Can we switch to open-source Rpi replacements like Radxa ROCK, Banana Pi and Orange Pi as a substitution? They are also notably cheaper.
You can work on supporting whatever device you want and own.
I’m personally also very much in favor of open-source RPI replacements.
Unfortunately the proprietary original one is still much more popular than the open alternatives,but that doesn’t neccessarily mean it needs to be supported first.
ARM is also more widely-used than RISC-V,yet we got a fully functional RISC-V port years earlier because the platform,being fully open-source,is easier to work with.
3 Likes
We don’t have a working ARM port at all. So, in the context of this thread, this is a bit silly. I expected that the “fully functional machine” would start from a machine where Haiku already works (there are many to pick from).
1 Like
What would be an ideally open source alternative then? Everything I look up seems to have some amount of non-free blobs and most other than Raspberry Pi don’t even try to mainline stuff (for Linux) although community efforts have apparently been fairly successful in getting some Allwinner SoCs mainlined despite Allwinner’s total lack of cooperation. Even RISCV stuff seems to frequently have driverless/non-open parts.
Have a look at the post above mine.
@daniel0366 recommended Radxa ROCK,Banana Pi and Orange Pi,and I’d like to add Pine64 devices.
I had a look at all four,most of their devices seem to use Rockchip and a few Allwinner.
I’m not an expert for ARM,but as far as I know Rockchip is rather open and good to support,having read some blog posts at Pine64 about their Rockchip-based devices in the past.
Pine64 also wrote about community efforts to mainline the Rockchip stuff if I remember right,but that doesn’t matter at all here,Haiku isn’t Linux.
What matters is that the hardware is well documented which seems to be the case for Rockchip from what I’ve read so far (again,I’m no expert and I’m happy with x86_64)
I had a look at Pine64. They also manufacture laptop Pinebook and tablet Pinetab. Working on Pine64, starting from rock64 would be a good entry to varieties of ARM devices.
Linux mainline does matter to the extent that it serves as proof that something actually is open enough to be supported and has a presumably well written open source implementation already available to study.
I did specifically look at the ones listed in the above post. Most or all of them had one or more of the issues I mentioned.
I have a Pinebook Pro. It was OK for someone who needs ARM, very inexpensive. Dead now - opened up the case one too many times. At present I doubt they’re on the market - where I looked, out of stock for a really long time.
If Pine64 were really into providing a platform, I think they could have gotten there pretty easy, but the way it was, if you wanted to branch out from the OS they shipped on it (Manjaro Linux), it could get kind of hairy. There were ways to get a boot menu, but it involved flashing some 3rd party stuff, of which there were various alternatives, because Pine64 apparently had no in house expertise or interest in that level of putting together a platform.
So … if the Pinebook were still on the market, maybe it could provide some Haiku developer with a usable ARM platform. It wouldn’t ever have anything to do with the subject of this thread, though.
Our biggest flaws are platform agnostic ones:
- bluetooth
- multitouch
- usb audio
- usb webcam
All of that won’t work even if Intel itself develop a platform for us. I also think we should have some reference hardware or team up with some vendor… But we are far from this
2 Likes
Just because there are a lot of supported devices does not mean that everyone has equal ability to obtain one that also meets their needs/requirements in other ways.
Even if they had perfect compatibility, I doubt anyone would really want to use a slow, under-resourced netbook…
1 Like
There are quite a lot of under-resourced people in the world, and an under-resourced netbook might be all that they can afford.
@draKo - Several possibilities exist to create a small sustainable group of computers for Haiku development and usage.
Use the HCLs (i.e BeSly, elsewhere) as guidebooks to this.
Most people can use native BeOS/Zeta/Haiku on newer machines like Apple M4 and AMD Rysen based computers.
As for drivers, that is another topic.
This topic was discussed in various threads from SoC to SFF to laptops/AIOs. You can have a manufacturer select a customized or premade computer very compatible for Haiku usage…
It is obvious that not everyone can afford his dream computer, me included 
But you can get used / refurbished Business notebooks like Thinkpads for low cost, these are quality devices with i5 and i7 cpus, not underresourced netbooks
Thats the kind of head space I’m wanting to focus on.
Make someting work right, and focus on other things later.
The way it feels using Haiku is its spread to thin, so it’s failing unnecessarily, on good equipment.
2 Likes
Since Haiku does not run on arm yet, i doubt anyone is running it on an apple m4 natively, but maybe in an emulator.
I don’t have any ryzen where i could test it on, but a lot if intel i5 and i7 different generations, thinkpads from x201,x220 x230 t430 t460 t470 yoga15 imac 27" 2012 imac27" 2017 all work (imacs no sound, unfortunatly)
Agreed. that would only restrict those who can buy or afford a haiku computer around the world. I see countless people praising they can run haiku on a old computer. Also how many countless people i have seen praising it can be ran on a historical computer of the 90s with a PIII (the 32-bit version).
I mean, this project is not trying i suppose to exclusively appease high-end users from niche cases in developed countries, right? 
, so you are right. it would not make sense.
2 Likes
Emulation.
Now, This gets into things like a SolidRun HoneyComb LX2 Mini ITX Workstation or the Apple M4 Mac Mini or Mac Studio computers. Best to run through emulation in most cases if going to attempt a supportable scenario and real daily driver use case.Old Apple Intel computers are still around - if you want a nice iMac or whatever else for a native install.
The request seems more like low cost supportable and compatible hardware for running Haiku natively on that hardware.
There are low cost computers available for a project of 5 - 8 main compatible and completely functional hardware machine setups for Haiku/BeOS. Just know that the actual Haiku developers and HaikuPorts maintainers/contributors have their own personal (or work) machines and those computers are not all the same maker and model. So, someone can pick up the axe/sword and do something like that. Not hard. Just do it…
1 Like
The really important thing to understand here is that Haiku is still very much a project that’s in-progress/under development and the availability of a usable build doesn’t mean you should expect the kind of support and polish that usually goes along with a “1.0” release.
1 Like
Totally agree, even tho it wont be fully polished as we wanted, the progress and feedback that is being given by current users group is helpfull to made high quality software decisions… I’ve been through some UX/UI stuff, and to polished install it isn’t that much to be improved 