The hardware is known and well documented has the best OS support.
If I have issues, I can just move the SD card to another one, or just reflash another SD with 0 downtime or fuss.
Availability is a big oof. We learned over the past few years that you shouldn't build anything of any importance on top of RPi because when manufacturing crunches hit "hobbyist" products are first to be sidelined.
1. VQFN package is a pain to deal with. You can't use a soldering iron, so its reflow-only and therefore requires hotplate + reflow hot air to deal with. Due to the difficulty of rework, you likely can't use the soldering iron+solder wick technique to fix solder bridges, for example... and therefore also require a soldering stencil to work with (gotta deposit the correct amount of solder-paste and hope everything works out good). Inspection requires X-Rays as well because everything is "under" the package, so you can't even visually inspect your joints.
I definitely prefer TQFP (seems to be the package with the biggest inventory in practice), and enjoy even the larger SSOP or even SOIC packages available from some competitors. All competitors offer VQFN as an option as well, in case you are truly size-constrained. But in my experience, a 9mmx9mm TQFP-32 is perfectly fine (7x7mm raw, but the leads take up another 1mm for total of 2mm added on the two sides). Your 32-traces are going to take up like half an inch of routing space anyway, so its not like the 5mmx5mm VQFN saves much space in practice IMO (IE: You'll need 6-layers or something... and/or to cut out lines to really take advantage of the space-savings from a VQFN)
Note: Hobbyists _should_ be using hotplates to do the bulk of their surface-mount soldering jobs. But accessibility to soldering iron for rework / touching up problems is a huge advantage to TQFP.
2. 0kB of on-board flash. RP2040 requires external components to boot, complicating the flashing / loading / bootup process.
3. Poor sleep specs, for a microcontroller. Your typical AVR DD (modern ATMega), or STM32G0, or Ti MSP430 / MSPM0 are all in the dozens or low-hundreds of microamps sleep (with some competitors in the ~hundreds of nanoamps for deepest sleep states). RP2040 sleeps at ~0.5mA to 1mA, an order of magnitude more power.
4. Poor power specs, for a microcontroller. Your typical AVR DD, SAMD, STM32G0, Ti MSP430 / MSPM0 are maybe single-digit mA (under 10mA across the board). RP2040 is like 20mA. I kid you not: an 8-bitter like AVR DD can be going full-tilt at like 4MHz or 8MHz and still have less power-usage than RP2040 *at idle*.
5. Lack of peripherals. RP2040 has PIO and SRAM as superpowers... but its got fewer timers, ADCs, DACs, OpAmps, Comparators, than all of its competitors.
---------------
RP2040's main superpower is its 264kB of SRAM. If you need gobs and gobs of SRAM, RP2040 is probably the right choice.
All other situations? I'm pretty sure you'd prefer a 12-bit differential ADC with 16x gain (ex: AVR64EA48), or 12-bit differential ADC (no PGA) + 3x OpAmps (ex: AVR128DB64) and multiple comparators and 4x onboard Vref, or 2x Zero-drift OpAmps + 1x general-purpose OpAmp (ex: TI's MSPM0 line).
Honestly, I'm not impressed by the Pico for general purpose use. The Pico works as a specialized compute-device (high SRAM, high MHz and dual-core), but this is simply impractical for many applications. You'd really rather have ADCs or Comparators in more typical electrical designs. Combo'd with the high-power usage (probably from all that redundant RAM the RP2040 has), and its difficult to recommend.
--------
Going in the other way: Cortex-M4F cores with a floating-point unit, even at 64MHz, will be much-much-much faster than a 200MHz RP2040. Why?
Because hardware float32 matter.
Why do you have 264kB of SRAM when you don't even have a floating-point unit? The RP2040 can't even be a good DSP-like system with all that compute power because you gotta software-emulate floats.
So if you _actually_ need compute power, you'd probably should be picking an STM32F4 (aka: Cortex-M4F) processor instead. RP2040 sits in this awkward zone where its got gobs of SRAM + MHz but is stuck on Cortex-M0+ (lacking floats, and other key instructions), so it doesn't actually have very fast compute in a very common case that matters...
So yeah, RP2040 works if you're willing to rewrite to integer-fixed point... if you don't need peripherals like ADCs or ACs, if you're willing to spend more power and complexity (due to the lack of on-board Flash), and are willing to use leadless VQFN packages. Its... quite a niche IMO.
I've been very surprised to not see other people discussing it.
The rpi3b is still around. You can also get the rpi zero, rpi zero 2 W, rpi pico, etc.
Get the right one for the job. If you need an M.2 SSD or actual 4k video playback, go for the rpi 5, if not, for sth. else.
Yeah, and it costs over $50 for just the SBC:
https://www.amazon.com/Raspberry-Pi-MS-004-00000024-Model-Bo...
I don't think that estimate takes into account the power savings you'd get from an ARM sbc versus a NUC or NUC-like minipic.
I don't think this is relevant. RPis cost is a critical factor in it's usefulness as a disposable toy computing platform, not web servers that need to run 24/7.
Moreover, that cost is further worsened by the need to buy a custom power supply.
For long term deployments, I think recycling what would become e-waste by reusing older enterprise units is better overall for the environment than letting it go to a dump and buying a lower power consumption unit.
TL;DR: TDP is not a factor for me, quick and lowish cost access to devices right away is way more important at the start
I gather all the manufacturers have similar offerings that are available dirt cheap on the used market.
I've owned every Raspberry Pi since the first one but I think the 4 was my last one. They are fun to tinker with with but not great for long-term use in my experience. I felt like I had to babysit it way too often, it absolutely wasn't a "setup an forgot" device, again in my experience. It'd be rock solid for 2-3 months then randomly stop working, rinse and repeat.
A little bit back I bought a cheap (~$250) mini pc to run Home Assistant on and I haven't looked back. The Raspberry Pi 4 did decent at that job but was a little flakey for my liking and I even had a special case with an m.2 adapter so I didn't have to use the SD card. And before "$XX < $200" (whatever these are going for now), I spent over $100 for my RPi 4 (top ram, I think 4? or was it 8?) especially since you need a case, power supply, etc. Also I had to buy a m.2 ssd and that special adapter.
EDIT: I don't doubt that some of you don't have stability issues and/or have a Raspberry 1 that's been running since release without issue. That just wasn't my experience. Maybe that's my fault, I don't know. I know that I bought "raw" boards, kits from multiple suppliers, and expensive cases that came with approved power supplies but without fail in a couple months it wouldn't be pingable/ssh-able and I'd have to power cycle it and sometimes the software on it would just lock up (same software wouldn't have issue on my home server). In the end maybe I'm an idiot or not smart enough to manage a RPi but especially after the stock issues not too long ago and who the RPi Foundation prioritized, coupled with my experience of the hardware, I'm not really interested in the platform.
Once an 8GB pi cost more than an i3 or i5 mini pc that is capable of being upgraded it became completely pointless even considering them.
The only things I now have running on Pi's is my PiHole on an old Pi 2 - it's been flawless for years, and a Pi 3 that handles humidity detection in the utility room and turns on an extractor fan.
But the Pi probably costs less once you factor in energy costs, no?
It absolutely is. I have 4 RPi devices in the house actively doing work and I only touch them when I need to change a feature or something like that.
1. Front end to my 3D printer - RPi 2.
2. Magic Mirror - RPi 4.
3. Front end to my ancient HP LaserJet to make it wireless - RPi Zero W
4. Detector for Dryer finishing and sending me a text message - RPi Zero 2 W
These require no babysitting whatsoever.
I have Pis that have been running for 3+ years now (and that haven't been running longer because I upgraded them to Pi 3 or 4 boards, although they are performing the same jobs), and it's not just the OS either - I have Raspbian on the Pi 3 that runs my desktop touchscreen monitor (which controls all my lighting, and has an uptime of 167 days), and Ubuntu on my Homebridge/Zigbee gateway (which has an uptime of 61 days because I upgraded it before Summer, but has been running for nearly 2 years now).
If you need more CPU power, the RK3588 boards (as long as they have proper Armbian support) are decent bets, but at that point an Intel N100 might be more interesting since those CPUs can average below 10W on light load and will boost to better overall performance. Yes, they will be more expensive, but the going rate is 16GB RAM and 512GB SSD for less than $200, so...
But if you want something that just works _forever_ at under 5W mean load, the Pi 4 is still it. And you can boot it off an USB SSD to great effect (I have a Proxmox instance running that way).
1. You can configure the zero2 so that when it is connected to the iPad with a usb cable, 2 machine will construct an ethernet over usb and you can mosh from iPad to zero2 without any external network. zero2 also get both the data and power from a single usb cable
2. zero2 is cheap to replace. Micro sd card is cheap and easy to be cloned. You don't worry about software or hardware loss.
3. zero2 is actually quite usable in headless/ssh mode. I used to do the same thing on zero1 and it's a little bit sluggish.
I actually have my entire setup stored in a mint tin box.
Where I've had problems with other Pis has mostly come down to power supplies that don't deliver enough power. Getting a properly sized power supply will likely solve a lot of your issues.
I have docker containers running on my local app server that I literally never think about and have been humming away without issue for years (aside from server restarts or updates which I rarely need to do and it's often as easy as a `docker pull`). That's what I'm looking for, if it requires any more interaction then I'd rather pay more. I'm not saying "my time is worth so much" exactly, it's not like every hour of my day is packed full and I couldn't do some more server admin but I don't want to do that. I have the money (which isn't really all that much more) to avoid it so I do.
It's the same way that in my early 20's I cobbled together local storage servers full of shucked hard drives to save money and now I pay for a Synology and raw drives. I just don't want to spend my free time doing that anymore.
My three Orange Pi 5's are in my home k8s cluster (mostly as local GitHub Actions arm64 builders, but also have m.2 storage for Longhorn and run the occasional one-off) will probably tick along without being looked at again for years.
If you have 3 of those, you can run a Proxmox cluster, which is a godsend for applications like Home Assistant, which require stability and uptime.
The only time I use a Pi these days is when I need to interface with something via the GPIO pins. Netbooting increases the reliability vastly.
With BirdNET-pi, I have had a few hiccups due to the SD cards getting worn out from constant writing.
I was surprised by just how much of the hardware and software of the Pi 5 was custom designed by (or for) the Raspberry Pi org. Custom MIPI implementation, custom IO chip, custom image signal processor, custom camera drivers (w/libcamera), custom video drivers (w/Mesa+DRM). They were able to cut out pretty much all of the Broadcom firmware and the VideoCore is no longer handling the bootup process.
Nope that part remains. The VideoCore VPU does almost nothing other than that and power management on the RPi5 though
If that's not an option AND your setup has no power issues (e.g. it has good power supply, externally powered usb hub etc), disabling journalling might also improve performance: tune2fs -O ^has_journal /dev/sdaXXX (can't do it on mounted fs, so either do it from initram consolee, do pivot_root to minimal fs in ram, or do it from another machine)
same for network
The web works again.
The broadcom thing made sense to begin with, they got a great deal on the pricing. But with the likes of the Rock Pro being pretty much on a price parallel now (and higher performing) it no longer makes sense.
I would expect the boards with the RK3588 to still be more pleasant to use than an RPi5, but the RPi5, used with an m.2 drive, should be significantly better.
You'd expect the experience to get better since people have been testing with prerelease software as well, but... we'll see.
All of them are either taking preorders or say they will start deliveries in late december or early january.
The only retailers that seem to have any stock at all are in the UK and buying from there is a nightmare due to brexit (assuming the boards ever arrive at all - they have a nasty habbit of taking people's money but never deliver).
[0] The test on the page looks like "AVAILABLE" and copies as " Available on Nov 03,2023 "
Radxa did have the Taco and Mediasonic does make compact 4/8 bay JBODs bur these are USB only.
There are also various NVME to SATA/SAS adapters but these mostly make use of port multipliers and speed won't be amazing. You'll probably also want some SATA backplanes to make wiring less nightmarish but these are annoying to source at a reasonable price.
My latest idea I had for something semi compact and performant was to try and use a USB4 SSD enclosure with adapters to connect a full size PCIE card. You'd then get a Thunderbolt DAS theoretically.
You can buy whatever else you want, even cheaper and better today, and after three years, getting a modern supported OS will be a total pain, due to binary blobs, custom drivers, custom "design solutions" etc.
RaspberryPi is currently the only SBC where you can take any revision of it and still get fully working(! yes, even wifi, bluetooth, etc.) software for it, even if it's not the newest and the greatest model.
If anything, support gets better with time, usually.
If you want to talk about bad support, let's talk about something like the NanoPi R5 series. Oof.
1. Raspberry Pi
2. X86 System
3. Barely functioning SOCs with no mainline support and major software issues until eternity.
A lot of the small/mini/micro form-factor systems (even ~five year old used corporate machines) are almost always the (significantly) better option at the price point. I bought a used Dell SFF system a couple of weeks ago for $119 that runs circles around the Pi in every way (16GB RAM, NVMe, real I/O, real USB, PCIe, six core i5, etc) while idling at 9 watts, which makes little difference in operating cost even in locales with the most expensive electricity in the world - especially as the Pi continues to be more power hungry and often requires a fan. The SFF system is even upgradable to 64GB of DDR4 and has a full length x16 PCIe slot and an additional x4 slot.
This machine does everything in my house, even functioning as a 2.5gb router which is already something the Raspberry Pi could never do. Add on a dozen docker containers, a VM or two, and the gap only widens. I did upgrade the RAM for $40 but the systems it replaced more than make it come out ahead on power and cost. You can still resell a Pi 4 for ~$70.
Perhaps most important - availability. There is no shortage of these systems on eBay and elsewhere (or new), there never has been and there likely never will be.
For many years the Raspberry Pi was a good default "go-to" for many applications but x86_64 has come a long way and at this point I struggle to come up with all but a few bespoke use cases for the Raspberry Pi.
A lot of recreational tinkerers are already giving up their free time, and accepting a slightly inferior product to the commercial options. To also spend more would be a step too far :)
Getting my 20-year-old mono laser printer network-enabled with $20 of hardware: Smart, e-waste-reducing move, even if it only supports 300dpi.
Doing the same thing with $100 of hardware: Why waste the time, when a brand new mono laser would cost less, offer better print quality, and have built in wifi?
Setting up a DIY networked camera with $50 of hardware: Fun little hobby project, maybe more secure than the sketchy cloud options and cheap imports that never get firmware updates.
Doing the same thing with $130 of hardware: Just buy from Axis or Pelco or whoever. They'll have figured out the waterproofing, and the night vision, and the PoE, and nobody will call you a dumbass for buying the "wRoNg tYpE oF mIcRoSdcArD"
When I had to buy a SBC last time I couldn’t bring myself to get the Pi 4 because it was missing core features (4K HDR decode) vs the alternatives. But I love the community around Raspberry Pi and it’ll definitely increase my options with this SBC.
I think the Pico/RP2040 works great for most basic electronics projects, and honestly you can get surprisingly far with its processor, up to even basic 3d graphics.[1]
[1] https://sites.google.com/site/tedrossin/home/electronics/ras...
If you want to run kiosk, Pi Zero is a much better choice. It already has all that stuff. It is $10 vs $4. It runs normal OS.
i figured that after a decade of selling millions of Pis, they could establish direct sales channels by now. but instead we are still stuck with re-resellers who charge a hefty markup on marketplaces people outside of electronics would ever visit.
this is like the major critcism that remains for me with their original mission. the price might have started being accessible, but to buy one has almost never been so.
I've had my eye on https://turingpi.com/product/turing-rk1/
