Hi,   
      
   Very interesting system!   
      
   Some questions about some of your test cases. I am trying   
   to replicate two of your test cases (*) in another system   
   to get some timing measurements.   
      
   But the system that I am using cannot solve either of   
   the two test cases, I get a memory overflow. So I am   
   thinking about how I could finally make the test case run.   
      
   The questions are:   
   - Is some horner schema used during evaluating res with   
     q1=s1, q2=s2, etc.. (Not needed for evaluating to zero,   
     but might use less memory and/or give more speed)   
      
   - Are the multivariate polynomial fractions GCDed during   
     intermediate results. (Not needed for evaluating to zero,   
     but might use less memory and/or give more speed)   
      
   - Do you have even slightly smaller test cases than   
     (res1, ss1, ss2, etc..) so that a comparison would be   
     possible.   
      
   Bye   
      
   (*)   
   http://home.bway.net/lewis/fermat/FerTest1   
      
   Am Dienstag, 21. Februar 2017 22:54:14 UTC+1 schrieb Robert Lewis:   
   > Fermat 6.0 includes vastly improved multivariate polynomial GCD algorithms.    
   There is a suite of tests vs Magma.   
   >   
   > Robert H. Lewis   
   > Fordham University   
   >   
   > http://home.bway.net/lewis/   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)