| |
Krill
Registered: Apr 2002
Posts: 2969 |
Please help testing: New loader release imminent
Hi,
before releasing the next version of my loader, i'd like to make sure it works alright on random actual hardware.
Basically i only need to know on which hardware combination it fails.
Some information about exotic hardware which works with it (anything that is not 1541/71/81) is also appreciated.
I'd be grateful for all reports.
The test images can be found here.
Thank you! |
|
| |
MagerValp
Registered: Dec 2001
Posts: 1074 |
My C64 setup is packed away at the moment, so I just did a quick check in VICE. d64 and d71 work as expected, but loadertest-c64.d81 is acting really strange. With a jiffydos kernal and vanilla 1581 roms I can't even list the directory with F1. Vanilla kernal can load and start LOADER TEST, but performance is abysmal (worse than kernal IEC routines). The IRQ is flickering too.
VICE bug, loader bug, or PEBCAK? |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
Strange at least, it works fine on my VICE setups.
Apparently the drive model isn't detected correctly, and what you see is KERNAL fallback on devices which are on the serial bus and don't come with custom KERNAL. (KERNAL fallback on IDE64 or netload of course doesn't disturb IRQs. The worse-than-plain-KERNAL-performance effect is probably due to all the overhead for the KERNAL GETCHR call, which is of no import with the fast custom KERNAL devices.)
But that you cannot even list the directory before the loader test tool is actually running makes me a bit suspicious that this setup is a bit broken to begin with..
Can you send me the ROMs? Thanks! |
| |
Clarence
Registered: Mar 2004
Posts: 121 |
loadertest-c64.d64 on c128dcr
Red flashing border, right away when it tries to install the loader
On top it writes:
@$74fe/$86($75,$23): Invalid memory configuration
On bottom:
PAL, #8/CBM1571CR, $0856
POLLBLOCK
When I try to re-run it's all the same, only the memory address on top is sometimes different fe. $7575 instead of $74fe |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
Clarence: I could reproduce the error in VICE. Should be fixed by tonight :) |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
Clarence: Alright, this stuff here should work on your C128DCR. Thanks! |
| |
Trurl
Registered: Mar 2002
Posts: 61 |
It doesn't like my 1541! The d64 fails on breadbox+1541, c64c+1541 and flat c128+1541. Loading the first pic brings garbled picture and then it says @$1342/$81 (numbers in brackets vary): Wrong end address.
However both the d64 and the d71 run fine on my C128D. Also works on IDE64 (both c128 and c64, dos 0.90), the hires/multi rasterswitch flickers though.
Couldn't test IDE64 pclink due to slashes in filenames... and unfortunately my 1541-II and 1570 are unusable atm :( |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
Trurl: Thanks for the report. I currently don't have a 1541, only several 1541 II. Are you possibly in for some remote debugging? I'd simply send you images to test until the bug is located and fixed. |
| |
Trurl
Registered: Mar 2002
Posts: 61 |
Sure, PM'd. |
| |
Crimson
Account closed
Registered: Jul 2005
Posts: 6 |
Krill, we should finally hook up! |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
Crimson: Indeed! And have a few beers, plus i can get that old 1541 back. Wonderful! :D |
| |
tlr
Registered: Sep 2003
Posts: 1787 |
Works on breadbox + 1541U-II 2.3tlr2 (AA). |
| |
Clarence
Registered: Mar 2004
Posts: 121 |
Thx Krill, the new d64 works fine with the c128dcr internal drive. Test ran OK, for 8 minutes or so.
However when I switch off the internal drive and use an 1541u on the c128d the test hangs after about 30-40 seconds at this point (screenshot):
 |
| |
vedos
Account closed
Registered: Jul 2003
Posts: 33 |
Hi Krill!
Few days ago I got a new C-64C and 1541-II. I decided to test the loadertest-v115.zip on the hardware, but it didn't run longer than ~8 seconds. I thought that the new hardware is the problem, but it seems that the loader doesn't work on any combination of my C= hardware. I tested it quickly with old breadbox, two new model C-64s and two different 1541-II drives. If I'll find my 1541 I'll test it also later.
Here's some error messages I wrote down.
--->
@$1342/$81($09,$4c): Wrong end address.
PAL, #8/CBM1541II, $0856
2.250 kB/s POLLBLOCK
---
@$1342/$81($f7,$43): Wrong end address.
PAL, #8/CBM1541II, $0856
3.409 kB/s POLLBLOCK
---
@$1342/$81($0f,$49): Wrong end address.
PAL, #8/CBM1541II, $0856
1.990 kB/s POLLBLOCK
<---
64ever Insomnia (v1.1) drive test passes fine. I also tested some demos and all seem to work ok.
I will do more thorough tests later.
BTW: Does anyone have older version of Insomnia (v1.0?) which is more sensitive on the drives it works. I'd like to have it on my "test kit".
|
| |
Krill
Registered: Apr 2002
Posts: 2969 |
Alright, this is confirmed to work fine with Trurl's 1541 now.
So, all for whom the former builds didn't work, please check if this one does now, thanks! (Also hi vedos!)
Clarence: Maybe this build works on your C128DCR+1541u setup. If not, i'll let it rest until all "real" drives work and then see. Not the first time 1541u is to blame for my loader not working alright with it.. :) |
| |
Total Chaos
Registered: Mar 2006
Posts: 74 |
v118 with no errors on C64c and 1581. |
| |
vedos
Account closed
Registered: Jul 2003
Posts: 33 |
loadertest-v118 works now on the 1541-II I'm using.
The another 1541-II doesn't work, but it's most probably because the drive heads aren't clean.
I have also 2x normal 1541 and one 1541 clone drive... I'll try to find 'em also to test. |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
vedos: Thanks!
Clarence: I have a suspicion that 1541u fails with extended disk images. I have prepared another image which doesn't use tracks 36-40 (it is a 40-track image, however). Can you tell me if this works? Thanks!
|
| |
The MeatBall
Registered: Dec 2001
Posts: 367 |
Quote: Hi,
before releasing the next version of my loader, i'd like to make sure it works alright on random actual hardware.
Basically i only need to know on which hardware combination it fails.
Some information about exotic hardware which works with it (anything that is not 1541/71/81) is also appreciated.
I'd be grateful for all reports.
The test images can be found here.
Thank you!
Should this work with IEEE drives aswell? If so I can test it on my SFD-1001 (unfortuanely my 2031lp seems to be broken :( ) |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
Meatball: I'm working on it, expect a SFD-1001 compatible image soon. Thanks for the suggestion, never checked that in VICE before. |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
This build *should* work on SFD-1001.
The MeatBall: Now you may test it, i'm excited! :) |
| |
The MeatBall
Registered: Dec 2001
Posts: 367 |
Quote: This build *should* work on SFD-1001.
The MeatBall: Now you may test it, i'm excited! :)
I'm away for the weekend, but will see if I can test it next week. Now, how to get a .d82 over to a floppy when I can't use my 1541U with my IEEE interface :) |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
You can use your favorite filecopy program to put all files on an SFD-1001 disk - the files are the same on all images.
Since the loader uses the KERNAL fallback for all drives it doesn't have custom code for, it's not faster than loading normally on an SFD-1001. |
| |
Clarence
Registered: Mar 2004
Posts: 121 |
Gunnar, yep, v120b works alright with c128dcr+1541u. |
| |
Hoxy
Account closed
Registered: Sep 2002
Posts: 5 |
Tested version 123 on:
Old vc-1541, worked
Oceanic-118, worked
Enhancer 2000, didn't work. Hung immediately on 'pollblock' |
| |
The MeatBall
Registered: Dec 2001
Posts: 367 |
Quote: You can use your favorite filecopy program to put all files on an SFD-1001 disk - the files are the same on all images.
Since the loader uses the KERNAL fallback for all drives it doesn't have custom code for, it's not faster than loading normally on an SFD-1001.
Not tested on my SFD yet, as for some reason my MSD IEEE interface won't let me use a serial device #9 together with a IEEE device #8 :( |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
Hoxy: "Enhancer 2000" is a weird beast, haven't known it until now. According to the internets it seems to fail with a lot of custom loaders, but i'd like to know why and maybe even get the loader to work with it.
Can you make a ROM dump of its firmware? Would you help me get the loader to work with it?
The MeatBall: The interface may have a way to switch #9 to serial. Some come with physical switches for that, but i suspect those who don't have a BASIC command for that or at least some magical POKE. Do you still have the manual? |
| |
Hoxy
Account closed
Registered: Sep 2002
Posts: 5 |
As you say, it's weird, that's why I just had to test your loader on it :)
If you really want to try to make it work on it, I'll do my best to help you.
I'll get a ROM dump to start with then. |
| |
The MeatBall
Registered: Dec 2001
Posts: 367 |
Quote: Hoxy: "Enhancer 2000" is a weird beast, haven't known it until now. According to the internets it seems to fail with a lot of custom loaders, but i'd like to know why and maybe even get the loader to work with it.
Can you make a ROM dump of its firmware? Would you help me get the loader to work with it?
The MeatBall: The interface may have a way to switch #9 to serial. Some come with physical switches for that, but i suspect those who don't have a BASIC command for that or at least some magical POKE. Do you still have the manual?
I've got the manual (http://c64disks.hvilket.net/MSD_C64_to_IEEE-448_Manual_and_Note..) and it has a small type-in that's supposed to let me use serial and ieee devices simultaneously, but I've had no luck in getting it to work :( |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
The MeatBall: Can you send me the driver code which is active at $9c00..$9fff after enabling, or the contents of the IEEE cartridge's ROM, or the contents of the disk the manual mentions? (The more, the merrier :D)
To work with the current loader test build, the driver code must be relocated to somewhere in the $c000..$cfff range ($cc00..$cfff, as in the manual, should be the safest option). Once it's working, i can add some code to automatically relocate the driver if need be.
Getting #9 to work as serial is needed to actully copy stuff to the MSD drive, but maybe i can help you there using VICE and the driver code. (As i understood, most if not all IEEE-488 interfaces look pretty much identical from the software side and only come with different driver code - so VICE might be able to emulate it.) |
| |
The MeatBall
Registered: Dec 2001
Posts: 367 |
Right, how do I save $9c00-$9fff from basic? My 1541U doesn't work with the CIE, so I need to do it from basic :)
Also, there is no disk, nor do I have the ROM dumped (but I'm pretty sure I saw it online somewhere when I was looking for the manual). |
| |
iAN CooG
Registered: May 2002
Posts: 3187 |
?peek(40959):poke43,0:poke44,156:poke45,0:poke46,160:save"1",8,1
note down the value printed because must be patched in the resulting file "1", the byte at $9fff will be replaced by the byte "1" (the file name) |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
I found something which might resolve the last issues.
This is the latest build and might fix the 1541u issue. The extended tracks were't the problem, they weren't even accessed in the failing older builds.
Clarence: Can you see it this build works with your 1541u? Thanks! |
| |
PAL
Registered: Mar 2009
Posts: 271 |
Krill... cant you just join offence so that we can start releasing demos on time? |
| |
Krill
Registered: Apr 2002
Posts: 2969 |
PAL: Hmm, and how would i influence your development procedures to the better? :) |
| |
PAL
Registered: Mar 2009
Posts: 271 |
You are a coding genius and a creative force and have done so much cool stuff and you know best what we know the less of... And the comment were also a fun one due to the LCP hassle... he he... we admire you! |
| |
The MeatBall
Registered: Dec 2001
Posts: 367 |
Quote: ?peek(40959):poke43,0:poke44,156:poke45,0:poke46,160:save"1",8,1
note down the value printed because must be patched in the resulting file "1", the byte at $9fff will be replaced by the byte "1" (the file name)
Um, and since I cannot have the 1541U plugged in when enabling the IEEE interface, and I don't have access to the serial devices when IEEE is plugged in, I'm having trouble actually saving this to media that I can access on a pc :(
|
| |
Perplex
Registered: Feb 2009
Posts: 255 |
Quote: PAL: Hmm, and how would i influence your development procedures to the better? :)
You'll swing the whip, of course!
We'll credit you as "cracking by Krill". |
| |
Clarence
Registered: Mar 2004
Posts: 121 |
v132 has no problem. |
| |
Luca
Registered: Apr 2002
Posts: 178 |
v132 on Plus/4 w. 1571 drive just tested.
It stucks in the installing loader screen with the counter locked at 00:00:00
|
| |
King Durin
Account closed
Registered: Oct 2007
Posts: 85 |
Flat NTSC 128 w/64K Video Ram, JiffyDOS 6, 1541U + RetroReplay Cart Image.
Running the 1581 tests, get the following error:
@$16a3/$80($48,$00): Intenral error.
NTSC, #9/generic, $0862
.097 kB/s GETCHUNK/CMP
|
| |
vedos
Account closed
Registered: Jul 2003
Posts: 33 |
I think I tested this also on Oceanic drive before I sent drive to Uneksija and it worked fine!
BTW: Still looking Insomnia v1.0. |
