Lion Wang reacted to Jens Bauer in Banana Pi M4
Cortex-A73 is by design (eg. ARM) using lower power and produces lower heat than Cortex-A72.
Cortex-A75 even lower power and quicker than Cortex-A73.
-So it will likely pay to choose the latter implementation over the former, even if the price of the CPU is higher.
For build-farms and quick data-processing, it's interesting having high-speed CPU cores and high speed network (this can be spread out on several GbE ports or just a single 10GbE port). 2GB or 4GB LPDDR4 would also be attractive for this kind of configuration. Native 6G SATA would be a huge advantage here as well.
For storage (eg. NAS), one could likely go with the old Cortex-A7, native 6G SATA support and 1GB to 2GB RAM (still 4GB will be interesting when you're using RAID configurations a'la FreeNAS, where each 1TB storage space requires 1GB RAM). Again as many (independent, full speed) GbE ports will be attractive for this configuration.
If the CPU you choose have PCIe, you can basically do anything you want; just please don't waste the PCIe on USB3. Adding PCIe switches would be interesting too.
As I've mentioned earlier, it's not easy to find an affordable board that has both native 6G SATA, GbE network and PCIe. I picked the EspressoBIN due to the low price and that it "technically" would cover my needs, but I've had many problems with it for several years. It still has problems when I make software-reboots (sometimes hangs), so that's a board I will not recommend. Some boards also have problems with the RAM being affected by EMI due to bad board design. The EspressoBIN was an empty promise; it can't be used as a router/firewall unless you add an external USB3-to-Ethernet adapter. The speed on the 3 ports is limited to 1Gbit for all three [eg. they share 1Gbps!], so I fail to see why they even bothered making the board more expensive by adding the Topaz switch.
(Perhaps so that other board designers, such as you, can learn from their mistakes?)
Lion Wang reacted to NicoD in Banana Pi M4
Isn't there a simular SoC with HDMI-in? I thought I red somewere this was the cheaper version of that SoC. Here-> https://www.cnx-software.com/2019/02/13/banana-pi-bpi-m4-rtd1395-board-raspberry-pi/
A small, cheap SBC with HDMI-in and Linux support for encoding it. That I'd buy and many of my viewers.
Then I wouldn't mind having 100Mbit/s ethernet and USB2.
Having an SBC for this would make life easier compared to the HDMI-capture box I'm using now. I never know what it is recording until I see it afterwards.
The W2 looks great, but it's too expensive not knowing if it'll work well for my goal.
Keep on making great stuff. I love all SBC's.
Lion Wang reacted to malvcr in BananaPi R2 (.csc mt7623 as new boardfamily)
This is a picture with an R2 located in a RACK assembly.
The set has an Orange Pi Zero as a "communication/firewall device", together with a Banana Pi M2+ as the Application Server and the Banana Pi R2 (v1) as a Database/NFS/ClamAV server. I am configuring a Moodle on the machine.
As a recommendation from TKaiser in a previous post, the R2 it is managing two 2TB hard disks with BTRFS and not RAID1 (really it is smoother and easier to configure than a software RAID). The disks power it is provided directly by the 430W power supply (not the R2 power connectors) ... I know this is as a 12 cylinder engine in a Beetle :-) ... The enclosure has 3 fan, so no machine arrive to 50C degrees.
It can be improved. There are details, but in the future things will be better. In this moment only the power led it is attached to something, and the Power Supply has a wire in the CPU connector to "simulate" something there (if not, the Power Supply will not work). The machines are located in a presentation cardboard piece painted with temperature resistance silver paint, as no screw post in the enclosure match the SBC machines holes. They are suspended on the board with metal posts.
One if my "nightmares" with this machine has been the internal networking. Basically the old style ifup-ifdown doesn't work. Armbian it is a complex thing because it keeps the ifup-ifdown together with the NetworkManager and the SystemD.NetworkD, all at the same time. So, sometimes you do something but it doesn't work because "the other guy" works against you and you really have no idea what it is happening, and things go wild when working three machines at the same time.
At the end, The M2+ and the Zero were configured with NetworkManager and the R2 with SystemD.NetworkD ... in this respect, they are not compatible in the Armbian setup methodology.
The Zero and the M2+ take their power from the R2 (later I will use the Power Supply directly). I remember when having the legacy 3 kernels and the zero was possible to use the USB cables for networking with the g_ether module. But I never was able to do this to work well with the 4 series kernels. In fact, now Armbian comes loaded with the g_serial instead. Could be interesting to recover that functionality, as this could reduce the complexity with this type of machine combinations. By now, I am relying in the old friend RG45.
Lion Wang reacted to Ryder.Lee in Banana Pi R64
The best news is that the final patchset version for HNAT framework was accepted ( kernel 4.16) and it's based on nftables. - https://www.spinics.net/lists/netfilter-devel/msg50973.html
We will try to add netfilter-base HNAT support for MT7622 & MT7623 in the future.
Lion Wang reacted to TonyMac32 in Banana PI BPI-W2
I think it's more like the dentist, always complaining that I'm not flossing enough... I don't want to hear it, but the dentist is right.
4.9 kernel, ok, 2015 u-boot, . Looks like an interesting board, but I worry about drivers and any hope of mainline support.
Lion Wang reacted to hjc in Banana PI BPI-W2
You're not alone, many board makers have to sign NDA with chip vendors. For example, when Dragonboard 820c was under development, they signed NDA with Qualcomm, and they must get approval before releasing Qualcomm kernel/bootloader sources.
However before they got the approval to release kernel/bootloader source code, they didn't even release the board, nor any GPL-licensed binaries. You should follow the rules, too.
Lion Wang reacted to Larry Bank in Banana Pi Zero
I just received my BPI-M2 Zero board and put it through its paces. Seems to perform similar to other H2+ boards. I added a Pin->GPIO map to my SPI_LCD and ArmbianIO projects (https://github.com/bitbank2).
1) The SPI driver has that same odd issue where it occasionally spits out errors and runs a little slower than it should.
2) It doesn't come with an IPX antenna, but you need to add one to use wifi unless you're sitting on top of your access point.
3) A small passive heat sink seems to be plenty when running the mainline (4.1x) kernel.
4) The default HDMI out resolution is 1920x1080p@60fps. This causes a "jumpy" display on my monitor. Switching to 1280x720 fixed it. I've seen this on other AllWinner H2/H3 boards, so it's probably something strange with the mainline kernel hdmi code.
Overall I'm happy with the board. It would be nice to use the bluetooth. Does anyone know how to enable it?
Lion Wang reacted to guidol in Banana Pi Zero
if the costs are rising to fast - how about installing a standard-eMMC socket like on the ODROID C2 or inside the Pinebook?
I think this would cost only a few cents?
For me that would be fine - then while I upgraded my Pinebook to 32GB eMMC I could use the old 16Gb eMMC on my ODROID C2
AND if the market will supply faster/bigger eMMC modules I can swap out the slower/smaller against a faster/bigger one
Lion Wang reacted to guidol in Banana Pi Zero
I think nobody has a complete "perfect" product....neither "Orange Pi" nor "Banana Pi" nor "Nano Pi" nor "Raspberry Pi".
Every product has some details that could be better (lower heat, better documentation, better linux-images, better cases, less binary blobs).
So I got some of every sort
But - I personally - like the wiki from FriendlyElec/FriendlyARM at
Most - NOT ALL - pages does include very interesting and valuable informations for me (PinOuts and HowTos, FAQ).
OK some should be more actual - but there are many pages for many diffrent models and if ne page hasnt the information then I could find mostly the information at a page of a other model.
The downside - their forum seems mostly supported from users to users. No technican seem to take a look here on a daily basis
Lion Wang reacted to lvmc in Banana Pi Zero
@Lion Wang SinoVoip's gitbooks has to be reviewed. Technical information is not consistent and English is not good enough, it has to be fixed urgently. Over the last years I have been seeing many Chinese companies doing good products or at least pursuing to build good products but failing on how to communicate and support customers, think about it. Never forget that customers don't care about how hard is to build a new product or how insane is to run a business, they only care about how good the product is.
@Tido gave a good suggestion... before releasing any new product and during the entire product lifespan do short technical / writing meetings with your technical team to consolidate knowledge.
The entire problem here seems to be about communication, as always. Most of Chinese companies have good engineers that doesn't even speak or write in English. If it is the case, hire someone that could help you to fulfil this gap, as soon as possible.
@TonyMac32, @chwe thank you for focusing on the real problems, not at person level discussions. In just a few questions I think you got the answers you were looking for. That is a good start point.
@tkaiser I would like to suggest you to just 'stop' buddy, please.
Let's give the last chance to SinoVoip understand, change, react and SOLVE the main issues. Everything said here is applicable for any hardware supplier.
Lion Wang reacted to TonyMac32 in Banana Pi Zero
Indeed you do, however it does not have "CON4" anywhere on the board silkscreen that I see, and being a 2x2 header alignment is still a bit of a question, especially when Pin1 is marked to be on the right-hand side, which is non-conventional compared to the rest of the board. Not saying anything is inherently wrong, the schematic is there, is labelled, but I for one, a senior electrical engineer, did not notice it, instead seeing an unlabelled 2x2 on the physical device. This should be on your page explicitly labelled, and probably a silkscreen adjustment made.
Now, were it a 2x3 with 2 of those pins being a V_in.... Now that would certainly be useful and meet the approval of many. Not to mention making orientation a non-issue as it would then be non-ambiguous. ;-)
Do the various test points exist in hardware, such as the "TV-out" (TP8?)? If not they need removed, simply because you are referring to the schematic as an ultimate resource (as it should be) That said, TV-out is a Pi Zero feature...
If you are open to a feedback, I would say a good variant would be one of these without wifi, but instead with the additional USB's available via header, a position for a barrel jack or direct-solder power input, and an eMMC.
Lion Wang reacted to chwe in Banana Pi Zero
Nice to get a fast response. But when you've ethernet there, this should be on your schematics! and described somewhere. Most people would expect that the connector might be for powering (or maybe the USB otg). And to improve things, make sure that this information is not only visible here in the armbian forum. It should also be in your documentation for this board.
Lion Wang reacted to chwe in Banana Pi Zero
Maybe not for your use cases but for others. Not everyone has the same needs... I see use cases where size and wifi matters... Since OPi0 has heat issues, and both other cheap small boards (opi0 and nanopi) have the annoying XR819 wifi.
don't start this again... Thomas and you will fight on a personal level again, and @Tido will close the thread again...
Doesn't make sense... If they have an interesting board, I'm sure people from the Armbian community will work on this board (maybe not Thomas but others.. ) And as @@lex said... Edit: Getting basic armbian support for this board should not be that much work. Starting with a similar board (1.1V/1.3 V regulation) add wifi and configuration of fex for the pinheader and you'll have a basic armbian image for this board.
I appreciate this, so let's start with the 4pin connector from my last post.
FYI: Short version of the post with the picture which is lost somewhere in the www nirvana during upload. I was too annoyed to write everything again...
Lion Wang reacted to lvmc in Banana Pi Zero
@tkaiser and @Lion Wang it is time to put an end point on this "fight".
In my option, things are getting too aggressive and boring from both sides and it is not good for both the Armbian community and SinoVoip business.
@Lion Wang the community & customers are not satisfied with the information you are providing about your products. As CEO you have to listen and act as soon as possible to correct what is wrong.
@tkaiser you are essential for Armbian community, but it has been useless to read your posts about SinoVoip.
My suggestion is to give the last official chance for SinoVoip with Armbian community. I suggest to port Armbian to BPI ZERO and during the porting process both sides tell everything about what is not going right or hasn't been done right. Documentation is not good? SinoVoip will fix it. We need some detailed information about the hardware? SinoVoip will provide it... and so on.
Let's do it, guys?
We all want to build an ecosystem between hardware and software and a friendly community.
Lion Wang reacted to zador.blood.stained in BPI-R2 Board Bring-up
I don't remember anyone linking the mainline DT for this board, so here we go:
It's interesting to see a CPU operating points table there (so DVFS should work already), looks like Ethernet (with DSA), USB and basic peripheral interfaces (I2C, CIR, I2S, PWM, GPIO, ...) should be supported too, but don't see anything that would represent SATA or PCIe yet.
Lion Wang reacted to malvcr in BPI-R2 Board Bring-up
It is good to know that the R2 it is being taken seriously. This machine has important things and, although there are very good alternatives, it has a market place.
I have been sending maybe 100 times by now an 825 megabytes ubuntu iso file from one R2 to another and between an R2 and a Mac Mini machine, testing different types of configurations (in the while I am creating the system I will use the R2 for).
Here there have some numbers that could be useful:
(AES 256 bit):
100% CPU max
- real 0m53.189s
- user 0m44.210s
- sys 0m8.030s
75% CPU max
- real 0m27.015s
- user 0m2.290s
- sys 0m17.750s
Checking at this test alone, it is clear that with the cryptodev driver active (and with the right openssl compiled for it), the machine it is faster processing. And then it is the top 100% capacity when using only the CPU ... I was trying to figure how to test that remaining 25% ... so, I made a multithread program that received the data (running in a R2), and executed 4 parallel sets of openssl+sending data from the other R2.
The "general "throughput" for all the "bundle" gives around 45.8 MB/s. This is much higher than the around 17 MB/s I can have with only one similar session.
The issue here is that the final speed can't be calculated only taking into consideration the crypto engine. A final test would need software designed for this, because when I cypher with openssl and then send the file on ethernet, I need to "write" the file to disk and to re-read it, and the DISK is a key factor on the overall transmission speed. An extra write is really heavy here.
So, if I like to see a wonderful speed without sacrificing the machine, the disk must speed up. The final numbers for secure transmission of data must involve all the key factors : CRYPTO+DISK+NET+CPU.
But ... in general, I think it is good enough for my purposes. When I have a better software platform to test all together (without punishing any of the factors), I will come to show my numbers.
Lion Wang reacted to malvcr in BPI-R2 Board Bring-up
I think I understand now this board.
- It is not an integrated board, but a "carrier" based one. You purchase the carrier board and the SOM (System on Module) to make it to works as you expect.
It uses only one line for mSATA interface, the same as the R2. Then, why the difference in speed?
It is necessary to add a mSATA to SATA adaptor to connect standard SATA drives, or to use mSATA drives. And uBoot must be modified to allow the PCIe slot to work as mSATA.
- SolidRun Armada SOM A388 with eMMC : $69
- With ClearFog Base Carrier $129
- With ClearFog Pro Carrier $189
Similar scenario than ClearFog (using SOM).
- Only has 1 ethernet
- Quad 1 GHz NXP i.MX6 version 2GB RAM and 8GB eMMC: $191
- Needs M.2 to SATA adapter
- No USB 3.0 (only 2.0)
For a multi-ethernet scenario with storage, the HummingBoard is limited by the lack of native multiple RJ45 and USB 2.0 (where the second ethernet could be attached).
When only needing two ethernet, the ClearFog with Base Carrier seems to be enough. Cost is around 50% higher than the R2. And the Pro around 100% higher.
ROCK64 ... not yet available for purchase (ships until November 3 if purchased in October 16 - Pine has their history on delays)
- $60.89 without shipping ( 2GB version + 16 GB eMMC + USB-SATA cable ) ~ still lacks a secondary ethernet.
- It only has one USB3 port, so the bandwidth must be shared between SATA and any secondary ethernet.
- $49 in amazon
- Dual Core
- Three Gigabit Ethernet ports (1 WAN, 2 LAN)
- Independent SATA interface with its own Power Supply
- Proper 12V barrel power connector
- Mini PCIe
- Has the place to add the eMMC but it seems must be soldered there
R2 still has a place in its price/availability/performance ratio.
If there is an ExpressoBin with eMMC included, Quad Core and 2GB, maybe it could cost around $80. In that case, would be a better option than the R2.
The ROCK64 lacks interfaces to provide good bandwidth when the multi-ethernet scenario is included.
It is not possible to have HummingBoard with multiple Gigabit Ethernet connections.
The Armada SOM is Dual core ... could be possible to use the HummingBoard SOM with the ClearFog carrier?
The main important elements are to determine if an improvement in the software can work the problems detected with the R2 for the full performance capacity.
In my particular case, although performance has some importance, it is not the main driver to choose one or another product. The processing unit is more important (hence 4 cores would be better than 2), together with the integrated eMMC (I can't deal with soldering these tiny things) and the availability for connecting devices.
Today all them are not so perfect (taking all factors into consideration) options. I am sure than in 6 months this will be a completely different world.