black/Yellow		Unimplemented Intruction


No Chipmem, Fastmem detected:
Flashing bars on screen when chars is printed. Text output on Serialport 9600BPS
Can also be forced by holding left mousebutton during poweron, if no fastmem, chipmem is used but no screen is turned on.

• Odpowiedz

@MariuszK, post #24

• Odpowiedz

@Jaro2k , post #23

• Odpowiedz

@ak47, post #26

• Odpowiedz

@lopezpb, post #27

• Odpowiedz

@ak47, post #28

Uruchomiłem wszystko.
DiagRom odpalił.





Pisałem do Johna.
Twierdzi, że RAM jest ok i wg niego uszkodzoną jest Alice.
Napisałem, że jest już w podstawce,wszystko jest sprawdzone.
John stwierdził ,że takie zachowanie może być spowodowane bardzo bardzo słabym sygnałem V i Hsync

Ostatnia aktualizacja: 18.05.2019 00:41:19 przez ak47

• Odpowiedz

@ak47, post #29

• Odpowiedz