<ullbeking>
my current plan is to convert an 20V laptop brick into a device that can power several of these things at a time, i.e., cut the plug off and the solder a number of male jacks in its place
<ullbeking>
i've got some dc-dc converters...
<ullbeking>
like step-down buck dc-dc voltage regulators (LM2596)
ninolein_ has quit [Remote host closed the connection]
<montjoie>
wow a 3G RAM variant!
<ElBarto>
nice !
<ElBarto>
really cool to have mini pci-e
ninolein has joined #linux-sunxi
<tllim>
hi wens
<tllim>
the 3G RAM still not 100% confirm yet, will test on next week
<tllim>
for miniPCIe, we will try out RTL-8111 GbE board and dual SATA ASM1061 board first
<montjoie>
on my pine64 I got python[24274]: unhandled level 2 translation fault (11) at 0x00000000, esr 0x92000006, this is reproductibbe any idea ?
<montjoie>
I tried as suggested by icenowy[m] to drop DRAM_CLK to 572
<wens>
hmm, the new wifi chip can do DDR50, which needs I/O voltage switching support
<rellla>
is H6 sun50iw3p1 or sun50iw6p1?
GrimKriegor has joined #linux-sunxi
<icenowy[m]>
6
camh has quit [Ping timeout: 248 seconds]
<rellla>
wens: then there is at least GPL2 code in the the H6-sdk for the display driver v3
megi has joined #linux-sunxi
jernej_ has joined #linux-sunxi
<jernej_>
wens: DE3 is mostly compatible with DE2, shouldn't be hard to adapt current logic
<jernej_>
it is interesting that they added another type of layer, along of vi and ui, fbd (I think), which supports only AFBC (Arm FB compression) formats
yann has joined #linux-sunxi
<jernej_>
there is no support in mainline for them yet
Langoor has quit [Ping timeout: 248 seconds]
Langoor has joined #linux-sunxi
tom_nov has quit [Ping timeout: 268 seconds]
<jernej_>
I wonder if they would open up DRAM code
<icenowy[m]>
jernej: I also wonder it ;-)
montjoie has quit [Ping timeout: 268 seconds]
montjoie has joined #linux-sunxi
<tllim>
I have request, seems like they didn't object
Andy-D_ has quit [Remote host closed the connection]
camh has joined #linux-sunxi
awais has quit [Ping timeout: 255 seconds]
awais has joined #linux-sunxi
<ullbeking>
willmore: thanks! That’s one of the best solutions I’ve seen so for.
Putti has joined #linux-sunxi
camh1 has joined #linux-sunxi
<ullbeking>
I did a search on Ali and found a store that sells a lot of good solutions for that plug type
<ullbeking>
will
afaerber has quit [Quit: Leaving]
camh has quit [Ping timeout: 268 seconds]
<ullbeking>
willmore: what are you using for your power source? I’m currently trying to build one using laptop power brick and a dc-dc converter at the end end to convert each terminal to 5V
<KotCzarny>
isnt dc-dc adding point of failure to the mix?
Putti has quit [Ping timeout: 255 seconds]
lkcl has joined #linux-sunxi
tkaiser has joined #linux-sunxi
afaerber has joined #linux-sunxi
tom_nov has joined #linux-sunxi
<tkaiser>
tllim: I would assume on Pine H64 USB2 is not present on pins 36/38 of the mPCIe slot? If so I would add a footnote to wiki and sales material so people know that the usual mPCIe WWAN modems won't work here.
<tllim>
@tkasier, the USB2.0 present on mPCIe pin 36/38
<tllim>
@tkaiser, I have link you up with JMicron, please check your email
lamer15107466370 has joined #linux-sunxi
lamer15107466370 has quit [Client Quit]
tkaiser has quit [Ping timeout: 255 seconds]
tkaiser has joined #linux-sunxi
ninolein has quit [Ping timeout: 240 seconds]
<tkaiser>
tllim: That's great. Sorry for the delay, in the meantime tested, responded and waiting to test out new firmware
ninolein has joined #linux-sunxi
hardfalcon has quit [Ping timeout: 260 seconds]
<tllim>
noted. please keep me update on your test so taht can factor into transformer project.
<tllim>
the transformer project currently in finalize stage.
<willmore>
ullbeking, I use an old ATX power supply that's been modified to be a bench power supply.
<willmore>
I don't really recommend that for most people unless you put some kind of current limiting in the cable. One tiny short on the board could easily fry the whole thing as the ATX supply has some 20A available.
ninolein has quit [Ping timeout: 268 seconds]
ninolein_ has joined #linux-sunxi
ninolein has joined #linux-sunxi
ninolein_ has quit [Ping timeout: 248 seconds]
scream has joined #linux-sunxi
<montjoie>
I use one without power limiting....
jernej_ has quit [Ping timeout: 260 seconds]
scream has quit [Remote host closed the connection]
<willmore>
montjoie, same here. I guess we live on the edge. I wouldn't recommend it to people whos electrical skills I'm not confident of.
<icenowy[m]>
tkaiser: do you know how is the LED next to UART connector on OPi R1 used?
<icenowy[m]>
tkaiser: when talking to tllim he says one USB controller is reused between USB Type-A and PCIE USB pins
<icenowy[m]>
so if PCIE WWAN is used one USB Type-A will be not usable
<icenowy[m]>
P.S. have you seen the OPi photos uploaded yesterday?
<montjoie>
willmore: what you would recommand for limiting power, just a simple fuse ?
<icenowy[m]>
the OPi 3 Plus and One 2 boards seem to have some problem
<icenowy[m]>
the USB port on One 2 seems to be only USB 2.0
matthias_bgg has quit [Quit: Leaving]
<icenowy[m]>
and the Ethernet on 3+ seems to be 100Mbps (the RGMII interface is muxed with CSI, and the HDMI IN seems to occupied CSI)
<icenowy[m]>
(then only EPHY is available)
matthias_bgg has joined #linux-sunxi
<tllim>
what happen to the USB 3.0 port?
<willmore>
montjoie, I'd like to use 12V to distribute power to the boards. Each board would have a polyfuse of the appropriate rating followed by a DC/DC converter which feeds the board.
<icenowy[m]>
tllim: I don't know
<icenowy[m]>
but if even USB3 is not wired it will be a hugh waste for H6 ;-)
nashpa has quit [Ping timeout: 250 seconds]
<tllim>
how you jump into the conclusion of not having USB 3.0 port?
<willmore>
icenowy[m], from looking at the pictures, simply a lack of space to put a USB3 connector in the vertical config.
<icenowy[m]>
tllim: see the USB port's length
<tllim>
I see
<willmore>
tllim, look at the pic of the third board.
<icenowy[m]>
how can 9 point sit in such small space ;-)
<willmore>
icenowy[m], they can ,but not for a vertical USB3.
<icenowy[m]>
and the USB on the dual-USB board (Lite 2) seems to be 3+2
<willmore>
The horizontal ones don't take up very much more space--just a few mm on the back. The vertical ones take up a lot more--as you can see from the pics.
<icenowy[m]>
yes ;-)
<willmore>
So, you can have ethernet or you can have USB3+wifi/bt?
nashpa has joined #linux-sunxi
<icenowy[m]>
yes
* willmore
really wants gigE and USB3 on a small board. Skip the wifi/BT. I was hoping the One 2 would do it that way. :(
<montjoie>
choice between ethernet and USB3...
<tllim>
@willmore, how about PCIe?
<icenowy[m]>
P.S. interestingly the H series IC usually have the biggest IO capability in a time
<tllim>
and also iommu
<icenowy[m]>
e.g. 4xUSB2 on H3/5
<icenowy[m]>
USB3 + PCIe on H6
<willmore>
tllim, Where would you put it on such a tiny board?
<icenowy[m]>
pine have no small size boards now :-)
<willmore>
Don't misunderstand me, please, I would love some PCI-E, but I'd need a Opi-PC sized board before I would expect there to be room for the connector.
<icenowy[m]>
instead there's the giant Pine A64
<willmore>
I'd be happy if they would start putting the SoC on the bottom of the boards so that it can heatsink to the case.
<tllim>
then the case need to be aluminum type
my123 has joined #linux-sunxi
my123 has joined #linux-sunxi
my123 has quit [Changing host]
<willmore>
tllim, or a heatsink could be put down there where it won't run into the inductors, capacitors, connectors, etc...
<tllim>
not a prefer way for Single Board Computer
<willmore>
I'm pretty impressed with the heatsinking of some of the newer friendlyArm boards. Their heatsink covers and protects a lot of the board.
<KotCzarny>
isnt it better for the heatsink to be on top?
<KotCzarny>
ie. convection
<willmore>
KotCzarny, Depends on how the board will be mounted.
<willmore>
If it's in a case, then it doesn't matter.
<willmore>
If it's in free space, then it could help to have it on 'top', but what side is the 'top'? Pick on or flip it over.
<tllim>
this is why I quote not a prefer way for a board
<willmore>
Most of these boards are going to be 1)in a case 2) mounted inside some other device.
<willmore>
In the first case, heatsinking off the bottom makes more sense. In the latter, the orientation of the board will be pretty much random, but likely inverted, so again, bottom heatsinking makes more senese.
Nyuutwo_ has quit [Ping timeout: 240 seconds]
<KotCzarny>
willmore, but it would require quite a tight fit
<KotCzarny>
ie. specialized case
<KotCzarny>
or heatsink
ykchavan has joined #linux-sunxi
<willmore>
KotCzarny, just a metal bottom plate or a bottom case portion.
<willmore>
Which is no more complex than a heatsink on top.
<willmore>
The board will be mounted with standoffs in the #2 case.
<tkaiser>
And it works really well. See ODROID HC1
<willmore>
Good example.
<willmore>
U3 as well.
tllim has quit [Quit: Leaving]
hardfalcon has joined #linux-sunxi
<icenowy[m]>
tkaiser: you have opi 0+ as well?
VargaD has joined #linux-sunxi
<willmore>
Is that the H5 variant with no/eMMC?
<tkaiser>
icenowy[m] Just arrived an hour ago
<tkaiser>
willmore: Check the wiki ;)
<icenowy[m]>
oh congrats ;-)
<icenowy[m]>
this board is finally not always ignored ;-)
<tkaiser>
Why should it be ignored? GbE, small and perfect companion to the NAS Expansion board.
<icenowy[m]>
yes but it seems that currently no many people use it ;-)
<icenowy[m]>
one of my friend purchased one, to run VPN
<icenowy[m]>
(not the same with the one who purchased R1 ;-)
<tkaiser>
I would want an OPi R2 with GbE and RTL8153 for this purpose
<willmore>
tkaiser, it's got a page now? Yay! ;) I can never tell these devices apart. So darn many devices named to crazily...
<icenowy[m]>
tkaiser: can H5 afford it? ;-)
<icenowy[m]>
dual GbE's
<KotCzarny>
tkaiser, there are 2 variants of opi0+2, h3 and h5
<tkaiser>
willmore: Yes, I wholeheartedly hate these Orange and NanoPi names
<montjoie>
willmore: thans for the polyfuse, will try with it
<willmore>
montjoie, you're welcome. Thanks for all of your work!
<tkaiser>
KotCzarny: Yes. And this page sucks. And I created a redirect to it.
<icenowy[m]>
my another (this time a new one) friend tried to use OPi 0+2 to play video
<icenowy[m]>
and they got picture broken
<icenowy[m]>
because of not enough DRAM bandwidth
<tkaiser>
KotCzarny: In the unmaintainable table the size of the new device is wrong: 45x48mm
<KotCzarny>
table is perfectly maintainable, and imo usable to navigate through instances of the boards
<KotCzarny>
no need to click multiple times to compare, choose
<montjoie>
willmore: any idea of the maximum allowed current ? 3A ?
jbrown has joined #linux-sunxi
<willmore>
montjoie, depends on the board. 3A seems safe. But, I suggest fusing *before* the DC/DC board. Because Polyfuses have a small voltage drop under load. Before the DC/DC it's not a problem. After the regulator it's more of an issue.
<willmore>
So, you'd need to spec that current to match the wattage you expect out. So, if the board can take 5V@3A, that's 15W. So, you'd need a 1A polyfuse on the 19.5V line going into the DC/DC. Or a 1.5A if using 12V.
<willmore>
Keep in mind the DC/DC converter will waste a little of the power, so you'll want to include that in your calculations.
awais_ has quit [Ping timeout: 240 seconds]
JohnDoe_71Rus has joined #linux-sunxi
ykchavan has quit [Ping timeout: 260 seconds]
<ullbeking>
icenowy[m]: what’s the story with OPi R1? I bought a couple because they were cheap and I am curious, but I haven’t used it yet
<ullbeking>
Are they bad?
<ullbeking>
In the same order I received plain OPi Zero, which I was excited until I discovered it is this weird Rev. 1.4, which also seems to be bad due to overheating problems :-(
<icenowy[m]>
not bad at all ;-)
<ullbeking>
Overheating is not bad? Or R1 is not bad?
<icenowy[m]>
R1
<ullbeking>
I also got a couple Zero+2, not sure about those either
<ullbeking>
It’s very comfusing mess
<[TheBug]>
ullbeking: they even overheat with lowered DRAM settings?
<ullbeking>
I haven’t yet tried with the latest kernel
<[TheBug]>
no im just saying lower DRAM setting in fex to lower mhz
<ullbeking>
Can you even get it on there if it’s so prone to overheating and freezing up?
<ullbeking>
Oh
<[TheBug]>
see if you still get overheat issue
<ullbeking>
I didn’t know you can do that :-)
<[TheBug]>
yeah was one of the biggest hurtles to running stable Android on the boards because a lot of the DRAM timings provided by OEM are set too high and were causing overheating -- Armbian also does a good job of testing DRAM and setting it more realisticly in their releases.
<ullbeking>
willmore: yes I heard mixed opinions on using ATX PSU
VargaD has joined #linux-sunxi
<willmore>
ullbeking, they have lots of power, but with great power comes great authority.
<willmore>
And with that....responsibility. ;)
<ullbeking>
Right!
Jagan has joined #linux-sunxi
<ullbeking>
That’s why I’m going for laptop charging brick instead
<ullbeking>
I work in a large company, have just written to asset management dept to ask if they have e-waste I can have
<ullbeking>
Somebody told me ATX supplies are fine as long as you know how to use terminal block covers properly, but I don’t even know what that is so I am leaving that alone until another day ;-)
<tkaiser>
All I find is /boot/orangepi/OrangePiH5orangepi.dtb on the Xunlong images and that's rather generic (wrong) stuff: http://sprunge.us/UHaP
<tkaiser>
That's the same rubbish settings from the Xunlong OS images for OPi PC2 (limiting cpufreq to 1008 MHz since they never figured out how to get voltage regulation working)
Jagan has quit [Ping timeout: 260 seconds]
lurchi__ has quit [Ping timeout: 248 seconds]
<ullbeking>
Haha, wow this OPi stuff really is bargain-grade computing, huh
lurchi__ has joined #linux-sunxi
<ullbeking>
I was chatting with a colleague yesterday about learning ARM (if that’s your reason for using OPi, etc) and we agreed that from that, educational, perspective, one may very well be better off simply investing in a proper dev kit with manuals and real data sheets, proper probes, etc.
<willmore>
ullbeking, what's wrong with using an emulator?
<ullbeking>
willmore: nothing, especially if you’re doing this as personal projects. (However I imagine that if you’re a professional hardware engineer designing a product for business reasons, then you probably use a real dev kit, however that’s a different use case with a different audience, etc)
<ullbeking>
tkaiser’s comment about voltage regulation prompted that thought
<willmore>
Ahh, when you said "learning ARM", I assume the instruction set, etc.
<ullbeking>
Well yes, but the whole ecosystem too, as well as everything else that goes along with designing hardware
<ullbeking>
Emulators are cool but when you put it onto a circuit board it’s different again
<willmore>
True.
<ullbeking>
So I’m kinda talking about two things at once
<ullbeking>
Mostly I’m just amused at how cheap and flakey these OPis can be, yet we still spend time on them because they are so much fun
reinforce has quit [Quit: Leaving.]
<willmore>
There are a lot of factors at work here. Let's start by saying that there isn't anything that prevents an Allwinner based board from being very reliable. There are, however, plenty of things that make that difficult.
<willmore>
Allwinner and their historical (and maybe now history) habit of not releasing decent docs make it a challenge to port an OS to their devices.
<willmore>
Board makers get better docs, but still not as good as they should get. They don't get enough support from Allwinner in their designs--often the designs are simple modifications of Allwinner reference designs.
<willmore>
Many board makers are new to designing boards or at least boards of this complexity. That leads to many n00b mistakes.
<willmore>
Then there is the code that Allwinner does release--the BSP--which is often of 'quick get it done and throw it out the door' quality.
<willmore>
If you have a board maker trying to stuff that software onto their poorly designed board, you're going to have a circus.
<willmore>
There isn't anything preventing a more mature board designer from making a very good board and from working with a community (like this one) to provide solid kernel support for it, but it's going to cost more money. Tkaiser has a saying about that...
<willmore>
Take Olimex as an example. They have a lot of experience making boards, but it still took them a bunch of tries to get things reasonably right.
<KotCzarny>
so.. back to the old story. i have old u-boot.bin but i assume its not enough to boot the device (a13), right?
msimpson has joined #linux-sunxi
vagrantc has joined #linux-sunxi
amasood has joined #linux-sunxi
<KotCzarny>
driver is not a problem
<KotCzarny>
oops, this goes to different channel. but question about uboot stands
Putti has joined #linux-sunxi
awais has quit [Ping timeout: 268 seconds]
Gerwin_J has quit [Quit: Gerwin_J]
chomwitt has joined #linux-sunxi
anarsoul has joined #linux-sunxi
<diego71>
KotCzarny: usually you need also the spl file
<KotCzarny>
how do i build/get it?
megi has quit [Quit: megi]
<diego71>
KotCzarny: look on the wiki for more information. Anyway spl file is build from u-boot source.
<KotCzarny>
i'm not talking about mainline but vendor source (2011), and i need it because of nand
<diego71>
The trick is than is not installed as a file, but written on the sd on a fixed position
tom_nov has quit [Quit: Leaving]
reinforce has joined #linux-sunxi
chomwitt has quit [Ping timeout: 240 seconds]
aballier has quit [Ping timeout: 240 seconds]
yann-kaelig has joined #linux-sunxi
megi has joined #linux-sunxi
aalm has joined #linux-sunxi
<tkaiser>
icenowy[m]: No idea about the additional led on OPi R1. Just booted it, everything works fine but this thing does nothing.
matthias_bgg has quit [Read error: Connection reset by peer]
<tkaiser>
But I'm currently using Allwinner's 3.4 rtl8152 driver...
matthias_bgg has joined #linux-sunxi
netlynx has joined #linux-sunxi
netlynx has joined #linux-sunxi
netlynx has quit [Changing host]
kloczek has quit [Remote host closed the connection]
msimpson has quit [Quit: Leaving]
lemonzest has quit [Quit: Quitting]
matthias_bgg has quit [Ping timeout: 240 seconds]
anarsoul|2 has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
amasood has quit [Ping timeout: 248 seconds]
awais has joined #linux-sunxi
yann has quit [Ping timeout: 250 seconds]
reinforce has joined #linux-sunxi
hardfalcon has quit [Ping timeout: 264 seconds]
<tkaiser>
KotCzarny: Slightly off-topic. But we now have a JMicron firmware backup/update tool. I was able to transplant ROCK64 SATA cable firmware to Xunlong's NAS Expansion board and now the latter is UAS capable
<tkaiser>
tllim introduced me to the JMicron guys working on this. They prepare a new firmware which also fixes SAT compliance issues (SAT --> SCSI / ATA translation)
afaerber has quit [Quit: Leaving]
<willmore>
That's important for smartctl, right?
<willmore>
Darn, I have to run. BBL.
lurchi__ is now known as lurchi_
hardfalcon has joined #linux-sunxi
<ullbeking>
willmore: indeed. And Olimex is an example of a board I’d like to explore, it just seems better quality and I hear decent things of it
<ullbeking>
tkaiser: I am confused as to how that NAS extension board could be called “NAS”.
<ullbeking>
Is it simply because it implants SATA natively?
<ullbeking>
implements*
hardfalcon has quit [Ping timeout: 240 seconds]
lurchi_ is now known as lurchi__
scream has joined #linux-sunxi
hardfalcon has joined #linux-sunxi
<ullbeking>
willmore: the thing that is amazing however is that new business and technology practices are such that new board makers are able to take a chip and create a functional board /at all/
<ullbeking>
it’s pretty incredible
<ullbeking>
almost to the point that I’m getting a bit of SBC overload
<ullbeking>
It’s still too hard to find something blob free and fully libre however
anarsoul|2 has quit [Remote host closed the connection]
<tkaiser>
smaeul: Nice, thank you. I remembered that $somewhere there was already a DT :)
<ullbeking>
smaeul: wow cool, thank you
<ullbeking>
It’s fascinating to see a new device added all in one go like that
<ullbeking>
Never knew it was so straightforward?! Is it?
<ullbeking>
Re those reg* structs,
<ullbeking>
Is that how you compensate for poor voltage regulation on the board?
hardfalcon has quit [Ping timeout: 240 seconds]
<tkaiser>
Huh? Which poor voltage regulation? The board can switch between 1.1V and 1.3V like almost all other small H boards. It's just not (yet) defined/active
<ullbeking>
I didn’t mean for that particular board
<ullbeking>
I meant the pattern of using those structs could be applied to boards where it is a problem, e.g. the OPi Zero Rev 1.4
<ullbeking>
I am just looking through the code and trying to figure it out
<tkaiser>
I got today an OPi Zero rev 1.4. Now I've one 1.0 from a year ago reporting low temperatures and this one most probably reporting high temperatures. But I lost interest... took Xunlong way too long to ship dev samples...
<ullbeking>
It just goes back to my earlier comments about using real dev kits vs cheap SBCs. The only reason I understand anything at all about these things is from independent research, experimentation, asking in IRC and forums, plus a lot of bumbling around and silly questions
<ullbeking>
tkaiser: yes the Rev 1.4 is really hot
<ullbeking>
I have a link at home about how somebody got it to cool down though
<ullbeking>
That was my issue, I ordered s bunch of them and I wish I’d gotten the pre 1.4 ones, this is just unnecessary hassle
<ullbeking>
And yes, likewise, after the +2E was out of stock for so long I started to lose interest too
<ullbeking>
And am now hesitant to order any in case of unannounced “improvements”
<ullbeking>
I asked Steven and he said they are the same but I am not sure how “same” is defined here
hardfalcon has joined #linux-sunxi
ninolein has quit [Ping timeout: 240 seconds]
scream has quit [Remote host closed the connection]
dave0x6d has quit [Quit: Connection closed for inactivity]
dave0x6d has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
Putti has quit [Ping timeout: 240 seconds]
Mr__Anderson has quit [Remote host closed the connection]
lkcl has quit [Ping timeout: 248 seconds]
ninolein has joined #linux-sunxi
jstein_ has joined #linux-sunxi
jstein_ is now known as jstein
<willmore>
Is the sy8106a in the PC2 controlled differently than in the PC?
lkcl has joined #linux-sunxi
lkcl has quit [Ping timeout: 248 seconds]
lurchi__ is now known as lurchi_
kilobyte has quit [Read error: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number]
kilobyte has joined #linux-sunxi
jstein has quit [Remote host closed the connection]
voxadam has joined #linux-sunxi
lurchi_ is now known as lurchi__
ninolein has quit [Remote host closed the connection]