It doesn't appear that all CPUs participating in the project are listed at https://boinc.multi-pool.info/latinsquares/cpu_list.php

Is this list not automatically generated? Or are some processors not recognized due to some obscure error (e.g. there are no AMD Threadripper processors on the list)

That statistic is generate when you click into the page by Boinc software looking inside the host table (it has all statistics about a PC attached to the project).

If I manually search for it I get:

mysql> select count(*), p_model from host group by p_model having p_model like '%Threadripper%';
+----------+--------------------------------------------------------------------------------+
| count(*) | p_model                                                                        |
+----------+--------------------------------------------------------------------------------+
|        2 | AMD Ryzen Threadripper 1900X 8-Core Processor [Family 23 Model 1 Stepping 1]   |
|        9 | AMD Ryzen Threadripper 1920X 12-Core Processor [Family 23 Model 1 Stepping 1]  |
|        1 | AMD Ryzen Threadripper 1950X 16-Core Processor  [Family 23 Model 1 Stepping 1] |
|      110 | AMD Ryzen Threadripper 1950X 16-Core Processor [Family 23 Model 1 Stepping 1]  |
|        2 | AMD Ryzen Threadripper 2920X 12-Core Processor [Family 23 Model 8 Stepping 2]  |
|        5 | AMD Ryzen Threadripper 2950X 16-Core Processor [Family 23 Model 8 Stepping 2]  |
|        2 | AMD Ryzen Threadripper 2970WX 24-Core Processor [Family 23 Model 8 Stepping 2] |
|       83 | AMD Ryzen Threadripper 2990WX 32-Core Processor [Family 23 Model 8 Stepping 2] |
+----------+--------------------------------------------------------------------------------+
8 rows in set (0,46 sec)

The Boinc code inside the php file, use some filter onto the results to limit them (as it was too slow in Seti):

function get_data() {
    $db = BoincDb::get(true);
    // get CPU model status in a special query;
    // enumerating hosts was too slow on SETI@home.
    //
    // Ideally a model's fpops should be the median over hosts of that model.
    // But SQL has no median function.
    // Instead, take the mean of plausible values
    //
    $x = $db->enum_fields('host', 'StdClass', 
        'p_model, count(*) as nhosts, avg(p_ncpus) as ncores, avg(p_fpops) as fpops',
        'p_fpops>1e6 and p_fpops<1e11 and p_fpops <> 1e9 and expavg_credit>'.MIN_CREDIT.' group by p_model',
        null
    );
    $m2 = array();
    foreach ($x as $m) {
        if ($m->nhosts < MIN_COUNT) continue;
        $y = new StdClass;
        $y->model = $m->p_model;
        $y->p_fpops = $m->fpops;
        $y->mean_ncores = $m->ncores;
        $y->nhosts = $m->nhosts;
        $m2[] = $y;
    }
    return $m2;
}

1) Takes only CPUs that has an expect average credits of min 10
2) Takes only CPU that has floating point capacity inside a given range
3) Takes only CPU that has at least 10 PC with it

Executing the above query (with point 1 and 2, but not 3), gives:

mysql> select p_model, count(*) as nhosts, avg(p_ncpus) as ncores, avg(p_fpops) as fpops from host where p_fpops>1000000 and p_fpops<10000000000 and p_fpops <> 100000000 and expavg_credit>'10' group by p_model having p_model like '%Threadripper%';
+--------------------------------------------------------------------------------+--------+---------+-------------------+
| p_model                                                                        | nhosts | ncores  | fpops             |
+--------------------------------------------------------------------------------+--------+---------+-------------------+
| AMD Ryzen Threadripper 1950X 16-Core Processor [Family 23 Model 1 Stepping 1]  |      8 | 31.6250 | 4536869382.474418 |
| AMD Ryzen Threadripper 2920X 12-Core Processor [Family 23 Model 8 Stepping 2]  |      1 | 24.0000 | 5031380453.020853 |
| AMD Ryzen Threadripper 2950X 16-Core Processor [Family 23 Model 8 Stepping 2]  |      3 | 32.0000 | 5298576471.103373 |
| AMD Ryzen Threadripper 2990WX 32-Core Processor [Family 23 Model 8 Stepping 2] |      7 | 55.1429 | 4328623771.943422 |
+--------------------------------------------------------------------------------+--------+---------+-------------------+
4 rows in set (0,29 sec)

If we apply points 3, as nhost are less than 10, no results pops up.

Point 2 is not the case here, but what makes it not appears is even point 1 as there are probably inactive CPU that has the expected result equals to 0 that cut off those contribution.

So the result is correct in the way Boinc is given those statistics

Thanks

Aha! That makes sense... thanks!