Opens profile photo
Follow
Click to Follow FritzchensFritz
Fritzchens Fritz
@FritzchensFritz
Watch neat Infrared photos or siliconpr0n on Flickr: flic.kr/ps/2ZnBdY
BerlinJoined January 2017

Fritzchens Fritz’s posts

NexGen Nx586 Processor Die-Size: 10,971mm x 11,219mm = 123,084mm² w/o scribelines 10,724mm x 10,963mm = 117,567mm² The NexGen processor was first manufactured in IBM's 0.5µm CMOS process node. This chip here was already manufactured in 0,44µm (440nm).
Image
Image
Image
Image
6
288
BE-S1000 from Baikal Electronics, 48 Core Arm Cortex-A75 CPU for HPC CPU manufactured at TSMC with 16nm. For the next few years, probably one of the last Russian CPUs with modern process node.
Image
Image
Image
Image
11
271
Well, thermal imaging to see bouncing cores escalated quickly. Ryzen 5 7600 IOD broke into 5 pieces, not sure why but temperatures and power consumption was okay. After 60 seconds of cinebench single core test, image artifacts appeared and then there was a crash. 🧵1/5
Image
Image
Image
16
230
I did not expect that the structures of N5 on Zen4 would be so barely visible (low contrast). pic number: 1. Camera with macro lens 2. Microscope 5x lens with external light 3. Microscope 5x epi illumination (brightfield DIC) 4. Microscope 20x epi illumination (brightfield DIC)
Image
Image
Image
Image
5
201
Rocketlake-S (i5-11400), preparation for some thermalcam tests. Copper spacer only for boot and windows start. TIM cleaning with HCl, liquid rubber doesn't protect the caps very well, anyway the CPU is still running without all these tiny caps :-)
Image
Image
Image
Image
9
93
twitter.com/FritzchensFrit Luckily, sent another Centaur CPU. Again problems with etching but then everything went well. Here are some pics. flic.kr/s/aHBqjzXMjT
Image
Image
Image
Image
Quote
Image
Image
Image
Image
There will be no more pictures of the Via Centaur Something went wrong and the chip lost its colors. The chip looked quite good in the meantime and could even have been put under the microscope but my impatience and the urge for perfection drove me to etch the chip again.. 1/2
1
91
There will be no more pictures of the Via Centaur Something went wrong and the chip lost its colors. The chip looked quite good in the meantime and could even have been put under the microscope but my impatience and the urge for perfection drove me to etch the chip again.. 1/2
Image
Image
Image
Image
6
83
Generated with DALL·E some shiny silicon wafers. Partly very strange pictures come out, these are the better ones - Hold on hand a 300mm silicon wafer - A person took a bite out of a 300mm silicon wafer - Closeup of a shiny and colorful silicon wafer with rectangle structures
Hold on hand a 300mm silicon wafer
A person took a bite out of a 300mm silicon wafer
Close up of a shiny and colorful silicon wafer with rectangle structures
2
57
Chips & Pizza? A lot of broken graphic cards has accumulated over time. So, today was desoldering day and in the basement is still a box with dozens of older cards.😱
Image
Image
Image
1
57
That Golden Finger Device works pretty well on the Athlon K7 500 (Argon core with 250µm, binned at 650, sold as 500). 900MHz is semi-stable and bootable into DOS@1.95V. More voltage results in more instability... The resistors on the device get quite warm🧐 #AMD #Retro #Athlon
Image
Image
Image
Image
I am completely surprised by sodium hydrogencarbonate (NaHCO3). Have never used it to etch microchips but together with H2O2 this gives me amazing easy etching results. GP100 was etched with it and Athlon64 with SOI process works also well.🙀
Image
Image
Image
3
40
Thanks to for his effort of die-shot annotations.
Quote
Another great project in regard to Russian microprocessor technology. This time the 28nm Elbrus 8S from MCST, which uses a VLIW ISA and x86 binary translation The chip was again provided by @RiddleRiderOne and @FritzchensFritz made the beautiful die shots: flickr.com/photos/1305612
Image
Image
Image
Image
1
40
Received today a turntable from Amazon. China product, power supply exploded immediately when plugged in and fall apart. I still feel the explosion in my hand but fortunately have not been injured. That has been a lesson to me, don't trust cheap products anymore.
Image
Image
5
40
Hello little Argon 500, would you like to reach 1Ghz? Not sure at all how much such an argon can achieve. I am very curious to find out. Hope a cloned golden finger device with a water block helps. Vcore goes up to 2.05V
Image
Image
3
38
Replying to
Cracked chip due to high temperature caused by failed overtemp protection or short circuit due to overvoltage caused by driver problems? A long time ago, didn't you show also a chip like this? Have found something like this only once on the PCB of a defective RX6800 from Ebay.
Image
1
36
Replying to
The Nx686 is the successor and was demonstrated in October 1995, but never reached the market. AMD bought NexGen and after a redesign the Nx686 appeared as AMDs K6. It would be nice to see what the Nx686 prototype looks like from the inside. We will probably never know.
Image
1
34