Using Integer Addition to Approximate Float Multiplication - Once the domain of esoteric scientific and business computing, floating point calc... - https://hackaday.com/2025/04/10/using-integer-addition-to-approximate-float-multiplication/ #softwaredevelopment #softwarehacks #floatingpoint #fpu #gpu

The Pentium Processor’s Innovative (and Complicated) Method of Multiplying by Three, Fast - [Ken Shirriff] has been sharing a really low-level look at Intel’s Pentium (1993) ... - https://hackaday.com/2025/03/08/the-pentium-processors-innovative-and-complicated-method-of-multiplying-by-three-fast/ #reverseengineering #retrocomputing #pentium #radix-8 #fpu

The Pentium Processor’s Innovative (and Complicated) Method of Multiplying by Three, Fast https://hackaday.com/2025/03/08/the-pentium-processors-innovative-and-complicated-method-of-multiplying-by-three-fast/ #ReverseEngineering #Retrocomputing #pentium #radix-8 #FPU

Intel 當年的 Pentium FDIV bug 居然還有新的發現...

三十年前的 Pentium FDIV bug 居然還有新的發現冒出來：「Intel's $475 million error: the silicon behind the Pentium division bug」。

先快速複習

https://blog.gslin.org/archives/2025/01/03/12177/intel-%e7%95%b6%e5%b9%b4%e7%9a%84-pentium-fdiv-bug-%e5%b1%85%e7%84%b6%e9%82%84%e6%9c%89%e6%96%b0%e7%9a%84%e7%99%bc%e7%8f%be/

Irregular reminder: floating point calculations are not exact.

Take your program to a different #hardware and your tests may start failing:

https://codeberg.org/libobscura/libobscura/src/branch/master/crates/conv/tests/convtest.rs#L40

https://codeberg.org/libobscura/libobscura/actions/runs/30#jobstep-7-431

#CIJI:
#FPU ruft zur Einreichung von Anträgen für die #Finanzierung von grenzüberschreitenden, gemeinschaftlichen #Recherchen auf. 10 #Zuschüsse i.H.v. je 10.000 EUR werden an #Journalisten-Teams vergeben
#Bewerbungen bis 9. September 2024 möglich

Infos hier
https://www.ciji.info/announcements/?pk_campaign=mailjet&pk_kwd=Monthly%20newsletter%20%23July&pk_source=newsletter#cross-border-grants-2024

AMD Zen 5 Execution Engine Leaked, Features True 512-bit FPU

Giving "Zen 5" a 512-bit FPU meant that AMD also had to scale up the ancillaries [..]. The L1 Data cache has been doubled in bandwidth, and increased in size by 50%. The L1D is now 48 KB in size [..]. FPU MADD latency has been reduced by 1 cycle. Besides the FPU, AMD also increased the number of Integer execution pipes to 10, from 8 on "Zen 4."

https://www.techpowerup.com/321201/amd-zen-5-execution-engine-leaked-features-true-512-bit-fpu

Ich versuche meine Artikel immer etwas "anzureichern", so dass auch Fortgeschrittene etwas davon haben.
In diesem "Cheatsheet"-Artikel über #Winkel und #Steigung gibt es am Ende auch ein kleines Demo wie man #Assembler-Code und #FPU-Befehle in C-Programme einbettet.

https://dd3ah.de/winkel-und-steigung/

How about we make this #FloatingPointFriday? Here's a gem I posted in another thread and that I regularly use to scare my students and to demonstrate to them the dangers of #floatingPoint. Every #IEEE 754 compliant #FPU will give the same answer. It allows me to talk about the fact that radix representations of a number may be finite in one base (10) and infinite in another (2). Then I throw in the fact that floating point numbers have #finitePrecision and that #machineEpsilon is a thing.

Home-Built CPU Runs With Home-Built Toolchain - A few years ago [Takaya Saeki] and fellow students of the University of Tokyo, wer... - https://hackaday.com/2022/11/22/home-built-cpu-runs-with-home-built-toolchain/ #computerhacks #custom #risc #cpu #fpu #isa