| ENBSeries http://enbdev.com/enbseries/forum/ |
|
| TES Skyrim 0.193 http://enbdev.com/enbseries/forum/viewtopic.php?f=2&t=2138 |
Page 32 of 51 |
| Author: | charlievoviii [ 23 Jul 2013, 16:52 ] |
| Post subject: | Re: TES Skyrim 0.193 |
Xilandro wrote: ENBSeries >> Tested with vsync off in enb - fps hit. Tested without enb. Hit. Changed in skyrimprefs.ini iPresentInterval=1 to 0 - hit Added line iPresentInterval=0 to skyrim.ini - 89fps Problem solved =3 Thanks for waypointing me) use FPS limit in the enblocal.ini instead to set your FPS limit. |
|
| Author: | Drallen [ 23 Jul 2013, 16:57 ] |
| Post subject: | Re: TES Skyrim 0.193 |
New test: disable ENB, run Skyrim with SPM. Results: FPS and stuttering improved (or back to acceptable levels) - 2 CTDs: 1 when changing locations, 1 which looks like >3.1Gb limit crash. Also, Skyrim Performance Monitor on-screen display is still dim and pixelated! Analysis: Clearly, ENB is not the problem. Maybe not driver/hardware problems either. I think there is something in skyrim.ini / skyrimprefs.ini that is conflicting both ENB and SPM. That would be my fault - I have been tweaking both of them wildly in hopes of stopping >3.1Gb CTDs (I have too many). ENB v0.193 resolves the >3.1Gb crashes but costs performance. Question: Are there any ini settings that should or should not be used with ENB? Or any ENB ini settings that were changed between 21/7 v0.193 and test36 (these may have fixed whatever was wrong with my inis before)? |
|
| Author: | charlievoviii [ 23 Jul 2013, 17:03 ] |
| Post subject: | Re: TES Skyrim 0.193 |
Drallen wrote: New test: disable ENB, run Skyrim with SPM. Results: FPS and stuttering improved (or back to acceptable levels) - 2 CTDs: 1 when changing locations, 1 which looks like >3.1Gb limit crash. Also, Skyrim Performance Monitor on-screen display is still dim and pixelated! Analysis: Clearly, ENB is not the problem. Maybe not driver/hardware problems either. I think there is something in skyrim.ini / skyrimprefs.ini that is conflicting both ENB and SPM. That would be my fault - I have been tweaking both of them wildly in hopes of stopping >3.1Gb CTDs (I have too many). ENB v0.193 resolves the >3.1Gb crashes but costs performance. Question: Are there any ini settings that should or should not be used with ENB? Or any ENB ini settings that were changed between 21/7 v0.193 and test36 (these may have fixed whatever was wrong with my inis before)? lol your test make no sense. So you tested without ENB and got better performance and less stutter ? DUHHH As for skyrim.ini and skyrimprefs.ini tweaks, it takes a lot of time. It took me weeks to get everything perfect and balance between visual and performance. Same goes for enbseries.ini settings. One way to test if it's your skyrim.ini and skyrimprefs.ini issue causing your CTD, copy them to a different location. Open the skyrim launcher and redo your settings inside the launcher. After run the game. If it stills CTD, than might want to check your load order, papyrus log for scripts error, or your mods for incompatibility. |
|
| Author: | ENBSeries [ 23 Jul 2013, 17:06 ] |
| Post subject: | Re: TES Skyrim 0.193 |
CruNcher Injector is harmless, you just don't know how to use it. When it injected mod in to the game, it can be unloaded, because no longer in use. I can't disable it or exit from it, because more than one process can run at single time or after each other. |
|
| Author: | ENBSeries [ 23 Jul 2013, 17:09 ] |
| Post subject: | Re: TES Skyrim 0.193 |
Drallen Nothing changed for test 36 in ini setting. And version which is on site is much better for AMD cards than 36, same stuttering, but less memory usage, the only cost is crash because of incompatibility with memory monitor. EDIT: can you run the game without patch at all? Or it's the case when everything toggled on to get crash at any cost? |
|
| Author: | CruNcher [ 23 Jul 2013, 17:11 ] |
| Post subject: | Re: TES Skyrim 0.193 |
ENBSeries wrote: CruNcher Injector is harmless, you just don't know how to use it. When it injected mod in to the game, it can be unloaded, because no longer in use. I can't disable it or exit from it, because more than one process can run at single time or after each other. ok so injector.exe detects tesv.exe start hooks enbseries.dll into it's win32 process and then injector.exe gets destroyed from enbseries.dll is not possible ? |
|
| Author: | HighPitchedMcFly [ 23 Jul 2013, 17:12 ] |
| Post subject: | Re: TES Skyrim 0.193 |
Hi all First thank you for all your work Boris and all comunity. so Im not an expert but I see you want some tester for the new code and i will be happy to help you if i can. i have an amd card perhaps , so if i can help with this hardwre i will be . with the 192 release the memory usage=true work well but hit my fps -20 I will test the 0.193 this night ... |
|
| Author: | ENBSeries [ 23 Jul 2013, 17:18 ] |
| Post subject: | Re: TES Skyrim 0.193 |
CruNcher Quote: gets destroyed from enbseries.dll is not possible ? Hard to understand for me. If you mean unload itself when injected - impossible, because skyrimlauncher.exe (at least) also must be injected. You need manually exit from injector after game started. Huh, i have an idea, but not 100% sure. May be send message from the mod after several seconds when it's started and running effects? HighPitchedMcFly Thanks |
|
| Author: | ENBSeries [ 23 Jul 2013, 17:23 ] |
| Post subject: | Re: TES Skyrim 0.193 |
No, it's bad idea to exit automatically from injector, found several conditions when such behavior is wrong. |
|
| Author: | CruNcher [ 23 Jul 2013, 17:35 ] |
| Post subject: | Re: TES Skyrim 0.193 |
Ok @ least you thought about it now if you see problems in your architecture with this approach that's another thing, i guess i will test too what these conditions would be Because im really lazy to always close enbinjector.exe so i let it run and get angry about it's cpu usage in the background letting my CPU Fan speed up Intels Speedstep and Fan Controll System is very proactive on such things i mean i know your statepoint about it since several posts now though still i find it not a thing to overlook ![]() ![]() i find this still a lot for a simple injection pooling, and Intel agrees with it, especially the kernel usage is massive, i mean you even beat firefox in its background consumption i surely wouldn't be proud about this 1st place, the cpu cycles run upwards like crazy, and that surprises me im used to very good efficient windows code from russian devs over the past, and your enb is that too but that enbinjector idle behavior is still crazy to me Im also not sure how one of the Kernel 3rd Party security layers should interfere here and might causing this overload nothing impossible of course. |
|
| Page 32 of 51 | All times are UTC |
| Powered by phpBB® Forum Software © phpBB Group https://www.phpbb.com/ |
|