Better WiFi/BT module?
#11
FM+GPS functionality would be awesome. But I dont know what are limitations comes from SDIO port. Also I think it would be difficult to make this kind of module because the pins of wifi expansion port maybe a limitation. For me I think it is not worth to try, needs lots of customization in kernel and I dont even mention about making a module from bga packaged chip  Confused
  Reply
#12
I checked availability of the chipset on several distributors and didnt see stock anywhere... a better approach is probably to just make an adapter board for an existing SDIO module...
Come have a chat in the Pine IRC channel >>
  Reply
#13
I dont understand, adapter board for existing SDIO module, what do you mean?
  Reply
#14
Well you get an SDIO stamp (like the one on the Pine module from Realtek) and make a breakout PCB to adapt it to the pin headers of the Pine board...
Come have a chat in the Pine IRC channel >>
  Reply
#15
The mentioned chipset was just an example, as that was the one with the highest possible bandwidth that supported SDIO. So it's not a necessity to have THAT chip, but basically anything that can get, say, 800Mbps+ on 5Ghz would be perfect.
  Reply
#16
(02-09-2017, 11:35 AM)fonix232 Wrote: The mentioned chipset was just an example, as that was the one with the highest possible bandwidth that supported SDIO. So it's not a necessity to have THAT chip, but basically anything that can get, say, 800Mbps+ on 5Ghz would be perfect.

Jomsmirl suggested AMPAK module and here is the link: http://www.ampak.com.tw/product.php?id=22. I will explore on next week when meet up with China vendors.
  Reply
#17
(02-12-2017, 10:43 PM)tllim Wrote:
(02-09-2017, 11:35 AM)fonix232 Wrote: The mentioned chipset was just an example, as that was the one with the highest possible bandwidth that supported SDIO. So it's not a necessity to have THAT chip, but basically anything that can get, say, 800Mbps+ on 5Ghz would be perfect.

Jomsmirl suggested AMPAK module and here is the link: http://www.ampak.com.tw/product.php?id=22. I will explore on next week when meet up with China vendors.

That AMPAK module, while looks interesting, it does not provide full speed - 433Mbps is pretty much what you'd expect from an N network, not AC.

However if you check the 356S (http://www.ampak.com.tw/product.php?id=23), it provides double the bandwidth at possibly similar pricing.
  Reply
#18
You have to keep in mind that SDIO is not PCIe or USB3.0 :-)

Does anyone know what the maximum bus speed for SDIO3.0 is on the A64?

http://files.pine64.org/doc/datasheet/pi...t_V1.1.pdf page 41

If I read this correct, the SDC1 controller supports either

DS,HS,SDR12,SDR25,SDR50,SDR104

DDR50

SDR104 has a bus clock of 150Mhz, whichs means that at 4 bit you would get theoretically 600Mbit/s raw bit rate minus the frame/protocol overhead,
so in practice probably quite a bit less?

Tyical SDIO3.0 Wifi modules have around 433Mbit https://boundarydevices.com/product/bd_sdmac_wifi/

You also need 1.8V signaling for the SDR104 mode... which is maybe possible because Port G seems to be connected to it's own DLDO/ELDO

http://files.pine64.org/doc/Pine%20A64%2...elease.pdf
http://files.pine64.org/doc/datasheet/pi...t_V1.0.pdf

ELDO1 is adjustable from 0.7-1.9V (Imax 400mA)
DLDO4 is adjustable from 0.7-3.3V (Imax 500mA)

Faster modules might still give a little bit more headroom for link degradation...
Come have a chat in the Pine IRC channel >>
  Reply
#19
There are 802.11n WiFi modules that support 5GHz, 300Mbps-450Mbps speeds and ALSO respect your freedom - don't need any closed source code to function. In example, Atheros ath9k family, and AR9462 module. Meanwhile, there are no 802.11ac which could work without any blobs. Hopefully the next revision of Pinephone 5 will pick AR9462 as its' WiFi module, instead of the crappy Realtek/Broadcom/Intel/Qualcomm blobbed alternatives.
  Reply
#20
(02-07-2017, 01:20 PM)fonix232 Wrote: Would it be possible for a 3rd party to create a better module?

I believe with the SDIO interface, it should be possible to implement a Broadcom BCM4356 (which supports SDIO v3.0 for WiFi and GPIO for BT4.0), and achieve WiFi speeds up to 867Mbps (not to mention this chip has a built-in FM receiver, which could also be used).

Also I understand that this would raise prices for such a wireless module, but for many, it would be a better choice (especially as there's no PCIe or USB3.0 port onboard). I just feel like 150Mbps is quite limited for such a versatile device.
Hello,

On my pinebook, i update the kernel with recent Debian bullseye

Code:
Linux pinebook-290 5.10.0-5-arm64 #1 SMP Debian 5.10.24-1 (2021-03-19) aarch64 GNU/Linux


and also with recent Debian firmware-realtek package...

Code:
ansible@pinebook-290:~$ dpkg -l |grep firmware
ii  alsa-firmware-loaders                1.1.7-1                            arm64        ALSA software loaders for specific hardware
ii  firmware-linux-free                  3.4                                all          Binary firmware for various drivers in the Linux kernel
ii  firmware-realtek                      20210208-4                          all          Binary firmware for Realtek wired/wifi/BT adapters
ansible@pinebook-290:~$

After inserted r8723bs kernel module, i have in console ....

Code:
[  150.053978] r8723bs: module is from the staging directory, the quality is unknown, you have been warned.
[  150.082005] RTL8723BS: module init start
[  150.085973] RTL8723BS: rtl8723bs v4.3.5.5_12290.20140916_BTCOEX20140507-4E40
[  150.093045] RTL8723BS: rtl8723bs BT-Coex version = BTCOEX20140507-4E40
[  150.103510] RTL8723BS: module init ret =0
[  150.104008] systemd-udevd[357]: rtl8723bs: Device (SEQNUM=2815, ACTION=add) is queued
[  150.107685] systemd-udevd[357]: Validate module index
[  150.107770] systemd-udevd[357]: Check if link configuration needs reloading.
[  150.112206] systemd-udevd[357]: Successfully forked off 'n/a' as PID 4072.
[  150.112481] systemd-udevd[357]: rtl8723bs: Worker [4072] is forked for processing SEQNUM=2815.
[  150.113173] systemd-udevd[357]: r8723bs: Device (SEQNUM=2816, ACTION=add) is queued
[  150.113523] systemd-udevd[4072]: rtl8723bs: Processing device (SEQNUM=2815, ACTION=add)
[  150.114573] systemd-udevd[357]: Successfully forked off 'n/a' as PID 4073.
[  150.114927] systemd-udevd[357]: r8723bs: Worker [4073] is forked for processing SEQNUM=2816.
[  150.116757] systemd-udevd[4073]: r8723bs: Processing device (SEQNUM=2816, ACTION=add)


and no network device is created, so that no WLAN interface  (!)

Any idea ?

Thanks in advance
  Reply


Possibly Related Threads…
Thread Author Replies Views Last Post
  [split] Better WiFi/BT module? moriel5 3 5,796 05-18-2022, 01:55 PM
Last Post: moriel5
  WIFI/BT Power Supply / VBAT-EXT xalius 3 5,323 08-12-2021, 06:41 AM
Last Post: Luke490
  Wifi / Bluetooth Board Update S265 2 4,601 07-30-2020, 04:53 PM
Last Post: S265
  Create a WiFi hotspot on Debian 8 JulianM 6 28,280 03-10-2019, 01:02 AM
Last Post: waqasnasir
  Schematics of Wifi/BT module gamelaster 7 10,837 09-11-2018, 02:02 PM
Last Post: gamelaster
  Wireless Module Driver Source code mido2018 0 2,224 08-08-2018, 04:26 PM
Last Post: mido2018
  Installing another WiFi module on one of the buses mido2018 1 3,950 07-22-2018, 07:15 AM
Last Post: tllim
  Unable to activate wifi on Linux smouton 10 17,281 02-21-2018, 09:56 PM
Last Post: dkebler
  Pine64 doesn't recognize Wifi/BT Module clobeca 3 6,677 07-13-2017, 08:58 PM
Last Post: MarkHaysHarris777
Sad wifi direct bj_jung 2 5,886 12-29-2016, 10:26 AM
Last Post: happytuna

Forum Jump:


Users browsing this thread: 2 Guest(s)