I think MS's biggest mistake was to not properly maintain and develop the Foundation Classes, basically a thin C++ wrapper library on top of the C API that retained most of the benefits of the Win32 API while eliminating a lot of the boilerplate code. Instead they went after Java with the .NET managed stuff, bloated and slow compared to the native API.
Qt is now the best "old school" UI framework by far.
> Now I run a S23 Ultra and after two years it still does everything I need.
Maybe, but it is fully under Google and Samsung's control, and is choke full of spyware. You couldn't pay me to use a stock (Googled) Android phone for this reason alone.
> I wish I could understand why it is so difficult to build an un-googled android image.
> It is so easy to install linux on a PC, yet I don't see the same happening for android while it's actually running a linux kernel, so it really begs the question.
It's not particularly difficult -- see Graphene and Lineage. The main issue is that there are few phones on which to run these custom builds. Ironically, Google Pixels allow to run other operating systems than the one they come with (the bootloader can be unlocked). Other than the Pixel and a couple of Chinese models, you are looking at low-end or ancient hardware. You can't just build a phone without OS and install Linux/Android like you would on a PC.
I think the conclusion here is that Raspberry Pis are now too pricey (especially when factoring in the various required accessories) and rarely make sense for typical desktop use vs. x86 mini-PCs. They make even less sense compared to various used thin clients that can generally be found on eBay.
You're paying a premium for physical compatibility with a ton of niche accessories. Whether or not they make sense depends on how important those accessories are to your use case.
That and the prices never really came back down to earth after the chip shortage hikes.
> You're paying a premium for physical compatibility
No. There are a bunch of alternatives with some to full pin compatibility. Some being many times faster [1]. No new projects should use a new Raspberry Pi.
Your video rates the PI as 10 for support, 10 for ease of use and 7 for performance. Just the support and ease of use is enough. You're paying for a mature ecosystem where you know things work and you don't have to waste time struggling.
Thanks for the video. I just bought a RPI5 and was curious if this was a mistake, but after watching the whole 'I love PI' video, I am still okay with my choice.
It is good to know that there are other boards with better multi-thread performance and AI capabilities. However, there are also a few things I disagree with in the test setup, such as rating only multithread performance and giving the best single-thread performance the lowest overall rating. In addition, concluding the AI tests without the extension board for the RPI5 seems a bit weird.
So thank you for the video, but I think it depends on what you are trying to achieve and it is not a simple there you get more bang for your buck.
It would make a lousy desktop computer even if it was 10x as powerful.
- high current 5V USB power supply you probably don't have
- HDMI micro port you have like 1 cable for
- PCIe through very fragile ribbon cable + hodgepodge of
adapters
- more adapters needed for SSD
- no case, but needs ample airflow
- power input is on the side and sticks out
GPIO is the killer feature, but I'll be honest, 99% of the hardware hacking I do is with microcontrollers much cheaper than a Pi that provide a serial port over USB anyways (and the commonly-confused-for-a-full-pi Pi Pico is pretty great for this)
We had a problem trying to bring up a couple of Pi 5, hoping they'd represent something reproducable we could deploy on multiple sites as an isolation stage for remote firmware programming. Everything looked great, until we brought one somewhere and untethered it from ethernet, and we started getting bizarre hangs. Turned out the wifi was close enough to the PCIe ribbon cable that bursts of wifi broadcasts were enough to disrupt the signal to the SSD, and essentially unmount it (taking root with it). Luckily we were able to find better shielded cables, but it's not something we were expecting to have to deal with.
I dunno, I brought a pi 500+ with an SSD, 16GB RAM, little screen, PSU, mouse and cables. It was around £300.
It's not super powerful but my young kids use it to surf the net, play Minecraft, do art projects, etc. (we are yet to play with the gpio).
I don't get on with the keyboard but otherwise would make a decent development machine for me, considering my development starts with me ssh'ing into some remote VM and running vim.
The whole lot is tiny and extremely portable, we pack it away in a draw when not in use.
All in it felt like good value for money for something that took about 3 minutes to get up and running.
That's actually about the same price as the pi 500+ without the screen. Except that one has 500gb Vs 256gb SSD, but doesn't have the snazzy led keyboard.
I never really understood how GPIO is a killer feature with them. There are so many ways to get GPIO, from $5 USB dongles to any microcontroller/dev board that's ever exists. What's special about Raspberry Pi GPIO that I'm missing?
The only case I can think of is very heavy compute that relies on low latency GPIO related to that compute?
The low latency is the reason why the PiStorm (Amiga CPU accelerator) project works so well on a Pi 2, 3 or 4. (Pi 5 is no longer suitable since the GPIO is now the other side of a PCI-E bus and thus suffers significantly higher latency than on previous models, despite being much faster in terms of throughput.)
In general the best part of the Pi is that it's so stable as a platform. Everyone has the same hardware and generally the same OS so guides online just work - it may be one of the if not the best Linux on the desktop experience I've used personally.
Along with that the gpio is there and ample so it's extremely easy to just start using it.
I do argue an esp 8266 or esp32 are better for a development microcontroller but you have to muck with cabling it up before you can even load a program on it which is a few more extra steps than a Pi
For desktop use cases, sure. But the Pi's target market is makers and educators who want small and efficient and can interface easily with peripherals like cameras and GPIO. Desktop users and low-end home labbers are a distant second.
Small yes, efficient no. It's more or less on par with comparable minipcs in power draw, needs an active cooler and idles at over 3W. That's more than a Pi 2 uses going flat out. They keep increasing power draw with marginal efficiency gains and don't bother to do much power management. It's absolutely atrocious compared to the average smartphone SoC.
I really hope there's some kind of battery oriented low wattage high efficiency version planned someday, because we're up to requiring a 5A power supply and it's getting absurd.
Aren't those Fritz!Box routers (common in Europe) precisely examples of "shitty routers with terrible security?"
The first thing I would do with a typical residential Internet connection is to ask the ISP to give me an ONT so that I can use my own router, a commodity x86 PC running Linux. Their underpowered plastic boxes simply won't cut it when it comes to complex firewall rules and high VPN throughput. I also don't want to deal with their shitty web UIs and would rather script the setup I want.
I have yet to find a security issue with it. I know German ISPs misconfigured their management network at some point, letting the Fritz!Boxes access each other, but that would've happened with any managed modem that was misconfigured like that.
I bought my Fritz!Box. My ISP has no control over it. TR-069 and other upstream management protocols have been disabled completely.
So far, I'm easily getting gigabit+ speeds across both IPv4 and IPv6. VPN is too much to ask (beyond emergency LAN access, I suppose) but that's what the home server is for.
The web UI is kind of nice, actually. Maybe not to everyone's taste, but the firewall management is a lot less of a clusterfuck than trying to properly configure simple port redirects over the command line. Heaps better than OpenWRT in my opinion. I've run my own Debian router box for a few years and I can say I'm doing just fine without.
I'd say a Fritz!Box is a good router for normal users. Easy interface. Good enough hardware. Stable modems. Some nice software features. Absolutely not a device for prosumers.
No really, they are pretty decent. I stopped running an old PC for router and firewall after I got a Fritzbox. It can traffic-shape, forward ports, configure fixed IP addresses and DNS names, provide limited guest access to the WiFi, analyze the WiFi spectrum (and show a graph) to choose uncongested channels, and do a whole bunch of things that I don't use but which are conceivably useful like VPN server, file server and such.
> Aren't those Fritz!Box routers (common in Europe) precisely examples of "shitty routers with terrible security?"
Not at all. They had security bugs, sure, but not constantly. Each device has a randomized admin password from the factory. Some changes require physical hardware access because one needs to press a button to confirm. They support the hardware for ages. Their 7490 model just got a feature firmware update. The model is 13 years old!
In Germany, if you ask someone where his router is he might not know what you talk about. But he understand if you asked about "your fritzbox". (Even in cases where they have something else.)
But enough of the glazing. In 2024 they got sold to private equity. Lets see how the enshittification will treat them.
Well, normal people generally can’t be trusted with cars: in one country of only 3.5% of the humans we kill two jumbo jets worth of people with them every day.
Tylenol is another example. Building materials is a third (building and fire codes are a relatively recent invention). Hell, even penicillin is by prescription only.
Letting the circumstance happen where median people can easily cause externalities through ignorance or carelessness is how we incinerated the planet and destroyed the biosphere as we know it with fossil fuel emissions, because it’s nbd (still even now in 2026, when we know about runaway polar greenhouse curves) to get in your ICE car and drive to the corner store.
When normal people had GP computers, we got botnets millions strong and DDoS in the Tbit/sec range and keyloggers on every hotel lobby computer hooked up to the boarding pass printer. Median people are way safer on the internet now than before.
> Well, normal people generally can’t be trusted with cars: in one country of only 3.5% of the humans we kill two jumbo jets worth of people with them every day.
If you mean Indonesia (the county closest to 3.5% of the human population) or the US (the nearest above 3.5% at 4.1%+) then you are high by an order of magnitude. Two jumbo jets are around 1000 people. US car deaths are around 100 a day and Indonesia is a little lower.
If you mean Pakistan (the next country after Indonesia at 2.9%) you are high by close to two orders of magnitude. They have around 15 deaths a day.
That seems like an untenable stance. Most people don't pick healthy foods to eat or exercise as much as they should. Should we dictate what they can and cannot eat etc?
In the US, in my experience, young people don't want to deal with cash at all. Older people do, but it's not always convenient to meet up.
Most banks charge a fee for sending a wire. Sending an ACH is free, but most restrict that to your own account. Revolut is the only one I've seen that lets you just spam ACH to anyone. In both cases, it isn't instant.
Zelle largely fixes those issues, but has its own issues, like a lot of banks not supporting it and/or arbitrarily low send limits.
I don't understand either. My contact surface with my bank is so small. I log in once a month to download transactions. What is everyone doing that they need constant immediate access on their phones? I'd probably debank before buying a special iPhone to access a bank account.
Let me give you a preview of a world coming to you, and present day reality in Ireland:
1. Your employer pays your salary by bank transfer, which requires you to have a conventional bank account.
2. You then want to spend that money, how do you do that?
Debit card? You need the phone app to retrieve the PIN when the bank first sends you the card.
Cash withdrawals in the branch? For amounts less than €10,000, the staff will direct you to the ATMs in the branch. These require an activated debit card to withdraw money, and activating that card requires the phone app.
Manual money transfers in the branch? Once again, for amounts less than €10,000, the staff won't do it - they'll instead direct you to the PCs in the branch. These are just loading the same website you can access on yours, which will ask you to the confirm with a 2FA push notification to log in.
Try another bank? The legacy banks all got the same auditor who advised them that app based 2FA is the easiest way to implement PSD2, and reduce the likelihood they get held liable when customers get scammed, so they all implemented that as the only option. The neobanks of course, are accessed solely by apps.
Yeah I wish we could do without a bank in modern life. When bitcoin first began I was really in support of it because I saw potential in freeing us from the dark stranglehold of the banking industry. Everyone just manages their own digital money.
But nope the cryptobros just turned it into another pyramid speculation scheme and the governments ruined the customer independence with their KYC stuff. Now it's just an online version of the old system where the exchanges are the new banks.
Qt is now the best "old school" UI framework by far.
reply