| |
6r6_Nostalgia
Account closed
Registered: Oct 2003
Posts: 14 |
Old fucked up c64 models
Anyone experienced games that would not run or crash for no reason on your C64 ?
I have two commodore 64 1984 models, and they both
crash on my disk version of Myth Cartridge.
Recently I got this game tested on someone else's C64
equipment with a oceanic disk drive and the game was running without any problem (both pal and ntsc).
So, I put up my c128d and tested the game there, and,
yes! it was running without any problem from my 1581 and 1571 drive. Tried running the game again on my two c64, and - yes IT WAS CRASHING again.
All my other games are running nicely on these 2 c64, so what is the problem ? I have heard about demos that wouldnt run because of the chips was too hot... So i let
the c64 heat for a few hours, and then tested the game,
crash! still no luck..
Now, I talked to Bamboo on IRC the other day and he
said he experienced similar problems on the game HOOK.
He said something about the scrolling routine would
crash the machine. And he got a 1982 UK model.
AFAIK Myth uses dd04/dd05 to do the scrolling.
We sort of came to a conclusion that certain old c64 models
would fuck up..
Well, any comments.. anyone ?
6R6 - N0stalgia
|
|
| |
raven
Account closed
Registered: Jan 2002
Posts: 137 |
From what i know through discussions
with other coders the problem is a
certain version of the VIC chip (cant
remember exact model).
For some reason it corrupts random
memory locations when $d011 is changed
rapidly.
$d011 is used massivly for FLI and VSP (hardware-
scrolling) which are usually the effects that
cause those C64's to crash.
I suggest you talk with WVL/Xenon. I think he tried
to find some pattern to the corrupted memory
locations, dunno if he succedded :)
|
| |
6r6_Nostalgia
Account closed
Registered: Oct 2003
Posts: 14 |
Just tested F4CG, Contex, Censor and Nato's version of
this game on my computer. They all crash after playing
for 30 seconds. |
| |
Stryyker
Registered: Dec 2001
Posts: 465 |
At one time Graham was doing research into VSP related issues with RAM refresh and so forth. I guess if he knew what needed memory was maybe not getting refreshed properly (I guess VSP can skip some of these cycles) a simple read might refresh them.
C64 RAM can go a little over 1 second before being refreshed? |
| |
6r6_Nostalgia
Account closed
Registered: Oct 2003
Posts: 14 |
Reading about demo crashes here :
http://groups.google.de/groups?dq=&hl=de&lr=&ie=UTF-8&threadm=P..
.. and runned the following program on my c64:
1000: lda $d011
1003: eor #$20
1005: sta $d011
1008: jmp $1003
After a while (from a few seconds up to several mniutes) the
machine froze...
Fucking Crap!
|
| |
6r6_Nostalgia
Account closed
Registered: Oct 2003
Posts: 14 |
oops! start of link here:
http://groups.google.de/groups?dq=&hl=de&lr=&ie=UTF-8&frame=rig.. |
| |
WVL
Registered: Mar 2002
Posts: 886 |
Well, I never found any patterns :) Also refreshing the RAM by coding it into the routine didn't help shit.. seems to be a bus-problem to me. So the VIC&CPU try to access the memory at the same time, or something like that... Maybe the cpu is even fed data being read by the VIC, so the code that's run is coming from the gfx data, but I don't think this is the case (I tried making the gfx data read only $ea)
Also the memory corruptions occur in every page in the memory, so if $0027 is corrupted, so will $0127, $0227 etc be..
weird. but I couldn't find a way to stop it. |
| |
raven
Account closed
Registered: Jan 2002
Posts: 137 |
@wvl:
You remember which exact VIC chip version does this?
|
| |
algorithm
Registered: May 2002
Posts: 702 |
I had a C64C which had the same problem. Any type of Demo which had Bitmap scrollers (VSP) would sometimes crash on my machine.
The only remedy would be to turn off the machine and turn it back on in the hope that it would work next time
But this problem was finally rectified when I used a new power supply on the same C64C machine. The crashes disappeared!!!!
|
| |
WVL
Registered: Mar 2002
Posts: 886 |
@Raven : no, it seems to happen on all VIC's, if you wait long enough. Especially 128's crash very quickly. |
| |
6r6_Nostalgia
Account closed
Registered: Oct 2003
Posts: 14 |
I dont have a normal power supply... I am using a REU power supply even though I dont have a REU attached... Is that supposed to be unhealthy ?
|
| |
Merman
Registered: Dec 2002
Posts: 140 |
I have noticed similar crashes in MAYHEM IN MONSTERLAND and CREATURES, presumably due to the same thing. |
| |
WVL
Registered: Mar 2002
Posts: 886 |
yes, it's due to the same thing.
There is a solution though. Get a different c64. |
| |
QuasaR
Registered: Dec 2001
Posts: 145 |
There's also a different ROM in some really old C64 (check AAY64). I have one and RED STORM/TRIAD isn't running on it but on my other (&newer) C64/C128 it works just fine. The AGSP-bug seems to have something to do with warm ICs (my C128 really hates these effects when it's running for some minutes... :( |
| |
raven
Account closed
Registered: Jan 2002
Posts: 137 |
@wvl:
well i have lots of 64's here & my main machine is a 128D.
Never had this prob on any of them but i know ppl
who have them all the time, doesnt matter if the
machine is cold or hot.
|
| |
WVL
Registered: Mar 2002
Posts: 886 |
@ Raven : I have one machine that has the bugs when it is hot. I can run f.e. Insomnia on it the first time when I turn it on, If I try again it will almost 100% crash. Same for other demos (especially Dutch Breeze)... it really matters if the machine is cold or hot. |
| |
Oswald
Registered: Apr 2002
Posts: 5017 |
my first c64 behaved pretty strange, it crashed for example always on IK+, but had no problems with dutch breeze.. :) |
| |
raven
Account closed
Registered: Jan 2002
Posts: 137 |
@Oswald:
Maybe it just didnt like tha game? ;) |
| |
Oswald
Registered: Apr 2002
Posts: 5017 |
yea, it wanted me to watch dutch breeze instead :D |
| |
WVL
Registered: Mar 2002
Posts: 886 |
No, it just wanted to get fingered. |
| |
Oswald
Registered: Apr 2002
Posts: 5017 |
:DDDDD |
| |
Sledge
Registered: Sep 2003
Posts: 102 |
Quote: I have noticed similar crashes in MAYHEM IN MONSTERLAND and CREATURES, presumably due to the same thing.
|
| |
Sledge
Registered: Sep 2003
Posts: 102 |
6r6:
No, running with a REU PSU is a good thing even if you don't use the REU. I guess they ment you should use a newer model PSU instead of the old, brown one. They tend to fuck up alot. A REU PSU is as reliable as a C128 PSU.
|
| |
fade
Account closed
Registered: Mar 2002
Posts: 290 |
you wouldnt have these problems on an emulator :) |
| |
Slator
Registered: Jan 2002
Posts: 273 |
Quote: you wouldnt have these problems on an emulator :)
still people like you would exist.... ;-D
bugger all... |
| |
HackZ0id
Account closed
Registered: Sep 2002
Posts: 6 |
Overheating of chips in the computer often result in these type of strange behaviour ( like some of you already mentioned ).
If you want to examine a "semi-working" c64 for overheating problems, buy some "cooler-spray" from your local electronic component shop. Here in sweden it can be bought at elfa and biltema. Biltema charges less that 30SEK for a bottle if i recall it right.
By spraying one chip at a time you can conclude which chip it is that starts to fail when reaching working-temperature.
Another tip is to mount transistor radiators ontop of the most often failing chips. I've done that on my SID, VIC-II, CIA's, PLA, ROMS, CPU.
Cooling the chips extends MTBF a LOT!
This must be done BEFORE your chips start to fail!
Anyone intressted in getting some instructions about cooling those vital chips inside your C= hardware? ( if enough people want this I might write a HOWTO about it )
Hz/Sid
|
