Raspberry Pi 2

[drmike]

@TheLinuxBug indeed.

I am waiting until someone here gets one and actually uses it  and tells us how much better it is.   The Model B perturbs me on sound quality (horrid) and the whole shared bus for USB + Ethernet is a buzzkill.   Wondering if they changed things on the newest model.

 

[trewq]

The Model B perturbs me on sound quality (horrid)

A $4 USB sound card gets around that. I have a B+ Pi in my car for playing music, I know how horrid the onboard sound is.

 

[drmike]

A $4 USB sound card gets around that. I have a B+ Pi in my car for playing music, I know how horrid the onboard sound is.

That's what I did... One of those Diamond sound tubes or whatever they call them... Dressed up $4 version meant to look like a radio tube... just being nostalgic in my senior years

 

[TheLinuxBug]

I will just put this here 

This actually looks pretty cool.. but a little bit of a higher budget needed.

Cheers!

 

[KuJoe]

I will just put this here 

This actually looks pretty cool.. but a little bit of a higher budget needed.

Cheers!

Thanks for the link. When my daughter gets a bit older I'll grab up one of these for her to build.

 

[TheLinuxBug]

I also found this of some interest. Seems you can likely power the PI with very little issue using solar power. Could think of some cool setups with one of those cellular wifi hot spots (Did someone say FreedomPop?) and a decent little case you could literally have the pi running at the top of a tree, in the real 'Cloud'    I think with 2.2 amps and a 10000mAH battery to hold the charge for you you could probably have the thing running pretty much anywhere that has access to the sun.  On top of a building, top of a tree, etc.

Edit: could be a good surveillance solution too if you added a camera to it...

Cheers!

 

[drmike]

I also found this of some interest. Seems you can likely power the PI with very little issue using solar power. Could think of some cool setups with one of those cellular wifi hot spots (Did someone say FreedomPop?) and a decent little case you could literally have the pi running at the top of a tree, in the real 'Cloud'    I think with 2.2 amps and a 10000mAH battery to hold the charge for you you could probably have the thing running pretty much anywhere that has access to the sun.  On top of a building, top of a tree, etc.

Edit: could be a good surveillance solution too if you added a camera to it...

Cheers!

You little LinuxBug you.... That CottonPickers site  My wallet.

http://ccgi.cottonpickers.plus.com/shop_raspberry.html

That timed controller with USB = AWESOME.  Been looking for such a solution. 

That's mega ideal for a lot of project using solar as the power source. No other way to pull such off on solar without limiting time/run, unless you live in very sunny place and/or have lots of solar + big battery.

10000mAH batteries are often sketch.  Marketing hype on the big number.  See this out of Asia dumped products right and left.  Ideal still solar setup = 1 charge controller + solar panels + sealed lead acid battery at probably 10-20AH @ 12V.   Going to need buck converters / down converted to 5V / USB, but pedestrian and common.

I'd love to see someone here do a solar build / off grid powered.  Ideally 10 years out such becomes commonplace rather than tinkerer / eccentrics / off griders.

 

[TheLinuxBug]

Well, some more useful links then:

http://www.maplin.co.uk/p/4w-solar-briefcase-n05hn

http://www.amazon.co.uk/N05HN-Solar-Briefcase-Battery-Charger/dp/B003UARDRK

(were linked from an article, would much rather find a US distributor, but for now this will just give you an idea what is possible)

The following link goes to an article on how to setup the solar panels to power your pi: here

Also thought I would include a link to this article, while the format of the article is awful, some of the stuff that it touches on is pretty cool.

Cheers!

 

[drmike]

Well, some more useful links then:

http://www.maplin.co.uk/p/4w-solar-briefcase-n05hn

http://www.amazon.co.uk/N05HN-Solar-Briefcase-Battery-Charger/dp/B003UARDRK

(were linked from an article, would much rather find a US distributor, but for now this will just give you an idea what is possible)

The following link goes to an article on how to setup the solar panels to power your pi: here

Also thought I would include a link to this article, while the format of the article is awful, some of the stuff that it touches on is pretty cool.

Cheers!

Appreciated! Some real interesting Pi projects   

The solar panel, meh $5 per watt.  Can do much much better these days.

Just rough modeling a Pi @ 3 watts x 8 hours = 24 watts+ an 8 hour day...

Need 3x solar watts and 3x+ battery to roll through cloudy days....

72 watts of panels and 10AH battery for an 8 hour run install on minimal use Pi solar with timer device.  Not too bad. Not cheap enough though.

 

[KuJoe]

I've been playing with my RPi B+, RPi2, and ODroid C1 for a good part of the night to see which one I like better and here's the results so far:

 

Power Consumption

Raspberry Pi Model B+ - Raspbian (Fresh install with LXDE no accessories/network)
2.2 watts on idle
3.1 watts with the single core maxed at 100%
 
Raspberry Pi 2 - Raspbian (Fresh install with LXDE no accessories/network)
1.5 watts on idle
2.9 watts with all four cores maxed at 100%
 
ODroid C1 - Lubuntu (Fresh install with LXDE running, no accessories/network)
1.9 watts on idle
4.1 watts with all four cores maxed at 100%

 

Network Tests

To and from my NAS going through wired 1Gbps ports, no OS tweaks.
 
Raspberry Pi Model B+ (100Mbps) = 93.7 Mbits/sec Download & 49.7 Mbits/sec Upload
Raspberry Pi 2 (100Mbps) = 94.0 Mbits/sec Download & 94.3 Mbits/sec Upload
ODroid C1 (1Gbps) = 442 Mbits/sec Download & 339 Mbits/sec Upload
 
To and from my NAS going over WiFi (Edimax EW-7811Un), no OS tweaks.
 
Raspberry Pi Model B+ = 50.8 Mbits/sec Download & 42.8 Mbits/sec Upload
Raspberry Pi 2 = 30.7 Mbits/sec Download & 30.6 Mbits/sec Upload
ODroid C1 = 11.6 Mbits/sec Download & 17.8 Mbits/sec Upload
 
Out of network tests, no OS tweaks.
 
*No tests from Model B+ due to it being in use during this test
Raspberry Pi 2 wireless = http://results.speedtest.comcast.net/result/811655704.png
Raspberry Pi 2 wired = http://results.speedtest.comcast.net/result/811662464.png
ODroid C1 wireless = http://results.speedtest.comcast.net/result/811659984.png
ODroid C1 wired = http://results.speedtest.comcast.net/result/811660464.png

 

Unixbench Results (no IO tests)

Raspberry Pi Model = Single Core: 74.1
Raspberry Pi 2 = Single Core: 130.8, 4 Cores: 386.8
ODroid C1 = Single Core: 166.8, 4 Cores: 774.6

 

Overview of each system

 

Raspberry Pi Model B+

The only things that made the Model B+ stand out compared to the other two was the location of the power input (much more convenient for how I mounted it under my desk), that it uses a regular SD card versus the MicroSD, and that it performed better than the rest in the wireless speed tests. The first two can be a pro or con for others so really the only positive thing was the wireless performance. I was planning on continuing to use mine for running MRTG and my IP camera but considering the RPi2 outperforms it in nearly every test, uses less power, and is the same price I'm not sure the benefit for using my Model B+ anymore unless wireless performance becomes a factor.

 

Raspberry Pi 2

The RPi2 doesn't look that good on paper compared to the ODroid C1 but blows the Model B+ out of the water in almost all of the tests. I like that it detected the WiFi adapter immediately without any issues and only required one file edit to get it to see a hidden SSID, IPv6 is not enabled out of the box so I didn't bother configuring it for this testing. Took some extra steps to get the correct monitor resolution and changing it requires a reboot. Installing Chromium and getting Pepper Flash installed was very simple and Flash ran surprisingly well when I used it. It was noticeably sluggish under normal usage (web browsing, updating and installing packages, compressing and uncompressing files, general desktop navigation, etc...). I connected to a VPS running Windows 7 (FreeRDP) that's about 11ms away and this is where the RPi2 took off. It was one of the smoothest RDP experiences I've ever had and comparable to when I RDP to the same VPS from my Windows 7 laptop. I was browsing websites, SSHing to other servers, running scripts, and it ran smoothly no matter how many things I was doing at once. Compared to the Model B+ and C1, this one wins in terms of remote desktop experience.

 

ODroid C1

The C1 performed the best in terms of raw performance and even though it hit under 50% of the 1Gbps port's potential, it still beat the 100Mbps ports on both Pi's. The wireless network was painful at best. The speed was severely lacking even though it showed 72Mbps in the GUI and it wouldn't maintain a connection for very long at all (I had to run the Speedtest page at least 10 times before it was able to complete without being disconnected). The wireless adapter was detected right out of the box, connecting to the hidden SSID network was extremely simple, and it grabbed an IPv6 address from my router without issue but that's the unstable connection would have me either ordering a different wireless adapter or grabbing an ethernet cable. After having so many issues on previous OS versions getting it to work with my monitor (HDMI to DVI adapter), it finally works perfectly on my monitor but now it's acting wonky with my TV and no matter what resolution I pick I can't get the whole desktop to display on it. I also did not like the smaller HDMI port versus the full-size HDMI ports on the Pi's and I'm not a fan of the unique power input either. This is the only one of the bunch that I would even consider using as a desktop replacement. Under normal usage it's very smooth and responsive, no jumping or slowness while browsing websites even when the page is loaded up with HD images. I was browsing vpsBoard without a hiccup and page loads were instant each time. I connected to my Windows 7 VPS over RDP (FreeRDP) with the same settings as the RPi2 and it was like night and day. It was usable but the display was off, there was lag between actions, and browsing the web was jumpy. If you're looking for a good solution for local usage then the C1 takes the cake. If you want to use it as a thin client then it'll work, but I would grab a RPi2 instead if I were you.

 

 

I apologize for any typos, I'm heading to bed now.

 

[drmike]

@KuJoe being a sweetheart.

That's the spirit!

The new quad Pi looks good on numbers all except wifi throughput, which is not relative to my use.  I care about wired NIC.

Lower power, faster.  Yeah ordering one.

 

[MannDude]

I did some benchmarks last night too!

I shall share.

(almost) Stock install of Raspian. Not overclocked.


UnixBench (w/ all processors) 436.8
UnixBench (w/ one processor) 173.2

   #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
   #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
   #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
   #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
   #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
    ####   #    #  #  #    #          #####   ######  #    #   ####   #    #

   Version 5.1.3                      Based on the Byte Magazine Unix Benchmark

   Multi-CPU version                  Version 5 revisions by Ian Smith,
                                      Sunnyvale, CA, USA
   January 13, 2011                   johantheghost at yahoo period com

Use of uninitialized value in printf at ./Run line 1378.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1378.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1378.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1378.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1588.
Use of uninitialized value in printf at ./Run line 1590.
Use of uninitialized value in printf at ./Run line 1588.
Use of uninitialized value in printf at ./Run line 1590.
Use of uninitialized value in printf at ./Run line 1588.
Use of uninitialized value in printf at ./Run line 1590.
Use of uninitialized value in printf at ./Run line 1588.
Use of uninitialized value in printf at ./Run line 1590.

1 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput  1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

1 x File Copy 256 bufsize 500 maxblocks  1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

1 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

1 x Process Creation  1 2 3

1 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent)  1 2 3

1 x Shell Scripts (8 concurrent)  1 2 3

4 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

4 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

4 x Execl Throughput  1 2 3

4 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

4 x File Copy 256 bufsize 500 maxblocks  1 2 3

4 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

4 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

4 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

4 x Process Creation  1 2 3

4 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

4 x Shell Scripts (1 concurrent)  1 2 3

4 x Shell Scripts (8 concurrent)  1 2 3

========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: ******************
   OS: GNU/Linux -- 3.18.5-v7+ -- #225 SMP PREEMPT Fri Jan 30 18:53:55 GMT 2015
   Machine: armv7l (unknown)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   CPU 1: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   CPU 2: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   CPU 3: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   06:36:38 up  6:35,  2 users,  load average: 0.43, 0.27, 0.42; runlevel 2

------------------------------------------------------------------------
Benchmark Run: Wed Feb 18 2015 06:36:38 - 07:04:46
4 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables        2953981.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                      498.0 MWIPS (10.0 s, 7 samples)
Execl Throughput                                363.4 lps   (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks         74992.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           21595.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        191527.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                              172661.9 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  31795.9 lps   (10.0 s, 7 samples)
Process Creation                               1271.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   1174.4 lpm   (60.1 s, 2 samples)
Shell Scripts (8 concurrent)                    327.0 lpm   (60.1 s, 2 samples)
System Call Overhead                         412200.9 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    2953981.6    253.1
Double-Precision Whetstone                       55.0        498.0     90.5
Execl Throughput                                 43.0        363.4     84.5
File Copy 1024 bufsize 2000 maxblocks          3960.0      74992.0    189.4
File Copy 256 bufsize 500 maxblocks            1655.0      21595.5    130.5
File Copy 4096 bufsize 8000 maxblocks          5800.0     191527.5    330.2
Pipe Throughput                               12440.0     172661.9    138.8
Pipe-based Context Switching                   4000.0      31795.9     79.5
Process Creation                                126.0       1271.1    100.9
Shell Scripts (1 concurrent)                     42.4       1174.4    277.0
Shell Scripts (8 concurrent)                      6.0        327.0    544.9
System Call Overhead                          15000.0     412200.9    274.8
                                                                   ========
System Benchmarks Index Score                                         173.2

------------------------------------------------------------------------
Benchmark Run: Wed Feb 18 2015 07:04:46 - 07:32:57
4 CPUs in system; running 4 parallel copies of tests

Dhrystone 2 using register variables       11822275.4 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     1990.4 MWIPS (10.0 s, 7 samples)
Execl Throughput                               1335.2 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        115723.9 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           32182.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        302906.4 KBps  (30.0 s, 2 samples)
Pipe Throughput                              689839.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 122806.9 lps   (10.0 s, 7 samples)
Process Creation                               2852.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2609.9 lpm   (60.1 s, 2 samples)
Shell Scripts (8 concurrent)                    344.4 lpm   (60.4 s, 2 samples)
System Call Overhead                        1573353.8 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   11822275.4   1013.0
Double-Precision Whetstone                       55.0       1990.4    361.9
Execl Throughput                                 43.0       1335.2    310.5
File Copy 1024 bufsize 2000 maxblocks          3960.0     115723.9    292.2
File Copy 256 bufsize 500 maxblocks            1655.0      32182.7    194.5
File Copy 4096 bufsize 8000 maxblocks          5800.0     302906.4    522.3
Pipe Throughput                               12440.0     689839.7    554.5
Pipe-based Context Switching                   4000.0     122806.9    307.0
Process Creation                                126.0       2852.1    226.4
Shell Scripts (1 concurrent)                     42.4       2609.9    615.5
Shell Scripts (8 concurrent)                      6.0        344.4    574.1
System Call Overhead                          15000.0    1573353.8   1048.9
                                                                   ========
System Benchmarks Index Score                                         436.8

ioping -c 10
request=1 time=1.0 ms
request=2 time=0.7 ms
request=3 time=0.8 ms
request=4 time=0.9 ms
request=5 time=3.4 ms
request=6 time=0.8 ms
request=7 time=0.8 ms
request=8 time=0.9 ms
request=9 time=1.0 ms
request=10 time=0.8 ms

10 requests completed in 9147.2 ms, 891 iops, 3.5 mb/s

I/O Seek Test (No Cache)ioping -RD
1175 iops, 4.6 mb/s
min/avg/max/mdev = 0.6/0.9/3.4/0.3 ms

I/O Reads - Sequential
ioping -RL
69 iops, 17.3 mb/s
min/avg/max/mdev = 14.4/14.5/15.8/0.2 ms

I/O Reads - Cached
ioping -RC
61038 iops, 238.4 mb/s
min/avg/max/mdev = 0.0/0.0/0.0/0.0 ms

dd if=/dev/zero of=sb-io-test bs=1M count=1k conv=fdatasync
83.9574 s, 12.8 MB/s

dd if=/dev/zero of=sb-io-test bs=64k count=16k conv=fdatasync
83.1907 s, 12.9 MB/s

dd if=/dev/zero of=sb-io-test bs=1M count=1k oflag=dsync
104.073 s, 10.3 MB/s

dd if=/dev/zero of=sb-io-test bs=64k count=16k oflag=dsync
226.245 s, 4.7 MB/s

Overclocked to 900Mhz and 450MHz SDRAM, 2 overvolt.

Saw now real major increase in results. I tested this after letting the Pi return to it's idle temperature after running the first test. Since the temperature at the peak of the test didn't exceed 54 degrees celcius the Pi remained well within the 'safe zone', as I believe 85c is the automatic turn off point. Didn't feel like running the bench again but may increase it to Turbo (1000MHz ARM, 500MHz core, 600MHz SDRAM, 6 Overvolt) later and run the tests again. I've got heatsinks on the Pi and can place it somewhere nice and cool to keep it at a safe temperature.

Anyhow, the results are similar to that of above.

UnixBench (w/ all processors)	436.4
UnixBench (w/ one processor)	173.7

   #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
   #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
   #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
   #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
   #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
    ####   #    #  #  #    #          #####   ######  #    #   ####   #    #

   Version 5.1.3                      Based on the Byte Magazine Unix Benchmark

   Multi-CPU version                  Version 5 revisions by Ian Smith,
                                      Sunnyvale, CA, USA
   January 13, 2011                   johantheghost at yahoo period com

Use of uninitialized value in printf at ./Run line 1378.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1378.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1378.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1378.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1588.
Use of uninitialized value in printf at ./Run line 1590.
Use of uninitialized value in printf at ./Run line 1588.
Use of uninitialized value in printf at ./Run line 1590.
Use of uninitialized value in printf at ./Run line 1588.
Use of uninitialized value in printf at ./Run line 1590.
Use of uninitialized value in printf at ./Run line 1588.
Use of uninitialized value in printf at ./Run line 1590.

1 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput  1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

1 x File Copy 256 bufsize 500 maxblocks  1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

1 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

1 x Process Creation  1 2 3

1 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent)  1 2 3

1 x Shell Scripts (8 concurrent)  1 2 3

4 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

4 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

4 x Execl Throughput  1 2 3

4 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

4 x File Copy 256 bufsize 500 maxblocks  1 2 3

4 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

4 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

4 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

4 x Process Creation  1 2 3

4 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

4 x Shell Scripts (1 concurrent)  1 2 3

4 x Shell Scripts (8 concurrent)  1 2 3

========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: ******************
   OS: GNU/Linux -- 3.18.5-v7+ -- #225 SMP PREEMPT Fri Jan 30 18:53:55 GMT 2015
   Machine: armv7l (unknown)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   CPU 1: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   CPU 2: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   CPU 3: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   09:10:50 up 39 min,  2 users,  load average: 0.37, 0.29, 0.39; runlevel 2

------------------------------------------------------------------------
Benchmark Run: Wed Feb 18 2015 09:10:50 - 09:39:03
4 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables        2954372.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                      498.7 MWIPS (9.9 s, 7 samples)
Execl Throughput                                370.7 lps   (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks         75253.3 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           21596.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        194900.8 KBps  (30.0 s, 2 samples)
Pipe Throughput                              173469.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  31712.0 lps   (10.0 s, 7 samples)
Process Creation                               1237.2 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   1185.1 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    329.9 lpm   (60.1 s, 2 samples)
System Call Overhead                         411475.5 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    2954372.5    253.2
Double-Precision Whetstone                       55.0        498.7     90.7
Execl Throughput                                 43.0        370.7     86.2
File Copy 1024 bufsize 2000 maxblocks          3960.0      75253.3    190.0
File Copy 256 bufsize 500 maxblocks            1655.0      21596.4    130.5
File Copy 4096 bufsize 8000 maxblocks          5800.0     194900.8    336.0
Pipe Throughput                               12440.0     173469.8    139.4
Pipe-based Context Switching                   4000.0      31712.0     79.3
Process Creation                                126.0       1237.2     98.2
Shell Scripts (1 concurrent)                     42.4       1185.1    279.5
Shell Scripts (8 concurrent)                      6.0        329.9    549.9
System Call Overhead                          15000.0     411475.5    274.3
                                                                   ========
System Benchmarks Index Score                                         173.7

------------------------------------------------------------------------
Benchmark Run: Wed Feb 18 2015 09:39:03 - 10:07:20
4 CPUs in system; running 4 parallel copies of tests

Dhrystone 2 using register variables       11822497.3 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     1990.1 MWIPS (10.0 s, 7 samples)
Execl Throughput                               1352.0 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        116461.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           32098.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        302035.6 KBps  (30.0 s, 2 samples)
Pipe Throughput                              690250.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 122434.7 lps   (10.0 s, 7 samples)
Process Creation                               2818.6 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2640.7 lpm   (60.1 s, 2 samples)
Shell Scripts (8 concurrent)                    336.6 lpm   (60.5 s, 2 samples)
System Call Overhead                        1575376.8 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   11822497.3   1013.1
Double-Precision Whetstone                       55.0       1990.1    361.8
Execl Throughput                                 43.0       1352.0    314.4
File Copy 1024 bufsize 2000 maxblocks          3960.0     116461.5    294.1
File Copy 256 bufsize 500 maxblocks            1655.0      32098.5    193.9
File Copy 4096 bufsize 8000 maxblocks          5800.0     302035.6    520.8
Pipe Throughput                               12440.0     690250.1    554.9
Pipe-based Context Switching                   4000.0     122434.7    306.1
Process Creation                                126.0       2818.6    223.7
Shell Scripts (1 concurrent)                     42.4       2640.7    622.8
Shell Scripts (8 concurrent)                      6.0        336.6    560.9
System Call Overhead                          15000.0    1575376.8   1050.3
                                                                   ========
System Benchmarks Index Score                                         436.4

ioping -c 10
request=1 time=0.8 ms
request=2 time=0.8 ms
request=3 time=0.8 ms
request=4 time=0.7 ms
request=5 time=0.8 ms
request=6 time=0.8 ms
request=7 time=0.7 ms
request=8 time=0.8 ms
request=9 time=0.8 ms
request=10 time=0.8 ms

10 requests completed in 9158.6 ms, 1276 iops, 5.0 mb/s

I/O Seek Test (No Cache)

ioping -RD
1255 iops, 4.9 mb/s
min/avg/max/mdev = 0.6/0.8/2.9/0.2 ms

I/O Reads - Sequential

ioping -RL
69 iops, 17.3 mb/s
min/avg/max/mdev = 14.4/14.5/15.2/0.1 ms

I/O Reads - Cached

ioping -RC
80803 iops, 315.6 mb/s
min/avg/max/mdev = 0.0/0.0/0.1/0.0 ms

dd if=/dev/zero of=sb-io-test bs=1M count=1k conv=fdatasync
86.2182 s, 12.5 MB/s

dd if=/dev/zero of=sb-io-test bs=64k count=16k conv=fdatasync
84.3241 s, 12.7 MB/s

dd if=/dev/zero of=sb-io-test bs=1M count=1k oflag=dsync
103.916 s, 10.3 MB/s

dd if=/dev/zero of=sb-io-test bs=64k count=16k oflag=dsync
224.614 s, 4.8 MB/s

 

[KuJoe]

I did some IO benchmarks, but those are more dependent on the SD card being used versus the Pi's performance.

 

[drmike]

UnixBench I always look at those and go huh? Unsure how accurate UnixBench is on ARM platform stuff. Assuming it isn't real suitable for it.

"I've got heatsinks on the Pi and can place it somewhere nice and cool to keep it at a safe temperature."

I have a place approaching absolute zero; it's called outside.

[SIZE=13.63636302948px]Overclocking a SoC, people like living on the edge  I'll just buy 2...  I know cause you can and all.   Amazingly simple on the Pi's.  I won't I'll burn mine up no doubt.   [/SIZE]

 

[rds100]

Yeah, 12 MB/s write speed seems low. USB sticks can push more than this.

 

[fixidixi]

Yeah, 12 MB/s write speed seems low. USB sticks can push more than this.

sure. for how long?

 

[rds100]

sure. for how long?

About a year so far without a single dead USB stick yet?

 

[KuJoe]

I decided to overclock the RPi2 and rerun some tests to see if I can get closer performance to the C1. I set the CPU to the "RPi2" settings (1000MHz ARM, 500MHz core, 500MHz SDRAM, 2 overvolt) and set the GPU memory from 64MB to 256MB.

I played some Minecraft for the first time and it was pretty fun (Minecraft Pi comes pre-installed on Raspbian).

Also I'm seeing a 4-6Mbps gain up and down over wireless with the overclocking for some reason, still nothing compared to wired or the Model B+ but every Mbps helps over the LAN.

 

[Coastercraze]

I put it together yesterday and installed Raspbian on it. Pretty neat little thing.

Didn't do any tests yet. I did play some Minecraft on it just to see how it was.

Will mess with it more on the weekend.

 

[KuJoe]

Just finished the Unixbench with the overclocked RPi2 and it closes the gap quite bit between the stock RPi2 and the stock C1. The temperature sensor in the RPi2 showed 53.5'C when the test completed, my infrared thermometer shows the CPU/GPU chip at 34'C while it was on the last few tests.

I only ran the 4 CPU benchmark and skipped the single CPU test for the sake of time.


========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)
 
   System: jdraspbian: GNU/Linux
   OS: GNU/Linux -- 3.18.7-v7+ -- #756 SMP PREEMPT Wed Feb 18 16:14:51 GMT 2015
   Machine: armv7l (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   CPU 1: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   CPU 2: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   CPU 3: ARMv7 Processor rev 5 (v7l) (0.0 bogomips)
          
   23:12:11 up 2 min,  1 user,  load average: 0.11, 0.11, 0.05; runlevel 2
 
------------------------------------------------------------------------
Benchmark Run: Thu Feb 19 2015 23:12:11 - 23:36:29
4 CPUs in system; running 4 parallel copies of tests
 
Dhrystone 2 using register variables       13115973.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2208.3 MWIPS (9.9 s, 7 samples)
Execl Throughput                               1555.2 lps   (29.9 s, 2 samples)
Pipe Throughput                              763366.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 136544.7 lps   (10.0 s, 7 samples)
Process Creation                               3225.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2893.0 lpm   (60.0 s, 2 samples)
Shell Scripts (16 concurrent)                   190.2 lpm   (60.6 s, 2 samples)
Shell Scripts (8 concurrent)                    382.8 lpm   (60.3 s, 2 samples)
System Call Overhead                        1759180.8 lps   (10.0 s, 7 samples)
 
System Benchmarks Partial Index              BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   13115973.6   1123.9
Double-Precision Whetstone                       55.0       2208.3    401.5
Execl Throughput                                 43.0       1555.2    361.7
Pipe Throughput                               12440.0     763366.7    613.6
Pipe-based Context Switching                   4000.0     136544.7    341.4
Process Creation                                126.0       3225.8    256.0
Shell Scripts (1 concurrent)                     42.4       2893.0    682.3
Shell Scripts (16 concurrent)                     ---        190.2      ---
Shell Scripts (8 concurrent)                      6.0        382.8    638.0
System Call Overhead                          15000.0    1759180.8   1172.8
                                                                   ========
System Benchmarks Index Score (Partial Only)                          548.2


Now I'm going to switch off my laptop and use my RPi2 for as long as possible to see what kind of desktop replacement it has the potential for.