I didn’t mean to imply otherwise. It was indeed sorely needed and I’m excited about the RPi 4. But I’m also willing to wait a bit to see how the dust settles.
But I do know if I ever need to replace my server machine, I’m going to do it with a few RPi 4s instead of a big black box machine I have running everything now. The biggest sticking point for me was getting good NAS performance out of them. With USB 3 and GigE that problem should be solved. And with 4 GB RAM I wouldn’t need more than three or four of them.
I was agreeing with you and thought it was worth repeating. I find it interesting that if I said the exact same thing 6 months ago I would get flamed big time and excuses like ‘you don’t really need any more than what a pi3 does’.
The same thing is still there with a pi4, you buy a SSD capable of so much speed as the price is falling rapidly on ssds, and then it gets crippled by usb3. But hey 640kB is enough
Just interesting to see the same cycle repeat over the years.
EDIT: I am not aiming the above at anyone, it is just my humour and musings looking at general trends.
I think I’ll hold off until the revision of the board is released. Hopefully they will be able to sort out the thermal issues by then as well. If I need to install a fan to keep it cool when it is in its casing then one of the most important selling points (no moving parts) for me is gone.
The original plan was to release the RPi4 in 2020. It all feels a bit rushed now. USB charging problems, heat issues, no PXE and USB boot, etc. Why release this early when there’s still plenty of time until the original planned release date?
Let’s learn from this and not rush to push out OH2.5.0.M2 due to community pressure
I wonder how much pressure the RPi Foundation actually feels, given that it’s a charity that’s focused on making low-cost computers more widely available around the world. They certainly have to feel some pressure, but at the same time their competitors are furthering the Foundation’s mission. Charitable organizations (good ones, at least) operate on the idea that “someone needs to make this happen,” not “we need to be the ones to make this happen.”
The simplest answer might be a flawed QA process. They must not have tested scenarios with a wide variety of off-the-shelf USB-C cables, or else they would have discovered that their charging circuit doesn’t work. This seems silly given that many end-users buy RPis without official chargers, but it wouldn’t be the first time in history that something simple and seemingly obvious was overlooked.
I just hope it doesn’t end up hurting the RPi brand. I don’t think it will, but it doesn’t take much to damage a reputation nowadays.
Upton went on to say, “I expect this will be fixed in a future board revision, but for now users will need to apply one of the suggested workarounds. It’s surprising this didn’t show up in our (quite extensive) field testing program.”
…
We reached out to Raspberry Pi about this issue and were told a board revision with a spec-compliant charging port should be out sometime in the “next few months.”
I find just the opposite. The best charitably organizations are the ones who go out and actually do something (fund research, provide housing for families of children in hospital, give away trees, provide counseling, etc). The charities who just promote “awareness” with the hope that someone will do something about it are not effective at anything except getting donations.
I wouldn’t be surprised if their field testing included a kit that came with the power adapter. The fact that the testers didn’t think to test it also using third party power supplies given the history of RPi power problems is surprising.
I think they have too much momentum for something like this to stop the brand, as long as they provide a fix for it quickly.
Thanks for the clarification! What I meant to say was that charities should be (in my opinion) focused on achieving their goals, both through their own efforts and the efforts of partners/competitors. I’m just not a fan of charities who behave as if the only correct solutions are the ones they put forward.
Fessing up and admitting there is an error, and publishing the workaround is also a big help. Silence or denial of things always make orgs look worse than simply saying, “Yes, this has an error, here is what you can do about it, we’ll fix it in the future.”
And what is the goal of the Raspberry Pi Foundation? (Hint: Is is NOT us using OpenHAB).
The goal is providing educational kits to teach computer design & programming.
Those of us using the Pi for any other purpose are not part of their primary goal.
I disagree, based on my interpretation of their About Us page.
Outreach to schools is absolutely an important part of what they do, but the larger goal is promoting widespread computing around the world. That includes hobbyists like us, many of whom have increased our computing skills significantly in order to set up and maintain our openHAB systems on RPis.
That’s why I think that the clone competitors also further the RPi Foundation’s goals. It doesn’t matter to the Foundation that a person uses an RPi…it matters to them that the person has a computer in the first place.
The disagreement probably has to do with what the goal of the RPi was originally and what it has become. When the RPi 1 was first released, it was only expected to be used to teach coding in UK schools. They were all surprised at the rapid adoption of RPi by the Maker movement.
2012 was an exciting year. The Maker community was starting to come into it’s stride, then a whole slew of new Arduino boards were released. Then this little charity in the UK came out with this SBC that had the ease of use of a regular computer but could interact with the real world through the GPIO pins. I think the first article about the RPi in Make: Magazine was in early 2013 but I remember it only took a few months from first mention until that’s all anyone can write about.
Over the years RPi became more and more maker friendly and then in 2014 they released their first compute module. The compute module really is only useful for more commercial industrial uses. At this point I’m pretty sure their original “promote computer programming education” fell by the way side and they became more broad in their mission statement.
I moved my OpenHab over to a new RPi-4B-4G a week ago now. Results so far
Originally I thought I had a big problem. It turned out I’d used a bad SD card… So that was swapped out on Sunday after noticing that the system spent 90% of its time with 100% waitIO on 2-4 cores.
It’s definitely quicker than the 3B I had previously.
I have stacked a POE card and the Razberry zWave card on it. POE means I don’t have to worry about the dodgy not quite USB-C power.
It’s loud. The POE fan runs pretty much all the time now. Might have to tune the hysteresis on the temp a bit for that
The extra memory is much appreciated. I run with about 50% free. So 2GB would be tight. The 1GB of the 3B was obviously far too little
I still get the occasional blip with those annoying Not Ready errors in the web interface…
It’s far more susceptible to the original broken POE issues than the 3B+ was.
I think the biggest problem I have is the fact that for some reason my zWave just seems to be pants. The extra CPU and memory isn’t help that sadly.
My first awareness of the RPi was in 2013 when I worked for the University of Waterloo’s Faculty of Engineering. Students were raving about how much difference they made for fourth-year projects and team competitions, because they could spend less time in setup and more time on the actual design challenges. There was a massive leap forward in both the speed and quality of their work.
To be honest, I didn’t really understand the potential of the RPi until I bought one on a whim two years ago. And even then, I didn’t do much with it until jumping into openHAB.
I think something like this will mitigate the POE issues for the Pi 4. It worked for me for Pi 3s as there was significantly less heat than the POE hat (official and unofficial types) . Heat throttles the Pi significantly and the performance loss is obvious.
I’m about to take my first steps with OH, and was wondering if the openHabian (and the openHabian installation docs) will work with the new Raspberry Pi 4? Or am I better off buying a Raspberry Pi 3?
Hi!
Wellcome to the community!
I don’t own a raspberry 4 yet, but I read in other topics that openhabian is not ready for the rpi 4.
But we have reports of people running it on rpi 4 with a “manual” installation : raspbian + repository install.
Maybe some rpi 4 owner can confirm.
As @alexxio says, a normal install of Buster (Download from Raspberry Pi) and then a manual install of OH.
sudo apt-get install openhab2
There is a bit of setup to do before hand but that is very well documented in the documents section.
I’ve been running it since getting the RPi4 (start of thread) and have not had any issues.
I do use a small heatsink which keeps it down to between 60 and 65 degrees.
Heat is my main concern about the RPi4, as I wouldn’t want to have a fan running all of the time.
The other thing was the USB-C power, but I’ve since read that this is only a potential problem if you’re using a C-to-C cable (though C-to-A cables might still be a little hit and miss).
If I were just starting out myself, I’d consider buying a 4 with a decent heatsink (per @CDriver’s comment) and an official RPi4 power supply (to avoid USB-C issues). But I’d likely end up getting a 3B+, because it’s the path of least resistance. Openhabian runs very well on it, and there are many of us who can support you as you figure things out.
