Gimme.
Still kind of beta. But if you want.
Isn’t everything kind of beta?
@ALan I’d love to check this out when you’re ready. Half of our team is already running Ubuntu clients, but would you mind sharing the details on the little NUC you’re using? Is this the same one from your UltraGrid video?
Nice work Alan!
Same Cube, but any cheap Intel N150 or above, based mini pc will work fine. Those cubes are really hard to find. I did track down the actual factory, and they wanted a minimum order in the hundreds.
I am in desperately in need of a quick solution. I’ve been so slammed I haven’t been able to move Teraguchi forward.
Would you be down for sharing?
I wouldn’t mind working on the Linux host / Mac client side of things.
I mean, I’d rather save the cash I was gonna spend on renewals and make a donation to the Alan Latteri personal improvement fund than HP Nowhere.
That’s kind of where I’m at as well. Rather chip in and fork rather than blow money elsewhere.
I’m out for the day, but I’ll address this later. Be advised of the following. I have ONLY developed for the following platforms:
Host:
RL 9.7
MacOS 27 Apple Silicon -beta
CLient:
Ubuntu 26.04 Desktop on Intel w iGPU.
MacOS 27 Apple silicon - this is still very beta
Please remember the system requirements. I have ZERO interest in supporting older software configurations, and I chose the Client OS/hardware because it is the cheapest and BEST for the task.
I’ll work on making the repo public, but I need to go thru and remove certain sensitive stuff, it may take time. Read the Notes file in the Dropbox. I’ll try to help where I can.
CODEX had to do a bunch of re-arranging to make this a safe Public repo. I’ve not validated everything yet. Better to use the dropbox builds for the moment, unless you just want to start vibing away.
@ALan thank you thank you thank you.
This is light years better than the original Teraguchi.
I’ll fork and contribute.
I’ll see if I can knock out a Mac 26.x client on the fork.
Starting with Sunshine/Moonlight as a base gave me a huge head start. But in fact, it was my 3rd attempt. The first two were starting from scratch with Claude and CODEX respectively, and neither got to a good place. If I was to do it again, I would have started with Kyber, and I actually have a PLANK V2 I am slowly working on which is basically that. Not sure it will make iit to release or be any better than what I already have.
The current PLANK V1 does replace the upstream Sunshine transport mechanism with Kyber based single port QUIC.
Mac Client is the least test, really almost not at all. Linux Host/Client are pretty solid.
yup. I can bang on that.
And Im right in assuming you want your original PLANK to be limited just to your personal use cases of hosts/clients, oui? Alan style?
Fuck me. Astra made quick work of this.
Nah, I made it public so the community can collaborate to make it great. Well constructed, solid and tested Pull Requests welcome.
Fuck HP. Fuck Teradici. Fuck DCV. They are all garbage.
That being said, I really wish I didn’t have to do this. I wish the commercial marketplace would provide us with modern performant tools. But it seems we are going to have to be not only artists, but also vibe coders now, for the whole spectrum of tools. If we look at the interesting things with Flame over the past year, a good portion is not coming from MTL, but the user base here. Anyway, it will all be over within 5 years.
I agree. Well said.
I had the same fuckery the day after spending all night vibe coding a shitty v1 Teraguchi.
I was like…really? There’s no way this is gonna be like this. I have to build the Flames/AI stack/storage/auth/blah blah AND now I have to build the fucking remote control component?
I was into it until you messaged me and then I quickly realized I have zero idea what you or people like you actually need. That with massive work deadlines meant I ignored it all until you resurfaced it.
Now I have 7 PCoIP licenses, 2 of which expire this week, and Im like..fuck. I am not sending any more money to those clowns if I can help it.
I literally have zero idea what I am doing, so, forgive the nonsense I bring, but hey. I’ll gladly throw my Astra credits on it and alpha/beta test.
Hey. I found my first bug. Quitting plank-client on Mac OS 26.x whilst Flame is connected does not quit plank-client.
Now, lemme figure out how to do a well constructed, solid pull request.