| |
tlr
Registered: Sep 2003
Posts: 1714 |
Help test x64sc, the next generation x64.
We've spent a month and a half rewriting x64 into x64sc and would like to have some feedback.
Please test this with as many things as you can and report any emulation bugs you find in here or alternatively on #vice-dev @ freenode.net.
Let's bring VICE from the 90's into the modern age!
Good:
- emulator rewritten, more accurate
- models:
Breadbox (default) x64sc -VICIImodel 0 -ciamodel 0 -gluelogictype 0
C64C x64sc -VICIImodel 1 -ciamodel 1 -gluelogictype 1
C128-ish x64sc -VICIImodel 1 -ciamodel 1 -gluelogictype 0 Bad:
- slow (will be faster)
- possible regressions, please test
- no NTSC (yet)
- no REU (yet)
- no snapshots (yet)
- mystery drive bug (disable virtual device traps)
- GUI not yet updated
Win32 binary: x64sc-r22294-win32.zip
Source code (subversion repository):
svn co https://vice-emu.svn.sourceforge.net/svnroot/vice-emu/branches/..
|
|
| |
MagerValp
Registered: Dec 2001
Posts: 1055 |
Well, it builds (though I had to work quite a bit on the build scripts for macosx), and it runs... is there anything in particular you'd like us to test? :) |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
everything, yes :) the whole emulation core has been touched, and the vic-ii is pretty much rewritten - so it is very likely that new bugs have been introduced. (although for 6 weeks, it works pretty damn well already) |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
@mv: sorry about that, no one on the x64sc team uses a mac currently. Care to share the patches?
As groepaz says, the emulator is more or less rewritten. Especially the VIC-II emulation.
This means that in-line graphics changes and more intricate in-line switching of modes is now supported.
We need help finding any high tech (or simple) demos/programs that show bugs or break in some way.
Some random examples:
- VIC-Check now works. Use -VICIImodel [0|1] to select.
- Edge of Disgrace shows the reported bugs for different machine configurations.
i.e white dots on the terminator part for 6569 (old vic), and glitches on the floor part on the C64C.
- Big Vic Show (part 4) no longer shows a graphics glitch.
- Starion-Intro Remake, no more bogus sprite glitch.
- Crest Avantgarde (The wall), Bug showing color one pixel too far left in the top and bottom border no longer there when selecting 6569 (old vic).
- Programs protected with +h Emu-Fuxx0r V2.0 runs. |
| |
Frantic
Registered: Mar 2003
Posts: 1627 |
Good work! Can't help smiling about the massive amounts of anal nerdery incorporated in this release :)
...a nerdy smile too: Hieh hihe hieh hieh... |
| |
Count Zero
Registered: Jan 2003
Posts: 1821 |
Whoever fucked this up did it on all trees it seems - we just figured:
- run x64 in any plain basic mode
- poke2,2:sys2
-> ALT+F4 ... rest is dead
I really really need to know who fucked this one up so tremendously :) --- that person owes me a large amount of time now.
l8r
Count Zero/CyberpunX/SCS*TRC
PS: "anal nerdery" ... shiver :) |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quote:- run x64 in any plain basic mode
- poke2,2:sys2
-> ALT+F4 ... rest is dead
What x64 revision are we talking about? Not x64sc?
Anyway I can't reproduce this on win32 nor for x64sc or the trunk x64 version. |
| |
Count Zero
Registered: Jan 2003
Posts: 1821 |
We are just discussing this on irc.
My recent build is linux 32bit on ubuntu 8.04 - another build is by groepaz on his 64 bit system.
Both are recent svn checkouts - mine was compiled with --enable-memmap - groepaz without. Both crash when entering basic again after the soft/hard requester. Also additional soft/hard resets using key combo didnt revive the emulation. On groepaz it crashed with 100% cpu usage after a few hard resets.
Similar behaviour was seen by someone else on a 2.1 (!!!) windows build - not as straight forward on the crash but reproducable. (and as I just hear: even loading a snapshot doesnt resolve the problem when at this point)
viceteam will have to look hard for this truely annoying issue.
# svn up (or clean checkout in case of sc)
# ./autogen.sh
# ./configure --enable-memmap --disable-ffmpeg --enable-fullscreen --enable-gnomeui && make clean && make && make install
revision is 22294
l8r
Count Zero/CyberpunX/SCS*TRC |
| |
Count Zero
Registered: Jan 2003
Posts: 1821 |
Pressing SHIFT fixes the issue! LOL - sometimes :)
Someone will throw this at sf.net I guess - too lazy myself now.
l8r
Count Zero/CyberpunX/SCS*TRC |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
So, to get back on track:
- Crest Light (sunkist) sometimes shows a small graphics glitch just above the scroller to the right. (6569 emulation)
- Crest Light (mirinda) shows a small glitch on the left at the bottom of the grey area. There is a bug in the same spot on a real 6569R3 box, but a different one.
Both are under investigation. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
- +H2K, circular screen wipe after C=64 texture part glitches.
This bug was related to reuse of sprites around the fetch cycles and has been fixed in r22308.
(win32 binaries provided on demand)
|
| |
algorithm
Registered: May 2002
Posts: 702 |
I know this may be offtopic. But is there any link where i can download the X64e (which is a modified version of vice that gives more solid refresh rate for tweaked 50hz video mode) This was posted in the lemon64 forums but link does not exist anymore |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
no, that version was only ever spread in binary form, as you said yourself, it disappeared. however, you can get the very same effect by enabling opengl sync, and then switching emulation speed to "unlimited". |
| |
Oswald
Registered: Apr 2002
Posts: 5017 |
great news, altho vice was accurate enough for me :) but I was never a perfectionist. its always nice to see these efforts. |
| |
booker
Account closed
Registered: Jul 2003
Posts: 333 |
x64sc doesn't crash on Aero (Windows 7) upon moving the window on the taskbar. yay! (most likely same will work for Vista) |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22330-win32.zip
- Repository should build on MacOSX.
- CIA model can be selected separately (-CIA[1|2]Model <0|1>)
- Fetch latency corrections.
- Fixes Box Check "Sprite A".
- Performance improvements.
|
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
Impressive, x64sc > hoxs64 |
| |
Frantic
Registered: Mar 2003
Posts: 1627 |
Didn't build on my Mac. However, I am quite lame when it comes to in-depth linux/gcc/makefile-stuff so it is probably just my fault. Anyway, I tried:
autogen.sh
./configure
make x64
...and after some time of compiling (with some warnings here and there) it finally said:
make[1]: *** No rule to make target `../src/vicii/libvicii.a', needed by `x64'. Stop.
I'll just wait for some binaries to appear, but perhaps it is still interesting for you to know.. ...or not. :) |
| |
Mr. SID
Registered: Jan 2003
Posts: 421 |
Frantic, try: http://www.lallafa.de/files/vice/snapshots/ |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quote: Didn't build on my Mac. However, I am quite lame when it comes to in-depth linux/gcc/makefile-stuff so it is probably just my fault. Anyway, I tried:
autogen.sh
./configure
make x64
...and after some time of compiling (with some warnings here and there) it finally said:
make[1]: *** No rule to make target `../src/vicii/libvicii.a', needed by `x64'. Stop.
I'll just wait for some binaries to appear, but perhaps it is still interesting for you to know.. ...or not. :)
Supposedly fixed in r22339. Thanks for reporting. |
| |
Ninja
Registered: Jan 2002
Posts: 404 |
Sounds very promising!
Looking really forward to be able to select the CIA-type; I needed that a couple of times and also looked at the vice-sources, but well...
BTW if you bring VICE into the modern age, could you also bring your SCM into a modern age? Like git or whatever? The svn update takes more than 10 minutes already, and I haven't even reached your branch :( |
| |
Count Zero
Registered: Jan 2003
Posts: 1821 |
Doh, Ninja - complete co takes a lot less than 10 minutes here - svn up is done in seconds usually? :)
/cz |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
i recommend getting rid of that us robotics modem .... =) |
| |
Ninja
Registered: Jan 2002
Posts: 404 |
Okay, guys, come here over to Tokyo and prove me wrong ;)
(I take this as no BTW, pity) |
| |
Count Zero
Registered: Jan 2003
Posts: 1821 |
You pay the ticket, Ninja? :)
Blame sf anyhow :)
/cz |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:You pay the ticket, Ninja? :)
yeah, was that an offer? i will even bring entire sourceforge on DVD for you then.... or sth =) |
| |
Ninja
Registered: Jan 2002
Posts: 404 |
It was a bit better today, so I finally have the source (but svn still feels a bit aged ;) (Yes, this is personal opinion. Do not reply!))
Back to topic: Compiling and running went smooth, will use it as my default emulator now and report whenever I notice something. |
| |
Krill
Registered: Apr 2002
Posts: 2839 |
Quote: Sounds very promising!
Looking really forward to be able to select the CIA-type; I needed that a couple of times and also looked at the vice-sources, but well...
BTW if you bring VICE into the modern age, could you also bring your SCM into a modern age? Like git or whatever? The svn update takes more than 10 minutes already, and I haven't even reached your branch :(
Sorry to be off-topic, but please don't mix up slow connection speed with any concept of version control "modernity". Unless we want to have several more or less independent VICE source trees, git et al. will give no advantage worth mentioning. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:Unless we want to have several more or less independent VICE source trees, git et al. will give no advantage worth mentioning.
you know.... the fact that i cant simply pull a recent trunk, work on it and then eventually push back my changes .... is pretty much exactly what stopped me working on the MMCR implementation. git ftw! =P |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22354-win32.zip
- 6569R1 emulation (-VICIImodel 2 or -VICIImodel 65691)
- Corrected SID read latency.
|
| |
Ninja
Registered: Jan 2002
Posts: 404 |
Krill: Sigh, I said "DO NOT REPLY"! Updating a git-repo is faster as its usage of bandwitdh is far more efficient alone. Scanning for differences between local copy and trunk is at least one magnitude faster (the reason that it worked for me later was not a better connection speed but the fact that I did a complete new checkout instead of updating my old repo). Other advantages? Having the complete history available without net-access, creating local branches on the fly, advanced merging-techniques, simplified patch creation... Your last sentence doesn't sound as if you have _really_ used git yet ;)
SVN is by no means useless, but for VICE I honestly think git would be the better alternative. Technically. If most people developing for VICE are not familiar with it, then it doesn't make sense yet, that's for sure, too.
|
| |
tlr
Registered: Sep 2003
Posts: 1714 |
x64sc has been merged to the trunk.
(svn co https://vice-emu.svn.sourceforge.net/svnroot/vice-emu/trunk)
There is still quite a bit of work before it's ready for release. |
| |
booker
Account closed
Registered: Jul 2003
Posts: 333 |
Quoting tlrx64sc has been merged to the trunk.
(svn co https://vice-emu.svn.sourceforge.net/svnroot/vice-emu/trunk)
There is still quite a bit of work before it's ready for release.
Where's win32 exe in this pls? Thx. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
there is none, you check out the source and then compile it =P |
| |
booker
Account closed
Registered: Jul 2003
Posts: 333 |
Thx :)
Would be najs to include .exe anyway x-) |
| |
Count Zero
Registered: Jan 2003
Posts: 1821 |
Yes, including the .exe into the source code repository makes a lot of sense.
What you are looking for is a nightly build and a link to it.
/cz |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22414-win32.zip
- First build since merge to trunk.
- Adjusted MCM bit switching latency. (Fixes Nine)
- Various changes. |
| |
Ed
Registered: May 2004
Posts: 173 |
I tested Nine with the new r22414 build but got this result:
It looks smoother from the earlier emulation, though. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quote: I tested Nine with the new r22414 build but got this result:
It looks smoother from the earlier emulation, though.
As far as I can tell this is correct emulation of the demo running on an old vic.
You need to switch emulation to new vic in the way specified in the first post in this thread. |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
x64sc.exe -VICIImodel 1
gives "argument 1 not valid for option -VICIImodel"
copying from commandline options
-VICIImodel <model>
Set VIC-II model (6569/6569r1/8565/6567/8562/6567r56a)
so must be explicit: -VICIImodel 8565
still the ini directive is
VICIIModel=1
|
| |
tlr
Registered: Sep 2003
Posts: 1714 |
@iAN: Ah, forgot about that, thanks for reminding me!
It should currently be like this: Breadbox (default) x64sc -VICIImodel 6569 -ciamodel 0 -gluelogictype 0
C64C x64sc -VICIImodel 8565 -ciamodel 1 -gluelogictype 1
C128-ish x64sc -VICIImodel 8565 -ciamodel 1 -gluelogictype 0 |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
and this shortcut is even better =)
-model <model>
Set C64 model (c64/c64c/c64old/c64ntsc/c64cntsc/c64oldntsc)
so -model c64c is enough. |
| |
Ed
Registered: May 2004
Posts: 173 |
Thanks. Now I got it working. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22426-win32.zip
- REU support added
- MMC Replay support added
- allow -sidstereoaddress 0xd420, etc |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
hex parameters should be everywhere, thx!
Where is the mmcr setting?
REU file path in REU settings UI is limited in length, should allow MAX_PATH chars. If I browse for it, the file is correctly opened, but I still can't edit the line past 40th character. This applies to both x64 and x64sc |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22446-win32.zip
- Cycle stealing repaired (got broken when REU was added back)
- 1 cycle shift of REU timing corrected. (fixes REU-Colors)
- some MMC64 and Expert cartridge fixes
|
| |
Lubber
Registered: Jan 2002
Posts: 26 |
Cucumber Juice 2 doesn't start (it hangs showing the Systemsetup, when detecting video and/or system as FloppyDOs detection seems to work) |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quote: Cucumber Juice 2 doesn't start (it hangs showing the Systemsetup, when detecting video and/or system as FloppyDOs detection seems to work)
Seems to work.
True drive emulation not on?
Virtual device traps? |
| |
Lubber
Registered: Jan 2002
Posts: 26 |
Both on...i tried a second time and now it worked. (without anything changed to x64 config.)
don't know why it didn't worked some minutes ago :? |
| |
Lubber
Registered: Jan 2002
Posts: 26 |
mmh, i tried it a third time and now it hangs again..
.... something seems to be instable or too much cycleexact i guess |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
Quote: Cucumber Juice 2 doesn't start (it hangs showing the Systemsetup, when detecting video and/or system as FloppyDOs detection seems to work)
works here, with either -model c64 or -model c64c, 1541 or 1541-II roms |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22470-win32.zip
- improved behaviour of $d016 MCM bit splits
- drive rotation code rewritten (correct rotation speed, unaligned data, weak bits)
- Prophet 64 support
- various cartridge fixes
- win32 UI for machine model selection
|
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22523-win32.zip (x64sc & x64)
- Different addressing glitches for splits with charrom fetches vs. ram fetches on 6569 now emulated. (fixes: Crest Light, parts mirinda and sunkist)
- CIA TOD frequency is handled better
- reSID-fp as default for x64sc
- monitor "list" command
- (mouse as) lightpen support added to x64sc
- 0x hex prefix for all int cmdline stuff
- cart system cleanup, a couple of new carts, bunch of monitor fixes (mostly breakpoints)
|
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
remove the damn buffer overflow msgbox in sound.c, no one needs to have the emulation stopped if it's loosing some sound. just by commenting this check it gets logged (as if it's really bothering to have it even logged)
/* removed by iAN CooG:
if (delta_t) {
return sound_error(translate_text(IDGS_SOUND_BUFFER_OVERFLOW_CYCLE));
}
*/
|
| |
Erhan
Registered: Feb 2005
Posts: 17 |
tested the beta binary @ forum64.de
svn 22527
x64sc
Crest Avantgarde Chars in all 4 borders part
glitches at rightmost character(in all c64 models selection)
with x64 it's ok |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22577-win32.zip (x64sc & x64)
- Gate $D023 glitch at MCM=0 -> 1 transitions by MC enable from cbuf. (fixes: Crest Avantgarde, part CIB upscroll)
Thanks for reporting erhan!
- Various changes and general clean up.
|
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22705-win32.zip (x64sc & x64)
- PAL-N (Drean) support (x64 & x64sc)
- Proper NTSC support in x64sc
- 1351 movement-in-both-ports bug fixed (hi TWW!)
- "native" screenshots in "Doodle" format
- Too many cart changes to enumerate
|
| |
TWW
Registered: Jul 2009
Posts: 541 |
Hi ;-]
It seems like the emulation is working. If only someone could test my "test-program" on the real thang I'm a happy customer !!! |
| |
Iapetus/Algarbi/Wood
Registered: Dec 2004
Posts: 71 |
Nice project. It would be awesome to have a good x128 version too. |
| |
TWW
Registered: Jul 2009
Posts: 541 |
is the 22705 the same binaries you can download from vice HP?
I am asking coz I want to run this on 64 bit os.. |
| |
null
Account closed
Registered: Jun 2006
Posts: 645 |
Btw, I've been having a strange issue (with both x64 and x64sc)... Namely warp mode doesn't seem to be used to it's full potential; I only get a speedup of about 300 - 400% on this machine (AMD Phenom II X4 955 @ 3.2GHz), but on my 2GHz Pentium 4 laptop, I easily get to 700% speed, if not much higher.
It doesn't bother me so much, but I'm still kind of curious what the problem might be. I know it happens under Windows Server 2003 and windows 7 ultimate (both 64-bit), haven't tested on Gentoo yet.
------------------------------------
http://zomgwtfbbq.info |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quote: is the 22705 the same binaries you can download from vice HP?
I am asking coz I want to run this on 64 bit os..
No, they are not the same. I only have a 32 bit msys/mingw setup.
We'd love to have someone set up some automatic builds of the trunk though, volunteers?
|
| |
Rubi
Account closed
Registered: Apr 2002
Posts: 48 |
Quote: Btw, I've been having a strange issue (with both x64 and x64sc)... Namely warp mode doesn't seem to be used to it's full potential; I only get a speedup of about 300 - 400% on this machine (AMD Phenom II X4 955 @ 3.2GHz), but on my 2GHz Pentium 4 laptop, I easily get to 700% speed, if not much higher.
It doesn't bother me so much, but I'm still kind of curious what the problem might be. I know it happens under Windows Server 2003 and windows 7 ultimate (both 64-bit), haven't tested on Gentoo yet.
------------------------------------
http://zomgwtfbbq.info
Enable "Render to DX primary surface" in the video settings. Should bring back some speed. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22731-win32.zip (x64sc & x64)
- NTSC sprite timing fixes (x64sc)
- Sound buffer overflow error replaced with warning
- Excessive drive cpu load during reset fixed
- Drive rotation improvements (fixes Snapshot)
- Weak bit emulations on 64-bit systems fixed
- VIA wrap handling rewritten
- Some cartridge fixes
Any improvements to NTSC emulation is dependent on further bug reports. |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
thanks for removing the sound buffer msgbox, about time ;)
-cartexpert now requires parameter (no way!), and can't be activated anymore from menu.
in monitor, io d000 in this build doesn't work but i see on SF there's already a patch for it so I will try to build it tomorrow.
and while at it, my email is iancoog(at)email.it if you can change it in the about box =) |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
"-cartexpert now requires parameter (no way!), and can't be activated anymore from menu."
that was unified, its correct. and the menu item is still there, it just moved (in gnomeui anyway, i give a damn about win32 =P). and first time for 20 years, expert cart actually even works =P
|
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
hrmf. I see now the source, changelog doesn't even mention it:
-cartexpert now requires a parameter for a binary, a new
-expert option is added to just enable it without providing a binary.
Anyway the fact that the option is in the menu doesn't mean that it works: selecting "expert/enable" it only gives "invalid cartridge", no matter if i specify -expert or not on the cmdline. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
the ui (assuming windows) is very broken and lagging behind atm (lots of things missing there).
however, using -expert it should work (and you dont need the menu item after that). |
| |
Grafitronix Plus
Account closed
Registered: Sep 2009
Posts: 3 |
I just tried to play Mars Saga (Ikati+Talent cracked version).
The r22731 version of x64, CPU became jammed with $AB39.
I jumped into the Monitor and it lists: ** Monitor 276 057
(C:$ab39)
I went into the "Gambling Establishment", chose "Laser Slots" and "Save my City Map" when it crashed. It was on the disk swap-- it did not make it to the prompt.
Not sure what else to do, but it ran fine up to that point. |
| |
Grafitronix Plus
Account closed
Registered: Sep 2009
Posts: 3 |
Forgot to mention:
Windows 7,
Intel Centrino Duo DualCore 2Ghz (both processors),
nVidia 8600GS M video card,
... in my HP Pavilion dv6000
(bios upgrade claims it's a dv6500).
The sc version gives warp mode of max:
349% 60fps
The x64 version gives fluctuating max:
1049% 65fps (every 30 or 40 seconds)
740% 62fps
694% 60fps
Strange, but it can be disheartening to know that the SC version is 'slower' than the non-SC version... (warp mode) and more compatible with the 1541 proper emulation.
IMHO, not bad. Working on a couple of other games to test, but will probably note something later in the week.
Godd going, so far... I know there's still work to be done. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
the 60fps hint that you were testing in NTSC mode ... are you sure the game in question works properly on a real NTSC setup ? |
| |
hyperactive
Account closed
Registered: Jul 2010
Posts: 5 |
Hi guys.
The original g64 game of "Mikie" from Action city will not load on this new version of vice (Version 2.2.5)
On vice 2.2, it loads without a problem, but on 2.2.5, I get a CPU jam error at 14b9. |
| |
hyperactive
Account closed
Registered: Jul 2010
Posts: 5 |
it's r22731 |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
I'm experiencing some weird slowdown after some minutes in warp mode (x64.exe). See a screenshot here
It's easily tested, attach a .crt (Action replay is fine, anything actually, I noticed this while trying Kompressmaster_v41.crt), disable options/sound playback, warp mode and do nothing. Drops from 3000% / 90fps (VS71 build, GCC build from TLR reaches 1600% / 70pfs) to 1% / 0fps here after 3-4 minutes.
I always disable sound to get max speed when trying native crunchers. Warp mode AND Options/Max speed/no limit get the same slowdown (unlike someone tried to tell me :P)
If it's of any help tracking down the problem, after I change borders from normal to full, hence reinitializing the video, speed returns normal, warp speed slightly reduced, but after some minutes I get back again 0 fps.
EDIT: seems like that when this slowdown occours, by deactivating True Drive Emulation, the speed goes back to normal, instantly. Actually now I retested and the slowdown doesn't occour at all if I disable TDE right after the loading finishes. So must be something related to TDE.
X64.exe from v2.1 doesn't have this problem, as well as x64sc.exe from latest build. will try also the official 2.2 later. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22786-win32.zip (x64sc & x64)
- drive rotation clock overflow issue fixed. No more slow down.
- grey dot emulation on 856x. (fixes Emusuxx0r+)
- fixed io dump monitor command, e.g "io d000".
- MACH5 cartridge emulation.
- misc cart fixes.
- CRT emulation added for NTSC and others.
Note: ColorGamma range changed. May need to set to 2200 manually.
EDIT: seems there is a rather fatal problem with the drive emulation in this build. Will post a new one soon. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22787-win32.zip (x64sc & x64)
- fixes drive breakage introduced in r22786.
|
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
Thanks, tried and so far so good, crunching always reaches the end in no time, warpmode seems even faster then before, possible? =)
I've left Vice in background for an hour in warp mode, checking now and then, and it's always around 3000%, no slowdowns. Cheers! |
| |
hyperactive
Account closed
Registered: Jul 2010
Posts: 5 |
Hi.
I have 4 games that don't work with this new version
Last Duel has a signature on track 25 that cannot be seen by the drive emulation.
the same goes for Miami vice with a signature on track 40.
Mikie also has signatures on track 17 or track 19 or both. curiously, this game works in vice 2.2, so something must have gotten broken in this new version.
Last but not least there's Road Runner and Wile E coyote, there is no protection but it uses a dodgy loader.
I am not able to attach files to this message, but if anyone wants to have a look at these games to see if they can patch up vice to get it working you're more than welcome. |
| |
TWW
Registered: Jul 2009
Posts: 541 |
2 Suggestions:
#1: Allow the 'inject PRG into RAM' to do actually that (Even in IO area). I know this wouldn't work on a real thing but for development purposes this makes things fast & easy.
#2: Allow to pass some argument which doesen't open a new x64 program if one already exist. Instead the emu can be reset and a new binary can be injected. This is alos for saving some seconds and development purposes.
Oh and I need this done before monday but that leaves you guys with aaaaallll weekend to figgure it out :-D (yeah that was a joke..) |
| |
Perplex
Registered: Feb 2009
Posts: 254 |
What TWW said.
(In VICE 2.2, I've tried the remote monitor feature in Linux, to issue commands to a running x64. It works well if you do it from a telnet session on the command line, but when I try to use a script to echo commands to the remote monitor port, x64 just crashes with no log output when my script is done and the connection is closed. Any ideas?) |
| |
hyperactive
Account closed
Registered: Jul 2010
Posts: 5 |
Hi there.
the current august 14 build of nibtools seems to have broken some of your weak-bit detection routines.
Some cyan loader titles after being converted from nib to g64 do not work any more, while others still do work. It looks like you'll need to do some more work on the drive emulation, and I'll have to reconvert all those cyan loader titles all over again and send them to Peter Rittwage. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
there is no "weak bit detection routine" in any emulator :)
what you describe sounds more like that recent change in nibtools is the core problem =P do these "new" g64 work in hoxs64 ? if they worked before in vice and hoxs, and now dont work in either, then pretty definately mnib is the problem :) |
| |
JCB
Account closed
Registered: Jun 2002
Posts: 241 |
There's a post from Pete Rittwage on C64pres from the end of last month saying the recent changes to nibtools had to be rolled back. He goes on to say the rips are probably ok just the conversion to G/D64 will need to be redone. |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
Inject to RAM fixed by tlr \o/
http://sourceforge.net/tracker/?func=detail&aid=3039021&group_i.. |
| |
TWW
Registered: Jul 2009
Posts: 541 |
*THUMBS UP!*
edit: Any chance of someone making a binary? |
| |
Stainless Steel
Registered: Mar 2003
Posts: 966 |
Any chance for snapshot functionality? |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r22983-win32.zip (x64sc & x64)
- don't access charROM on Ultimax mode VIC-II fetches (fixes hidden part in Pain)
- remote monitor line ending fixes (should work better on win32/OSX now)
- various fsdevice & cart fixes
- on-the-fly ddraw/DX9 switching
- autostart mode Inject-to-RAM does what it says
- IDE64 4.1 support
- true aspect ratio for win32
- x64sc snapshot support (please test!)
- CIA timer B bug disabled by default (use -ciamodel 2 to get the classic behaviour)
Note 1: snapshot format has changed, and may change until the release.
Note 2: x64 previously used to emulate the 6526 CIA with timer B bug.
We tested various machines but didn't find one with the timer B bug yet.
These tests can be used to find out: cia-int-r04.zip
If you have a combination of CIA1 | CIA2 | CPU different from...
MOS/6526/3884 | MOS/6526 R4/3283 | MOS/6510 CBM/3184
MOS/6526/4682 | MOS/6526R-4/2484 | MOS/6510 CBM/2484
MOS/6526/1685 | MOS/6526/1685 | MOS/6510/1385
CSG/6526A/3590 216A | CSG/6526A/3590 216A | CSG/8500/3390 24 ...then please run the tests and report back!
I'm especially interested in the 8521R0. |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
great! Already tested Pain/Agony hidden part and works only in x64sc+AR4.2. Isn't really possible to fix it for x64 too?
Freeze also works but not saving the attached cart, you have to attach AR4.2 before loading a freeze from that hidden part, oh well =)
In x64, trying to attach the AR4.2 I got a signal 11 a couple of times but I can't reproduce it systematically :( Seems to occour if I start x64 and I attach it while the screen is black, but not always. I hate random bugs =)
|
| |
Durandal
Registered: May 2006
Posts: 30 |
Good to finally see Drean emulation on X64! :)
You can count on me if there's any test that needs to be done on a real Drean machine.
My Drean machine (CIA1|CIA2|SID|VIC|CPU):
MOS/6526/4485 | MOS/6526/4485 | MOS/6581 R3/0886 | MOS/6572R0 | MOS/8500 R3/5185
|
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quoting The_WOZYou can count on me if there's any test that needs to be done on a real Drean machine.
Thanks, PM sent. ;)
|
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Anyone in NTSC land that would care to run a few test programs?
Please PM me. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:great! Already tested Pain/Agony hidden part and works only in x64sc+AR4.2. Isn't really possible to fix it for x64 too?
it is fixed in x64 too (cart system really behaves exactly the same in x64 and x64sc) - thing is that the whole code is very unstable, and starting the hidden part often doesnt work (just try a few times).
the AR this is supposed to work with is the 7.3 "atrax" hack btw :) |
| |
hedning
Registered: Mar 2009
Posts: 4595 |
I can run tests on my Drean C64C, silver badge c64, C64G and my SX-64 if that helps anyone. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quote: I can run tests on my Drean C64C, silver badge c64, C64G and my SX-64 if that helps anyone.
Thanks, PM sent! |
| |
Frantic
Registered: Mar 2003
Posts: 1627 |
I may very well have been dreaming but I *think* Goto80 mumbled something about recently getting a Drean 64.. |
| |
Cruzer
Registered: Dec 2001
Posts: 1048 |
Will it have the ability to remember that I want true drive emulation every time I open a D64? |
| |
4mat
Registered: May 2010
Posts: 63 |
Is 1571 emulation broken in this? When I try to attach a d71 file in both .exes I get an error box that says "Recieved signal 11 ()".
|
| |
Durandal
Registered: May 2006
Posts: 30 |
I took a look at the current X64 sources and noticed that the Drean clock parameters are based on the patch I submited to VICE Team back in 2006, but I had made a mistake and the clock value was wrongly stated as 1.014 MHz.
2 years later, in 2008 I noticed the error and, with better data, found the correct clock frequency to be 1.02344MHz
Thus, c64.h must have
#define C64_PALN_CYCLES_PER_SEC 1014000
modified to:
#define C64_PALN_CYCLES_PER_SEC 1023440
See here:
http://www.solid-state.com.ar/index.php?entryid=11
and here:
http://www.solid-state.com.ar/index.php?entryid=86
for the data backing up this clock value. |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
Quote: Is 1571 emulation broken in this? When I try to attach a d71 file in both .exes I get an error box that says "Recieved signal 11 ()".
If I attach a d71 with 1541 selected all I got is a drive error while trying to load the dir. Selecting 1571 and attaching d71 works here. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
New version: x64sc-r23120-win32.zip (x64sc & x64)
- distribute restore key presses randomly across a frame
- Normal/Full/Debug borders switchable on the fly (only x64sc)
- Improved accuracy of NTSC emulation (sprite/light pen coordinates)
- Improved accuracy of PAL-N emulation (sprite/light pen coordinates, clock frequency)
- as usual, too many cart fixes to mention
Special thanks to the people who volunteered to run test programs that provided crucial information:
- hedning, The_Woz and Thierry for Drean (PAL-N) testing.
- Jason Compton for testing a great many NTSC machine configurations.
All test programs (PAL/NTSC/NTSCOLD/PAL-N) are available in our subversion repository for other emulator writers to use:
http://vice-emu.svn.sourceforge.net/viewvc/vice-emu/testprogs/
Please consider contributing your findings. |
| |
Frantic
Registered: Mar 2003
Posts: 1627 |
@tlr: Are there any particular test programs that would be helpful for you if people tried out?
//FTC |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
This still applies:
Quote:Note 2: x64 previously used to emulate the 6526 CIA with timer B bug.
We tested various machines but didn't find one with the timer B bug yet.
These tests can be used to find out: cia-int-r04.zip
If you have a combination of CIA1 | CIA2 | CPU different from...
MOS/6526/3884 | MOS/6526 R4/3283 | MOS/6510 CBM/3184
MOS/6526/4682 | MOS/6526R-4/2484 | MOS/6510 CBM/2484
MOS/6526/1685 | MOS/6526/1685 | MOS/6510/1385
CSG/6526A/3590 216A | CSG/6526A/3590 216A | CSG/8500/3390 24 ...then please run the tests and report back!
I'm especially interested in the 8521R0.
Is the 8521R0 really the mysterious timer B bug device?
And does it behave like a 6526 or a 6526A in other respects? |
| |
Fungus
Registered: Sep 2002
Posts: 616 |
Is there a 64 bit build of this I can test? |
| |
TWW
Registered: Jul 2009
Posts: 541 |
Quote: No, they are not the same. I only have a 32 bit msys/mingw setup.
We'd love to have someone set up some automatic builds of the trunk though, volunteers?
In reply to: Is there a 64 bit build of this I can test? |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
It was quite long since anyone dropped a build so here's one: x64-x64sc-xvic-r25693-win32.zip
Changes are too numerous to be listed. |
| |
The Human Code Machine
Registered: Sep 2005
Posts: 110 |
Thx for sharing! After invoking the integrated monitor it's not possible to close it and it won't work after trying to close. Any idea? I tested both x64 and x64sc. |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
I wonder when the win32 port will be fixed from this WRONG behaviour, if it works only for groepaz, let the window open only in the damn linux port. :P |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
THCM, iAN: There's a bug on that issue here, comment if you agree on its importance: http://sourceforge.net/tracker/?func=detail&aid=3463496&group_i..
(iAN already did)
And please open new bugs for any other anomalies. Be sure to cite the revision somewhere. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
ian: the supposed fix is right in the bugreport.... now if anyone would bother trying if it actually works, then it maybe could get applied =P |
| |
TWW
Registered: Jul 2009
Posts: 541 |
Any chance of someone compiling a Win 64bit version? |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quote: ian: the supposed fix is right in the bugreport.... now if anyone would bother trying if it actually works, then it maybe could get applied =P
I've tested it somewhat and applied it:
x64-x64sc-xvic-r25705-win32.zip
Please test extensively! |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
x & alt-f4 now close the window, but in stepping mode (z or n) it's causing a fast flicker, window close/reopen, very annoying for me and I don't understand the whole change what's for, annoying people? I don't think so. please explain the need of this change and why it should affect normal operation. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quote: x & alt-f4 now close the window, but in stepping mode (z or n) it's causing a fast flicker, window close/reopen, very annoying for me and I don't understand the whole change what's for, annoying people? I don't think so. please explain the need of this change and why it should affect normal operation.
@iAN: I wasn't involved in the change but I assume there is some purpose on other platforms at least.
Win32 breakage must be a side effect but GUI code was never my forte.
Please file a bug. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:I don't understand the whole change what's for, annoying people?
yes, only to piss you off.
Quote:please explain the need of this change and why it should affect normal operation.
if you dont want changes, just keep using vice 1.7.
other than that, the idea is to finally clean up the whole mess a bit and actually make all guis adhere the internal variables correctly and maybe even work like they should. (the windows gui always identified itself as "i can keep the monitor always open" - although it appearently can not do that at all) and yes, that will break intermediate development versions from time to time. deal with it. if after the proposed fix it behaves like you say, the windows gui is actually even more broken than suspected. (and uses the internal variable wrong at other places too - else the patch should have reverted everything to the old behaviour completely) prepare for more breakage in future intermediate versions - or become the dedicated win32 gui maintainer who will fix them right when they pop up.
here is how its supposed to work, btw: https://sourceforge.net/apps/mediawiki/vice-emu/index.php?title.. (it may be interesting to know how exactly you think it should work, please be precise)
|
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
Hi,
I have tried my release of robocop in VICE version R25705 win32 and the game crashed in the second level (the confrontation with the hostage holding the girl) when I holding fire button and then go fast left then right still firing..
RoboCop +2
It dosn´t happened with CCS64 emulator v3.8. In some other version of VICE (don´t know version exactly - some older maybe 2.2 or 2.1) it doesn´t happened. I would like to know if there is a bug in game or in the VICE emulator and what exactly happened.
Can someone solve this? .)
Thanks
eLK |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
does it work on the real thing then? :) |
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
Sorry but I couldn´t test it on real machine :) |
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
Aha :) Problem solved :o)
It is because of the emulation of joystick. The problem is that the game didn´t count with firing and have both the directions holding - as it is possible in emulation yet. I think so it is the bug in emulator :) that it is possible to have the both the dirrections of joystick holding together just that is on real joystick impossible :) |
| |
MagerValp
Registered: Dec 2001
Posts: 1055 |
Actually, it's uncommon but perfectly possible on real hardware. I'm sure there were cheap joysticks where the switches stuck. If you want to reproduce, just open a WICO joystick and squeeze the leaf switches. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
a) yes its very possible on the real thing. on many if you press down the stick into the base, you can easily close opposite switches.
b) if that happens in vice, you have configured it to allow it. in the joystick settings there is "allow opposite directions" - disable it (which is also the default). |
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
Okay - sorry - you are right! It helped .)
But the behavior is then useless. If you are holding fire and go right then press left still holding right and fire robo turns left, but when you release the button left still holding right and fire robo doesn´t turn right and you have to press it again.. This is useless if you are playing games like this. I like the CCS64 behavior, it works perfectly! Try the differences :) This is why I don´t like VICE and because of the terrible "monitor". :( |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
for playing games i just use my c64. or at least a real joystick.... no problems then whatsoever. (and ccs was cool in 1996, not so much today) |
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
Not exactly the true :) and it is not an argument why it works wrong. I think every emulator has its own.. If the programmers they coding them should learn together it does exist the perfect emulator yet. But - even the VICE is very advanced, it has a long way to perfection :o) The cracked things works right but the originals in G64 or NIB format don´t work right at all. Some games don´t load well. NIB can be only used in new german MICRO64 which has the best drive emulation I have ever seen. :) So I think it has not as perfect emulation for example of 1541 drive even it has many settings there then e.g. old CCS64 even if it is the obsolete thing yet as you already said. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:The cracked things works right but the originals in G64 or NIB format don´t work right at all. Some games don´t load well. NIB can be only used in new german MICRO64 which has the best drive emulation I have ever seen. :)
the patch from bero (micro64) which fixes all that is about to be merged into trunk this night - it actually beats hoxs64 in drive emulation now =P |
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
Great! WOW! Oh - then - it could happen to be my favorite again :o)
Please, do something with the joystick then and - I think it should be nice to extract the drive sound from the exe and put it somewhere to make exe smaller and to be possible to change in case someone have better :) |
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
And avoid the fix in Micro64 build 670 that causes MICRO64 crashes when opening NIB. I have posted it to MICRO64 forum yet. I think it wasn´t good idea to strictly force the maximum value because of better compatibility. But I am not the coder and I only see what it does before and then.. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:Please, do something with the joystick then
if you think there is a bug, please report it on the tracker. |
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
Quote: Quote:Please, do something with the joystick then
if you think there is a bug, please report it on the tracker.
I don´t have an account on sourceforge and when I tried to create it no activation e-mail came back.
I will try it later again but if you can add the report what I discovered, send it please to the tracker to be modified. Maybe it is not a bug at all, but it is surely freaky behavior and now it is useless to playing games like this one. It could happened that you press both the directions together on keyboard but now then you have to press it again if it happened.
Better behavior should be like:
You pressed fire and left, then you add right - so the emulator could test the last direction pressed and if it is the opposite emulator release the left and press the right (still holding fire so - no changing it). You can then release left and - you have pressed nothing so nothing should happen, still holding right and fire and so on.. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:but if you can add the report what I discovered
to be honest, i dont quite understand what the problem is at all - and since i am not using windows at all either and thus can not even reproduce what you are saying, that would be like adding random stuff i know nothing about to me, so i rather dont =P |
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
I am succesfully registered and have activated account to sourceforge yet, so I will there describe the problem better. But also thanks for report :) |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:x & alt-f4 now close the window, but in stepping mode (z or n) it's causing a fast flicker, window close/reopen
this is actually correct - if the gui can not keep the monitor always open, then it will have to close it (its the same in GTK now, except the flickering is caused not by close/reopen but by transfering focus forth and back to the emu window). conclusion: we need some special case handling for singlestepping - something other than the horrible hack that is gone now though :) |
| |
AmiDog
Registered: Mar 2003
Posts: 97 |
Quote: Quote:The cracked things works right but the originals in G64 or NIB format don´t work right at all. Some games don´t load well. NIB can be only used in new german MICRO64 which has the best drive emulation I have ever seen. :)
the patch from bero (micro64) which fixes all that is about to be merged into trunk this night - it actually beats hoxs64 in drive emulation now =P
Silly question perhaps, but what about the KryoFlux patches? Are their changes basically the same as bero's, or are they conflicting, or do they complement each other? |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
bero has investigated the SPS patches and took the bits that he found useful (and added quite a bit on top). according to our tests there is nothing that works on the SPS version that doesnt work in trunk.
in other news, i fixed some of the monitor annoyances the other night. it does properly not close when singlestepping now and should basically behave as expected also in windows now :) |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quoting Groepazin other news, i fixed some of the monitor annoyances the other night. it does properly not close when singlestepping now and should basically behave as expected also in windows now :)
Win32 build: x64-x64sc-xvic-r25789-win32.zip
Seems to work nicely, good work! |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
way to go, had the time to test it now, and acts as expected. thanks for your time. I have no time to test it throughly now, HVSC release is imminent and I should focus on it only =)
Something for next weeks |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
tlr: you could undo the previously suggest console_can_stay_open = 0 (and set to 1 again) and see wether it still does.... (x should close the monitor as before, g should exit but keep the monitor open. additionally changing the "keepmonitoropen" resource to 0 should completely revert to old behaviour). |
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
Quote: Quoting Groepazin other news, i fixed some of the monitor annoyances the other night. it does properly not close when singlestepping now and should basically behave as expected also in windows now :)
Win32 build: x64-x64sc-xvic-r25789-win32.zip
Seems to work nicely, good work!
In this release joystick emulation on PC keyboard works much better :o) and is much more usable ;) but - still not exactly perfect ;)
If you press the left while still holding right it doesn´t change anything but leave right holding (it is not as wrong - but better it was if it is left because you changed to left - so you like to go left) But it happen if you then release right..
///This behavior causes problems when you holding right and then you press down (plus holding fire) - sometimes it is still only right (plus fire) but it should be right-down (plus fire). And to me it happen randomly.///
text in /// not exactly describe the problem - I found that it only reacts slow to me but it works and is now usable.. But still better to behave like I described it upper.
And I made a table where I describing how it should work. You can download it here:
http://www.uschovna.cz/zasilka/J1YY5KJDDCTXJVHM-PGU |
| |
Erhan
Registered: Feb 2005
Posts: 17 |
hi,
i think this has some problems after winvice 2.3 stable and these x64sc betas..
win 7 x86, local=TR
file - attach disk image - choosing a disk image above; below, image contents are visible for the first time.
again, file - attach disk image, previous disk image is selected but image contents are empty (for windows XP it's normal)
|
| |
6R6
Registered: Feb 2002
Posts: 244 |
Vice true drive emulation bugs on my filename loaders.
When flipping disk sides you need to wait 1-2 seconds before pressing space to continue loading.
To test :
Download this zip.
http://www.n0stalgia.org/forum/download/file.php?id=93
Move the even (00 02 04 etc) filenames to one disk.
And move the odd (01 03 05 etc) filenames to another disk.
This way you will have to flip the disk every time you load a new pic. Notice you have to wait 1-2 seconds before it can load the file.
This doesn't happen on real hw.
|
| |
6R6
Registered: Feb 2002
Posts: 244 |
Another bug:
When I have a d71 or d81 image attached to vice (and 1571 or 1581 drive emulation selected).
Then I set true drive emulation to OFF because I want to load with "virtual drive emulation".
Now when I do a M-R to detect the drive, before starting loader, vice will in 90% of my tries return data from the 1541 drive rom. Making my loader belive the drive is a 1541 drive and that my 2bit fastloader is supposed to be installed.
Vice shouldn't do this I believe. It should only return the error channel message "73, virtual drive emulation....".. not data from the 1541 drive rom.
Everything works fine when i do the same thing using D64 and 1541 drive emulation.
|
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:Now when I do a M-R to detect the drive, before starting loader, vice will in 90% of my tries return data from the 1541 drive rom.
thats a known problem (also dont spill bugs into this thread please, add them to the tracker. the one who is fixing the drive stuff doesnt read this thread at all, for example.) |
| |
ϵʟʞ
Registered: May 2012
Posts: 26 |
Can someone compile the newest v.2.3.19 release 25859 and put here link to Windows build, please?
Thanks |
| |
dink
Account closed
Registered: Mar 2012
Posts: 30 |
I can't get it to compile, ./configure errors out on /src/lib/p64 .. any ideas? |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quote: I can't get it to compile, ./configure errors out on /src/lib/p64 .. any ideas?
This might work:
$ cd vice/
$ make distclean
$ ./autogen.sh
$ ./configure
$ make
$
|
| |
hedning
Registered: Mar 2009
Posts: 4595 |
I'm still willing to test stuff on my Drean if people need it, for making emulation better etc. Just PM me. Or post here.
I have it hooked up. |
