-
-
Notifications
You must be signed in to change notification settings - Fork 42
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Most MiSTer jtcores break experimental DV1 support and put MiSTer into a bad state #535
Comments
Additional notes: Does not repro when I follow the same steps but use an official DV1-updated core (ex: "Tropical Angel" arcade) instead of "Quartet". Does not repro when I follow the same steps but use an official non-DV1-updated core (ex: NeoGeo) instead of "Quartet". Does not repro when I follow the same steps but use the latest After entering the bad state by loading "Quartet", loading a core that has been updated to the latest official framework (and which normally works with |
This comment from @paulb-nl is likely helpful for jotego: |
That explains a lot! Thanks! |
Nice catch! |
I have applied the change in f32092a |
If it will help, I can test a private build against my RetroTink-4K and report back. |
Hey @jotego , if you like, I'm more than happy to send you a RT4K. Thanks a lot for taking a look at this. |
Thanks, @mikechi2, I am expecting to borrow one next week. Are you involved in the development of RT4K? |
Yes you could say that ;) I just sent you a DM on Twitter since we follow each other. |
All MiSTer files updated today in JTBIN. Let me know how it goes. |
Looks great on my end! Thank you! |
This is a reliable repro, starting from a powered-off state, using
MiSTer_unstable_240203_16710d
and the latest stablemenu.rbf
(from 2023).MiSTer.ini
hasdirect_video=1
andvideo_mode=8
in the default section, but it hasdirect_video=0
for the[menu]
section.Repro Steps:
menu.rbf
loads. RetroTink-4K OSD correctly displays 1920x1080. Good.MegaDrive_unstable_240127_16710d.rbf
and load Sonic the Hedgehog. RetroTink-4K OSD correctly displays240p (MegaDrive)
and correctly auto-decimates and auto-crops the image. Good.Actual Result:
RetroTink-4K does not detect that the video source has changed (its OSD still displays
MiSTer 240p (MegaDrive)
and still displays the same SPD Infoframe data), and it does not correctly auto-crop or auto-decimate the image. It appears that MiSTer is still sending the SPD infoframes for the MegaDrive core, despite it no longer being the running core.Expected Result:
RetroTink-4K should detect a video source change, see a new SPD infoframe, and stop applying the MegaDrive core's
DV1
settings.Repro Steps (continued):
menu.rbf
.Actual Result:
RetroTink-4K detects a video source change and displays
HDMI 1080p
in its OSD, but the image is cropped and scaled incorrectly. RetroTink-4K appears to still be applying the MegaDrive core'sDV1
settings.Expected Result:
RetroTink-4K should stop applying any auto-decimation or auto-cropping because the
menu.rbf
core is explicitly configured to not use DirectVideo.Repro Steps (continued):
menu.rbf
.Actual Result:
RetroTink-4K detects a video source change and displays
HDMI 1080p
in its OSD, but the image is cropped and scaled incorrectly. RetroTink-4K appears to still be applying the MegaDrive core'sDV1
settings.Expected Result:
RetroTink-4K should stop applying any auto-decimation or auto-cropping because the
menu.rbf
core is explicitly configured to not use DirectVideo.Final observation:
Most of the jtcores cause MiSTer to end up in a permanently bad state in which it is "stuck" using stale
DV1
data. The only way I've found out of this state (so far) is to hard power cycle MiSTer. I have not (yet) been able to repro this same issue with any non-Jotego cores.The text was updated successfully, but these errors were encountered: