!UNISOC Tiger 117
---
agk's phlog
6 Aug 2021 @ 0255
---
written on Pinebook Pro in kitchen
while stressed about school, work, and family
---
I bought a candybar-style 4G feature phone with
VoLTE from Nokia to prepare for the end of 2G GSM
cellular in my country. I don't love it. It's
just okay. It only cost $50.
4G featurephone tech
---------------------
My new Nokia 225 4G is representative of VoLTE-
capable featurephones Nokia, Maxwest, Itel, AGM,
CAT, and other brands started selling this year.
Non-KaiOS featurephones are almost all UNISOC
system-on-chip T117 devices running Mocor RTOS
(real-time operating system).
Tiger 107 and 117 4G LTE featurephone chips were
announced at India Mobile Congress in Oct 2019.
The 32-bit Arm Cortex-A7 chips run at 1.0 GHz.
They support dual SIM, Bluetooth 5, FM radio/MP3
listening, SD storage, and a bad camera. No 802.11
radio: no wifi calls, data, nor hotspot.
The chips have different LPDDR2 single-channel RAM
capacity and camera resolutions. T117 supports up
to 512 MB RAM and 2 megapixel photos. T107 supports
up to 384 MB RAM and 0.3 MP.
Non-T117 4G featurephones cost more. Punkt MP02 is
a Qualcomm Snapdragon 210, Sunbeam F1 a MediaTek
MT6739---both low-end smartphone chips. KaiOS phones
are Snapdragon 205 or 210s. Mudita Pure is a Quectel
LTE EG25-G, an IoT chip running a FreeRTOS fork.
Nokia 225 4G realities
----------------------
My old Nokia has 1/20th the processor clock speed
and 1/128th the RAM of the new one. New one uses
vast power for VoLTE and Bluetooth. It has no other
capability over phones of old Nokia's specs.
Old Nokia is 4 x 10 (40 cm2). 4G candybars are 5 x
12 (60 cm2) or bigger. Old Nokia has 4 cm diagonal
display; new ones are at least 6.
New phone copies android industrial design: Flat,
glossy black, big, covered with fingerprints. My
thumb can't find a key on the flat keypad by touch.
The curved matte black back features branding and
chrome-outlined lens. Old Nokia fits in my hand. I
use its textured keys without looking.
New display is colorful. Icons open dumb games and
send URLs to Opera Mini (Facebook! Oxford English
Dictionary!). Can't delete or hide icons. Can't use
Opera Mini without 4G data plan. Maybe UNISOC will
learn digital detoxers are a market. Why not offer
monochrome display option? Why not allow hiding
garbage icons in a folder?
That's all, really. I assigned key shortcuts to
calculator and SMS messenger, moved calendar and
alarm clock icons to handy spots, disabled the
camera, and the phone's fine. Better for my needs
than a smartphone, KaiOS flip phone, or big, heavy,
or costly feature phone. It'll do.
Dumbphone dreams
----------------
People hack smartphones: Apps (termux, linphone,
conversations, launchers), root and ADB access,
firmware and ROMs: LineageOS, GrapheneOS, CalyxOS,
Linux (PostmarketOS/SXMO, Phosh on Mobian/Manjaro
ARM). Even hardware (Pinephone). What could
featurephones be with love from mobile hackers?
Look what Ben Wilson and Aamir Niaz are doing with
the ESP32-based WiPhone: Mod the open firmware and
OS with Arduino sketches and Python. Call on wifi
via SIP. Text via SIP or LoRa, or write an XMPP
client. $90, modular, extendable. The UI is elegant.
A 4G LTE expansion board is planned. For hackers,
not a consumer device, but a good step. I want to
see Mudita's RTOS ported to it.
Albert Gajsak's Chatter ESP32 LoRa communicator
kit is for STEM education. Build two for $100.
Cardenas et al made a bot to relay Telegram mess-
ages through a gateway to LoRa. Pine64's PineDio
is such a gateway. Bunnie Huang's Precursor pocket
FPGA dev kit might produce good things. Justine
Haupt is making a 4G rotary cellphone kit.
Diverse small, cheap, tough, low-spec, non-distract-
ing, no-telemetry, ergonomic, hackable phones are
possible. I'm not the only one to remember and dream
about them. What would good phones do with low specs,
low cost, minimal sensors, productive work supported,
and screentime minimized?
-----------------------
Cardenas AM et al. (2019). A Low-Cost and Low-Power
Messaging System Based on the LoRa Wireless Techno-
logy. Mobile Networks and Applications.
doi:10.1007/s11036-019-01235-5
