Reply 1840 of 2419, by bruenor41
I think it is related to changes in svn 4130
I think it is related to changes in svn 4130
You're mounting the image as if it has a FAT filesystem, but it has none, which is why it is rejected by the added sanity checks. If you specify drive 2 instead of C and add the "-fs none" switch then mounting should work fine for your purpose. Not the first time someone has gotten hung up by this, so perhaps the error message could be improved to suggest specifying no filesystem when mounting.
Hi there new here with a quick question .. i usually use dosbox on my crt monitor and it switches resolutions correctly ie 70hz dos mode
was playing with dosbox-x but i cant seem to get it to play ball with my crt all i can get is 60hz scaled and it looks terrible on a crt i have tried every setting no matter what i get fluffy looking screen in 60hz .... anyone running on a crt without issues ? thanks
wrote:You're mounting the image as if it has a FAT filesystem, but it has none, which is why it is rejected by the added sanity checks. If you specify drive 2 instead of C and add the "-fs none" switch then mounting should work fine for your purpose. Not the first time someone has gotten hung up by this, so perhaps the error message could be improved to suggest specifying no filesystem when mounting.
Ok, so 2 issues.
1st. this breaks the workflow of going from a blank drive, to partitioned to formatted.
2nd. why is it OK to have a blank drive with no filesystem, but not a partitioned drive with no filesystem?
Perhaps instead of rejecting such a case, it should just give a message saying that no filesystem, or no supported filesystem was found, and may still need to be created. And attach the device anyway.
But thanks for the suggestion, I will change my config to specify drive 2 instead of C as suggested.
wrote:why is it OK to have a blank drive with no filesystem, but not a partitioned drive with no filesystem?
I'm not sure. I believe you must include the "-fs none" switch for a blank image as well, at least in official source where the added sanity checks were adopted from. In any case, it will always be problematic if DOSBox is using the FAT drive backend on an image that does not have a FAT filesystem.
I guess you would like the FAT or no FAT choice to be automatic, and perhaps it could be, but image mounting currently assumes FAT when a drive letter and no -fs switch is specified.
I am still scratching my head... DOSBox-X (or any other DOSBox patched with NE2000) only gets to the internet when I use my ethernet connection. It fails with wifi. I know someone has told me before the reason why, but it just doesn't look right to me. Could someone explain in simple terms why it doesn't work with wifi? Because, there is a scenario where this doesn't look right to me:
When I use DOSBox-X in a VM. The guest linux has only ethernet, and the host macOS has only wifi. DOSBox-X compiled and running from the guest machine can connect to the internet despite of this configuration. Strange....
"Design isn't just what it looks like and feels like. Design is how it works."
JOBS, Steve.
READ: Right to Repair sucks and is illegal!
Patch was applied to only one of the two key sets:
void KEYBOARD_AddKey2
case KBD_printscreen:
extend=true;
if (pressed) { ret=0x12; ret2=0x7c; }
else { ret=0x7c; ret2=0x12; }
void KEYBOARD_AddKey1
case KBD_printscreen:
KEYBOARD_AddBuffer(0xe0);
KEYBOARD_AddBuffer(42 | (pressed ? 0 : 0x80));
KEYBOARD_AddBuffer(0xe0);
KEYBOARD_AddBuffer(55 | (pressed ? 0 : 0x80));
wrote:I am still scratching my head... DOSBox-X (or any other DOSBox patched with NE2000) only gets to the internet when I use my ethernet connection. It fails with wifi. I know someone has told me before the reason why, but it just doesn't look right to me. Could someone explain in simple terms why it doesn't work with wifi? Because, there is a scenario where this doesn't look right to me:
When I use DOSBox-X in a VM. The guest linux has only ethernet, and the host macOS has only wifi. DOSBox-X compiled and running from the guest machine can connect to the internet despite of this configuration. Strange....
Winpcap doesn't support wireless. npcap does. I don't know if the NE2000 patch has been updated with npcap support. When using DOSBox in a VM then it's not using wireless inside the guest it's using wired.
According to the website npcap only works on Windows 7+. Not sure how true that is would need to do some tests but considering NDIS 6 started with Vista then possibly Vista would be the minimum for npcap.
Winpcap works on 9x-8.1
Mabye the patch can be updated to support both assuming the NE2000 patch doesn't work with npcap.
Thank you for the explanation! So, no “love” for macOS in this NE2000 patch? Well...
"Design isn't just what it looks like and feels like. Design is how it works."
JOBS, Steve.
READ: Right to Repair sucks and is illegal!
wrote:Hi there new here with a quick question .. i usually use dosbox on my crt monitor and it switches resolutions correctly ie 70hz dos mode
was playing with dosbox-x but i cant seem to get it to play ball with my crt all i can get is 60hz scaled and it looks terrible on a crt i have tried every setting no matter what i get fluffy looking screen in 60hz .... anyone running on a crt without issues ? thanks
Video modesetting in fullscreen is disabled in DOSBox-X.
Hi, I think I come across some bug in the dosbox-x debugger.
I want to write CF opcode (IRET instruction) in some memory address, so I put in the debugger window:
SM xxxx:yyyy CF
However, the byte in memory address xxxx:yyyy changes to 00 instead of CF
No problem with other opcodes so far.
I'm using dosbox-x 0.82.22 && Debian testing.
Thank you and keep up the good work.
wrote:Hi, I think I come across some bug in the dosbox-x debugger. […]
Hi, I think I come across some bug in the dosbox-x debugger.
I want to write CF opcode (IRET instruction) in some memory address, so I put in the debugger window:
SM xxxx:yyyy CF
However, the byte in memory address xxxx:yyyy changes to 00 instead of CF
No problem with other opcodes so far.
I'm using dosbox-x 0.82.22 && Debian testing.
Thank you and keep up the good work.
You may be using a binary release made before the bug with SM and xxxx:yyyy memory addresses was fixed. It seems to work here with the latest commit.
Hi, window title says "dosbox-x 0.82.22" so I'm pretty sure it's the latest version.
Anyway I've downloaded again https://github.com/joncampbell123/dosbox-x/ar … v0.82.22.tar.gz and I've compiled it from a clean directory and problem remains.
I just realized that the byte dosbox-x overwrites seems to depend on the memory address. For example
sm 1000:772C CF -> byte becomes 00 instead of CF
sm 03C0:01D0 CF -> byte becomes 01 instead of CF
If you need some other tests just let me know. Thank you! 😀
wrote:Hi, window title says "dosbox-x 0.82.22" so I'm pretty sure it's the latest version. […]
Hi, window title says "dosbox-x 0.82.22" so I'm pretty sure it's the latest version.
Anyway I've downloaded again https://github.com/joncampbell123/dosbox-x/ar … v0.82.22.tar.gz and I've compiled it from a clean directory and problem remains.
I just realized that the byte dosbox-x overwrites seems to depend on the memory address. For example
sm 1000:772C CF -> byte becomes 00 instead of CF
sm 03C0:01D0 CF -> byte becomes 01 instead of CF
If you need some other tests just let me know. Thank you! 😀
I fixed the bug after the binary release, compile the latest commit from master.
Great! No hurries, I can wait for the next official release.
wrote:Great! No hurries, I can wait for the next official release.
Clearly if you can compile the source code from the release section, you can also git clone from the project page as well. The latest commit to master has the latest fixes and improvements.
git clone https://github.com/joncampbell123/dosbox-x.git
Sure, but I generally prefer to avoid master branch.
I can use DOSBox debugger in the meantime. Every month or so a new release of DOSBox-X is out so no hurries for me.
I'm happy to see this have already been fixed. Thank you!
Does anyone know why, when I boot from a disk image, dosbox shows me a boot menu like:
"Booting from drive C..."
"F1 . . . Dos"
"F5 . . . disk 2"
"Default: F1"
Then after a while it boots my image. F5 does nothing. I would like to know why it happens and if there is a way to remove so i can boot straight to my image...
"Design isn't just what it looks like and feels like. Design is how it works."
JOBS, Steve.
READ: Right to Repair sucks and is illegal!
I have been using dosbox X for a little while now, and it works great. Thank you!
Is it possible to attach two soundblaster devices for use by Sequencer Plus Gold? The software supports multiple soundblaster cards by loading its FM synth driver, SAPIFM1, twice. Here is an example:
SAPIFM1 /ADDR:230 /PORT:3
SAPIFM1 /PORT:2