RomRaider

I figured out the proper vKS scaling, but it doesn't appear like vKS and vKSR differ ever. So not really sure on why they have two sets of RAM parameters that appear to always be 1:1. Personally, knock strength doesn't necessarily need extreme 32-bit accuracy. Anything under the knock strength threshold (to trip a knock count) is technically just noise. But the useful bits are when it's ABOVE the knock strength threshold. Based off that magnitude, you could potentially see just how severe a confirmed knock event was. So I dunno if the accuracy loss would be worth it in the end. Personally, logging 32-bit RAM parameters takes up too much space to be of much use with the OEM SIDs. So only logging specific bytes might be the only real option.

Sorry for bumping an old thread, but I'm sure you guys have figured it out at this point..

How reliable is vKNCNT reading at higher revs?
Can it show values other than "1"? (that's what I've seen)
Does it even register knock events outside a knock window?

Any help greatly appreciated

Without using knock headphones or having something to cross reference, it's hard to say for certain. It certainly can pick up false knock events at higher RPMs, but it seems to be pretty well filtered from my testing.


Yes, it will increment with every detected knock event and reset back down to 0 after enough time has passed. It will continuously be recording knock events regardless of the knock window. The knock window is just where it will actively retard ignition timing if knock is occurring, it doesn't affect the knock sensor logic.

Thanks, man, you've finally come back