Ivan Nikitin

Hi, Standard version of EqualizerAPO does convolution using 32-bit float numbers. It uses FFTW in 32-bit mode and performs matrix multiplication using 32-bit numbers. I simply changed internal processing to use 64-bit numbers and used double-precission build of FFTW. There is still some 32-bit processing: Windows send data in 32-bit format. So, new version of EqualizerAPO converts these single precision data to double precision, does all the processing and then converts back to 32-bit to send data...

Hi, I modified Equalizer APO source code to use double precision 64 bit numbers instead of standard precision 32-bit numbers it currently uses. The difference in sound quality is subtle, but I performed an ABX test and missed just once out of 8 tries Here is the same song processed with 32-bit and 64-bit version of Equalizer APO. 32-bit: https://yadi.sk/d/pDTHz_0F8BlcoQ 64-bit: https://yadi.sk/d/TnzcDoGnYUN4fg I used the configuration generated with HeSuVi for my headphones. If it sounds interesting,...

Thank you for the reply! Is there any software to generate reference samples to test the equalizer? For example, if I need to check how good the LP filter is.

HelloIt sounds like complex chained GraphicEQ commands reduce sound quality. If I join the two commands into one, it sounds better.Is there a way to measure distortions caused by the processing?-- Ivan Nikitin