raidenii wrote: ↑Tue Feb 09, 2021 9:29 amIf you delete the remaps, does that issue persist?
Also try https://filetrip.net/gba-downloads/home ... 32301.html to see if the button test is correct - unfortunately I don't have OGS so I can't reproduce this.
Can you check again, it's called MPV in the Core Downloader.
Can you elaborate on the wifi issues? Were you able to see ssids? Is the problem that you were having trouble to connect? If so latest image fixes connecting to WiFi endpoints with special characters either in the name or passphrase.silverthornne wrote: ↑Tue Feb 09, 2021 11:58 pmThis is a very good distribution for the OGS but it was giving me issues with my Wifi dongle that I don't experience with the official release, so it's not my daily driver. Will keep an eye out for updates though!
It was more a matter of inconsistencies. The kind of thing that I know drives developers insane because I'm a developer
Yeah I know how annoying it can be with spotty Wifi. RRVL latest should have the same drivers as stock kernel.silverthornne wrote: ↑Wed Feb 10, 2021 1:43 amIt was more a matter of inconsistencies. The kind of thing that I know drives developers insane because I'm a developer
The first time I booted it up, it recognized my Wifi dongle and connected to my network without issues so I connected via SSH and configured it fine. However, it's been a very frustrating issue after that first reboot. It sees networks sometimes, it doesn't most of the time. I have to remove the dongle and reconnect it, refresh the network list, even hard reboot to get it to work now. Sometimes it connects, I SSH in, and then it suddenly disconnects and when I go to check on the Network, it has an IPV6 IP address instead of an IPV4 address. It can't do manual updates when that happens, so I think that IPV6 address is just a placeholder because it actually lost communication with the dongle somehow. That's with a Wifi only dongle; I also have a BT+Wifi dongle that works great with the reference image but has never been recognized by this image.
My network has an _ in the name; not sure if that's a special character in this context and the reason why I am having issues. If that's the case, I'll reflash with an image that fixes that and try again as I really did like this image.
Actually, in other cores (Flycast atm) dpad right also does not work on ogs
Sounds like issues with keymap in OS/RA level rather than individual cores.
To complete the description: I have hotkey (f2) bindings set for vol up to dpad right and save slot -1 to L2, and they work fine.
I remapped the input in Settings -> Input -> Port 1 Controls pressing L2 and R2, here's what the values are for me on OGS. (As for L3 / R3, these are leftmost and rightmost F buttons, I did not actually check i the remap was needed):
Man, I tried to find out how to donate you a drink. It took a few tweaks but your changes fixes the dwc2 driver in my 4.19.161 kernel too. I tried it about 8 times, watched it over UART, tried a jumpdrive, headset, camera, and it seems to be fixed fixed. Bravo. I'm really happy about this.valadaa48 wrote: ↑Sat Feb 13, 2021 11:58 amRRVL kernel 19 (latest update) fixes 2 kernel bugs:
1) dwc2 not working on resume without rmmod'ing it
2) rk817_codec resetting volume to max on resume
I'll pass these to HK but I figure you all can test this and let it bake in to be sure. The result is much faster suspend/resume as mentioned before and also the volume level shouldn't glitch on resume.
You're welcome. Did you need to apply any additional patches to get it working? I still consider it experimental although it's working well for me. It seems as though esp8089 from the BE still has to be rmmod'd as it itself appears to have a suspend bug...notime2d8 wrote: ↑Sat Feb 13, 2021 1:40 pmMan, I tried to find out how to donate you a drink. It took a few tweaks but your changes fixes the dwc2 driver in my 4.19.161 kernel too. I tried it about 8 times, watched it over UART, tried a jumpdrive, headset, camera, and it seems to be fixed fixed. Bravo. I'm really happy about this.valadaa48 wrote: ↑Sat Feb 13, 2021 11:58 amRRVL kernel 19 (latest update) fixes 2 kernel bugs:
1) dwc2 not working on resume without rmmod'ing it
2) rk817_codec resetting volume to max on resume
I'll pass these to HK but I figure you all can test this and let it bake in to be sure. The result is much faster suspend/resume as mentioned before and also the volume level shouldn't glitch on resume.
No, the only changes I had to make were due to naming changes.valadaa48 wrote: ↑Sat Feb 13, 2021 1:56 pmYou're welcome. Did you need to apply any additional patches to get it working? I still consider it experimental although it's working well for me. It seems as though esp8089 from the BE still has to be rmmod'd as it itself appears to have a suspend bug...notime2d8 wrote: ↑Sat Feb 13, 2021 1:40 pmMan, I tried to find out how to donate you a drink. It took a few tweaks but your changes fixes the dwc2 driver in my 4.19.161 kernel too. I tried it about 8 times, watched it over UART, tried a jumpdrive, headset, camera, and it seems to be fixed fixed. Bravo. I'm really happy about this.valadaa48 wrote: ↑Sat Feb 13, 2021 11:58 amRRVL kernel 19 (latest update) fixes 2 kernel bugs:
1) dwc2 not working on resume without rmmod'ing it
2) rk817_codec resetting volume to max on resume
I'll pass these to HK but I figure you all can test this and let it bake in to be sure. The result is much faster suspend/resume as mentioned before and also the volume level shouldn't glitch on resume.
I am getting a report from a 8821cu wifi user that their wifi has become spotty recently, not sure if it's related to this or something else. It's quite difficult to troubleshoot people's wifi issues if you don't have their devices and can't reproduce locally!
So in the debugging with 송재환 it seems the wifi issues weren't due to the kernel changes but rather that a new update ran retroarch in realtime via chrt -rr 99 and apparently there was some starvation or something. I reverted that change and all is good.notime2d8 wrote: ↑Sat Feb 13, 2021 2:19 pmNo, the only changes I had to make were due to naming changes.valadaa48 wrote: ↑Sat Feb 13, 2021 1:56 pmYou're welcome. Did you need to apply any additional patches to get it working? I still consider it experimental although it's working well for me. It seems as though esp8089 from the BE still has to be rmmod'd as it itself appears to have a suspend bug...notime2d8 wrote: ↑Sat Feb 13, 2021 1:40 pmMan, I tried to find out how to donate you a drink. It took a few tweaks but your changes fixes the dwc2 driver in my 4.19.161 kernel too. I tried it about 8 times, watched it over UART, tried a jumpdrive, headset, camera, and it seems to be fixed fixed. Bravo. I'm really happy about this.valadaa48 wrote: ↑Sat Feb 13, 2021 11:58 amRRVL kernel 19 (latest update) fixes 2 kernel bugs:
1) dwc2 not working on resume without rmmod'ing it
2) rk817_codec resetting volume to max on resume
I'll pass these to HK but I figure you all can test this and let it bake in to be sure. The result is much faster suspend/resume as mentioned before and also the volume level shouldn't glitch on resume.
I am getting a report from a 8821cu wifi user that their wifi has become spotty recently, not sure if it's related to this or something else. It's quite difficult to troubleshoot people's wifi issues if you don't have their devices and can't reproduce locally!
And yes, I'm on my 3rd driver trying to figure which one works best. But WiFi signal clutter, interference, and slow downs are hard to diagnose over the internet.
are your bios files in place? It's working here.NukeLassic wrote: ↑Sat Feb 13, 2021 3:42 pmI'm not sure if it's this update, or if I've missed something, but attempting to launch Spring PSX RrARMed-32 Bit as my PlayStation core will briefly show a black screen then eject me back to the RetroArch menu. Some PlayStation games used to have this error, but not all of them seem to.
Using the most recent RRVL version, all PlayStation files in .chd file format.
Just to be clear, this is on the OGA-BE. These were games I had previously played on this same machine, and worked previously. BIOS files appear to be in place and unchanged (unless RRVL is looking in a different place now than previously? My BIOS files are still in /roms/bios ), but I'm still getting dumped back to the Retroarch menu when I try to launch.valadaa48 wrote: ↑Sat Feb 13, 2021 3:57 pmare your bios files in place? It's working here.NukeLassic wrote: ↑Sat Feb 13, 2021 3:42 pmI'm not sure if it's this update, or if I've missed something, but attempting to launch Spring PSX RrARMed-32 Bit as my PlayStation core will briefly show a black screen then eject me back to the RetroArch menu. Some PlayStation games used to have this error, but not all of them seem to.
Using the most recent RRVL version, all PlayStation files in .chd file format.
Also, if you're using chd you may want to update duckstation and try that. I find it to be pretty good especially if you overclock to 1.5GHz.
You can try running it from ssh after doing `sudo sv stop retroarch` something like `retroarch32 -v -L /usr/arm/opt/libretro/pcsx_rearmed_libretro.so <rom>` and check the output to see what's going on.NukeLassic wrote: ↑Sun Feb 14, 2021 6:20 amJust to be clear, this is on the OGA-BE. These were games I had previously played on this same machine, and worked previously. BIOS files appear to be in place and unchanged (unless RRVL is looking in a different place now than previously? My BIOS files are still in /roms/bios ), but I'm still getting dumped back to the Retroarch menu when I try to launch.valadaa48 wrote: ↑Sat Feb 13, 2021 3:57 pmare your bios files in place? It's working here.NukeLassic wrote: ↑Sat Feb 13, 2021 3:42 pmI'm not sure if it's this update, or if I've missed something, but attempting to launch Spring PSX RrARMed-32 Bit as my PlayStation core will briefly show a black screen then eject me back to the RetroArch menu. Some PlayStation games used to have this error, but not all of them seem to.
Using the most recent RRVL version, all PlayStation files in .chd file format.
Also, if you're using chd you may want to update duckstation and try that. I find it to be pretty good especially if you overclock to 1.5GHz.
On further experimenting, looks to also be the case with Spring Flycast, but not Spring Mugen64Plus.
I'll try DuckStation with the overclock. There are a few games that didn't seem to play too nicely with it at stock speed, but hopefully overclock addresses the bulk of those. (Also, anyone familiar with how to convert saves from PCSX ReARMed's format to Duckstation's would be appreciated.)
Agreed it may run a bit better on 64bit + OC. For chances, I'd say it's up to finding someone willing to port it over. =)ptrm wrote: ↑Thu Feb 18, 2021 8:46 pmWhat are the chances for Xash3d in the near future? I see it ported already (viewtopic.php?p=319574#p319574) and use it successfuly with a dedicated batocera sd card just to run half life. But special effects make the fps drop to 20 and 1.5GHz overclock available on RRVL would be of great advantage
If you just update and reboot it should go away.NukeLassic wrote: ↑Sun Feb 21, 2021 12:04 pmI'm not sure if it's an issue with just me, but my screen will occasionally black out randomly. I can't find a particular cause for it, it happens in all emulators and even just in Retroarch menu. Putting the OGA in standby and coming out of it will fix the black screen, so it's only a minor issue, but seems worth bringing up on the off-chance it isn't just me and an issue affecting others too.
Someone else also has this issue. EmuELEC includes the out of tree module separately and apparently works. Void Linux has that same module as a DKMS package which is the preferred way of installing it but as you already noticed the kernel headers aren't 100% intact to be able to build it. I spent an hour trying to clean it up in kernel -23 but it still needs some more work. You can try moving over or extracting the kernel source and trying to get it to compile but I agree it's a lot of work. I think the better option is if HK placed the module with the out of tree version that works better so each OS doesn't need to build this separately.loaidheach wrote: ↑Mon Feb 22, 2021 7:17 pmI got my OGS a few days ago and installing RRVL worked like a charm, but for the life of me I can't get my WIFI running (although I can get online via USB tethering with my phone). It's a TP-LINK AC600 T2U Archer nano, which has a RTL8811au chipset (which I thought should work). Since I can't connect via ssh and have the WIFI stick connected, debugging is a bit annoying.
lsusb lists the stick, but the rtl8812au module isn't loaded. Manually loading it doesn't change anything, the device is not detected. On my desktop I installed the the driver from https://github.com/aircrack-ng/rtl8812au. I tried building that (directly on the OGS) against 4.4.189_RRVL-23, but that also failed finally because of a missing elfconfig.h in the kernel-headers package (the elfconfig.h from the 4.4.189_8 doesn't work).
Is there any way to get this stick running on RRVL at this moment? Am I maybe just missing something obvious?
//edit: I am not hell-bent on the T2U, I'm open to buy another stick that's known to working (I saw someone on reddit suggest the TP-Link TL-WN725N for the OGA on stock ROM)
I don't know why I haven't thought of the full kernel sources. I got it working with RRVL-headers, some stuff from the vanilla kernel and ignoring an error with openssl. I will post the full steps here once I redid it from a fresh install.valadaa48 wrote: ↑Tue Feb 23, 2021 1:42 amSomeone else also has this issue. EmuELEC includes the out of tree module separately and apparently works. Void Linux has that same module as a DKMS package which is the preferred way of installing it but as you already noticed the kernel headers aren't 100% intact to be able to build it. I spent an hour trying to clean it up in kernel -23 but it still needs some more work. You can try moving over or extracting the kernel source and trying to get it to compile but I agree it's a lot of work. I think the better option is if HK placed the module with the out of tree version that works better so each OS doesn't need to build this separately.loaidheach wrote: ↑Mon Feb 22, 2021 7:17 pmI got my OGS a few days ago and installing RRVL worked like a charm, but for the life of me I can't get my WIFI running (although I can get online via USB tethering with my phone). It's a TP-LINK AC600 T2U Archer nano, which has a RTL8811au chipset (which I thought should work). Since I can't connect via ssh and have the WIFI stick connected, debugging is a bit annoying.
lsusb lists the stick, but the rtl8812au module isn't loaded. Manually loading it doesn't change anything, the device is not detected. On my desktop I installed the the driver from https://github.com/aircrack-ng/rtl8812au. I tried building that (directly on the OGS) against 4.4.189_RRVL-23, but that also failed finally because of a missing elfconfig.h in the kernel-headers package (the elfconfig.h from the 4.4.189_8 doesn't work).
Is there any way to get this stick running on RRVL at this moment? Am I maybe just missing something obvious?
//edit: I am not hell-bent on the T2U, I'm open to buy another stick that's known to working (I saw someone on reddit suggest the TP-Link TL-WN725N for the OGA on stock ROM)
FWIW, I've used https://www.amazon.com/gp/product/B07M6YB838 from the beginning and it's never failed me.
Nice work! Can you perhaps post the ko so the other users can give it a try? It should work for everyone as long as I don't update the kernel. Also can you confirm if it needs any firmware that isn't available? modinfo should report it.loaidheach wrote: ↑Tue Feb 23, 2021 3:24 amI don't know why I haven't thought of the full kernel sources. I got it working with RRVL-headers, some stuff from the vanilla kernel and ignoring an error with openssl. I will post the full steps here once I redid it from a fresh install.valadaa48 wrote: ↑Tue Feb 23, 2021 1:42 amSomeone else also has this issue. EmuELEC includes the out of tree module separately and apparently works. Void Linux has that same module as a DKMS package which is the preferred way of installing it but as you already noticed the kernel headers aren't 100% intact to be able to build it. I spent an hour trying to clean it up in kernel -23 but it still needs some more work. You can try moving over or extracting the kernel source and trying to get it to compile but I agree it's a lot of work. I think the better option is if HK placed the module with the out of tree version that works better so each OS doesn't need to build this separately.loaidheach wrote: ↑Mon Feb 22, 2021 7:17 pmI got my OGS a few days ago and installing RRVL worked like a charm, but for the life of me I can't get my WIFI running (although I can get online via USB tethering with my phone). It's a TP-LINK AC600 T2U Archer nano, which has a RTL8811au chipset (which I thought should work). Since I can't connect via ssh and have the WIFI stick connected, debugging is a bit annoying.
lsusb lists the stick, but the rtl8812au module isn't loaded. Manually loading it doesn't change anything, the device is not detected. On my desktop I installed the the driver from https://github.com/aircrack-ng/rtl8812au. I tried building that (directly on the OGS) against 4.4.189_RRVL-23, but that also failed finally because of a missing elfconfig.h in the kernel-headers package (the elfconfig.h from the 4.4.189_8 doesn't work).
Is there any way to get this stick running on RRVL at this moment? Am I maybe just missing something obvious?
//edit: I am not hell-bent on the T2U, I'm open to buy another stick that's known to working (I saw someone on reddit suggest the TP-Link TL-WN725N for the OGA on stock ROM)
FWIW, I've used https://www.amazon.com/gp/product/B07M6YB838 from the beginning and it's never failed me.
At this point it's all a bit hacky, but this should be enough to get the headers kind of into shape to dkms. Since the dkms driver is a different package it would be ok, but should not be installed by default. I agree, option (3) would be the best option, because it would benefit all distros on the OGA/OGS.valadaa48 wrote: ↑Tue Feb 23, 2021 5:18 am
Nice work! Can you perhaps post the ko so the other users can give it a try? It should work for everyone as long as I don't update the kernel. Also can you confirm if it needs any firmware that isn't available? modinfo should report it.
I guess for integrating it we can do a few things: 1) somehow pull in that tree when RRVL/Void builds the kernel and copies the ko to the modules dir (this is what EE does) 2) Just use the dkms as it's officially supported by Void Linux. We'd have to fix the headers package so it can build. This will also cause dkms recompiles each time the kernel is upgraded for everyone whether or not they use that module... 3) Try to slipstream this driver into the HK kernel
I think (3) might be the best option.
/lib/modules/4.4.189_RRVL-23/kernel/drivers/net/wireless/88XXau.ko . sudo -i, but you can also put sudo in front of every command.Code: Select all
xbps-install -y bc odroid-go2-rrvl-kernel-headers gcc python
Code: Select all
KERNEL_VERSION=$(uname -v) # e.g. 4.4.189
KERNEL_RELEASE=$(uname -r) # e.g. 4.4.189_RRVL-23
KERNEL_FAMILY=v${KERNEL_VERSION%%.*}.x # e.g. v4.x
$KERNEL_VERSION has not changed. Instead of linking via ln you can probably just move the files over, but I haven't tried it.
Code: Select all
cd /usr/src
wget https://mirrors.edge.kernel.org/pub/linux/kernel/${KERNEL_FAMILY}/linux-${KERNEL_VERSION}.tar.gz -O- | tar xz
cd kernel-headers-${KERNEL_RELEASE}
mkdir arch/arm/include
ln -s $PWD/arch/arm/* arch/arm/include/
cp -r /usr/src/linux-${KERNEL_VERSION}/arch/arm64/kernel arch/arm64
cp -r /usr/src/linux-${KERNEL_VERSION}/tools .
cp /boot/config-${KERNEL_RELEASE} .config
modules_prepare step will fail due to a missing openssl header (openssl-devel is not available), but the part relevant for us is done.Code: Select all
cp /boot/config-${KERNEL_RELEASE} .config .
make oldconfig
make modules_prepare
-j4 parameter tells make to use four compile threads to use all cores. If cross-compiling, adjust this to your core count. If the build process fails for any reason, use make clean and run make without any parameters before reporting any bugs.
Code: Select all
cd /usr/src
git clone https://github.com/aircrack-ng/rtl8812au.git
cd rtl8812au
make -j4 && make install
Most of the discussion about RRVL is done in the discord chat and since people complained that we don't have a way to dynamically choose the right size logo, we removed the logo altogether. Someone added a wiki entry on how to add custom logos though. You can try finding it.stellarvelocity wrote: ↑Thu Mar 04, 2021 11:24 amI think I have might have an issue with the latest update.
I turned on my OGS today to play some PSX (after using it fine the last few days) and I noticed that my custom logo.bmp didn't load, and it was trying to display the OG-A scaled logo. Then it wouldn't load the Spring PSCX core at all. I did a couple of update checks, and when it rebooted I get the HARD KERNEL logo and then it drops straight into RA. No Logo.BMP at all. Upon checking via SFTP, the file has been deleted from the SD card!
I still couldn't get the PCSX Spring Core to work, but I tested all other cores were working and everything was up-to-date. I reset RVVL with the .sh in the rom folder, and it still didn't fix the issue.
As a last-ditch I deleted the cores and every Spring_Core I had, to be sure where the error for PCSX issue was - and now I can't redownload them! I went into Core Downloader, and I can download any core (example, Rick Dangerous Core) and it downloads quick and installs. However, all of the Spring_Cores hit 50% exactly and stop. I have four of them loaded on screen right now, and they're all stuck at 50% even after several power cycles and SD Card pulls.
Any idea what's happening? I would suspect a bad SD but all the other roms and cores are working fine, and downloading correctly. I read elsewhere on the forum that Moonlight PulseAudio was causing issues with PCSX, and I did try to get Moonlight working last week, but I haven't had any issues since then. Any thoughts?
FYI we believe we found the issue with pcsx_rearmed. You can do an update and it should work now. If you lost your spring files, then just run `sudo xbps-install -Sf libretro-core-info rr-base` and you should be good.valadaa48 wrote: ↑Thu Mar 04, 2021 1:28 pmMost of the discussion about RRVL is done in the discord chat and since people complained that we don't have a way to dynamically choose the right size logo, we removed the logo altogether. Someone added a wiki entry on how to add custom logos though. You can try finding it.stellarvelocity wrote: ↑Thu Mar 04, 2021 11:24 amI think I have might have an issue with the latest update.
I turned on my OGS today to play some PSX (after using it fine the last few days) and I noticed that my custom logo.bmp didn't load, and it was trying to display the OG-A scaled logo. Then it wouldn't load the Spring PSCX core at all. I did a couple of update checks, and when it rebooted I get the HARD KERNEL logo and then it drops straight into RA. No Logo.BMP at all. Upon checking via SFTP, the file has been deleted from the SD card!
I still couldn't get the PCSX Spring Core to work, but I tested all other cores were working and everything was up-to-date. I reset RVVL with the .sh in the rom folder, and it still didn't fix the issue.
As a last-ditch I deleted the cores and every Spring_Core I had, to be sure where the error for PCSX issue was - and now I can't redownload them! I went into Core Downloader, and I can download any core (example, Rick Dangerous Core) and it downloads quick and installs. However, all of the Spring_Cores hit 50% exactly and stop. I have four of them loaded on screen right now, and they're all stuck at 50% even after several power cycles and SD Card pulls.
Any idea what's happening? I would suspect a bad SD but all the other roms and cores are working fine, and downloading correctly. I read elsewhere on the forum that Moonlight PulseAudio was causing issues with PCSX, and I did try to get Moonlight working last week, but I haven't had any issues since then. Any thoughts?
So we finally rm -rf'd 32bit completely now that we have opks working. Literally the only thing using 32bit is pcsx rearmed (and you should really just switch to duckstation, it's much better anyway). I suspect something broke there. I tried reproducing it on a newly flashed image but I couldn't. Maybe others can chime in with their experience with pcsx rearmed and if it's broken for them? One user said it is working for some but not other games.
I ran the script, and updated, and the Spring Command/Cores are back. I still can't get PCSX Rearmed to launch, and Duckstation doesn't work with my .pbp format images; and when I load a bin/cue performance is really really bad even with 1x Internal Resolution and Geometry Correction turned off. When I try to load a ROM manually or from playlist with PCSXR it just drops me back on the RA menu.valadaa48 wrote: ↑Thu Mar 04, 2021 2:39 pmFYI we believe we found the issue with pcsx_rearmed. You can do an update and it should work now. If you lost your spring files, then just run `sudo xbps-install -Sf libretro-core-info rr-base` and you should be good.valadaa48 wrote: ↑Thu Mar 04, 2021 1:28 pmMost of the discussion about RRVL is done in the discord chat and since people complained that we don't have a way to dynamically choose the right size logo, we removed the logo altogether. Someone added a wiki entry on how to add custom logos though. You can try finding it.stellarvelocity wrote: ↑Thu Mar 04, 2021 11:24 amI think I have might have an issue with the latest update.
I turned on my OGS today to play some PSX (after using it fine the last few days) and I noticed that my custom logo.bmp didn't load, and it was trying to display the OG-A scaled logo. Then it wouldn't load the Spring PSCX core at all. I did a couple of update checks, and when it rebooted I get the HARD KERNEL logo and then it drops straight into RA. No Logo.BMP at all. Upon checking via SFTP, the file has been deleted from the SD card!
I still couldn't get the PCSX Spring Core to work, but I tested all other cores were working and everything was up-to-date. I reset RVVL with the .sh in the rom folder, and it still didn't fix the issue.
As a last-ditch I deleted the cores and every Spring_Core I had, to be sure where the error for PCSX issue was - and now I can't redownload them! I went into Core Downloader, and I can download any core (example, Rick Dangerous Core) and it downloads quick and installs. However, all of the Spring_Cores hit 50% exactly and stop. I have four of them loaded on screen right now, and they're all stuck at 50% even after several power cycles and SD Card pulls.
Any idea what's happening? I would suspect a bad SD but all the other roms and cores are working fine, and downloading correctly. I read elsewhere on the forum that Moonlight PulseAudio was causing issues with PCSX, and I did try to get Moonlight working last week, but I haven't had any issues since then. Any thoughts?
So we finally rm -rf'd 32bit completely now that we have opks working. Literally the only thing using 32bit is pcsx rearmed (and you should really just switch to duckstation, it's much better anyway). I suspect something broke there. I tried reproducing it on a newly flashed image but I couldn't. Maybe others can chime in with their experience with pcsx rearmed and if it's broken for them? One user said it is working for some but not other games.
Thank you so much for the cores ...any update?Slaminger wrote: ↑Wed Feb 10, 2021 2:02 amhttps://github.com/Retro-Arena/ARES-Binaries
The Retro Arena and ARES prebuilt emulator and core binaries are open to the public, if anyone ever needs to grab a core to test something quick. N2 cores work for aarch64 and xu4 for armhf or arm32 testing.
1. Is it possible to connect a keyboard and launch a terminal directly on the device? Specifically for vim, file management, running commands etc.newodroid wrote: ↑Fri Mar 05, 2021 4:40 amCurrently waiting on my OGA BE and had a few questions about RRVL (and I guess the device in general) if you guys don't mind me asking:
1. Is it possible to connect a keyboard and launch a terminal directly on the device? Specifically for vim, file management, running commands etc.
I'm considering carrying around a small device with either a wired/2.4ghz keyboard or a bluetooth one and using it for taking down quick notes while I'm away from home, using either emacs or vim.
I bet the OGA is powerful enough to compile simple latex pdfs or run groff at the very least. Would also be cool to play scummvm games with a mouse and keyboard.
2. Would I be able to use sshfs to mount a remote filesystem directly onto the OGA?
3. I'm familiar with bash scripting, void and arch linux, but I've never done something like run a .sh script through retroarch. Is it just a matter of making the script executable and adding it to the retroarch menu somehow? Specifically I would write scripts to rsync saves and files from the OGA to the desktop and vice versa. And also for the above mentioned sshfs command to mount/unmount when needed.
4. This relates to the above terminal question: is it possible to use the dpad and buttons to control things like ncurses programs in a terminal, or would I always have to have a keyboard attached?
My main use case for this would be scrolling through a file using vim or less, and controlling programs like MPD + ncmpcpp or MOC for music.
It would be cool to be able to mount some music from my desktop using sshfs and play it back on the OGA directly. I do this on my laptop and it's very handy to save local storage space.
5. Can I use mupdf for simple document viewing? I know it's a low resolution screen for reading but it would be handy for just quickly viewing something. And can I use the dpad for scrolling?
6. Is it possible to use a USB DAC for high quality audio + amplifier with the current RRVL kernel and alsa?
7. finally how is the battery life on RRVL? I know you can overclock a bit for better performance in games but is it possible to underclock for better battery while doing less demanding things like listening to music or playing GB/C?
Long post I know, I'm excited for this little linux handheld. Thanks!
I wrote a simple/inefficient keymapper for use with ncmpp: https://github.com/mad-ady/antimicro-nonX4. This relates to the above terminal question: is it possible to use the dpad and buttons to control things like ncurses programs in a terminal, or would I always have to have a keyboard attached?
My main use case for this would be scrolling through a file using vim or less, and controlling programs like MPD + ncmpcpp or MOC for music.
It would be cool to be able to mount some music from my desktop using sshfs and play it back on the OGA directly. I do this on my laptop and it's very handy to save local storage space.
The joypad is not mapped as a keyboard so you'll need some kind of userspace translator. Ask shanti and the others here with what works for them. It's also not difficult to write one in C, python, rust, etc.. There isn't anything built-in to RRVL to do this though.
This is amazing!!I wrote a simple/inefficient keymapper for use with ncmpp: https://github.com/mad-ady/antimicro-nonX
Have fun with it!
I usually start a systemd unit with it while running ncmpp and turn it off after. So you can have a profile for each program you run...
Users browsing this forum: BrianFitz and 0 guests