Notes from from Janos Szabad, Albert Szent-Gyorgyi University:
This is the "instruction manual" that some people may look at before generating gynandromorphs with HorkaD.
While using ldsHor-D for the generation of gynandromorphs, it is good to know that the frequency of gynandromorphs is highest in the offspring when the ldsHor-D/TM3, Sb Ser males are taken from the strain mwh ldsHor-D e/TM3, Sb Ser/T(1;3)OR60 (42293 in Bloomington) and these males are mated with y v f mal virgins (strain 180 in Bloomington). Cultures should not be overcrowded. Gynandromorphs tend to hatch with a 2-3 day delay.
Please notice that (as described in Genetics 139, 1585, 1995) the frequency of gynandromorphs is in the range of 1-2% when other than y v f mal virgins are used and is 5-6% when the ldsHor-D/TM3, Sb Ser males are taken from outcrosses and are mated with y v f mal virgins.  
Some of the offspring of the y v f mal females and the ldsHor-D/TM3, Sb Ser males appear as males with red (wild type) eyes. (Eyes of the regular y v f mal/Y and the y v f mal/0 sibling males are creamy orange.) These flies are crypto-gynandromorphs and develop due to non-autonomy of the v mutation; although eyes of the crypto-gynandromorphs are y v f mal/0, some tissues in the body are y v f mal/X and produce a compound that overcomes the v mutant eye phenotype. 
To generate gynandromorphs with one red and one white eye, use e.g. y w homozygous females and ldsHor-D/TM3, Sb Ser males.
Some of the ldsHor-D descendants have long and thick as well as short and thin bristles. They are diplo-4//haplo-4 mosaics.. (For details see Genetics 139, 1585, 1995.)
If someone would like to make preparations of the gynandromorphs, here is a source how to proceed: Quick preparation of Drosophila for microscopic analysis. Drosophila Information Service 53, 215, 1978.