Overview Features Coding ApolloOS Performance Forum Downloads Products Order Contact

Welcome to the Apollo Forum

This forum is for people interested in the APOLLO CPU.
Please read the forum usage manual.
Please visit our Apollo-Discord Server for support.



All TopicsNewsPerformanceGamesDemosApolloVampireAROSWorkbenchATARIReleases
Running Games and Apps.

WHDLoad Games That Are Not Working In GOLD 2.7

Sean Sk

Posts 488
11 Mar 2018 06:31


I had posted in the IRC channel already for one of the games mentioned here. But it does get busy in there sometimes and for posterity I would mention it here.

The following games worked in GOLD 2.5 and also on a vanilla non-Vampired Amiga but not in GOLD 2.7. All games run on WHDLoad 18.4.5889 in Workbench 3.1. Latest slaves used. All games were ran with Cache's disabled. I'll also include dumps just in case.

Interphase:

There are two versions of this game, the retail version and the Amiga Format magazine version. This relates to the retail version. The Amiga Format version crashes two minutes in actually playing a level, but this is not related to the Vampire and should be disregarded.

The retail version will sometimes show 1 of two different errors. The dump refers to error message 1.

Error message 1: Exception "CHK,CHK2 Instruction" ($2018) at $CCCCCE4 occured.

Error message 2: Exception "Line 1111 Emulator" ($2C) at $CCCCCCE occured.

************************* 08-Mar-18 12:21:02 ************************18.4.5889***
Slave='Interphase.slave' (3152 bytes)
ShadowMem  832E400 -  8386400 ( 360448) AbsolutMem    58000 -    80000 ( 163840)
Resload    F8EEAB0 -  F8F674C (  31900) at F8EEAB0  GL=$F8F6858
Slave      F8FE738 -  F8FF364 (  3116) at F8FE738  BaseMemSize=$80000
attn=200F(80) fc=-1 kn=0 cs=34F3 rw=1 zpt=0 ep=0 ei=0
setcpu=0(SNCS,ENCS,BNCS)

Exception "CHK,CHK2 Instruction" ($2018) at $CCCCCE4 occured.

$0cccccac move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccb0 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccb4 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccb8 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccbc move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccc0 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccc4 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccc8 move.l      ($2a2a,a2),d5                  ;$0007664e
$0ccccccc move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccd0 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccd4 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccd8 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccdc move.l      ($2a2a,a2),d5                  ;$0007664e
$0ccccce0 move.l      ($2a2a,a2),d5                  ;$0007664e
›1m$0ccccce4 move.l      ($2a2a,a2),d5                  ;$0007664e
›22m$0ccccce8 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccec move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccf0 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccf4 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccf8 move.l      ($2a2a,a2),d5                  ;$0007664e
$0cccccfc move.l      ($2a2a,a2),d5                  ;$0007664e
$0ccccd00 move.l      ($2a2a,a2),d5                  ;$0007664e
$0ccccd04 move.l      ($2a2a,a2),d5                  ;$0007664e
$0ccccd08 move.l      ($2a2a,a2),d5                  ;$0007664e
$0ccccd0c move.l      ($2a2a,a2),d5                  ;$0007664e
$0ccccd10 move.l      ($2a2a,a2),d5                  ;$0007664e
$0ccccd14 move.l      ($2a2a,a2),d5                  ;$0007664e
$0ccccd18 move.l      ($2a2a,a2),d5                  ;$0007664e
$0ccccd1c move.l      ($2a2a,a2),d5                  ;$0007664e

exception stackframe:
$000685F8 20010CCC CCE42018 0CCCCCE2
regular stack:
$00068604 CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC
$00068624 CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC
$00068644 CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC
$00068664 CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC
$00068684 CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC CCCCCCCC

  ----0---- ----1---- ----2---- ----3---- ----4---- ----5---- ----6---- ----7----
Dx    7FFF        0        0  FFFF07FF  FFFFF800  FFFB0000        0  656FFFF
Ax      622    73BFC    73C24      1938      4D6A        0    73D00

                  TTSM III  XNZVC
›1mPC= CCCCCE4›22m  SR %0010000000000001  USP=7FC00  ›1mISP=685F8  ›22mMSP=0
VBR=F8F7858  SFC=5  DFC=5  CACR=0  PCR=4400001  BUSCR=0
TC=0  URP=0  SRP=0  DTT0=0  DTT1=0  ITT0=0  ITT1=0

          MEDRAAAABVCPSDT          YZ  PMRCBSDAAAA
intena=%›1m0100000000101000›22m dmacon=%›1m0010011111010000›22m adkcon=$1500
intreq=%›1m0000000000000000›22m
vposr....A000  vhposr...02A7  joy0dat..0000  joy1dat..0000  clxdat...8001 
pot0dat..0306  pot1dat..2B2B  potinp...5500  serdatr..3800  dskbytr..8000 
deniseid.FFFC  hhposr...FFB8 
    ciaa:    PI ROPS    ciab:    PI ROPS
  cra=›1m00000000›22m  ta=21FF<21FF    cra=›1m00000001›22m  ta=  29<  36
      AII ROPS      AII ROPS
  crb=›1m00000000›22m  tb=21FF<21FF    crb=›1m01000001›22m  tb=EC19<FFFF
      10RTWCLO  parallel    DRCCDSOY      M3210HDS
  pra=›1m11111100›22m  prb=›1m11111111›22m      pra=›1m11111111›22m  prb=›1m11111111›22m
        ddra=›1m00000011›22m ddrb=›1m00000000›22m    ddra=›1m11000000›22m ddrb=›1m11111111›22m
        event=000018  icr=00  sdr=00    event=001D59  icr=01  sdr=00

The Ultimate Ride:

This works fine until you start a race. Then the screen will go black and do nothing until you press the fire button which will then exit with the following error message:

Exception "Line 1111 Emulator" ($2C) at $4 occured.

************************* 08-Mar-18 12:24:38 ************************18.4.5889***
Slave='UltimateRide.Slave' (1560 bytes)
ShadowMem  832E3F8 -  83883F8 ( 368640) AbsolutMem    5A000 -    80000 ( 155648)
Resload    F8EF720 -  F8F73BC (  31900) at F8EF720  GL=$F8F74C8
Slave      F8FF3A8 -  F8FF99C (  1524) at F8FF3A8  BaseMemSize=$80000
attn=200F(80) fc=-1 kn=0 cs=34F3 rw=1 zpt=0 ep=0 ei=0
setcpu=33D(DC,IC,SCB,ECB,BNC)

Exception "Line 1111 Emulator" ($2C) at $4 occured.

$ffffffcc move.l      ($2a2a,a2),d5                  ;$000168db
$ffffffd0 move.l      ($2a2a,a2),d5                  ;$000168db
$ffffffd4 move.l      ($2a2a,a2),d5                  ;$000168db
$ffffffd8 move.l      ($2a2a,a2),d5                  ;$000168db
$ffffffdc move.l      ($2a2a,a2),d5                  ;$000168db
$ffffffe0 move.l      ($2a2a,a2),d5                  ;$000168db
$ffffffe4 move.l      ($2a2a,a2),d5                  ;$000168db
$ffffffe8 move.l      ($2a2a,a2),d5                  ;$000168db
$ffffffec move.l      ($2a2a,a2),d5                  ;$000168db
$fffffff0 move.l      ($2a2a,a2),d5                  ;$000168db
$fffffff4 move.l      ($2a2a,a2),d5                  ;$000168db
$fffffff8 move.l      ($2a2a,a2),d5                  ;$000168db
$fffffffc move.l      ($2a2a,a2),d5                  ;$000168db
$00000000 ori.b        #0,d0
›1m$00000004 line-f      $f000
›22m$00000006 ori.b        #0,d1
$0000000a ori.b        #0,d0
$0000000e dc.w        $19fe
$00000010 ori.b        #??$fe,d0
$00000014 ori.b        #??$b6,d0
$00000018 ori.b        #0,d0
$0000001c ori.b        #0,d0
$00000020 ori.b        #0,d0
$00000024 ori.b        #0,d0
$00000028 ori.b        #0,d0
$0000002c ori.b        #0,d0
$00000030 ori.b        #0,d0
$00000034 ori.b        #0,d0
$00000038 ori.b        #0,d0

exception stackframe:
$0006BCFC 20040000 0004002C
regular stack:
$0006BD04 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
$0006BD24 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
$0006BD44 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
$0006BD64 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
$0006BD84 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

  ----0---- ----1---- ----2---- ----3---- ----4---- ----5---- ----6---- ----7----
Dx        0  FFFFFFFF        FF        B6        0      2BC    3000D      FFFF
Ax    63940    639F6    13EB1    634EE    63FFE    DFF0A0    DFF000

                  TTSM III  XNZVC
›1mPC=      4›22m  SR %0010000000000100  USP=7FC00  ›1mISP=6BCFC  ›22mMSP=0
VBR=F8F84C8  SFC=5  DFC=5  CACR=0  PCR=4400001  BUSCR=0
TC=0  URP=0  SRP=0  DTT0=0  DTT1=0  ITT0=0  ITT1=0

          MEDRAAAABVCPSDT          YZ  PMRCBSDAAAA
intena=%›1m0110000111101010›22m dmacon=%›1m0000011111000001›22m adkcon=$1100
intreq=%›1m0000010000000000›22m
vposr....A000  vhposr...02AE  joy0dat..0404  joy1dat..0804  clxdat...8023 
pot0dat..0306  pot1dat..3434  potinp...0500  serdatr..3800  dskbytr..8000 
deniseid.FFFC  hhposr...FFBE 
    ciaa:    PI ROPS    ciab:    PI ROPS
  cra=›1m00001000›22m  ta=8BD8<8BD8    cra=›1m00001000›22m  ta=50BC<50BC
      AII ROPS      AII ROPS
  crb=›1m00000000›22m  tb=21FF<21FF    crb=›1m00000000›22m  tb=FFFF<FFFF
      10RTWCLO  parallel    DRCCDSOY      M3210HDS
  pra=›1m11111100›22m  prb=›1m11111111›22m      pra=›1m11111111›22m  prb=›1m11111111›22m
        ddra=›1m00000011›22m ddrb=›1m00000000›22m    ddra=›1m11000000›22m ddrb=›1m11111111›22m
        event=000DB1  icr=00  sdr=76    event=10BD69  icr=00  sdr=00

Indianna Jones and the Last Crusade: The Adventure

Speed is erratic and will sometimes stall during opening sequence. Worked fine in GOLD 2.5.




Szyk Cech

Posts 191
11 Mar 2018 14:00


sean sk wrote:

All games run on WHDLoad 18.4.5889 in Workbench 3.1.

Why you didn't use ApolloOS?!? Many times the Team claim that older versions are not compatible...


Sean Sk

Posts 488
11 Mar 2018 14:37


These issues have nothing to do with the OS.
 
  Why wouldn't WB 3.1 be compatible? Haven't had any troubles with it up until now. Specifying that Apollo OS is necessary for the Vampire is completely ludicrous. I don't have a large enough CF card nor do I wish to install an OS so large. Although it looks nice, I don't need all the bloat that comes along with it. Plus I own a license of WB 3.1.


Vojin Vidanovic
(Needs Verification)
Posts 1916/ 1
11 Mar 2018 16:01


sean sk wrote:

These issues have nothing to do with the OS.
    Why wouldn't WB 3.1 be compatible?

It can with some components. Fresh 3.1 install needs a lot of updates ro get to somewhere. Also, sometimes older WHDLoad versions work better. As always, a bit of experiments help.

But thanks for the reports, both team and WHDLoad bug reporting should help to fix the issues.



Manfred Bergmann

Posts 226
13 Mar 2018 19:10


I'm having some trouble getting Elite Advanced working via WHDload on Vampire.
Anyone has that working?

Basically the game works until you leave the station, then it crashes when entering the station exit tunnel.
WHDLoad says: "Exception 'Integer Devide by Zero'".

Manfred


Manfred Bergmann

Posts 226
15 Mar 2018 17:43


Manfred Bergmann wrote:

  I'm having some trouble getting Elite Advanced working via WHDload on Vampire.
  Anyone has that working?
 
  Basically the game works until you leave the station, then it crashes when entering the station exit tunnel.
  WHDLoad says: "Exception 'Integer Devide by Zero'".
 

 
  Just tried the game (Elite Advanced) with Core 2.0 and 2.5 where it works.
  So something in Core 2.7 has changed that makes the game incompatible.


Manfred Bergmann

Posts 226
15 Mar 2018 19:02


Manfred Bergmann wrote:

Manfred Bergmann wrote:

  I'm having some trouble getting Elite Advanced working via WHDload on Vampire.
    Anyone has that working?
   
    Basically the game works until you leave the station, then it crashes when entering the station exit tunnel.
    WHDLoad says: "Exception 'Integer Devide by Zero'".
 

 
  Just tried the game (Elite Advanced) with Core 2.0 and 2.5 where it works.
  So something in Core 2.7 has changed that makes the game incompatible.

Tested on AmigaOS 3.9 BB2 with WHDLoad 18.3 and the latest 18.4 beta.


Sean Sk

Posts 488
02 Apr 2018 03:40


Another WHDLoad game that was working in GOLD 2.5 but no longer working in GOLD 2.7 and onwards is Lotus Esprit Turbo Challenge.
When starting a race screen is black showing only the HUD. Sound of the car revving can be heard, but cannot start the race.


Gunnar von Boehn
(Apollo Team Member)
Posts 6207
09 Apr 2018 09:35


sean sk wrote:

I had posted in the IRC channel already for one of the games mentioned here. 

Sean,

WHDLOAD games is a complex topic.
In a nutshell, some Amiga games are "unclean" written.
And they can not work on faster AMIGAs.

WHDLOAD does fix some of these games, but not all.
WHDLOAD sometimes uses "tricks" to make the AMIGA slower, which as side effect helps to run some games - without fixing them.

These WHDLOAD tricks slow down by a certain factor - as faster your AMIGA as less they help.

With GOLD 2.7 the VAMP got a lot faster this is what you see now.

The dirty games can all be fixed.
As soon as the VAMP does fully control the AMIGA Blitter/Chipset most of these games will run fine - even with faster CPU.

As you know we work on this.
And we try to take care of those games.
But as you will easily understand this is a big task which needs lot of time for us to finish.


Manfred Bergmann

Posts 226
09 Apr 2018 10:07


Gunnar von Boehn wrote:

 
sean sk wrote:

    I had posted in the IRC channel already for one of the games mentioned here. 
   

    With GOLD 2.7 the VAMP got a lot faster this is what you see now.

 
  Shouldn't the 'turtle' mode than slow down the Vampire so that the games will work again?
 
 
  Manfred


Gunnar von Boehn
(Apollo Team Member)
Posts 6207
09 Apr 2018 10:14


Manfred Bergmann wrote:

Shouldn't the 'turtle' mode than slow down the Vampire so that the games will work again?

Yes, "Turtle mode" fixes a number of bad games.

The main game problem is buggy programming of the Blitter.
The games did simply not fully to programming rules by Commodore.

Slowing down the CPU does help here.
But the real cure is taking full control of the Blitter.
We are working on this - we see this.
Believe us this makes the real difference.




John William

Posts 563
09 Apr 2018 18:28


Gunnar von Boehn wrote:

Manfred Bergmann wrote:

  Shouldn't the 'turtle' mode than slow down the Vampire so that the games will work again?
 

 
  Yes, "Turtle mode" fixes a number of bad games.
 
  The main game problem is buggy programming of the Blitter.
  The games did simply not fully to programming rules by Commodore.
 
  Slowing down the CPU does help here.
  But the real cure is taking full control of the Blitter.
  We are working on this - we see this.
  Believe us this makes the real difference.
 
 

Heheh. Can I be honest? I am sadistic by nature. When I see Commodore point list of rules that I should not do when writing games I will do all opposite and break all the rules! BWAHAHAHAHAHAHHH!

Now comes the real question. Is that what the programmers did? Did they find pleasure in seeing these rules to break them or they are just morron with no ill intention and really was just bunch of lazies who just wanted fast programming to get fast buck?


Gunnar von Boehn
(Apollo Team Member)
Posts 6207
09 Apr 2018 18:47


John William wrote:

Now comes the real question. Is that what the programmers did?

Using the AMIGA Blitter needs some special coding rules.
If you not follow them, it might still work on your System but might fail on a faster AMIGA.

Its very simple the coders wrote the game without noticing their coding bugs - as they at that time did not had faster systems like 68040 to verify the game on.




Sean Sk

Posts 488
10 Apr 2018 06:58


Gunnar von Boehn wrote:

  The dirty games can all be fixed.
  As soon as the VAMP does fully control the AMIGA Blitter/Chipset most of these games will run fine - even with faster CPU.

Thanks heaps for your reply Gunnar, and, as always, I appreciate what you're doing here. :)

From what I understand, in order to benefit from taking full control of the Blitter, you would have to include and use a new Blitter built into the core itself (GOLD 3 and onwards as you have mentioned in the past). This would mean all video would have to be output through the DIGITAL-VIDEO port. This in itself is an excellent feature!

The only concern I have is that I would like to still be able to use the RGB video out into a very large CRT monitor, since I still enjoy the look of games on CRT. I also have an A1200 which is going to be reserved for the V1200 and that will use DIGITAL-VIDEO to a LCD monitor and all the extra benefits that come along with SAGA. Nevertheless it would still be nice to be able to use a CRT on at "least" one of my machines. And before anyone says anything, NO I'm not taking the V600 out of the A600 since once you've had one in there it's impossible to go back. :)

Manfred Bergmann wrote:

Shouldn't the 'turtle' mode than slow down the Vampire so that the games will work again?

I have used turtle mode for those affected games but it does not seem to help. It seems there are greater issues, such as the Blitter.

In such a situation Gunnar, is it at all possible to provide a legacy environment for non working WHDLoad games rather than relying on WHDLoad itself to provide the environment since, as you say, WHDLoad can be ineffectual in slowing down the system enough for some games?



Gunnar von Boehn
(Apollo Team Member)
Posts 6207
10 Apr 2018 07:35


sean sk wrote:

The only concern I have is that I would like to still be able to use the RGB video out.

I think we can make this possible.

Sean many thanks for your reports.

How about joining our tester team?
You can then help testing and improving new features before they are released.


Sean Sk

Posts 488
10 Apr 2018 14:05


Thanks Gunnar, that would certainly be fun! :) Although I'm not sure how much help I can be, since I'm not so technically minded, and time can be limited at times. All I can say is: this worked, that didn't, etc. :)
  Anyway, if you need more testers, let me know how I can sign up to this outfit. :D

posts 16