Dear Andreas, I agree with you that the ADSA tool documentation statement is misleading. I think it should be rewritten to more accurately state what the ICW data does reflect. Sincerely, Tim -----Original Message----- From: Andreas West [mailto:ahnen@awest.de] Sent: Sunday, December 13, 2015 1:39 AM To: Tim Janzen; genealogy-dna@rootsweb.com Subject: Re: [DNA] Are FTDNA's 1cM matches shown in ICW as well? Thanks Tim and yes, I have two phased sets for each of my parents at GEDmatch. Let us all be thankful for them providing the tools that all those multi-million (or even billion) companies are not giving us. Also good to hear that you agree with my points. Do I understand your "disagree" with the ADSA tool documentation (the statement I posted) as that you agree with me that it's misleading and can't be done with FTDNA data? Andreas

I think we should all be glad for the work that the guys at Gedmatch have done to allow us to run triangulations against such a large number of matches. If triangulation would be as easy as some mistakenly understand it (those that believe ICW is triangulation) then more websites would support us in doing so and I'm pretty sure FTDNA would provide us with a proper triangulation tool (again, it would be nice if they would be more clear that their ICW tool isn't triangulation but hey, it's all about marketing and keeping the impression you do). It's by no means a small feature and requires an enormous amount of disc space and computing power to run all possible combinations. Just to illustrate, with 1,000,000 customer (both 23andme and Ancestry have more now) you have 499,999,500,000 possible combinations to check for. Given you have to check for atDNA and X-DNA (two different functions/procedures most likely given they are based on different minimum matching criteria) that means those two companies would have to run currently more than 1 quadrillion (had to look that name up) of checks if they would automate triangulation (if they start from scratch, obviously they have done a lot of triangulations already). Hence we're only provided with a list of matches and it's up to us to find out which ones triangulate and which not. For matching they probably have some optimization along the way but still we're talking about comparing 800k+ SNP's per customer (again with optimization steps in between like described in Ancestry's algorithm) to identify who's matching who and leave it then to the customer to decide when to do triangulation of up to 5 people vs one other person (in the case of 23andme). I hope that people now understand what a huge task that is and why only very few are undertaking the task of identifying matches (with Gedmatch being the only company run by just a couple of people), let alone run triangulations for you. Andreas > On 14 Dec 2015, at 03:36, Tim Janzen via <genealogy-dna@rootsweb.com> wrote: > > Dear Andreas, > I agree with you that the ADSA tool documentation statement is misleading. I think it should be rewritten to more accurately state what the ICW data does reflect. > Sincerely, > Tim > > -----Original Message----- > From: Andreas West [mailto:ahnen@awest.de] > Sent: Sunday, December 13, 2015 1:39 AM > To: Tim Janzen; genealogy-dna@rootsweb.com > Subject: Re: [DNA] Are FTDNA's 1cM matches shown in ICW as well? > > Thanks Tim and yes, I have two phased sets for each of my parents at GEDmatch. > > Let us all be thankful for them providing the tools that all those multi-million (or even billion) companies are not giving us. > > Also good to hear that you agree with my points. > > Do I understand your "disagree" with the ADSA tool documentation (the statement I posted) as that you agree with me that it's misleading and can't be done with FTDNA data? > > Andreas > > > > ------------------------------- > To unsubscribe from the list, please send an email to GENEALOGY-DNA-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message

23andMe (and GEDmatch) let us compare two Matches to complete the Triangulation process. FTDNA could offer a feature to just confirm which ICW Matches (which the already list) are on the same segment. This would require some additional computing, but not so much as you calculate. Jim - www.segmentology.org > On Dec 14, 2015, at 12:59 AM, Andreas West via <genealogy-dna@rootsweb.com> wrote: > > I think we should all be glad for the work that the guys at Gedmatch have done to allow us to run triangulations against such a large number of matches. > > If triangulation would be as easy as some mistakenly understand it (those that believe ICW is triangulation) then more websites would support us in doing so and I'm pretty sure FTDNA would provide us with a proper triangulation tool (again, it would be nice if they would be more clear that their ICW tool isn't triangulation but hey, it's all about marketing and keeping the impression you do). > > It's by no means a small feature and requires an enormous amount of disc space and computing power to run all possible combinations. > > Just to illustrate, with 1,000,000 customer (both 23andme and Ancestry have more now) you have > > 499,999,500,000 > > possible combinations to check for. > > Given you have to check for atDNA and X-DNA (two different functions/procedures most likely given they are based on different minimum matching criteria) that means those two companies would have to run currently more than 1 quadrillion (had to look that name up) of checks if they would automate triangulation (if they start from scratch, obviously they have done a lot of triangulations already). Hence we're only provided with a list of matches and it's up to us to find out which ones triangulate and which not. > > For matching they probably have some optimization along the way but still we're talking about comparing 800k+ SNP's per customer (again with optimization steps in between like described in Ancestry's algorithm) to identify who's matching who and leave it then to the customer to decide when to do triangulation of up to 5 people vs one other person (in the case of 23andme). > > I hope that people now understand what a huge task that is and why only very few are undertaking the task of identifying matches (with Gedmatch being the only company run by just a couple of people), let alone run triangulations for you. > > Andreas > >> On 14 Dec 2015, at 03:36, Tim Janzen via <genealogy-dna@rootsweb.com> wrote: >> >> Dear Andreas, >> I agree with you that the ADSA tool documentation statement is misleading. I think it should be rewritten to more accurately state what the ICW data does reflect. >> Sincerely, >> Tim >> >> -----Original Message----- >> From: Andreas West [mailto:ahnen@awest.de] >> Sent: Sunday, December 13, 2015 1:39 AM >> To: Tim Janzen; genealogy-dna@rootsweb.com >> Subject: Re: [DNA] Are FTDNA's 1cM matches shown in ICW as well? >> >> Thanks Tim and yes, I have two phased sets for each of my parents at GEDmatch. >> >> Let us all be thankful for them providing the tools that all those multi-million (or even billion) companies are not giving us. >> >> Also good to hear that you agree with my points. >> >> Do I understand your "disagree" with the ADSA tool documentation (the statement I posted) as that you agree with me that it's misleading and can't be done with FTDNA data? >> >> Andreas >> >> >> >> ------------------------------- >> To unsubscribe from the list, please send an email to GENEALOGY-DNA-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message > > > ------------------------------- > To unsubscribe from the list, please send an email to GENEALOGY-DNA-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message