Raspberry Pi 3 (can i double the processor power?)

[ZXBasicKID]

Hello,

Admins, I couldn't find a section to post this in so please feel free to move it if required.

So, my question is this....

Is there a way to connect more than 1 Raspberry Pi 3 to boost its processing power. I ask because despite owning a couple of laptops i find myself using my Raspberry Pi 3 as my primary computer and whilst browsing speeds are pretty good, I'd love to double up the processor speed. I also use my Pi3 to write Python code (still a relative beginner)

Any help would be appreciated.

Cheers

Scott

[jondallimore]

http://www.southampton.ac.uk/~sjc/raspberrypi/

Its possible... although personally I would just use a laptop and install Linux on it.

[B.Goode]

You might like to follow this recent and still evolving forum discussion.


viewtopic.php?f=49&t=199994

But the short version is that there isn't a trivial way to meaningfully increase the cpu power of your existing RPi.

[W. H. Heydt]

Short answer: No.

Longer answer: No, because the SoC (GPU+CPUs) is soldered to the board and it would take a professional grade electronics rework station replace it. And even if you remove the SoC, what would you replace it *with*? This is not the x86 world where you can pop down to an electronics retailer a buy any one of several pin-compatible CPUs that can be put in the same socket.

Your only recourse if you want a more capable Pi is to wait until the RPT releases one. When will that happen? No one who talks about it knows, and those that know don't talk about it. The only hint that has been given was by Eben Upton when he said after the launch of the Pi3B in 2016, "Don't expect a new Pi for 2 to 3 years." So...in theory (I wish I lived in Theory. Everything works in Theory), we *might* see a Pi4B this year, and--if so--most likely in late February or early March (the RPF/RPT tend to release products around the time of year of the original Model B release anniversary), but don't count on it. Could be next year... Or the year after. Or....you get the idea. We won't know until there is a sudden flurry of interviews and announcements. Then everyone will get all excited and the initial production will sell out in a couple of days while RS and Farnell struggle to keep them in stock for a few months. After that, speculation will begin about what the Pi5B will bring and old timers will start rolling their eyes about it.

One other point... The "low hanging fruit" has already been picked. I wouldn't expect dramatic performance improvements from the next Pi. The current Pi is pushing the thermal envelope for passive cooling with no heatsink. The likelihood that the next Pi will use a 28nm part (instead of 40nm) should ease that problem, but not if there is a significant increase in clock speed. There is no particular reason to think that the Pi4B will go to an ARM core other than the current A53, so the IPC value will probably remain the same. Likewise core count. More RAM is *possible*, but may be precluded due to cost considerations (the RPF is pretty insistent on keeping the retail price the same).

[fruitoftheloom]

Hello,

Admins, I couldn't find a section to post this in so please feel free to move it if required.

So, my question is this....

Is there a way to connect more than 1 Raspberry Pi 3 to boost its processing power. I ask because despite owning a couple of laptops i find myself using my Raspberry Pi 3 as my primary computer and whilst browsing speeds are pretty good, I'd love to double up the processor speed. I also use my Pi3 to write Python code (still a relative beginner)

Any help would be appreciated.

Cheers

Scott

Depends on what software you intend to use, but joining 2 RPis together does not make them twice as fast.

https://www.pidramble.com/

[ejolson]

Admins, I couldn't find a section to post this in so please feel free to move it if required.

So, my question is this....

Is there a way to connect more than 1 Raspberry Pi 3 to boost its processing power.

Although Raspberry Pi clusters have come of age, there is still no forum area devoted to parallel processing.

[DougieLawson]

... there is still no forum area devoted to parallel processing.

That's because there's only one user on the forum who's interested in that stuff.

[Douglas6]

These posts always make me wonder if car manufacturers get customer support questions like "Can I put two Ford Fiestas together to get a 300 horsepower car?"

[B.Goode]

These posts always make me wonder if car manufacturers get customer support questions like "Can I put two Ford Fiestas together to get a 300 horsepower car?"

Maybe not, but back in the day there seemed to be petrol-heads who would discuss arcane projects like retro-fitting a little Ford Anglia with the power plant from a Mustang. Maybe similar things still go on?

[HawaiianPi]

Maybe not, but back in the day there seemed to be petrol-heads who would discuss arcane projects like retro-fitting a little Ford Anglia with the power plant from a Mustang. Maybe similar things still go on?

People stuff V8 engines in Mazda Miatas, and surprisingly it doesn't upset the 50/50 weight balance much. There are even companies who will do it for you if you don't have the tools/skills to do it yourself. V8 powered Miatas are insanely powerful and fast, but there's at least one person who thought just one V8 wasn't enough. https://www.youtube.com/embed/RP3VZ1l4qgU

And on the subject of double the processing power, there are other single board computers much more powerful than the Raspberry Pi3, and they don't cost much more (in fact, the ROCK64 is cheaper). But they don't have the support, and their software is not as stable or optimized. I have an ASUS Tinker Board, which is about twice as fast as my Pi3, but TinkerOS is a steaming pile of poo compared to Raspbian, and ASUS provides almost no support (they don't even have an official support forum). Raspbian OS and the great support system is why the Pi has done so well, while all the so-called "Raspberry Pi killers" have faltered.

[MaxK1]

+1/2.
The tinkerboard isn't bad if all you want to run are benchmarks. The support is very thin and seems to depend a lot on the pi forums. Tinker OS needs a bit work but it's sort of ok.

You can try overclocking, but it may be a bit of a crapshoot. One of my pi3 is capable of running, sustained at 1440,with a fan and heatsink. That is +20%. The other three pi3's only made it to 1350.

[jamesh]

Tiger made a twin engined 7 style car, two fireblade engines. Quite exciting apparently.

[W. H. Heydt]

Maybe not, but back in the day there seemed to be petrol-heads who would discuss arcane projects like retro-fitting a little Ford Anglia with the power plant from a Mustang. Maybe similar things still go on?

People stuff V8 engines in Mazda Miatas, and surprisingly it doesn't upset the 50/50 weight balance much. There are even companies who will do it for you if you don't have the tools/skills to do it yourself. V8 powered Miatas are insanely powerful and fast, but there's at least one person who thought just one V8 wasn't enough. https://www.youtube.com/embed/RP3VZ1l4qgU

People used to stuff large V8s into VW bugs. First step was to remove the back seat. But what I think is probably pretty much the ultimate "more engine than car" was done in the Netherlands after WW2 when some guy built himself a car around a Rolls-Royce Merlin engine. Apparently he was told to take it easy on the highways as he tended to scare the beegeebers out of other drivers when he blew past them. Especially at speeds in excess of 200mph.

And then for the "that was *fast*" (for the time) was the guy who set the land speed record in 1906 (a bit over 127mph) and set out to break it in 1907. There are two accounts of how fast he was going when he hit a bump in the sand on Daytona beach. One report said "over 180mph". The other was"195mph". He was using a over-sized (twice the normal displacement) Stanley Steamer engine and a stock boiler with the safety valve wired down. He actually survived the crash, though he was badly hurt. Pretty impressive speed for the time.

[ejolson]

These posts always make me wonder if car manufacturers get customer support questions like "Can I put two Ford Fiestas together to get a 300 horsepower car?"

It is certainly possible to hitch horses and then count the number of horses in the team. Apparently the right kind of horses sense what the others are doing and they all pull together. Although locomotives can also be hitched together to pull a train, Ford Fiestas were engineered according to more restrictive constraints and can not be hitched into a team--the same with chickens. Quoting Seymour Cray

If you were plowing a field, which would you rather use: two strong oxen or 1,024 chickens?

The question then becomes whether a Raspberry Pi is more like a chicken or an ox. Is one Pi able to sense what another Pi is doing and compute in the same direction?

Just as oxen and chickens have eyes and ears, a Pi requires some sort of communications hardware such as a network connection to sense what the other Pi is doing. The built-in 100-Mbit Ethernet of the Pi has twice the latency and 1/10 the bandwidth of a typical desktop computer. It is also a factor of 1000 slower than the interconnects used in typical computing clusters. Imagine driving a team of partially bind horses that can't hear.

On the other hand, chickens can see and hear fine, but generally have no desire to pull in the same direction. This is likely a software issue that is sometimes described by the phrase bird brain. Likewise, most software people run on one Raspberry Pi doesn't even try to leverage the help of another Raspberry Pi to increase performance.

Having said all of this, it is generally agreed that increasing parallelism either through quantum computing, GPUs, multiple cores or clustering is the way forward from a performance point of view. Lots of people are performing tasks that are trivial in the sense that they fit within the limitations of a single traditional processor thread. At the same time, with the exception of quantum parallelism, the Raspberry Pi provides a cheap but effective environment for learning about parallel computing.

[Douglas6]

[T]he Raspberry Pi provides a cheap but effective environment for learning about parallel computing.

Ah, but that was not the question that was asked (and is rarely asked). The OP asked "Can I connect two Pi's together so that I can download web pages faster?" That question is asked frequently, and the answer is a simple "No".

[ejolson]

The OP asked "Can I connect two Pi's together so that I can download web pages faster?"

Web browsing should be one of those computationally trivial tasks that only require a single CPU core; however, it has become one of the most computationally intensive tasks that many people perform. Modern web browsing software has been written to take advantage of multi-core CPUs, but does not take advantage of distributed-memory cluster computing. Creating a web browser that enjoys parallel speed-up when running on a computer cluster could lead to web pages that require a computer cluster to render--maybe not such a good idea.