Ice’s Falcon BMS Profile Updated for BMS 4.35
-
:bowd:
Again, i’m perfectly happy with just the map functioning correctly! the FR issue with the MFDs may be related to the Lilliputs being 60hz while my displays are set to 120hz (or not!). For a test I set the MFDs to “hidden” in Helios and then ran RTTClient64 alongside just for the MFDs, they are smooth as silk so i’m going to use this combination as a solution until someone tells me different…
Have a look on the Binding of the Data and Maps buttons you can easily switch so that your profile starts with the MAPS Panel active instead of the DATA Panel
-
is there a way to change the refresh rate for the MFDs in Helios? Using YAME64 or RTT the MFDs are silky smooth but I detect a bit of a low frame rate/mild stutter using Helios to display them.
The updating of data from sharedmemory (gauges / MFDs, etc …) is based on a fixed polling frequency that is at the core of Helios. The last time I looked into it I believe we discovered that it is 60hz. I noticed this small delay when I first implemented the Texture extraction of MFDs, DED, etc … in the Helios Falcon Interface. It’s not perfect but its workable in most cases. I do want to make it better but it has to be done without breaking the rest of Helios. YAME64 was smooth because I believe they extracted from DX9 and not from SharedMemory depending if you added the Direct3D hook dll or not. That was by far the fastest extraction method that I’ve seen besides the recent RTT implementation.
-
The Direct3d hook is not going to work as of 4.34. So with 4.34 YAME has to read from Sharedmemory as well
-
Quick (maybe) Helios question. I set the MFDs to “hidden” in my Helios profile so I could compare RTTClient64’s export to it. I’m sticking with RTT for the MFDs, should I remove the MFDs from the Helios profile or are they just being ignored using the “hidden” switch and not using CPU/GPU cycles or memory?
-
Shared memory reading is active as soon as you add the FalconInterface to Helios otherwise none of the gauges nor the CPD would be functional as they all are feed by the shared memory reader. I would have to dig into the code but i thing neither setting them to hidden nor to remove them completely from the Profile would stop reading the shared memory itself
-
Shared memory reading is active as soon as you add the FalconInterface to Helios otherwise none of the gauges nor the CPD would be functional as they all are feed by the shared memory reader. I would have to dig into the code but i thing neither setting them to hidden nor to remove them completely from the Profile would stop reading the shared memory itself
My concern is, is there a possible conflict if Helios is still pulling the MFD data while RTT is pulling/displaying it, even if Helios is not displaying it (MFDs)?
-
There is no conflict directly with having RTT and Helios polling for data in shared memory. If you hide the Helios mfds that should be enough mitigate any performance hit you might experience on the rendering side. If your not going to use the Helios ones then just remove them. It’s eady enough to add them back later if you want to switch back.
Sent from my iPhone using Tapatalk
-
There is no conflict directly with having RTT and Helios polling for data in shared memory. If you hide the Helios mfds that should be enough mitigate any performance hit you might experience on the rendering side. If your not going to use the Helios ones then just remove them. It’s eady enough to add them back later if you want to switch back.
Sent from my iPhone using Tapatalk
Thank you Wheelchock, as RTT handles the extraction much “smoother” i’m going to continue using the combination so I will remove them from Helios for now. Maybe when I find a cheap RTX 3090 the issue will not be present… ;D
-
Great, great work. Thank you so much for your effort.
-
Update 4.35-2 for BMS 4.35 U1