No Raspberry Pi 4 in 2019.
-
Kind of weird responses from Upton... I was not expecting the Pi 4 this year, but some of his answers made me raise my eyebrows, especially these:
"I think we kind of understand what featureset we want [and] what would be involved in getting that featureset. I don’t think we have a defined plan for turning that into a product yet."
Upton told us that he and his team haven't yet decided on a process node.
So he wants us to believe that since the Pi 3 was released in 2016, they might have gained somewhat of an understanding of the features they want but they don't even have a plan for how to turn it into a product. Considering the challenges they face, i.e. the need for a completely new SoC, I'd say that means the Pi is at least two years away from launch. The chance of them going from "there's no defined plan" to having designed/manufactured/tested a new SoC, implemented it on a new board and written launch ready software/drivers for it in one year are slim, to say the least. I'm a programmer and embedded systems designer myself, so I have at least some experience to draw from.
All of the above is if Upton is completely honest in his answers. There's at least some chance he's just trolling us all.
-
@Brunnis said in No Raspberry Pi 4 in 2019.:
There's at least some chance he's just trolling us all.
One can hope
-
with eric anholt (the guy who wrote the open source driver for the pi's VC4 gpu) about a year deep into writing the open source driver for the broadcom VC5 gpu (see https://anholt.github.io/twivc4/) i think it's a fairly good bet that the RPI4 will have VC5 (gles 3.1, vulkan, etc), which i assume dictates the kind of cpu/chipset/memory configs it can have, etc.
there's also been engineers on the RPI forum saying they know what it's going to be (i can't find the link), so i think upton is being a little bit evasive for whatever reason. i'm pretty sure they know the design at this point.
-
@dankcushions I seem to remember also something like this - maybe it was this topic ?
-
@mitu said in No Raspberry Pi 4 in 2019.:
@dankcushions I seem to remember also something like this - maybe it was this topic ?
yes, that's it - thanks! specifically: https://www.raspberrypi.org/forums/viewtopic.php?p=1409794#p1409794
Everything we have done so far has evolved the original design. I know the specs of the Pi4. It's not an evolution, it's a revolution and a HUGE step up.
-
@dankcushions Ahh, yes, I remember seeing this. I forgot to mention it in my first post, but that’s one of the reasons I said Upton may be trolling us. It’s interesting that it’s mentioned to be a revolution. I have always expected that, but many people still seem to believe that the Pi 4 will just be a slightly higher clocked A53 based design, maybe with more cores. I’d say that there’s a very high probability of them going with another symmetrical four core design, with conservatively clocked high-performance cores. I definitely expect them to try to lower the power consumption as well.
-
@BuZz said in No Raspberry Pi 4 in 2019.:
He got a 2 day ban for making it personal. As said - read the whole thread.
Just a friendly reminder: That was 4 days ago and he still appears to be banned. :)
-
@Clyde not sure. It says the ban is temporary. Maybe he has to log in to disable it and he hasnt. I can manually remove but I suspect that's the case.
To help us help you - please make sure you read the sticky topics before posting - https://retropie.org.uk/forum/topic/3/read-this-first
-
@BuZz A okay, just noticed it by accident. :)
-
That went from 0 to 60 faster than a Rimac C_Two. It's sure easy to do. Maybe it was just a bad day.
It seems store bought emulation is a far more enticing prospect than building from scratch from the popularity of all the "mini" consoles being sold by the millions.
Some interesting numbers from the foundation as of December 2018 and the emulation scene didn't even get a honorable mention. I think it would be interesting to see actual numbers but from poking around it sure looks like emulation is a smaller percentage in the big Pi picture as an educational board.
A lot happened in 2018, the Raspberry Pi folks has some interesting stats, marketshare / units and more.
-
23 million + Raspberry Pi computers sold.
-
250,000+ young people engaging with our clubs, competitions, and programmes every week.
-
Tens of thousands of young people learning digital skills through our partnerships with the Scouts, the National Citizen Service, and the European Space Agency.
-
30,000 + volunteers mobilised.
-
50,000+ teachers and volunteers have taken part in our online courses.
-
Over 1 million projects completed on our new projects site.
-
-
@Riverstorm That are interesting numbers and yes the Pie is a huge educational platform.
Haha ;) Maybe @BuZz can show other stats.
How many users registered each year (from 2016-2018)
How many posts were done during the years and so on ... I think the whole forum will not outstand the educational factor the foundation gives.The Pie is a great multi-purpose mashine!
Yes it is
-
@Clyde said in No Raspberry Pi 4 in 2019.:
@thelostsoul said in No Raspberry Pi 4 in 2019.:
For me its ok not having a new generation of Raspi this year. :-)
Same here. My Pi 3b (without +) runs all the systems that I need in my upright arcade cabinet, and I have too many games to play anyway. For anything else I have my Linux PCs (1 Tower PC that I seldom use nowadays, 1 Mini PC, and 1 Laptop).
Right - For Retropie my 3B+ is all I need/wanted for now...
I had hoped for a new Raspi with native HEVC and (extra wish) with 10bit decoding for h264/h265 (no, i am not asking about 12bit), but well - that can wait (as HD HEVC decoding in Software with the latest OSMC/KODI installs is as smooth as it can get (no dropouts/lags/etc))And as said, the Raspi is a Tinker Board... Playing around with a Breadboard and some sensors/actuators/whatsoever is addictive as retrogaming and more so if you have kids -> gaming can be fun (for kids: building your own ones in scratch or similiar visual programming languages is even more fun), but building your own devices that are controlled via the raspi is so... sooo much more an experience (and yes education) - and comparing both communities from outside their respective filter bubble, I am not sure which one has more "followers", but my bet would be on side of the non-emulation tinkers (with a huge overlap of people belonging to both sides education/edutainment/tinker and retrogaming/tinker EDIT: ok, media-center only people could be considered as another group (I know at least 2 People who only bought their Raspi to be used as a MediaCenter (KODI) and nothing else))...
-
@Ashpool said in No Raspberry Pi 4 in 2019.:
comparing both communities from outside their respective filter bubble, I am not sure which one has more "followers", but my bet would be on side of the non-emulation tinkers
Mine would be, too. Just looking at all the projects that are featured on https://www.raspberrypi.org, I'm astonished every time what people can and will do when they're given such an open little allrounder. Even many scientists use it for their research!
-
You also cant compare Odroid to the Pi as they are for two separate markets.
The Pi has multiple options and its used in thousands of different applications.
Obviously here, people are about the gaming, but that is a very small part of the pi.
Another huge difference is the fact the pi uses its memory quite well when compared to the Odroid. People speak a lot about specs, but those who have programmed or worked on games knows that specs don't mean a damn if the programming cant talk to the system correctly.
In speed tests, the Odroid in its native Ubuntu runs some apps slower then Pi on its native OS.
-
@UDb23 this made me sad :(
-
https://coral.withgoogle.com/products/dev-board/
Is this Google's RPi4?
Edge TPU Module (SOM)
NXP i.MX 8M SOC (Quad-core Cortex-A53, plus Cortex-M4F)
Google Edge TPU ML accelerator coprocessor
Cryptographic coprocessor
Wi-Fi 2x2 MIMO (802.11b/g/n/ac 2.4/5GHz)
Bluetooth 4.1
8GB eMMC
1GB LPDDR4
USB connections
USB Type-C power port (5V DC)
USB 3.0 Type-C OTG port
USB 3.0 Type-A host port
USB 2.0 Micro-B serial console port
Audio connections
3.5mm audio jack (CTIA compliant)
Digital PDM microphone (x2)
2.54mm 4-pin terminal for stereo speakers
Video connections
HDMI 2.0a (full size)
39-pin FFC connector for MIPI-DSI display (4-lane)
24-pin FFC connector for MIPI-CSI2 camera (4-lane)
MicroSD card slot
Gigabit Ethernet port
40-pin GPIO expansion header
Supports Mendel Linux (derivative of Debian) -
@dankcushions said in No Raspberry Pi 4 in 2019.:
there's also been engineers on the RPI forum saying they know what it's going to be (i can't find the link), so i think upton is being a little bit evasive for whatever reason. i'm pretty sure they know the design at this point.
I found an old thread on the RPi forums from November where jamesh ("Principal Software Engineer at Raspberry Pi (Trading) Ltd.") writes a number of interesting comments regarding the Pi 4:
Regarding whether the Pi 4 will be evolutionary or revolutionary:
https://www.raspberrypi.org/forums/viewtopic.php?t=226958#p1392095Revolutionary. The Pi4 looks pretty good to me! 5 and 6 also have impressive specs, although subject to change this far ahead, and there's a lot of work required for them.
Meanwhile, enjoy the current range, it's a while yet before the Pi4 comes out.
https://www.raspberrypi.org/forums/viewtopic.php?t=226958#p1392129
We also have a roadmap for the next 5 years. Which could change, but not the Pi4, that's pretty much set in stone.
I doubt anything wil change on the Pi4, too far along in development, so the sword wil remain firmly installed in the stone.
As response to a comment quoting Upton as saying the following back in 2017: "It's a long road to get to Pi 4," Upton said, "but we'll get there eventually."
https://www.raspberrypi.org/forums/viewtopic.php?f=63&t=226958&start=75#p1399467Eventually. Lots of work to do...
I'd say this paints a slightly different picture compared to the Upton quotes from Toms Hardware. I didn't really get the impression from Upton that the Pi 4 is pretty much set in stone. To be honest though, I would be quite surprised if it wasn't set in stone at this point in time.
It's pretty clear that we're unlikely to see the Pi 4 for yet a while, though. I'm guessing for Pi Day in a year's time (but obviously still hoping to be shocked by an announcement this coming Thursday :D).
-
jamesh on the Raspberry Pi forums March 26 2019:
Pi4 feature set IS set in stone. Of course, I'm not going to say what that feature set is, but it's pretty cool.
-
-
@mitu Or they’ve gone for a 12/14/16 nm class manufacturing process, enabling them to drastically lower power consumption while still providing a rough doubling of performance.
One can dream.
EDIT: To expand slightly on that. People widely believe 28 nm will be used for the Pi 4. I'm not so sure about that. 28 nm is rather old now and has been used for cheap SBCs for years. 40nm was just ~3 years old when the first Pi launched. 28 nm is now ~7 years old already and will probably be at least 8 years old when the Pi 4 launches. Now that the RPi Foundation takes the rather large step to design a completely new SoC for the Pi 4, I'd guess that they'd try to be as forward looking as they can afford, in order to be able to build a couple of future SoC variants on the same process. They can pretty much guarantee some rather large chip volumes to the foundry as well.
Also, they'll probably want to lower the power consumption on the Pi 4 compared to the Pi 3 B+ or even Pi 3. Power consumption has ballooned because they've been stuck on 40 nm and because they've wanted to squeeze every bit of performance out of the SoC. That has lead to power consumption figures that I'm guessing they're not all that happy with. Besides pushing the USB Micro-B capabilities for power delivery, it is incredibly easy to make the SoC throttle without additional cooling (two integer heavy threads will do the trick within minutes). Even with 28 nm it may be hard to combine both a decent power consumption reduction and a large performance increase.
28 nm is probably a lot cheaper than the more recent processes, though, so it's entirely possible that it is the only feasible process in order to meet the ~$35 target.
Contributions to the project are always appreciated, so if you would like to support us with a donation you can do so here.
Hosting provided by Mythic-Beasts. See the Hosting Information page for more information.