ejolson
Posts: 4019
Joined: Tue Mar 18, 2014 11:47 am

Re: Best Raspberry Pi Alternatives 2019

Tue Oct 08, 2019 5:41 pm

jcyr wrote:
Tue Oct 08, 2019 4:55 pm
ejolson wrote:
Tue Oct 08, 2019 4:00 pm
jcyr wrote:
Tue Oct 08, 2019 2:52 pm

Mine arrives tomorrow. It'll be interesting to see how well my X86 Ubuntu numerical (non video) applications using power hungry video cards port to this integrated ARM host based SBC.
It's worth noting that the GM20B Maxwell GPU in the Nano is primarily designed for machine-learning workloads. In particular, the peak floating-point performance is

Code: Select all

GPU          FP16  FP32  FP64
Jetson Nano   472   236   7.4 GFLOPS
which makes the Nano's GPU slower at double-precision than the Cortex-A72 CPUs on the Pi 4B. Of course the Nano also has some ARM CPUs. I wonder how close they are in speed?

If you are able, I would be very interested to compare the relative performance of the ARM CPUs on the Nvidia Jetson Nano to the Raspberry Pi 4B by running this Pi pie chart program on the Nano after it arrives.
The algorithms I've developed in CUDA are cryptographic using integer math. Does the test you suggest offload computation to the CUDA cores? I'm not that reliant on the performance of the ARM cores.
I thought about making a GPU-accelerated version, but never did--maybe over Christmas vacation.

Currently the Pi pie chart programs are OpenMP only and do not offload anything to the GPU. Thus, the resulting pie chart would only compare the quad-core Cortex-A57 on the Nano to the Quad-Core Cortex-A72 on the 4B. Although, the main point of the Nano is having a CUDA-enabled GPU, comparing the ARM cores is still a little bit interesting.

I don't have much experience using CUDA for integer arithmetic. How much faster do your CUDA-accelerated encryption routines perform compared to equivalent CPU versions?

alnaseh
Posts: 63
Joined: Thu Jun 23, 2016 5:12 am

Re: Best Raspberry Pi Alternatives 2019

Tue Oct 08, 2019 6:13 pm

Im my case i managed to get 140ms glass to glass latency with the max resolution 3280x2464 with 21fps on newer codec h265. They have very tight integration with gstreamer.

User avatar
Gavinmc42
Posts: 4194
Joined: Wed Aug 28, 2013 3:31 am

Re: Best Raspberry Pi Alternatives 2019

Wed Oct 09, 2019 1:29 am

I've been ignoring the main guys while living in Piland.
Interesting "Fanless" embedded stuff coming from Intel and AMD.
Some big heatsinks though :D
They say for embedded Display market etc?
Even triple HDMI boards, I wonder why?.

Is doe300 is still around and working on OpenCL for Pi4/VC6?
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

ejolson
Posts: 4019
Joined: Tue Mar 18, 2014 11:47 am

Re: Best Raspberry Pi Alternatives 2019

Thu Oct 10, 2019 2:19 am

jcyr wrote:
Thu Oct 10, 2019 12:42 am
ejolson wrote:
Tue Oct 08, 2019 4:00 pm
If you are able, I would be very interested to compare the relative performance of the ARM CPUs on the Nvidia Jetson Nano to the Raspberry Pi 4B by running this Pi pie chart program on the Nano after it arrives.
Nice! Familiar 64-bit Ubuntu on Jetson.

piechart results:

Jetson-nano

Code: Select all

[email protected]:~/pichart-30$ ./pichart-openmp -t jetson-nanno -n 4
pichart -- Raspberry Pi Performance OPENMP version 30

Prime Sieve          P=14630843 Workers=4 Sec=0.62325 Mops=1499.12
Merge Sort           N=16777216 Workers=4 Sec=1.55443 Mops=259.036
Fourier Transform    N=4194304 Workers=4 Sec=1.17645 Mflops=392.173
Lorenz 96            N=32768 K=16384 Workers=4 Sec=0.713005 Mflops=4517.82

The jetson-nanno has Raspberry Pi ratio=25.5737
Making pie charts...done.
[email protected]:~/pichart-30$
PI4B+

Code: Select all

[email protected]:~/pichart-30 $ ./pichart-openmp -t pi4b+ -n 4            
pichart -- Raspberry Pi Performance OPENMP version 30

Prime Sieve          P=14630843 Workers=4 Sec=0.54815 Mops=1704.51
Merge Sort           N=16777216 Workers=4 Sec=1.51346 Mops=266.049
Fourier Transform    N=4194304 Workers=4 Sec=1.70926 Mflops=269.926
Lorenz 96            N=32768 K=16384 Workers=4 Sec=0.620327 Mflops=5192.79

The pi4b+ has Raspberry Pi ratio=25.0723
Making pie charts...done.
[email protected]:~/pichart-30 $
Odd thing about your benchmark... it sometimes thinks the Jetson has 8 ARM cores?
The Pi chart program knows there are four cores, but tests various numbers of worker threads including overprovisioning with double the available cores and reports the fastest time.

Thanks for running the tests. The results are surprisingly similar and yet different.

ejolson
Posts: 4019
Joined: Tue Mar 18, 2014 11:47 am

Re: Best Raspberry Pi Alternatives 2019

Thu Oct 10, 2019 5:18 am

jcyr wrote:
Thu Oct 10, 2019 3:53 am
ejolson wrote:
Thu Oct 10, 2019 2:19 am
The Pi chart program knows there are four cores, but tests various numbers of worker threads including overprovisioning with double the available cores and reports the fastest time.
Ah, ok. The merge sort and Fourier tests ran faster with 8 threads.

Code: Select all

[email protected]:~/pichart-30$ ./pichart-openmp -t jetson-nanno
pichart -- Raspberry Pi Performance OPENMP version 30

Prime Sieve          P=14630843 Workers=4 Sec=0.623114 Mops=1499.45
Merge Sort           N=16777216 Workers=8 Sec=1.22402 Mops=328.959
Fourier Transform    N=4194304 Workers=8 Sec=0.917132 Mflops=503.061
Lorenz 96            N=32768 K=16384 Workers=4 Sec=0.707082 Mflops=4555.66

The jetson-nanno has Raspberry Pi ratio=28.9537
Making pie charts...done.
These are all compute bound tests. I fail to understand why 8 threads would be faster than 4 on a 4 core processor.
My suspicion is that over provisioning results in faster execution times because it allows the Linux scheduler to keep all cores busy by time slicing when the parcels of work aren't quite the same size and wouldn't otherwise divide the problem in an optimal way. Generally, however, it's a bit mysterious.

In this case it looks like the overall improvement is about 10 percent.

If you still have the time, it would be interesting to install gcc version 6.x to see whether the merge sort runs significantly faster, as it does on the Pi when using the older compiler.

User avatar
MikeDB
Posts: 166
Joined: Sun Oct 12, 2014 8:27 am

Re: Best Raspberry Pi Alternatives 2019

Sat Nov 09, 2019 10:09 pm

Surprised no-one has mentioned the LattePanda range. The entry product is twice the price of a Pi but has around twice the performance. It also uses a 14nm process rather than the 28nm of the Pi4 so uses a lot less power per instruction. And you can run Windows as well as Linux if you wish.

But despite having an Arduino co-processor, it doesn't have the I/O capability of the Pi which is it's big drawback.

Another alternative is the STMicroelectronics STM32MP1 which is a bit more expensive than the Pi and is slower but has a separate ARM M4 I/O processor which is useful in some applications.

Not saying any of these is the best, and I've used all three. It all depends on your application.
Will the forthcoming MIDI-2 spec at last allow us to set the volume to 11 !!

User avatar
Imperf3kt
Posts: 3081
Joined: Tue Jun 20, 2017 12:16 am
Location: Australia

Re: Best Raspberry Pi Alternatives 2019

Sun Nov 10, 2019 12:44 pm

MikeDB wrote:
Sat Nov 09, 2019 10:09 pm
And you can run Windows as well as Linux if you wish.
A slight nitpick, but this should be reversed.
"You can run Linux as well as Windows if you wish"

The LattePanda comes with a single Windows license and was designed for use with Windows 10 - only the latest and most expensive 'Alpha' model supports Linux.
55:55:44:44:4C
52:4C:52:42:41

Heater
Posts: 14222
Joined: Tue Jul 17, 2012 3:02 pm

Re: Best Raspberry Pi Alternatives 2019

Sat Nov 30, 2019 10:46 am

It's a commonly known fact that when you need a Pi only a Raspberry Pi will do. There are no alternatives.

Of course if you need something else then you are spoiled for choice in these modern times.
Memory in C++ is a leaky abstraction .

Heater
Posts: 14222
Joined: Tue Jul 17, 2012 3:02 pm

Re: Best Raspberry Pi Alternatives 2019

Wed Dec 18, 2019 12:18 pm

jcyr,
These are all compute bound tests. I fail to understand why 8 threads would be faster than 4 on a 4 core processor.
That is a bit of a mystery.

My suspicion when I read your post was as follows.

1) Some job gets split into many little tasks that need doing.

2) Somehow those tasks are getting distributed over multiple cores.

3) But those tasks likely have a very variable run times depending in the data they are given. Some complete faster than others.

4) That can lead to threads sitting idle. They are blocked waiting on some other thread to complete so that they can deliver their results or get new work. Or perhaps in some phase of the job all threads have have to complete, synchronize, before their results can be combined and used, that depends on the longest running thread, the others being idle.

In short threads can get blocked waiting for others to complete. If there is only one thread per core that means the core is idle.

Now, if we over provision with more threads than cores then a core can get on with running a second or third thread to handle another little task, whilst that first thread is blocked.

The reason I'm replying here now is that I found this presentation about this very problem with C++ with millions of threads and cores in high performance computing on super computes. Hartmut Kaiser "Asynchronous Computation in C++" : https://www.youtube.com/watch?v=5xyztU__yys

In it Hartmut Kaiser describes how simulations of galaxies, weather models, etc, running on super computers can suffer from having their cores idle for 30% of the time. His description of the problem is much better than mine above of course.

But Hartmut goes further. Over provisioning with threads can help, but threads are heavy weight and have overheads of their own. The suggestion then is to use 'async' programming to run multiple of those little tasks on one thread.
Memory in C++ is a leaky abstraction .

cjan
Posts: 760
Joined: Sun May 06, 2012 12:00 am

Re: Best Raspberry Pi Alternatives 2019

Wed Dec 18, 2019 11:25 pm

there are no 'Best Raspberry Pi Alternatives' because the price $35~$55, not SOC board itself.

W. H. Heydt
Posts: 11243
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: Best Raspberry Pi Alternatives 2019

Fri Dec 20, 2019 4:59 am

AmeliaR wrote:
Thu Dec 19, 2019 7:04 pm
Hi,
The NanoPi NEO Plus2 is less than half the size of the Raspberry Pi, but it packs a performance and storage punch.
Specs are much closer to a Pi3B+ (4 x Cortex-A53 vs. 4 x Cortex-A72 on the Pi4B). Faster Ethernet, but fewer USB ports. Same RAM (1GB).

Oddly enough, their default OS (Ubuntu) is set up to have a default user of "pi". One might wonder where they got *that* idea...

User avatar
mahjongg
Forum Moderator
Forum Moderator
Posts: 12514
Joined: Sun Mar 11, 2012 12:19 am
Location: South Holland, The Netherlands

Re: Best Raspberry Pi Alternatives 2019

Fri Dec 20, 2019 10:19 am

AmeliaR wrote:
Thu Dec 19, 2019 7:04 pm
Hi,
The NanoPi NEO Plus2 is less than half the size of the Raspberry Pi, but it packs a performance and storage punch.
Doesn't seem to have any kind of video output (HDMI or Composite). at least I see no connectors for video out.

ejolson
Posts: 4019
Joined: Tue Mar 18, 2014 11:47 am

Re: Best Raspberry Pi Alternatives 2019

Tue Dec 24, 2019 6:10 pm

mahjongg wrote:
Fri Dec 20, 2019 10:19 am
AmeliaR wrote:
Thu Dec 19, 2019 7:04 pm
Hi,
The NanoPi NEO Plus2 is less than half the size of the Raspberry Pi, but it packs a performance and storage punch.
Doesn't seem to have any kind of video output (HDMI or Composite). at least I see no connectors for video out.
It seems we are replying to a robot who has subsequently added spam links to their earlier post.

Aside from that, my impression is the engineering behind the single-board computers produced by Friendly ARM is pretty good. Prices are slightly higher than others and there are no significant third-party add-ons. At the same time, the manufacturer provides a surprisingly complete selection of accessories on their own. The manufacturer's forum is not as popular or well moderated as this one, but email led to amazingly helpful replies directly from the manufacturer.

The numerical methods course I teach was recently moved from computer science to mathematics. Since it is no longer possible for the enrolled students to use the file servers in the college of engineering, I've been looking into setting up a small NAS to hold student work during the course. The Pi 4B is a natural choice, but this more expensive RK3399 based SBC from Friendly is tempting as it brings the PCIe port out to a header and then provides an interface card to directly attach NVMe m.2 SSD storage.

LTolledo
Posts: 2395
Joined: Sat Mar 17, 2018 7:29 am
Location: Anime Heartland

Re: Best Raspberry Pi Alternatives 2019

Tue Dec 24, 2019 9:10 pm

the item linked by ejolson looks interesting,
would like to add that to my Nanopi systems collection (my main/majority SBC range is from RPF, 2nd from FriendlyElect , 3rd from Pine64.org)
but alas it not available (yet) at my favorite local suppliers, and outrageously expensive on amazon (listed price is about 5x compared to SRP) :shock:

...and the best Raspberry Pi alternative (for those who still insist on using W10) is to get Ryzen based boards! :mrgreen:
"Don't come to me with 'issues' for I don't know how to deal with those
Come to me with 'problems' and I'll help you find solutions"

Some people be like:
"Help me! Am drowning! But dont you dare touch me nor come near me!"

Return to “Off topic discussion”