I can post your message on the ISOGG Facebook page. The programmer can probably add a graceful way to stop at 30 and retain what has been analyzed to that point. However, it may not be practical to triangulate such large groups, and the programmer might want to throttle the number to an even lower number. The number of pairwise comparisons required grows exponentially, and the FTDNA servers would get hammered fetching all that data. Ann Turner On Wed, Oct 25, 2017 at 10:02 AM, Marleen Van Horne <[email protected]> wrote: > As a retired systems analyst who specialized in designing, building and > analyzing the results of test systems for computer software, I think I have > found a defect in the FTDNA Triangulator. > > Since I have 2895 Family Finder matches, I chose not to use the chromosome > browser for triangulation, instead, I chose to use the IN COMMON WITH tool > and run the Triangulation tool on the entire ICW group. This works without > any problems, if your ICW group has 30 members or less. This is the number > of matches displayed on the FTDNA page. > > If, however, your ICW group has more than 30 sets of test results, the > Triangulation tool stops working, it hangs up, you cannot close the tool, > the only way to gain control of your screen is to back out of the page > completely. You can then go back into Family Finder, but you have no > results for the triangulation run. > > The Triangulation tool knows how many records have been selected, but > under "Matches to triangulate" the most it lists is 31. It stops with the > 30th record, even though you have 80 records in the ICW group. There is no > way to triangulate part of an ICW group, so that you can copy the parts to > Excel and sort them. > > This is not a lack of understanding on how to use the tool, the tool just > stops running when you get to the31st record. I do not use Facebook, so I > do not know where to report the problem. > > Marleen Van Horne > > > ------------------------------- > To unsubscribe from the list, please send an email to > [email protected] with the word 'unsubscribe' without > the quotes in the subject and the body of the message >

Ann, Thank you. Yes, I realize it is heavy useage, but that is what computers are for, and it sure beats triangulating 2895 matches, five at a time. Even if it was designed to triangulate an ICW group one page at a time, the results could be merged in an Excel spreadsheet. It would be better for the user if they could all be done at once. Thanks again, Marleen Van Horne

Here is the response from Goran Runstrom: "Hi Ann & Marleen! I'll be brief here, as my current Internet connection is not the best. My email address is in the About tab of the triangulator, in case Marleen wants to email me directly. I will have a look at the problem with the tool not working when doing ICW without selecting any matches. My intention was to limit the number of triangulated matches to 31 (The main person plus the limit of 30 matches on a page). Perhaps it somehow allows more users to be triangulated, which is unintentional and not tested properly. The number of FTDNA server requests per number of triangulated matches is not square but n(n-1)/2, so 31 users would mean 465 requests. In order to not hammer the FTDNA servers all request parallelism is disabled and there is a 1 second delay between requests, so a triangulation run of 31 users should take around 10 minutes. You can triangulate a set of more than 5 matches by running the triangulator tool once, which will unlock the FTDNA selection limit of 5 users. You can then close the tool window and select up to 30 matches and run the triangulator again. Will post an update when I've had the time to do some testing myself." The reference to "square" is in response to a comment in my post. What I meant to say was that if a run with 30 kits takes 10 minutes, then a run with 60 kits would take 40 minutes (not 20 minutes). Ann Turner On Wed, Oct 25, 2017 at 11:28 AM, Ann Turner <[email protected]> wrote: > I can post your message on the ISOGG Facebook page. The programmer can > probably add a graceful way to stop at 30 and retain what has been analyzed > to that point. However, it may not be practical to triangulate such large > groups, and the programmer might want to throttle the number to an even > lower number. The number of pairwise comparisons required grows > exponentially, and the FTDNA servers would get hammered fetching all that > data. > > Ann Turner > > On Wed, Oct 25, 2017 at 10:02 AM, Marleen Van Horne <[email protected]> > wrote: > >> As a retired systems analyst who specialized in designing, building and >> analyzing the results of test systems for computer software, I think I have >> found a defect in the FTDNA Triangulator. >> >> Since I have 2895 Family Finder matches, I chose not to use the >> chromosome browser for triangulation, instead, I chose to use the IN COMMON >> WITH tool and run the Triangulation tool on the entire ICW group. This >> works without any problems, if your ICW group has 30 members or less. This >> is the number of matches displayed on the FTDNA page. >> >> If, however, your ICW group has more than 30 sets of test results, the >> Triangulation tool stops working, it hangs up, you cannot close the tool, >> the only way to gain control of your screen is to back out of the page >> completely. You can then go back into Family Finder, but you have no >> results for the triangulation run. >> >> The Triangulation tool knows how many records have been selected, but >> under "Matches to triangulate" the most it lists is 31. It stops with the >> 30th record, even though you have 80 records in the ICW group. There is no >> way to triangulate part of an ICW group, so that you can copy the parts to >> Excel and sort them. >> >> This is not a lack of understanding on how to use the tool, the tool just >> stops running when you get to the31st record. I do not use Facebook, so I >> do not know where to report the problem. >> >> Marleen Van Horne >> >> >> ------------------------------- >> To unsubscribe from the list, please send an email to >> [email protected] with the word 'unsubscribe' without >> the quotes in the subject and the body of the message >> > >