Hacker News Reader

BarsMonsteron Nyquest NY8A051H – 1.5 cent microcontroller: weekend die-shot

No, it's field-programmable. But you can only do it once.

BarsMonsteron Nyquest NY8A051H – 1.5 cent microcontroller: weekend die-shot

Hehe, I was also tricked into thinking it's an 8051-compatible :-)

BarsMonsteron Programming Intel 87C51 – first high-volume integrated microcontroller (1980)

Today we are used to luxury of fully integrated microcontrollers - all key components are conveniently integrated into single reliable part: non-volatile memory, SRAM, CPU core, PLL, ADC/DAC, PWM, serial ports, e.t.c It was not like that in the past and embedded systems typically required lots of chips, until Intel 8048 (MCS-48) was released in 1976 on n-MOS technology. Intel expected that 8048 will have limited product lifetime, and in 4 years, in 1980 it was replaced by 8051 (MCS-51) which conquered the world. It was first high-volume product to integrate 4KiB of PROM, 128 bytes of SRAM, GPIO, serial port as well as 8-bit core in a single crystal. 87C51FC variant was using 32KiB EPROM non-volatile memory instead of PROM's, double SRAM size (256 byte), C-version was manufactured on CMOS process - which makes it exceptionally modern for the time. It was not particularly fast - simplest commands took 12 clock cycles to execute, so even at 20Mhz it was doing just over 1 million operations per second, also - no 16-bit division commands. Modern 8051-compatible cores are much faster and often do single-cycle command execution.

Recently I got my hands on D87C51FC-20, and decided to experience the old ways of embedded software.

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

USB hub chip, USB<>SATA adapter - eating power for sure... SoC voltage scaling - need to be checked.

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

Page-up/down keys... There are likely other ways )

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

I was originally from Belarus, but as you mention - I am indeed in Zürich for some years.

Still, I can assure you, 99% of people in Belarus do work on x86, and cutting edge x86 hardware is being sold freely. Technological sanctions failed many years ago and at some point I will need to write an article about that...

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

It's just a step. We are not there yet. But we have to continue one step at a time.

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

Nearly all Linux packages you might want to install - work with nothing special required. Migration to ARM earlier this century made vast majority of open source software portable. Precompiled x86-only surely will not work.

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

USB 3.0 can eat alot of power due to high speed, if it's always active...

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

Do you imply that chip is fake and does not work? :-)

Do you imply that there are no fake products on Amazon, which is also making chips (AWS Graviton)? Exactly the same issue: https://www.youtube.com/watch?v=y83BS_mK9GE

There is a reason why Aliexpress and Amazon suffer form the same curse.

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

Not (yet)

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

Damn, best products are always in the future :-) Oasis looks very capable and juicy, looking forward to get my hands on it.

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

My tests show it's slightly slower than Pi 4 and significantly faster than Pi 3 on CPU performance.

1T Coremark: Pi 3: 3504 Lichee: 6900 Pi 4: 7938

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

In kernel config I see: CONFIG_EFI_STUB=y CONFIG_EFI=y

If that's what identifies it

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

https://mirror.iscas.ac.cn

BarsMonsteron Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

First tests of RISC-V laptop reveal a number of software and hardware issues, but show a path to freedom.

BarsMonsteron Casio-F-91W die-shot

Not... yet :-)

BarsMonsteron Casio-F-91W die-shot

I tried it at the beginning, but found it quite complicated. A the moment Hugin is used, which works nearly perfectly, especially on XY stage where you can disable frame rotation optimization.

BarsMonsteron Casio-F-91W die-shot

Power consumption and cost. 10 years on a battery is very hard to achieve for microcontroller.

https://sorwatch.net shows that : with much more advanced microcontroller (much more modern tech.node, low power features) - it only works for 1-2 years on a single battery after extensive power optimizations.

In custom silicon when seconds tick - it's maybe 100 transistors change their state. In microcontroller - it's roughly 10'000-100'000 transistors changing state to handle interrupt and update display content...

BarsMonsteron Casio-F-91W die-shot

I've seen some made in Thailand. I would expect chip itself & quartz oscillator made in Japan.

BarsMonsteron Casio-F-91W die-shot

Yep, this is my understanding.

Also, world practice shows that noone ever got into troubles for posting chip pictures, and in 10 years noone ever contacted me to remove any photos.

But companies do get intro troubles for trying to do copycat chip production, although it is not always easy to prosecute.

BarsMonsteron Casio-F-91W die-shot

Some time ago I posted exactly about that : https://zeptobars.com/en/read/how-to-open-microchip-asic-wha...

My latest microscope is this one, but Casio photo was made by temporary microscope I assembled from Thorlabs & Olympus parts: https://3.14.by/en/read/olympus-bx60-fluorescence-metallurgi...

BarsMonsteron Disc rot is a challenge for both archivists and collectors (2017)

I am the author of these photos... If you have a sample of rotten disk - you can send it to me and I'll take a look. Contacts are on the site.

Now my hardware is better, CD and DVD are easy to observe without physical damage. I also can see data on BD disk, but it is on the border of what is possible.

I also tried to decode data from the image, but it will require more collaboration.

BarsMonsteron How to open a microchip and see what’s inside

They are indeed enhanced, but about 2x, not 10x. At the time of the writing I was increasing saturation on all photos a bit in postprocessing. Now I am typically making photographs with RGB illumination which increases saturation optically (without noise), so only contrast/brightness range needs to be tuned.

Colors are mainly due to light interference caused by different glass thickness over the chip area.

Source: I am the author.

BarsMonsteron Ti NE555 – real vs fake: weekend die-shot

Wow, nice memory! Even I missed that. That means HA17555 is most likely also fake.

BarsMonsteron Ti NE555 – real vs fake: weekend die-shot

These fakes chips were labeled as Ti on the plastic package... They tried to make it look like real deal...

BarsMonsteron Soviet PCs

As far as I understand, common practice was lapping. Yield was always acceptable.

It is still being done now, for open-source hardware too, like here : http://zeptobars.ru/en/read/KR580VM80A-intel-i8080-verilog-r...

BarsMonsteron Soviet PCs

They were much less behind than it is now (1-2-5 years max). The problem is that leadership of USSR wanted to take advantage of using existing software - hence demanded 100% exact copy of hardware (starting from IBM-360). By many these actions were considered a treason, especially in retrospective.

There were independent developments, they worked but never went widespread (because they needed an "order" from the state).

BarsMonsteron Mikron 1663RU1 – first Russian 90nm chip: weekend die-shot

90nm is probably too good for microcontrollers :-) STMicroelecronics make most of (cheaper) microcontrollers on 180nm, AVR's are closer to 500nm... There are lots of Russian microcontrollers already available (based on Cortex-M3, M0, M4 cores), but at the moment they are manufactured overseas (X-Fab is favorite fab in Russia). Domestically-manufactured microcontrollers are much simpler (AVR or old PIC-level, there is even Russian AVR clone, which is not very widely used).

BarsMonsteron Mikron 1663RU1 – first Russian 90nm chip: weekend die-shot

There is some controversy here. In Russia all domestic chips authorized to be used in military equipment must be packaged in ceramic case. But the reasons for this lost in time when plastic packages were not good enough. These days it seems it is just a waste of money. In reality plastic packages offer some benefits for military applications over ceramic ones (like better G-shock/vibration resistance).