Slashdot Mirror
New Multi-Core Raspberry Pi 2 Launches
First time accepted submitter MicroHex writes Coming in at the same $35 price-point that has come to be expected from the Raspberry Pi, it looks like the new Model 2 will be packing a quad-core ARM processor with a GB of RAM. From the article: "The Raspberry Pi Foundation is likely to provoke a global geekgasm today with the surprise release of the Raspberry Pi 2 Model B: a turbocharged version of the B+ boasting a new Broadcom BCM2836 900MHz quad-core system-on-chip with 1GB of RAM – all of which will drive performance "at least 6x" that of the B+."
Man, you guys sure do edit harshly =p I don't see a word I wrote in there.
Maybe you should take two minutes and read the FAQ. The Raspberry Pi's primary design goal was to be low cost. There are a hundred other companies now selling more powerful (and expensive) boards. This was designed to be a learning tool for students and hobbyists, not a set top multimedia box.
Only the State obtains its revenue by coercion. - Murray Rothbard
Well, when you design a board with all the Pi's features, with your choice of SOC, that can be effectively sold at a $35 price point, you let us know. Until then, why don't you just accept that there are various products out there, with various strengths and weaknesses (and various prices!), and accept that some people have managed to do some pretty cool things with the original Pi, and no doubt they will do more cool things with this version.
Good grief is the naming scheme tiresome.
Did anyone think about problems this goofy naming scheme causes? The ease of searching supplier's catalogs, googling, etc? Hell, just talking to another person? "Oh yeah, I've got the Pi 2 Model B plus", versus "I've got a Model D." Did anyone concern themselves with the fact that a lot of resellers may not ID the revision at all? How are you supposed to google for an issue you're having with the latest model?
Please help metamoderate.
Phones come with a touchscreen, speakers, microphones, WiFi, Bluetooth, GSM, and an accelerometer, which also cost money. Probably more money than a few connectors on a board.
I'm laughing at those Windows 8 users posting here complaining that a friggin GIG of RAM isn't enough. Most rPI projects are also done on Arduinos and similar, with a 20Mhz clock and RAM measured in bytes. Typical Pi programs are hundreds of bytes. 1024 bytes is 1024 small variables; how many do you need to turn lamps on and off, or position a servo?
Running your 200 byte program on top of a Linux kernel is just a convenience. It's not made to run Microsoft Office on it all day, it's designed for reading a few switches, turning on a motor, and lighting an led - which requires about 24 bytes of RAM.
Of course some people use them as entertainment media centers. That's kind of the one oddball use that needs a thousand times the resources of most things people use their Pi for.
Yes, it's underpowered and possibly overpriced in comparison to (x, y, z,...)
But the Raspberry Pi has a large and growing ecosystem behind it -- developers (hardware and software), users, and more.
The Arduino is a similar beast -- underpowered, overpriced, and with a tremendous ecosystem, approachable and available to new classes of users.
As an example, look at what Adafruit is doing with Arduinos and the Raspberry Pi -- making them available, accessible, and useable by a wide audience, not just those tho are comfortable rebuilding kernels.
Look at other historical examples -- the underpowered 6502 (Apple ][) or that atrocity with 640k is good enough for anybody, right?
If you insist that 4 USB ports and an array of GPIO pins are the main selling features, then let me present to you the Odroid C1: http://www.hardkernel.com/main...
It's everything the Pi and Pi 2 is, and everything the Pi and Pi 2 never will be, for the same $35.
I think they're trying to maintain compatibility with the existing ecosystem. The GPU did not change, they just added 3 more cores and another half gig of RAM. This is a drop-in replacement to keep their product competitive without breaking anything too drastic with their existing product line. Sort of along the lines of why the iPhone 5 had a taller screen and iPhone 6 actually had a usable sized screen. Baby steps. Those Chinese phone sellers don't have to support that product after they wrap it in bubble wrap and drop it in the mail; the RPi organization has industrial customers who have standardized on their hardware as a Long Term Solution and make up a sizable portion of their business (they're forcasting approx 20% of their business in 2015 will be industrial customers). So there's that.
The Raspberry Pi 3 in 2017 or so should be pretty amazing, between that phone you linked to, and the new ESP8266 it's clear we've only waded hip-deep in to the era of ultra low energy, high powered wireless devices. In the mean time this is a very acceptable bump in performance to what originally was an educational toy.
moox. for a new generation.
The Raspberry Pi series is an awesome hobbyist device at an impossibly low price point.
I'm glad they are finally offering more memory and multi-core processors. That way I don't need to get a BananaPi or other copycat. This way, I can continue to support the vitally important Raspberry Pi foundation and their goals.
Thanks for finally offering more memory and multi-core. Next time let's also choose a truly open framebuffer, or let's pressure Broadcom to open their VideoCore architecture once and for all.
Kriston
No it isn't. It's a quad-core Cortex A7.
It's a quad-core Cortex A7, so very firmly ARMv7.
Ignoring the actual Pi debate -
Darn those people who still use 68HC11 and 6502 controllers. You seem experienced enough to likely know yourself that if you've got a chip that's cheap enough not to ruin your BOM, that is available/in-production, does the job, has a solid toolchain, and coupled with years of development experience globally and in-house, then you don't just throw that all in the trash because something newer/faster/smaller/cheaper comes out.
Half the time I think a lot of people jump to the newest stuff because they don't like having their exclusivity eroded. Using older stuff makes sense when your product doesn't need cutting edge and you want to have a wealth of experience / dependability to draw from.
As for me, I'm still enjoying the AVR Tiny4/5/9/10 series, it's like the modern 555 ;)
There's just no rationalizing away the fact that they have been grossly underpowered regardless of context.
I have to disagree there. They are plenty powerful for learning basic coding on. They're plenty powerful for a basic web server for a local network. They're plenty powerful for controlling various bits of hardware via the GPIO port. They're plenty powerful for plenty of things.
The BCM2835 which is present on all the previous Pi boards contains a half-baked USB controller core which is the cause of all the USB event dropout problems. It expects realtime response from the ARM11 to handle USB's split transactions within the required 1ms response window of USB. The ARM11 cannot always meet that response spec, and so the USB user experiences a dropout.
More details are given in this post and there are plenty of threads on the raspberrypi.org forum in which the Raspberry Pi Foundation's engineers confirm the hardware fault inside the BCM2835 SoC.
The Pi range of boards have had many other USB-related problems fixed in recent versions, especially those associated with the very poor power supply circuitry of the first release. On the whole the situation is much better, but the core USB dropout problem is not fixable because it's part of the BCM2835 chip.
Hopefully the new BCM2836 in RPi 2 does not use the same half-broken USB controller core as the BCM2835.
Keep your fingers crossed.
The Raspberry Pi has the hardware to be very cheap while still being able to connect to a general lab setup and powerful enough for a lot of nice little projects.
Yes, but can you get proper English language documentation for them? An AllWinner CPU is a good performer but the datasheet is only available in Chinese. The reason it's so cheap is that they didn't spend any money getting it translated or providing support overseas.
Also, Broadcom parts have a longer production life than the really low cost ones. They will be able to buy the same or fully compatible parts in five years time and keep making the same Raspberry Pi boards, providing a fixed hardware platform for people to work with. The low cost parts tend to get replaced fairly quickly. The people making those cheap phones will first look for some cheap SoCs and LCDs, then build the phone around them and make 100,000, and that's it. Parts are now obsolete and no longer manufactured, and the whole process repeats.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
It's everything the Pi and Pi 2 is, and everything the Pi and Pi 2 never will be, for the same $35.
Except for the documentation and support. The Pi is an educational computer, there are loads of tutorials, books, accessories and datasheets available for it. Your kids won't find their school offering classes for it. Good luck getting support on driver bugs, or even diagnosing why your stuff doesn't work. The Odroid might be more powerful but it isn't really suitable for n00bs.
It's the same with the Arduino. People laugh because performance is crap and it's over-priced, but it's also much easier than anything else out there and hence very popular.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
They released all the docs for the GPU, drivers are on their way http://dri.freedesktop.org/wik... https://www.youtube.com/watch?...
Did anyone notice this: http://dev.windows.com/en-us/featured/raspberrypi2support
Apperantly at zero cost... Might get interesting...
When using an Arduino, I can realize a whole project in just hours, including setting up the (very simple) IDE, starting a template project, searching and installing some helper libraries (Timer, I2C, Serial, LCD) and filling in the glue code on the position marked in the template. When using a different target, even setting up the tool chain can take days. I would have to buy hundreds or even thousands of chips until the investment in a different tool chain and the development (and debugging) of the missing libraries would pay of.
When I get stuck with an Arduino, I can find lots of documentation, lots of working (!) example code and even working (!) step-by-step tutorials (even video-tutorials). This seriously limits the risks when developing with Arduino. In the embedded world, it is very easy to find surprising show stoppers for a certain approach on a given platform.
So there are several good reasons to use Arduino (or Raspberry Pi) for home grown or semi-professional projects, even when there are other options with lower cost per chip.
The point of a Raspberry Pi isn't to replace an Intel i7 clocked at 3GHz, it's to replace a 6502 clocked at 2MHz: to provide kids with a system to hack on. You don't need shedloads of performance to develop great software, and, indeed, the less resource you have, the more inclined you are to code tightly and efficiently. As a learning tool, less really can be more.
I'm old enough to remember when discussions on Slashdot were well informed.