So has OpenAL been unaffected all this time?
Development Thread The 1h30 audio issue mystery
|
07-10-2013, 02:58 AM
07-10-2013, 03:00 AM
It is my understanding that all backends experience the issue so I am not sure why the time stretch was needed. delroth or degasus may have to be the ones to tell us that
i5-8600K @ 4.5 GHz
GeForce GTX 1080 SC2 ASRock Z370 Taichi LGA 1151 CORSAIR Vengeance LPX 16GB 2400 DDR4 SAMSUNG 970 EVO M.2 500GB PCIe SSD Indigo Xtreme TIM Phanteks Pro ES614P Black Steel Case EVGA SuperNOVA 750W PSU Windows 10 64-bit 07-10-2013, 03:24 AM
Time stretch isn't needed at all. The broken part was the resampling in Mixer.cpp. I only try to add time streting because resampling is a free operation with time streting. In fact, both can be done with the SoundTouch library (already used by openAL).
So we either reimplement our resampler, or we include SoundTouch for it and get time streting for free. btw: SoundTouch could also be used without time stretching. In theory it should be possible without unneeded latency. But time stretching always has latency (eg 50-100ms) which will be added to the backend latency :-( 07-10-2013, 03:53 AM
Is testing for OSX still needed? I tried to do it yesterday, but wii-mote got disconnected, so when i came back is said reconnect wi-mote.
Desktop HTPC:
Gigabyte Z97X-Gaming5- i7-4790k MSI-B75-E33, i5-2500k Corsair Vengeance 16GB RAM Crucial 8GB RAM Gigabyte GTX780 GHz Edt. ASUS GTX960 Strix Win 10 Pro / OS X 10.10 Win 10 Pro
degasus,
It sounds like the best option would give folks the freedom to select whether they want time stretching enabled or to disable it. Wait...since we brought back OpenAL and it already has time stretching implemented, is there a pressing reason to even add time stretching? I am just really hoping to have no latency on those sound effects if it is possible I just thought of a side note. Wouldn't this officially close all those similar issues open about this one problem? I could try to get a list together of all those if it makes it easier (assuming that is even something the devs are interested in).
i5-8600K @ 4.5 GHz
GeForce GTX 1080 SC2 ASRock Z370 Taichi LGA 1151 CORSAIR Vengeance LPX 16GB 2400 DDR4 SAMSUNG 970 EVO M.2 500GB PCIe SSD Indigo Xtreme TIM Phanteks Pro ES614P Black Steel Case EVGA SuperNOVA 750W PSU Windows 10 64-bit 07-10-2013, 10:03 AM
Just use the first build he posted with the 1h30m bug fixed but with no timestretch.
07-10-2013, 10:13 AM
I am
I was just *hoping* that the time stretching wasn't going to make it into master as a requirement for all backends.
i5-8600K @ 4.5 GHz
GeForce GTX 1080 SC2 ASRock Z370 Taichi LGA 1151 CORSAIR Vengeance LPX 16GB 2400 DDR4 SAMSUNG 970 EVO M.2 500GB PCIe SSD Indigo Xtreme TIM Phanteks Pro ES614P Black Steel Case EVGA SuperNOVA 750W PSU Windows 10 64-bit 07-11-2013, 04:30 AM
degasus said something along the lines of "fuck timestretching, I'll just fix the bug for now" earlier today, so I think you'll be fine. :D
<@skid_au> fishing resort is still broken: http://i.imgur.com/dvPiQKg.png
<@neobrain> dafuq <+JMC47> no dude, you're just holding the postcard upside down ---------------------------------------- <@Lioncash> pauldachz in charge of shitposting :^) 07-11-2013, 05:06 AM
Best quote ever
i5-8600K @ 4.5 GHz
GeForce GTX 1080 SC2 ASRock Z370 Taichi LGA 1151 CORSAIR Vengeance LPX 16GB 2400 DDR4 SAMSUNG 970 EVO M.2 500GB PCIe SSD Indigo Xtreme TIM Phanteks Pro ES614P Black Steel Case EVGA SuperNOVA 750W PSU Windows 10 64-bit 07-12-2013, 01:42 AM
Quick question,
Are there any estimates on when this change will be added to master (assuming it will be)? I understand it may take time, but once 4.0 comes out, I wanted to make sure it had this type of fix in it
i5-8600K @ 4.5 GHz
GeForce GTX 1080 SC2 ASRock Z370 Taichi LGA 1151 CORSAIR Vengeance LPX 16GB 2400 DDR4 SAMSUNG 970 EVO M.2 500GB PCIe SSD Indigo Xtreme TIM Phanteks Pro ES614P Black Steel Case EVGA SuperNOVA 750W PSU Windows 10 64-bit |
« Next Oldest | Next Newest »
|
Users browsing this thread: 1 Guest(s)