Igor

https://www.armbian.com/logbook/ Prime is in a better shape than Win.

Everything is fixable but ... you have to decide what is best for you. For debugging you need a serial console and some experiences. This part we could try to solve if you are able to provide full logs ... but it might take more time than starting from scratch, preferably on a new SD card ... since they die sooner or later. Then copy your data from old sd card here.

Try this way: https://docs.armbian.com/User-Guide_Advanced-Features/#how-to-unbrick-the-system And. Bad SD card is a common reason for a failed upgrade since this job is a stress for it. A lot of small files has to be written ...

Hummingboard v1 has mSATA and mPCI so this information is correct Now let's think how to store those features best way possible. Exactly what is useful to store and how? Currently, we get some data from /build/boards, then board properties are attached to WordPress articles(board download pages) as TAGs and I have one exception table for boards without the unique config, which is used when building download pages. Now we shell have all at one place/sucked to the database, but since one board config covers more boards, what would be the best way to proceed. Cubox-i is first such example. It's one image and covers 3 boards, HB1, 2 and Cubox. Shall we have those alike boards in the main config or we add subfolder in build/boards/xxxx and keep those configs there? We only need them for building this database and when in the database, we can do filtering.

Need to be investigated. If you use the one from the download section, then it's supported. If you build on your own, check/get involved in https://forum.armbian.com/forum/20-development/ Just after the last release we moved up with u-boot and things might be partially broken at the moment.

If you need a silent u-boot, use this patch: https://github.com/armbian/build/blob/master/patch/u-boot/u-boot-sunxi/remove-boot-messages-from-hdmi.patch.disabled and recompile u-boot. We don't have the intention to provide silent u-boot by default.

Not an issue. Reboot. Wrote on mobile

Try without dkms.

As I understand you lost network connection when logging out?

Than a twit feed would be just fine. Now I was using a box below the menu to put out the news. Such place is a must and IMHO location is right. Twit is already multiuser ready so I don’t need to do it alone. Here we talk about visually exposing torrents, to produce more clicks. Technical reality does not count and we can limit download speed later. Torrents can also be tuned up a little more. I don’t know why sometimes takes a lot of time before the download actually starts. Is logo home for www or forum or whatever? It’s a good question and perhaps we test how this works for new users. We already have hardwired navigation, whether is correct or not. This can be prepared to match both scenarios and adjust according to new users natural behavior later when the new page is ready to roll. A 2-3 week test should tell. Imagine that we are setting up a web page from scratch. Or that we set up a future-proof website from the current perspective: “What is Armbian now. What is represent?” Current web presentation is still only some enhanced version of my personal project. There are some good points too and those we should keep, transform into the new experience. We need to properly link things together, add what’s missing, remove what’s pointless and adjust the design. I agree that things are not O.K. - that’s why this topic exists. That is why we seek out for help and I am not possibly able or willing to do this alone. 1. dl.armbian.com is a file server and nothing else. For experts only, for those who need little to no additional info. We only need to add a common header & footer, unify fonts and it’s more or less done. On a related topic. We have links with .7z in the name which are linked further anyway. Without .7z we could change to any compression method ... 2. forum.armbian.com is something you described. A mixture and the heart of the project. Perhaps we could rather say, content is the heart of Armbian. Overall. Not just forum content, but the project per see. There are many silent users which will never say anything or little. But they are around. 3. www.armbian.com is ugly and overloaded. Agree. That’s why we have this debate. It is not what it should be. Since the key element is content … page design does not matter so much, UX does and this is what we are fixing in a phase one. The value what we give is worth many troubles and people do come back. Scientific data tells that there is no vivid difference between forum (58%) and www (56%) in returning VS new visitors in 2017. IMHO we need to give www a central role to the project. Without adding junk. It should be a place where you find news about development progress, features marketing and meaningful links to our product (images), content(documentation) or services(help/forum). IMHO this should be placed at /download as it is now … somehow it is natural that where we see all devices to have an (advanced) option to narrow down what do we see. By their properties. Technical background is that we need to prepare a better database, actually only add to board.conf those missing data. This way two things are solved – we are all used to enter data via Github and we don’t need to code separate entry form … which means anyone with commit rights can change any board property on the main website. Now not everything came from there - I still needed to add some data (image, name, which file to include) via WP. The only technical difference is that Github data will be sucked to WP database and anything search-wise can be done in WP. This part will be done by those guys. We certainly don’t need content creator since we have plenty of it. Unique. If little and big executions fall mostly on me, I have two options: push out when it's ready which means "never" or push it out with compromises and fix on the spot - check how many releases Debian put out after their first new stable release. And Debian as a synonym for conservative approach also seems to be only a myth by now ... Perfectionism in software development can backfire. Waiting, sprinting, staying hight on alert before release can’t last long. It’s harmful to us ...

Install packages from here: https://beta.armbian.com/pool/main/l/linux-4.13.14-sunxi64/ you can leave headers out if you don't need them. Or via apt-get. Hmm. We could implement this to armbian-config, to make rollback simpler. Yes, freeze in armbian-config or manually.

If you have internet access please provide the output of: armbianmonitor -u

I guess enabling nightlies was perhaps too fast (I need to test repository management) but at least a problems were found. Since you have a temporal network, roll back to previous nightly version ... and don't update kernel until this is fixed.

Well, this particular not but they design new ones with the same Soc: https://www.cnx-software.com/2017/11/12/nanopi-fire2a-fire3-boards-released-with-samsungnexcell-quad-octa-core-processors

Linux kernel development is anarchistic and our work is involved, dependant and related to this organizational type. Adding too strick wireframe and being harsh where is absolutely no need might be more damaging ... and a waste of time if you like that kind of word form. Scaring people not to come out with their ideas because somebody will bark at them how stupid they are might be also a waste of time. And damaging. Most people around the project are volunteers after all and demanding ultimate compliance won't work. Being a pro/cash has little effect on this either. Talking and discussing every possible matter consumes human resources which, frankly we don't have in indefinite capacity. Nobody has. Somebody has to lead and coordinate this web redesign and until this is mostly on me, I will resolve minor disputes instantly or I will not deal with this in any way. This is not a military precision stuff! Why do we need to panic? It's plenty of time, this is not print media and mistakes can be fixed. And I will for sure do more mistakes in the future. Perhaps even repeat them, which I hope not. If you think everything is waste of time, take a break from extreme specifics ... even this can be considered as a waste of time from a radical point of view. You have possible future boards in the dedicated section and there is a place for discussion if needed. There is no rush - take time. It's your right to write down that all boards sux and is pointless to deal with it. Should this change anything? Not necessarily and it should be this way. Can we now stay on this topic?

Navigation is wrong and my only thought was to do this either now or at relaunch. I put it back until then. ok?

WIP https://libre.computer/products/boards/roc-rk3328-cc/ Pro: - yet another RK3328 Cons: - micro USB powering

[WIP] https://libre.computer/products/boards/all-h3-cc/ Schematics Pros: - yet another H3 Cons: - micro USB powering

[WIP] Schematics Pros: - board vendor actively participates (listens to the community, provides information including schematic and cares about correctness, tries to bridge developer community and chip vendor) - board vendor take care of their board configuration and submit patches - board vendor provides dev samples - yet another A64 - good powering method Cons: - none specific to this board

[WIP] Schematics Pros: - board vendor provides dev samples and documentation - working with latest mainline kernel - many variants by the same vendor - official SDK/BSP currently on 4.4 - no unreliable shitty Micro USB for DC-IN but GPIO by default Cons: - new platform needing more initial work - not many active developers

"discussion = the action or process of talking about something in order to reach a decision or to exchange ideas." and it this topic we have to reach it and direct how WWW should look like. The first draft is how they understand us, mostly my bubbling. It can be far from our consent. It can be completely wrong. I don't like it either but something must be there, perhaps some latest new or twits instead? Fine for me, what others think? Don't really understand a point of this? Yes, better. Wordpress is CMS. Some boards have desktop some only server. I think there are those two "recommended downloads" with a torrent icon (well perhaps it doesn't look very well yet) following the direct link behind a name. Here are more options - text, icons or combination as it is now ... we want more torrent downloads and this way could be better in that direction. It was considered but it's hard to tell without seeing stats. I can only trust their UX experiences here. Now, a lot of things are wrong and even it's wrong we are used to those wrong actions. I have no idea to what you are referring to? With re-design, we haven't got anywhere yet and if your bubbling was overheard by a person in charge of a design perhaps you need to point it out. He understands our pain the way it was designed and can easily be reshaped around at this point. If it doesn't go in the right direction, speak up. "Everything is completely wrong" and "I already told you something completely opposite" tells nothing. Start to change this is not joyful for me and I would rather not deal with it but I have no other options. If you like to lead this web transformation, focus, sum ideas and communicate with the design team, welcome. I have other things to do. We are redesigning page for newbies. We can handle it in any case.

What exactly you want to install?

We don't support 3rd party hardware. Not possible. If this problem was caused by an upgrade, you need to find why and possible provide a fix for it. It's most likely a banal reason since there were not many changes to the old legacy kernel. Note that script.bin is overwritten on each upgrade. If you use custom, save it on different name and link to your script.bin to prevent from overwriting.

Unrelated to armbian. Wrote on mobile

First draft. Comments and ideas over are welcome. UX related only at this stage.