Now that AC Road Rumble has finally been implemented by a third-party app, this is the next big thing that I hope gets attention in SimHub.

Changing the audio driver via this link does make a decent difference - claws back around ~80-100ms worth in my case:

https://neosmart.net/wiki/fix-realtek-audio-delay-lag/

And it makes the latency good!   And useful in-seat for actual car information in my opinion.

But it could indeed be better with a lower-latency API software-side. 

It would require a significant rework, I'd imagine - but it'd surely be worth it. 

Cheers for any consideration.

Old->In the game (Assetto Corsa) I'm repeatedly dropping the car with a 'reset 2 track' mod, closing my eyes so the visuals don't confuse me, and listening to the sound from my Simucube DD wheel (and button rattle) compared to the signal of each sound device.

I've got the audio interface with ASIO coming through headphones, and the soundcard coming through the chair. Both are playing Simhub's signal. Testing each separately.

The headphones are pretty much right on with the DD force feedback.

The rumble sound from the chair is significantly later. I can visualize a "p-tshh" or actually like "throw" with the wheel noise being "th" and the shakers being "row". That's what I feel when I go over a road change on a track!!! By imitating this into audacity I am guessing it is around 100ms.

I don't think I can do that with the other signal. Maybe it's a little after the wheel, but it's a single moment to a much greater degree. That's why I think I'm not just tricking myself. Driving, it does feel more direct. (I am using 'true fullscreen' mode (alt+enter after alt-tabbing) for most direct image).

Testing for a third time now... ... blind sound test and driving... And, I am pretty convinced. When I watch and feel the bumps, if I press alt+enter to not use the above-mentioned 'true fullscreen,' I introduce a bit of monitor lag. The signal from the chair seems direct with that visual. And that is not the case with the signal from the headphones. It actually feels worse with that visual and direct on the true fullscreen (direct) visual. So, I am gonna say it works.

SO for someone who might be confused: It should be pretty easy to get a cheaper USB audio interface that lets you do this, and send the output to the shaker amp. It just has to have an ASIO driver, and preferably let you choose the buffer setting or at least see that it's running ASIO in a control panel thing. I'm not sure off the top of my head how the sub-$50 ones work.

But yeah, this was a good idea from the OP Escher89  :)

Old->I tried using the sound device with the ASIO driver compared to the computer sound card output, and the latency was the same.

I don't know if it's a problem of the sound latency. In synthesizers and music programs, it has a buffer setting for extra delay to add to prevent crackling when the sound driver is generating all kinds of things, and you can lower that buffer with the faster ASIO setting. With this, I don't know if the delay is being caused by a buffer like that. I don't see delay between seeing the light on the test button come on and hearing the tone. I don't know if this kind of buffer is coming into play. It could be a delay in Simhub somehow, or, looking at the graph of the effects and how it updates at a similar rate to the delay, maybe Simhub's update rate makes it have lag.

Or, maybe something could improve it.

I think bass shakers can profit from low latency audio via WASAPI or ASIO which are used in music production so the shaking is more direct.

Would be cool to have that feature so I can test if my assumption is right :P