Admixture Calculator vs Local Ancestry Inference , two different methods to dna analyses

[Solitude]

A person posted in reddit a interesting topic about this method analysis called local ancestry inference , is really too better than admixture , basically it split each chromossome in 50 pieces for example and analyze piece by piece , one by one , this method is better than admixture because analysis is done individually , i think the most updated method called LAI is being used by 23andme , Ancestrydna and DNAGenics , another companies like ftdna or genomelink i dont know if it still using outdated versions of LAI , problem with dnagenics is their database , their reference panels need still a lot of improvements , a lot of people for example from iberian sometimes get Italy , Balkan or French instead only Iberia , this is due their reference panels , admixture calc to people more outliers is not good , sometimes a iberian can get a slavic , germanic , balkan samples with no reason with LAI analysing piece by piece of their chromossomes this not happen , another note is speculative mode in shared modern origins is not too good just ignore it , sections more broadly like intermediate and speculative a lit bit of asian appear aswell

topic about this method : https://www.reddit.com/r/MyHeritage/comments/163kab7/explaining_the_difference_between_the_companys/

a good example would be my intermediate result North Iberia and West Iberia is good but should be labeled as Portugal instead Spain in West Iberia , NW Europe is not good it should be Netherlands instead norway and the percentage is low in this intermediate result , Amerindian in intermediate appear Central America.

https://i.postimg.cc/jjR8B4SY/dnagenics.png