Hallo,
ich habe jetzt nach mehreren Versuchen LinuxCNC auf meinem alten Rechner installiert. Die großen Distros gingen nicht weil es nicht möglich war vom Stick zu booten und für DVD waren sie zu groß. Ich habe dann MiniOs gefunden, das konnte ich problemlos installieren und auch
den Rt-Kernel und LinuxCNC, aber es war nicht möglich im Setup ein deutsches Tastaturlayout zu installieren. Dann habe ich MX Linux 
gefunden, damit konnte ich alles und auch Mesaflash installieren. Jetzt ist im Installationsprogramm von LCNC die 7I96s vorinstalliert,
ich habe aber die 7I96, wie bekomme ich die 7I96 ins Installationsprogramm, oder kann ich einfach die 7I96s nehmen?
MfG
Hans

tommylight wrote:

AMD graphic cards do not work with new RT kernels, 4.19 is the last one that worked.
Does the CPU have an integrated graphic?
Intel works fine with RT kernel, NVIDIA works but causes issues so is unusable, AMD goes to black screen, al this with 6.13 RT kernel.
Had no time to test the 7.nn kernels yet.
 

There is some bug in newer amdgpu driver, but works.
My Ryzens work with all Debian13 RT kernels (computer I write from is Ryzen 5500U, kernel 6.12.95+deb13-rt-amd64).
My AMD A10-7800 work with 6.12.63+deb13-rt-amd64 RT kernel.
With newer than 6.12.63+deb13-rt-amd64 works too, but only with low or medium resolution, enough for Axis or Gmoccapy.
The solution for blank screen is boot with nomodeset kernel parameter (gpu runs but without hardware acceleration) and change GRUB's and system's screen resolution to low, restart and test higher resolutions.

the CPU is a xenon 5650

It’s the same situation for me; I solved it by switching to an AMD "display-only" card.

oh, that explains it i guess. And no, MOBO does not have integrated graphics. guess im buying another graphics card

1. try other pins, looks like you are on two of the HDMI pins (there are filters on the board)
2. set an inverter/modifier to the step pin: (Pins -> Step -> Modifier + -> invert)
3. check that Pulse_LEN and Dir_Delay set to default (4.0 / 0.7)


I have several machines running on this, and there are a few other users as well, all with different drivers and motors; so far, it’s always worked.

EDIT: what is on the breadboard ?

EDIT2: I’ve gone ahead and ordered one of those DM542s just to be on the safe side,
it’s due to arrive on Tuesday or Wednesday

systemctl get-default

it reply's with graphical.target

oh, by the way, im using inductive switches.

I was thinking the same thing about if it was "normal" I would have been able to find those posts. So i just assumed maybe I just didnt set something up right yet (like the slow parameter you mentioned, Ill give that a try). thanx

DIR/PULSE issue resolved, thank you!! (2026-08-19)
I tested with the Branches dev from GitHub and the Tang Nano 9k, but the motor still couldn't return to the zero position after stopping. You must use the same DM542 stepper motor driver to reproduce this problem.

 
 
 
 

 
 

glxgears runs fine from a terminal. i was using a couple Nvida cards early on, but they had other issuse with scrambling the screen. so I was told an AMD card is prefered for Linuxcnc anyway, because the proper drivers are baked in. Thats why i got that amazon one.
I'll try and download the ISO again and use a different USB. but i cant for the life of me find anything about secure boot in my bios.

rubes@x3cnc:~$ sudo dmesg | grep gpu
[sudo] password for rubes:
[ 3.922158] [drm] GART: num cpu pages 524288, num gpu pages 524288
[ 3.948192] radeon 0000:03:00.0: fence driver on ring 0 use gpu addr 0x0000000100000c00
[ 3.948194] radeon 0000:03:00.0: fence driver on ring 1 use gpu addr 0x0000000100000c04
[ 3.948196] radeon 0000:03:00.0: fence driver on ring 2 use gpu addr 0x0000000100000c08
[ 3.948197] radeon 0000:03:00.0: fence driver on ring 3 use gpu addr 0x0000000100000c0c
[ 3.948198] radeon 0000:03:00.0: fence driver on ring 4 use gpu addr 0x0000000100000c10
[ 3.948609] radeon 0000:03:00.0: fence driver on ring 5 use gpu addr 0x0000000000075a18
[ 3.968652] radeon 0000:03:00.0: fence driver on ring 6 use gpu addr 0x0000000100000c18
[ 3.968654] radeon 0000:03:00.0: fence driver on ring 7 use gpu addr 0x0000000100000c1c
[ 11.149261] amdgpu: Virtual CRAT table created for CPU
[ 11.149286] amdgpu: Topology: Add CPU node
rubes@x3cnc:~$

I wouldnt begin to know how to decipher that.

Shared home/limit switches are pretty common so I would expect more
complaints about this if this always happened.

I wonder if this is possibly  caused by contact bounce.

Have you set the switch input inm pins to "slow" ?