OT AI

[Jindrich Fucik]

Jako hezký způsob testování AI mi přijde tato neověřená legenda na 
vláčkovém fóru:
http://diskuze.modely.biz/viewtopic.php?f=28&t=1825&start=1230#p332566

Dne 18.02.2024 v 10:02 Martin Záruba napsal(a):
> Vy na něj jdete vědecky.... Já zkusil jinou "pitomost":  Přelož do 
> angličtiny a zpátky větu:
> 
> Fialová srnka pije kafe z hrnka.
> 
> Výsledkem bylo:
> 
> Purpurová laň pije kávu z kelímku.
> 
> Na první pohled je to hrůza, ale když se nad tím zamyslíte, dodržel 
> přesně stejný smysl jen jinými slovy. Je to tak hrozné?
> 
> Martin Záruba
> 
> Dne 17.2.2024 v 19:09 Jan Waclawek napsal(a):
>> Toto je OT, lebo z AI z principu nemoze vypadnut nic rozumne. Ale zabava s
>> tym byt moze.
>>
>> Takze som sa hral s gemini od gugla a dal som mu tu otazku co sme tu
>> preberali o tom trojuholniku co ma len parne (sude) harmonicke.
>>
>> V podstate zacal spravne:
>>
>> Generating a triangular waveform with only even harmonics is not directly
>> possible in the traditional sense. A pure triangular waveform, by
>> definition, contains both odd and even harmonics.
>>
>> Kebyze tu prestane, normalne by som povedal, fajn, odpoved ako z prvej
>> stranky co gugl najde, vsetko sedi. Ale nie, on blbec to ma nutkanie to
>> rozviest.
>>
>> However, there are different approaches to achieve a waveform that
>> resembles a triangular wave but only consists of even harmonics:
>>
>> Fajn, ze by to teda mohlo pripominat trojuholnik, no preco nie. Skusme si
>> to vypocut.
>>
>> 1. Modified Square Wave:
>>
>>      Generate a square wave with 50% duty cycle (equal on and off times).
>>      Filter out all odd harmonics using a bandpass filter tuned to pass only
>> even multiples of the fundamental frequency.
>>      The resulting waveform will appear close to a triangular wave, but not
>> perfectly symmetrical due to limitations of perfect filtering.
>>
>> Cize to mame tvrdy vyhadzov z elementarnej matematiky. Podme dalej, mozno
>> sa nieco zaujimave este dozvieme
>>
>> 2. Sum of Even Harmonics:
>>
>>      Calculate the amplitude of each even harmonic using its inverse
>> harmonic number (1/2, 1/4, 1/6, etc.).
>>      Sum the individual even harmonics at their respective frequencies.
>>      This approach requires more complex processing but offers potentially
>> better control over the shape of the resulting waveform.
>>
>> https://www.desmos.com/calculator/g5ixmigqfx  nie, toto trojuholnik
>> nepripomina ani zdaleka (znova, je tam verzia zahrnujuca fundamentalnu a
>> aj bez nej)
>>
>> 3. Digital Synthesis:
>>
>>      Use a digital synthesis technique like DDS (Direct Digital Synthesizer)
>> to generate a lookup table containing discrete samples of a desired
>> waveform shape.
>>
>> atd. ani ciarka o tom, ako by sa dalo dosiahnut, aby tam tie neparne
>> (liche) harmonicke neboli.
>>
>> 4. Analog Synthesis:
>>
>>      Use analog circuit components like integrators and summers to create a
>> circuit that approximates the integration of a square wave, resulting in a
>> triangle-like waveform.
>>
>> Tiez sa tu nijako neriesia tie parne harmonicke, na to uz tipek davno
>> zabudol; a najma to  nie je analogova *synteza*, len uprava signalu -
>> syntezu nam tu predviedol pan kolega Lukesh ako fakt asi jedinu rozumnu
>> metoda ako dosiahnut "trojuholniku podobny priebeh", proste generovat
>> sinusy tych harmonickych a scitavat ich pomocou hejblatok dovtedy, kym
>> vysledok okometricky nebude trocha trojuholnikovity.
>>
>> No nic, zazraky sa nekonaju.
>>
>> wek
>>
>> _______________________________________________
>> HW-list mailing list  -  sponsored bywww.HW.cz
>> Hw-list na list.hw.cz
>> http://list.hw.cz/mailman/listinfo/hw-list
> 
> _______________________________________________
> HW-list mailing list  -  sponsored by www.HW.cz
> Hw-list na list.hw.cz
> http://list.hw.cz/mailman/listinfo/hw-list