Turl changed the topic of #linux-sunxi to: Allwinner/sunxi /development discussion - did you try looking at our wiki? https://linux-sunxi.org - Don't ask to ask. Just ask and wait! - https://github.com/linux-sunxi/ - Logs at http://irclog.whitequark.org/linux-sunxi
[7] has quit [Ping timeout: 272 seconds]
TheSeven has joined #linux-sunxi
<lennyraposo> pretty kewl apritzel
<lennyraposo> was checking it out
<lennyraposo> so if I am getting this right
<lennyraposo> thsi closer to mainline uboot
<lennyraposo> the only thing holding things from that end is the blob being used to initialize things
<lennyraposo> just making sure I got it right
<lennyraposo> and oh
<lennyraposo> hi all
<apritzel> lennyraposo: yes, you need libdram atm
<lennyraposo> yay
<lennyraposo> I understood it
<lennyraposo> :D
<apritzel> or you use boot0 (which also contains libdram, but you don't need to link it)
<lennyraposo> and you already have started dram initialzation function foor later development
<apritzel> and then you can even ship it and it still have upstream U-Boot
<apritzel> finishing up that README detailing the various options we have at the moment
<lennyraposo> from my understanding boot0 was initialted then drma init then back to boot0 correct?
<lennyraposo> originally I mean
<lennyraposo> sheesh my typing is horrendous today
* apritzel is still parsing ...
<lennyraposo> so to further my education here
<apritzel> boot0 does similar things like the SPL and it contains libdram as well
<lennyraposo> ok
<apritzel> it's just not OpenSource, so you can ship without violating the GPL
<apritzel> but it does things badly and is not very flexible
<apritzel> so we want to get rid of it
<apritzel> but atm it's the only "legal" way of initializing the DRAM
<lennyraposo> with the requirement of the Boot0 code
<lennyraposo> without
<apritzel> "legal" as in: can be distributed without legal issues
<lennyraposo> which causes the redistribute issue
<lennyraposo> nice way of getting around a problem
<apritzel> more nasty, I'd say
<lennyraposo> the binary for libdram is perfectly legal to redistribute
<lennyraposo> ?
<lennyraposo> is illegal i mean
<apritzel> don't know if "perfectly" is the right way, but reportedly Allwinner does not mind people redistributing boot0
<apritzel> and libdram is part of their BSP package, AFAIK
<lennyraposo> yes it is
<lennyraposo> next week I will have a few boards in that I can play around with
<lennyraposo> 8 in total
<lennyraposo> 2 x 2gb 6 x 1gb boards
<apritzel> well, how does that help?
<lennyraposo> I am setting aside 1 of ram size board for furthering my educaiton on these things
<lennyraposo> Wanna try some of the things listed on the wiki
<apritzel> as in?
<lennyraposo> usb
<lennyraposo> probing it
<lennyraposo> figuring it out
<lennyraposo> I figure can't hurt to go through the paces
<lennyraposo> contribute to devlopment
<lennyraposo> testing
<lennyraposo> etc
<apritzel> well, in my experience one board is enough for that ;-)
<lennyraposo> the other boards are for something else atm
<lennyraposo> ;)
<lennyraposo> something I am oding over at the local school my kids attend
<lennyraposo> doing*
<apritzel> does the BSP kernel actually support USB-OTG?
<lennyraposo> not sure
<lennyraposo> but according to tllim
<lennyraposo> The lower port can be configure as host or OTG by software. Default is host.
<apritzel> no, it's the upper port
<lennyraposo> that's what I thought
<apritzel> and I know what the hardware can do, but I am interested if you can tell Linux to do its OTG magic
<lennyraposo> you were the first to do so
<lennyraposo> FEL
<lennyraposo> I do not know
<lennyraposo> haven't ouched the thing in about 2 weeks due to work schedule
<apritzel> AFAIK this is recent Linux feature, so chances are that 3.10.ancient does not support it
<lennyraposo> touched*
<lennyraposo> I knwo that longsleep rolle din a bunch of things
<lennyraposo> but haven't had time to look into it
<lennyraposo> LXD was the last thing I noticed up on github
<lennyraposo> for longsleep's build
<lennyraposo> not a containers person myself
<lennyraposo> I can try later if you like
<apritzel> not so important, was just curious, cause it's a cool feature
<lennyraposo> I have an A to A cable
<lennyraposo> will give it a go when I am finished work
ganbold has joined #linux-sunxi
Florian- has quit [Ping timeout: 260 seconds]
Florian- has joined #linux-sunxi
KotCzarny has quit [Ping timeout: 260 seconds]
oliv3r has quit [Ping timeout: 260 seconds]
oliv3r has joined #linux-sunxi
KotCzarny has joined #linux-sunxi
ninolein has quit [Ping timeout: 250 seconds]
ninolein has joined #linux-sunxi
<willmore> ssvb, do you have a link to your Chinese supplier of SPI flash? I need a few for some ESP projects.
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
nashpa has quit [Quit: Going away]
nashpa has joined #linux-sunxi
staplr has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
<jrg> phm
<jrg> hm.. not really sure what freq the m3 can run at and still keep all 8 cores on
<jrg> 1.2GHz doesn't do it. it still drops to 4 under load with a heatsink
<jrg> guess it's better to just leave it at 1.6GHz and lose the cores
<jrg> at least it is stable for the most part at 1.6GHz
<jrg> 22:14:11 up 23:37, 5 users, load average: 6.99, 5.90, 5.77
IgorPec has joined #linux-sunxi
cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft has joined #linux-sunxi
kaspter has joined #linux-sunxi
IgorPec has quit [Ping timeout: 244 seconds]
kaspter has quit [Ping timeout: 260 seconds]
scream has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
apritzel has joined #linux-sunxi
IgorPec has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
jernej has joined #linux-sunxi
wave_ has quit [K-Lined]
ricardocrudo has joined #linux-sunxi
avph has quit [Ping timeout: 264 seconds]
ricardocrudo has quit [Remote host closed the connection]
avph has joined #linux-sunxi
Shirasaka-Hazumi has quit [Ping timeout: 260 seconds]
jernej has quit [Ping timeout: 276 seconds]
Shirasaka-Hazumi has joined #linux-sunxi
staplr has quit [Ping timeout: 240 seconds]
mosterta|2 has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
ajeandet has joined #linux-sunxi
ajeandet has quit [Client Quit]
ajeandet has joined #linux-sunxi
mosterta|2 has quit [Ping timeout: 260 seconds]
tyler-baker has quit [Quit: Shazbot]
al1o has joined #linux-sunxi
tyler-baker has joined #linux-sunxi
JohnDoe_71Rus has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/]
<ssvb> willmore: I was just stupid enough to initially order a somewhat overpriced spi flash module on aliexpress before finding a cheaper option, I can't really recommend it
tyler-baker has quit [Client Quit]
JohnDoe_71Rus has joined #linux-sunxi
tyler-baker has joined #linux-sunxi
apritzel has joined #linux-sunxi
enki_ has joined #linux-sunxi
IgorPec has quit [Ping timeout: 272 seconds]
mosterta|2 has joined #linux-sunxi
<longsleep> apritzel, lennyraposo I have not tried USB OTG whatever support and also not enabled it in my Kernel builds
<longsleep> but there seems to be some otg in the USB driver
<longsleep> root@pine64:/sys# find -name "*otg*"
<longsleep> ./bus/platform/drivers/otg manager
<longsleep> ./devices/soc.0/usbc0.5/otg_role
<longsleep> ./devices/soc.0/1c19000.udc-controller/udc/1c19000.udc-controller/is_otg
<longsleep> ./devices/soc.0/1c19000.udc-controller/otg_ed_test
aballier_ has joined #linux-sunxi
jernej has joined #linux-sunxi
<longsleep> apritzel, lennyraposo i have no idea and all but the sysfs toggles for USB OTG seem to work - see http://paste.ubuntu.com/16763669/
reinforce has joined #linux-sunxi
Nyuutwo has quit [Ping timeout: 240 seconds]
Nyuutwo has joined #linux-sunxi
IgorPec has joined #linux-sunxi
jrg has quit [Ping timeout: 264 seconds]
jrg has joined #linux-sunxi
IgorPec has quit [Quit: Nettalk6 - www.ntalk.de]
IgorPec has joined #linux-sunxi
yann|work has joined #linux-sunxi
jernej has quit [Ping timeout: 244 seconds]
Netlynx has joined #linux-sunxi
Netlynx_ has joined #linux-sunxi
Netlynx_ has quit [Client Quit]
apritzel has quit [Ping timeout: 244 seconds]
nove has joined #linux-sunxi
jernej has joined #linux-sunxi
paulk-collins has quit [Quit: Leaving]
raknaz has joined #linux-sunxi
premoboss has joined #linux-sunxi
yann|work has quit [Remote host closed the connection]
raknaz has quit [Client Quit]
yann|work has joined #linux-sunxi
ricardocrudo has joined #linux-sunxi
ricardocrudo has quit [Ping timeout: 260 seconds]
jernej has quit [Ping timeout: 252 seconds]
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
mosterta|2 has quit [Ping timeout: 244 seconds]
mosterta|2 has joined #linux-sunxi
ricardocrudo has joined #linux-sunxi
Nyuutwo has quit [Ping timeout: 240 seconds]
gzamboni has quit [Read error: Connection reset by peer]
enki_ has quit [Remote host closed the connection]
Nyuutwo has joined #linux-sunxi
jernej has joined #linux-sunxi
yann|work has quit [Remote host closed the connection]
yann|work has joined #linux-sunxi
apritzel has joined #linux-sunxi
al1o has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
yann|work has quit [Remote host closed the connection]
yann|work has joined #linux-sunxi
mosterta|2 has quit [Ping timeout: 240 seconds]
ajeandet has quit [Ping timeout: 260 seconds]
staplr has joined #linux-sunxi
yann|work has quit [Remote host closed the connection]
yann|work has joined #linux-sunxi
mosterta|2 has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
Amit_t_ has joined #linux-sunxi
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
gzamboni has joined #linux-sunxi
ajeandet has joined #linux-sunxi
cnxsoft has quit [Quit: cnxsoft]
jernej has quit [Ping timeout: 276 seconds]
ricardocrudo has quit [Ping timeout: 244 seconds]
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
apritzel has joined #linux-sunxi
Amit_t_ has quit [Ping timeout: 250 seconds]
Nyuutwo has quit [Ping timeout: 240 seconds]
IgorPec has quit [Ping timeout: 244 seconds]
Nyuutwo has joined #linux-sunxi
ricardocrudo has joined #linux-sunxi
<willmore> ssvb, okay, thank anyway. You're in Europe?
<ssvb> yes
<willmore> Okay, so local suppliers aren't going to be 'local' to me. :( I'll find something.
scream has quit [Remote host closed the connection]
jernej has joined #linux-sunxi
gzamboni has quit [Ping timeout: 240 seconds]
gzamboni has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
premoboss has quit [Ping timeout: 246 seconds]
yann|work has quit [Ping timeout: 264 seconds]
aballier_ has quit [Quit: leaving]
Amit_t_ has joined #linux-sunxi
apritzel has joined #linux-sunxi
ricardocrudo has quit [Ping timeout: 272 seconds]
IgorPec has joined #linux-sunxi
yann|work has joined #linux-sunxi
<phiplii> I glad to see that Armbian has the FTDI drivers built in...
<phiplii> because I really should have checked /before/ I spent the whole day moving a small webserver onto it from an old broken ("official") orangepi install of debian
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
yann|work has quit [Ping timeout: 260 seconds]
kaspter has joined #linux-sunxi
apritzel has joined #linux-sunxi
<longsleep> I am just doing some thermal testing of the Pine64 - Linux SoC temperature shows as about 20C more than the temperature taken from the outer surface of the A64 chip. Anyone got a clue if that large difference is normal?
<longsleep> i mean its 90C vs 70C
<apritzel> longsleep: doesn't sounds unreasonable to me, also the temperature conversion could be off (sensor not calibrated)
<longsleep> apritzel: mhm ok - i wonder if that means that heat sinks are not so effective
<KotCzarny> so you are measuring temp. of the heatsinks and not the soc?
<longsleep> KotCzarny: no heat sink yet - directly on the SoC surface
<KotCzarny> what are you using to measure? ir termo or something else?
<longsleep> KotCzarny: multimeter sensor cable
<KotCzarny> try different points on soc?
<longsleep> KotCzarny: yeah did that already, did not find big differences
<longsleep> i mean the farther from the middle the cooler, but only about 5C less at the corners
<longsleep> hottest in the middle
<KotCzarny> do you have any other thermo to confirm readings?
<KotCzarny> also, keep in mind, the soc's plastic casing isnt the best one to conduct heat, so it might be hotter inside
<longsleep> yes i have two multimeters and both show about the same
<longsleep> sure, but the SoCs plastic casing is where the heat sink connects
<longsleep> i received a heat sink from Pine64 today
<phiplii> longsleep : with a plastic casing vs metal contacts on the underside, I'd guess most of the heat leaves through the base
<longsleep> anyhow - to give you some numbers - Pine64 at 23C ambient creates 3.15 khash/sec at stable 89C with 4 cores at 1056MHz without heat sink
<longsleep> phiplii: yeah i measured temp at the bottom and its much cooler there (around 55C), and the board lies flat on the table
<phiplii> The temperature that matters is the highest bit
<phiplii> likely to be right where a core is, or perhaps the GPU
<phiplii> it will be a tiny area inside
<phiplii> everything else is going to be lower
<longsleep> phiplii: right, i am trying to get a feeling how useful a heat sink would be - the GPU is not doing anything in this kind of workload
<phiplii> I was looking at something similar with thermoelectric generators the other day
<phiplii> it wasn't my work so I didn't dig far
<phiplii> but there should be ballpark calcs on tinternet for heat rejection from heatsinks
<phiplii> accurate calcs are tricky because some of the heat will be radiated, and to calculate that precisely you need to know about the environment (what you're radiating to)
<phiplii> but the transfer to air directly should be a bit better
<longsleep> right, what is the usual environment people run their sunxi boards?
<longsleep> are shitty cases common?
<phiplii> heh
<KotCzarny> yup
<phiplii> are OPi OEM cases transparent to IR?
<phiplii> :)
<KotCzarny> my boards run either caseless amd vertical or in metal box
<KotCzarny> plastic cases are usually isolators when it comes to heat
<phiplii> metal boxes are better by a long way than plastic
<phiplii> as KotCzarny says
<longsleep> well my boards usually run in a 19" rack with steady cool air flow
<phiplii> for the price, a heatsink does a lot of good
<longsleep> but hey, in my home i have a shelf which has my audio and video machine inside and its easily 60C in there
<longsleep> how does one type the degree symbol on an english keyboard
<phiplii> with difficulty
<phiplii> you need to use alt-gr and type the thingy number
<phiplii> which I forget
<longsleep> alt-gr err
<longsleep> do not have that either
<phiplii> 248
<longsleep> no windows
<phiplii> same
<phiplii> on a mac it is opt-shift-8
<KotCzarny> 60C is perfectly readable
<KotCzarny> dont overdo :P
<phiplii> I usually just write 18*C
<longsleep> heh ok
<longsleep> i suppose i can do that
<phiplii> or google the wikipedia page and copy it out...
<phiplii> which I so don't do
<longsleep> KotCzarny: yes, but 60*C ambient temp will have a huge impact how effective a heat sink is
<KotCzarny> i love it how seamlessly openwrt upgraded (12 to 15)
<longsleep> Ɉ
<phiplii> °
<KotCzarny> wrote flash, yet it maintainet configs/files
<phiplii> if you need one now
<longsleep> phiplii: no i just found it lame to write 90C all the time here in the chat
<KotCzarny> if your ambient temp is >30C, then you are doing something wreong
<longsleep> KotCzarny: inside shelf with my switch wifi psu and two arm boards its around 60˚C
<longsleep> hah!
<phiplii> Fan would fix it fairly easily - I can keep the whole conservatory <30 with one 120mm fan and a slightly open window...
<longsleep> got it
<phiplii> *<30
<longsleep> STRG+SHIFT+2da = ˚
<longsleep> (on Linux)
fredy has quit [Excess Flood]
<longsleep> one can write any unicode character that way
<longsleep> just need to know the hex code :{
<longsleep> phiplii: i have a no fans policy in my apartment :)
fredy has joined #linux-sunxi
<KotCzarny> longsleep: still, if its 60C you are doing it wrong :P
<longsleep> KotCzarny: well there is no air flow at all in there
<KotCzarny> you should make some strategically located holes to make air move
<phiplii> or use some metal to wick the heat out
<longsleep> KotCzarny: it works just fine that way - if it would have wholes i would have to do something about all the freaking LEDs in there
<phiplii> blutac
<KotCzarny> hehe
<phiplii> the friend of people who have their computer in their bedroom's best friend
<phiplii> sigh
<longsleep> yeah, but hey it works and - but any Allwinner board would probably not
<KotCzarny> longsleep: but blinkenlichts ist fun!
<phiplii> delete words until that makes sense
<longsleep> one should not have a compute in the bedroom
<longsleep> +r
<KotCzarny> you mean tablet/phone? :P
<KotCzarny> air conditioner controller? :P
<longsleep> i really mean any electronics at all
<KotCzarny> you cant get away from e-m waves
<KotCzarny> at least thos sbc boxes are low power
<longsleep> yeah .. i am ok with it if the stuff is invisible :P
<longsleep> anyhow, somebody else has numbers for their SoC how many cpuminer hashes/sec per core ? Pine64 does 0.8 per core no heat sink
<ssvb> longsleep: regarding cpuminer, for best hash/sec results it needs to be compiled as 32-bit code and also explicitly with -mfpu=neon in CFLAGS
<KotCzarny> does it really matter? those boards are made for diy, not performance/$$$,watt
<ssvb> longsleep: as for the results from the other hardware - https://litecoin.info/Mining_hardware_comparison
<longsleep> ssvb: i uses the settings from tkaiser
lynxis has quit [Quit: beaming to the mars]
lynxis has joined #linux-sunxi
<ssvb> longsleep: the cpuminer code just does not have 64-bit assembly, so it does not reach peak performance if you compile it as 64-bit code
<ssvb> longsleep: I have two static 32-bit binaries (with and without neon) here - https://people.freedesktop.org/~siamashka/files/20160312-litecoin-cpuminer/
<longsleep> ssvb: got it - i just wanted to see if that is in reasonable range which according to your link it seems to be
<ssvb> KotCzarny: the point was that we want to run a reasonably optimized realistic code for power consumption tests
<phiplii> longsleep : When I was a student I didn't have much choice where I kept my computer - I only had one room of my own
<KotCzarny> ssvb: uhum
<ssvb> KotCzarny: while mining digital coins on a CPU is not a very practical task, it is at least somewhat "realistic" when compared to a purely synthetic cpuburn
<phiplii> shame it isn't all free : http://www.heatsinkcalculator.com/
<KotCzarny> ssvb: decoding/transcoding media is quite realistic usage
ricardocrudo has joined #linux-sunxi
<ssvb> KotCzarny: yes, or "openssl speed"
<KotCzarny> or both
<longsleep> ssvb: yeah i just use it to get something to compare with/without heatsink
<KotCzarny> compiling things as another category
<longsleep> compiling zfs Kernel module on the 512M pine64 took 30 minutes
<ssvb> KotCzarny: cpuminer is quite good because it provides the khash/s feedback, which is sensitive to throttling
<ssvb> KotCzarny: compiling is a relatively lightweight workload for the CPU
<longsleep> ssvb, KotCzarny https://imagebin.ca/v/2ijmvj1Gok9b Pine64 at 23˚C ambient no heatsink
<ssvb> longsleep: regarding the 20C difference between the sensor (junction temperature?) and the outer surface, the A64 datasheet provides some numbers
<ssvb> longsleep: it has "Junction-to-ambient thermal resistance" listed as 21.6 °C/W
ricardocrudo_ has joined #linux-sunxi
ricardocrudo has quit [Quit: Leaving]
<ssvb> longsleep: but for the outer case surface we would probably need "Junction-to-case thermal resistance", which is listed as TBD
<longsleep> means they have not measured it or leave it to their customers
<ssvb> still having some difference between the outer case surface temperature and the junction temperature is quite normal
<longsleep> ssvb: yes sure, i felt that 20C is a little high though
<longsleep> ssvb: i am trying to get an idea what i should expect from a heat sink
<ssvb> without a heatsink we could estimate the SoC power consumption as (90˚C - 23˚C) / 21.6 °C/W = ~3.1 W
<longsleep> yes
<ssvb> or something like 620 mA from a 5V PSU
<ssvb> can you also measure the current draw?
<longsleep> ssvb: not at the moment, i am in the living room watching soccer :)
<longsleep> ssvb: but in general sure
<ssvb> but of course there are other components on the PCB consuming power, not just the SoC alone
<longsleep> sure
<longsleep> i have ethernet connected, do the specs include the GPU (as its on the SoC)?
<longsleep> i have UART connected which draws some power, no USB no HDMI
al1o has joined #linux-sunxi
Mr__Anderson has joined #linux-sunxi
staplr has quit [Ping timeout: 264 seconds]
Amit_t_ has quit [Quit: Page closed]
staplr has joined #linux-sunxi
IgorPec has quit [Ping timeout: 260 seconds]
tithrion has joined #linux-sunxi
tithrion has quit [Client Quit]
tithrion has joined #linux-sunxi
tithrion has quit [Client Quit]
kaspter has quit [Ping timeout: 260 seconds]
JohnDoe3 has joined #linux-sunxi
JohnDoe_71Rus has quit [Ping timeout: 260 seconds]
<longsleep> All right, interesting data when adding the Pine64 heatsink while on full load https://imagebin.ca/v/2ikE3dyOzTxl
<longsleep> heatsink had ambient temperature when mounted, took about 4 minutes for the system to reach 80˚C again
<longsleep> (which is the first trip point)
<KotCzarny> now add some micro fan
<longsleep> no fan :P
<KotCzarny> and be amazed :P
<KotCzarny> while i agree with you on the no-fan policy, they work wonders
<longsleep> yes i know
<KotCzarny> and for the test you can add them temporarily
<longsleep> no i want to test the heat sink
<KotCzarny> make some heatpipe then
<longsleep> and it just failed to keep the system from reaching the first trip point at 23*C
<longsleep> its the heat sink people might want to buy from pine64, not something i want to make
<longsleep> the question is if it makes sense to tell people to get it or not
<KotCzarny> well, it lowered the temp by a few C
<phiplii> how did you bond the heatsink to the soc?
<longsleep> i mean the system is faster now 0.84khash/s per core vs 0.80 without (as the temp is now 10*C lower)
<longsleep> phiplii: it came with some thermal glue pad and i just placed it on the SoC as a normal user would do :)
<phiplii> what do normal users look like? I don't think I've seen one?
<longsleep> phiplii: it is pretty shitty though, as the heatsink is also covering the dram which have a sligtly less height - so its not even
<KotCzarny> normal users are those that install images from pine64.com site
<KotCzarny> ;)
<KotCzarny> etc
<longsleep> havent they linked my images?
<longsleep> err image, as i do only have one these days
<KotCzarny> werent they the ones that ignored your work because 'images were too little' ?
<longsleep> yes ..
<phiplii> ?!
<KotCzarny> :)
<phiplii> heh
<KotCzarny> yup, that kind of 'normal' users
<KotCzarny> clueless kind
<KotCzarny> as opposed to hackers
<longsleep> but the heatsink they sent me is pretty decent, its 45*C everywhere
<longsleep> while the SoC now shows 80*C
<longsleep> 10*C less, only shortly touching the first trip point (which is at 80*C)
<KotCzarny> longsleep, put the board vertically?
<phiplii> Do you have access to a milling machine?
<phiplii> you could take a skim off to fix the SOC / DRAM heigh issue
<longsleep> phiplii: yes, but i will not do hardware myself
<longsleep> phiplii: that should be done by them
<phiplii> where's the fun in that?
<longsleep> time constrains - one cannot do everthing :{
<phiplii> :(
<longsleep> hum i touched the board to place it vertical and it rebooted :/
<longsleep> maybe the micro USB plug is falling off soon
<phiplii> Someone at work got offered a free Pine64, but they didn't see the email until a couple of days after they'd had another that said "if you don't answer within the day, we withdraw the offer"
<phiplii> No free stuff for you
<longsleep> lol
<KotCzarny> :>
<longsleep> i think the heatsink should be larger
<longsleep> the height could be doubled so it comes even with the height of the ethernet port
<longsleep> then it might be enough to stay below 80C at full speed
<KotCzarny> yup, the more, the merrier
<phiplii> Is all this just running at 1.2ghz?
<longsleep> 1152.00 Mhz
<longsleep> not even 1.2
<longsleep> for whatever reason they chose 1152 to be the maximum while advertising for 1200
ajeandet has quit [Ping timeout: 260 seconds]
<longsleep> probablue due to undervoltage issues
<phiplii> so... where does it gain over the H3, or is it a one trick pony? (64bit)
<longsleep> well i just wanted it for arm64 and armhf compile platform and it is awesome for that
<phiplii> have they changed the specs since they first announced themselves? I swear it looked a bit more interesting the first time I read about it
<phiplii> although I did realise that all the videos talked about features that aren't on the base model
<longsleep> phiplii: not that i am aware of, 2GB 1000M NIC arm64 quad core is all what i care about
<longsleep> KotCzarny: vertical is just 1*C difference, i almost cannot tell the difference
<KotCzarny> uhum
nove has quit [Quit: nove]
jernej has quit [Quit: Konversation terminated!]
jernej has joined #linux-sunxi
<ssvb> longsleep: how does this heatsink look? got a picture?
<ssvb> longsleep: I'm using a rather tall one, which is quite a bit higher than the ethernet port - http://linux-sunxi.org/File:Pine64_board_booted_over_SPI.jpg
<phiplii> ssvb : and black for extra radiation! :)
ajeandet has joined #linux-sunxi
<longsleep> ssvb: let me take a picture
Mr__Anderson has quit [Remote host closed the connection]
kaspter has joined #linux-sunxi
kaspter has quit [Remote host closed the connection]
tkaiser has joined #linux-sunxi
<tkaiser> KotCzarny: Prior to 9:20 OPi Plus 2E lying flat on the table, after 9:20 upright position. Difference regarding voltage switching cpuminer performance: http://kaiser-edv.de/tmp/Gn4wTB/Bildschirmfoto%202016-05-26%20um%2009.44.51.png
<longsleep> tkaiser: i received the heatsink from pine64 - results at https://imagebin.ca/v/2ikE3dyOzTxl
<longsleep> ssvb: the pine64 heatsink should be the height of yours, i think then it might prevent it from reaching 80*C
<ssvb> longsleep: but they probably want to be able to use some shields attached to gpio expansion headers
<ssvb> I can imagine that a tall heatsink can get in the way
<longsleep> right, well i sent my suggestions to them so lets see what happens :)
<longsleep> other wise the heat dissipation of the thing is quite good
al1o has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
ajeandet has quit [Quit: Leaving]
reinforce has quit [Quit: Leaving.]
<tkaiser> longsleep: If it's only 10C under load then the heatsink from Pine64 folks isn't efficient. Any cheap 50 Cent 20x20mm heatsink with 5mm height (google for 'cubie 20x20 heat sink') does better. Covering both SoC and DRAM with different height might be the culprit.
<ssvb> tkaiser: the load was not too heavy in longsleep's test
<longsleep> tkaiser: well, it cooled down pretty good after mounting it
<longsleep> tkaiser: so i assumed it connected well with the SoC
<tkaiser> The cheap 20x20 mm heatsink here was responsible for a drop from 90C to 75C: https://github.com/longsleep/build-pine64-image/pull/3#issuecomment-195401168
<tkaiser> longsleep: Do you have another Pine64 to test with?
<longsleep> tkaiser: yes i have 3
<tkaiser> longsleep: I would believe a heatsink that covers only the SoC will show better results.
<tkaiser> Due to different height
<longsleep> tkaiser: i think i can mount this heatsink only on the soc
<longsleep> ssvb tkaiser also when testing with cpuburn-a53 runs at 92C with 960 and 1008Mhz with the heatsink mounted as in the picture
<longsleep> heatsink heats up to 50*C then
<tkaiser> longsleep: Will test tomorrow. I sent the Pine64 with the heatsinks applied to Mikhail (who did the patching up to 3.10.110 back then).
<tkaiser> So will have a look what's lying around (ordered by accident a 'few' heatsinks a few weeks ago), test and report back.
<tkaiser> BTW: Size/material of the PCB seems to be important also. Since I have no idea why OPi Plus 2E and Lite do perform way better than BPi M2+: http://kaiser-edv.de/tmp/Gn4wTB/Bildschirmfoto%202016-05-25%20um%2018.14.54.png
<tkaiser> Left OPi Plus 2E (no heatsink), middle BPi M2+ (cheap heatsink), right OPi Lite (no heatsink)
<longsleep> tkaiser: yes i saw that - but well there is only one Pine64 PCB as of now
vagrantc has joined #linux-sunxi
Netlynx has quit [Quit: Leaving]
<longsleep> tkaiser: also the heatsink thermal glue pad is dissolving - i probably should replace it with good quality thermal glue
<longsleep> tkaiser: i also did a quick try when mointing that heatsink only on the SoC which works but no visible difference
<tkaiser> longsleep: Sure, I'm curious how OPi 3 and PC 2 (or whatever weird names Steven/Xunlong chose for their H64 based boards) will perform regarding throttling. Pine64 seems to be Allwinner's reference design minus some stuff (which might explain the large size)
<longsleep> tkaiser: yeah that will be interesting - did they announce release plans/schedule already?
<tkaiser> longsleep: Not that I know. But at least a prototype should exist since one picture was posted in OPi forums a while ago.
tkaiser has quit [Quit: jIRCii - http://www.oldschoolirc.com]
vagrantc has quit [Ping timeout: 260 seconds]
mosterta|2 has quit [Ping timeout: 272 seconds]
vagrantc has joined #linux-sunxi
JohnDoe3 has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/]
ricardocrudo_ has quit [Ping timeout: 252 seconds]
jernej has quit [Quit: Konversation terminated!]
ricardocrudo_ has joined #linux-sunxi
ricardocrudo_ has quit [Ping timeout: 264 seconds]
keh has joined #linux-sunxi
libv_ has joined #linux-sunxi
libv has quit [Ping timeout: 264 seconds]
libv has joined #linux-sunxi