TES Skyrim 0.266
Forum rules
new topics are not allowed in this subsection, only replies.
new topics are not allowed in this subsection, only replies.
- Author
- Message
-
Offline
- Posts: 16
- Joined: 31 Jan 2014, 01:40
Re: TES Skyrim 0.266
Ok im kinda scared guys, I don't know what happened but this update literally double my fps, from 45 to 90 and i have everything maxed with like 500 mods, does this sound normal; im thinking its too good to be true. I enabled fix lag and zprepass or whatever it was called. I've only tested for like 10 min but this just seems crazy, also i had terrain parallax disabled so im thinking thats what was killing my fps, any thoughts?
-
Offline
- Posts: 6
- Joined: 29 Sep 2014, 12:38
Re: TES Skyrim 0.266
Are you using Fraps by any chance ? I was about to report that FixLag=true causes Fraps to report a wrong framerate number, like you said, twice as high as the value ENB fps counter gives.HizFather wrote:Ok im kinda scared guys, I don't know what happened but this update literally double my fps, from 45 to 90 and i have everything maxed with like 500 mods, does this sound normal; im thinking its too good to be true. I enabled fix lag and zprepass or whatever it was called. I've only tested for like 10 min but this just seems crazy, also i had terrain parallax disabled so im thinking thats what was killing my fps, any thoughts?
Steam fps counter reports the right value.
I don't seem to get any issues with ZPrepass though.
This is off-topic, but massive thanks for this skysan. I've had the crash on loading a save issue for a while, looked it up, found nothing valuable, tried the most conservative papyrus settings possible and couldn't fix it, it was very annoying. I'm really glad to finally learn of a workaroundskysan4298 wrote: I always CTD when I tried to load a save from the main menu because I have a few mods with heavily scripted. I always coc to a cell like whiterunbreezehome to initialize all scripted mods then load.
-
Offline
- Posts: 16
- Joined: 31 Jan 2014, 01:40
Re: TES Skyrim 0.266
I am not using fraps, im using MSI Afterburner, i know for a fact that the fps is showing correct but it seems very wierd that it just doubled my performance. Also, if your having problems loading a save then you probably need the clean your save and if cleaning it doesnt fix it then start disabling mods one by one and see if you can load saves, ive had that problem before and found a mod that was causing me to be unable to load saves though i dont remember what mod it was. http://www.nexusmods.com/skyrim/mods/52363/? is a good save cleaner thats easy to use.ayvis wrote:Are you using Fraps by any chance ? I was about to report that FixLag=true causes Fraps to report a wrong framerate number, like you said, twice as high as the value ENB fps counter gives.HizFather wrote:Ok im kinda scared guys, I don't know what happened but this update literally double my fps, from 45 to 90 and i have everything maxed with like 500 mods, does this sound normal; im thinking its too good to be true. I enabled fix lag and zprepass or whatever it was called. I've only tested for like 10 min but this just seems crazy, also i had terrain parallax disabled so im thinking thats what was killing my fps, any thoughts?
Steam fps counter reports the right value.
I don't seem to get any issues with ZPrepass though.
This is off-topic, but massive thanks for this skysan. I've had the crash on loading a save issue for a while, looked it up, found nothing valuable, tried the most conservative papyrus settings possible and couldn't fix it, it was very annoying. I'm really glad to finally learn of a workaroundskysan4298 wrote: I always CTD when I tried to load a save from the main menu because I have a few mods with heavily scripted. I always coc to a cell like whiterunbreezehome to initialize all scripted mods then load.
-
Offline
- *blah-blah-blah maniac*
- Posts: 1498
- Joined: 31 Mar 2012, 15:06
- Location: France
Re: TES Skyrim 0.266
Parallaxed terrain can't harm your performance at that point. I mean dividing your FPS by 2.HizFather wrote:Ok im kinda scared guys, I don't know what happened but this update literally double my fps, from 45 to 90 and i have everything maxed with like 500 mods, does this sound normal; im thinking its too good to be true. I enabled fix lag and zprepass or whatever it was called. I've only tested for like 10 min but this just seems crazy, also i had terrain parallax disabled so im thinking thats what was killing my fps, any thoughts?
Same goes for what you're describing, such a FPS boost is pretty crazy... What are your specs plz ?...
Boris
off-topic
Sorry if that has already been answered, but would it be possible to have two sets of settings for Reflections, Amount and Power, Interior and Exterior ?....
_________________
Lian Li PC011 Dynamic, Corsair AX 1500i PSU, i9 10850K @5.0 Ghz, Aorus Z490 Ultra, RTX3090 MSI Gaming X Trio, 32GB Corsair Vengeance Pro RGB RAM@3600, Corsair MP600 1TB NVME System Drive, 10 TB Storage, W10 Pro 64, Custom Hard Tubing Watercooling Loop
Lian Li PC011 Dynamic, Corsair AX 1500i PSU, i9 10850K @5.0 Ghz, Aorus Z490 Ultra, RTX3090 MSI Gaming X Trio, 32GB Corsair Vengeance Pro RGB RAM@3600, Corsair MP600 1TB NVME System Drive, 10 TB Storage, W10 Pro 64, Custom Hard Tubing Watercooling Loop
-
Offline
- *sensei*
- Posts: 289
- Joined: 08 Dec 2012, 23:05
Re: TES Skyrim 0.266
Care though, skyrim is known to screw up some of it's internal data while loading a save from an already running game. Known stuff: issues fiexd by enchantment reload fix, also a known issue when skyrim fails to use a proper skeleton if you load a save where your character has a different gender from what is already loaded(does not happen all the time, reasons unknown).ayvis wrote:This is off-topic, but massive thanks for this skysan. I've had the crash on loading a save issue for a while, looked it up, found nothing valuable, tried the most conservative papyrus settings possible and couldn't fix it, it was very annoying. I'm really glad to finally learn of a workaroundskysan4298 wrote:I always CTD when I tried to load a save from the main menu because I have a few mods with heavily scripted. I always coc to a cell like whiterunbreezehome to initialize all scripted mods then load.
_________________
i5 2500k@4.0ghz - 24Gb RAM - R9-290 4Gb Tri-X - SSD 240gb Intel 520 - Win7x64
i5 2500k@4.0ghz - 24Gb RAM - R9-290 4Gb Tri-X - SSD 240gb Intel 520 - Win7x64
-
Offline
- Posts: 42
- Joined: 19 Jan 2014, 14:14
- Location: Hanoi, Vietnam
Re: TES Skyrim 0.266
Well, a common recommendation of many mod authors is to do a "clean install" when updating their mod, which is a series of steps:ENBSeries wrote:keithinhanoi
Memory leaks are d3d resources not deleted, but if this happen, internal game engine also occur. I don't have much mods installed and they do not cost much memory, but probably heavy mods will have bigger leaks. I found that when loaded game and quit to main menu several times, while watching for resource usage. With valid saved games everything is cleared, but saves with removed mods increase stats of d3d objects.
1) disabling the old plugin of the mod,
2) loading the last save game (which will give you the warning message because the plugin is missing)
3) make a new game save and quit Skyrim
4) install the new plugin version of the mod, and then
5) run Skyrim and load the last save (from step 3)
Since these steps involve quitting out of Skyrim, that means the d3d device(s) is/are released and I assume everything starts fresh when you run Skyrim in step 5.
But I'd really like to test this theory out. My question is what do you use to monitor resource usage, especially the 3d3 objects / references?
PS., 0.266 (not the CTDtest) appears to work fine on my rig - and at least in an interior so far, I don't notice differences with EnableZPrepass enabled. I'll report back when I get outside. As for FixLag, it did not seems to make any noticeable difference, again in an interior, though I don't use windowed mode. I will keep testing when I have any chance.
_________________
EVGA GTX 760 4GB | i5-4590 Quad 3.3Ghz | ASRock Z87E-ITX | 16GB Crucial DDR3-1600 | 240GB SSD + 3TB HDD | Win7 x64
EVGA GTX 760 4GB | i5-4590 Quad 3.3Ghz | ASRock Z87E-ITX | 16GB Crucial DDR3-1600 | 240GB SSD + 3TB HDD | Win7 x64
-
Offline
- *blah-blah-blah maniac*
- Posts: 17557
- Joined: 27 Dec 2011, 08:53
- Location: Rather not to say
Re: TES Skyrim 0.266
skysan4298
Multithread rendering was already used in the game (last patch fixed it to be on one separate thread), but this leads to bsods and ctds. This is bad idea in general, because videocard is not cpu, it's busy most of the time and application can't do anything about that, no sense to call other drawing functions if previous are still in work. I don't think that dx11 implementation is something better, just driver level, not videocard. But game have at least two unused periods of time when render thread do nothing (videocard still proceed data like ssao or any other heavy on background) and these timings are very huge, 1.5-2 ms before drawing shadows and at the end of frame about 4-12 ms. Again, these delays are not noticable much when vsync enabled and videocard is busy enough with enbseries effects processing, let's say cpu do the job faster than gpu. If you see in some games that many cpu cores are busy, it's definetly not because of rendering.
Uriel24
Show screenshot with full list of processes for all users and what is in your tray.
prod80
Depth in bloom have full screen resolution and bloom code was not changed. I gave you in pm how TempParameters computed, all the same.
To find out any issues, make bypass code in some shaders, simply tex2D read at the end, it's easier to understand where the bug is.
Give me entire bloom shader, bug could be in samples filtering, no offsets in vertex shader too.
TechAngel85
I don't know which parameters affect zprepass, but those parameters must be like i wrote them, which are default.
Oyama
I don't know what is the logical point to make reflection parameters separated for interior and exterior, but as ssao was split like that by request, perhaps will do that too for reflection.
keithinhanoi
For resourse usage i'm activating some mod debug mode features.
FPS is not doubled with FixLag=true, it's negative side effect of some external software. Use internal fps counter of the mod.
Multithread rendering was already used in the game (last patch fixed it to be on one separate thread), but this leads to bsods and ctds. This is bad idea in general, because videocard is not cpu, it's busy most of the time and application can't do anything about that, no sense to call other drawing functions if previous are still in work. I don't think that dx11 implementation is something better, just driver level, not videocard. But game have at least two unused periods of time when render thread do nothing (videocard still proceed data like ssao or any other heavy on background) and these timings are very huge, 1.5-2 ms before drawing shadows and at the end of frame about 4-12 ms. Again, these delays are not noticable much when vsync enabled and videocard is busy enough with enbseries effects processing, let's say cpu do the job faster than gpu. If you see in some games that many cpu cores are busy, it's definetly not because of rendering.
Uriel24
Show screenshot with full list of processes for all users and what is in your tray.
prod80
This is not full screen bloom then.Also no issues with aliasing as I use a wide Gaussian blur which is very soft in it's falloff anyway.
Depth in bloom have full screen resolution and bloom code was not changed. I gave you in pm how TempParameters computed, all the same.
To find out any issues, make bypass code in some shaders, simply tex2D read at the end, it's easier to understand where the bug is.
Give me entire bloom shader, bug could be in samples filtering, no offsets in vertex shader too.
TechAngel85
I don't know which parameters affect zprepass, but those parameters must be like i wrote them, which are default.
Oyama
I don't know what is the logical point to make reflection parameters separated for interior and exterior, but as ssao was split like that by request, perhaps will do that too for reflection.
keithinhanoi
For resourse usage i'm activating some mod debug mode features.
FPS is not doubled with FixLag=true, it's negative side effect of some external software. Use internal fps counter of the mod.
_________________
i9-9900k, 64Gb RAM, RTX 3060 12Gb, Win7
i9-9900k, 64Gb RAM, RTX 3060 12Gb, Win7
-
Offline
- *sensei*
- Posts: 372
- Joined: 28 Jul 2013, 23:26
Re: TES Skyrim 0.266
Boris;
Pretty sure it has to do with offsets, but I'm not getting that in 265 binary, which is odd. Maybe bug is there in 265 too, but due to higher resolution it is not visible. Would appreciate you having a look at the shader. Any input for texture filtering would be helpful too... I struggle a bit with that part because basically I have no knowledge on it and I can't use your shader as a template because of using completely different code.
PM underway
Pretty sure it has to do with offsets, but I'm not getting that in 265 binary, which is odd. Maybe bug is there in 265 too, but due to higher resolution it is not visible. Would appreciate you having a look at the shader. Any input for texture filtering would be helpful too... I struggle a bit with that part because basically I have no knowledge on it and I can't use your shader as a template because of using completely different code.
PM underway
-
Offline
- *blah-blah-blah maniac*
- Posts: 1498
- Joined: 31 Mar 2012, 15:06
- Location: France
Re: TES Skyrim 0.266
Boris
Different interactions with lighting, IBL, direct lighting and such.
That's why I'd like to set their amounts and powers differently.
If you can separate these vars for exterior / interior, I would really find it usefull
Different interactions with lighting, IBL, direct lighting and such.
That's why I'd like to set their amounts and powers differently.
If you can separate these vars for exterior / interior, I would really find it usefull
_________________
Lian Li PC011 Dynamic, Corsair AX 1500i PSU, i9 10850K @5.0 Ghz, Aorus Z490 Ultra, RTX3090 MSI Gaming X Trio, 32GB Corsair Vengeance Pro RGB RAM@3600, Corsair MP600 1TB NVME System Drive, 10 TB Storage, W10 Pro 64, Custom Hard Tubing Watercooling Loop
Lian Li PC011 Dynamic, Corsair AX 1500i PSU, i9 10850K @5.0 Ghz, Aorus Z490 Ultra, RTX3090 MSI Gaming X Trio, 32GB Corsair Vengeance Pro RGB RAM@3600, Corsair MP600 1TB NVME System Drive, 10 TB Storage, W10 Pro 64, Custom Hard Tubing Watercooling Loop
-
Offline
- *blah-blah-blah maniac*
- Posts: 17557
- Joined: 27 Dec 2011, 08:53
- Location: Rather not to say
Re: TES Skyrim 0.266
prod80
I don't know what you planned to make, but i hate these lines:
In general, code used for cycles have big chance to not compile properly, because shader compiler from MS is awful and it's very hard to debug result for the bloom - too many passes which you can't see.
Reading depth as cut off factor is bad idea, because with small resolutions this basically aliasing == lines. And the lines themselves even without depth appear, because blurring range is too low. Per pixel is not enough, because pixels are square, you need to blur bigger range at least to make smooth image. Partially possible to implement bicubic filtering in postpass shader for low resolution textures, huge quality boost, but still with artifacts noticable when rotating camera on small bright objects, so bigger blurring range must be used anyway. I suggest to disable depth in cycles of two texture passes, then modify scaling and watch how it looks for SamplerBloom4 only output (link tempF1.x variable for simpler scaling). Another problem is separate blurring for horizontal and vertical lines, skipping any pixels leads to kinda moire lines. They starts to appear after certain scaling, so when bloom was 768 it was visible for resolutions like 3k*1.5k. I don't understand how amount of samples computed for cycles:
but this is culprit. Just do not skip pixels in split hor/ver blurring or use linear filtering specific range scaling and skip with it half pixels carefully, before moire appear.
I don't know what you planned to make, but i hate these lines:
Code: Select all
bloom.xyz = saturate( bloom.xyz );
Reading depth as cut off factor is bad idea, because with small resolutions this basically aliasing == lines. And the lines themselves even without depth appear, because blurring range is too low. Per pixel is not enough, because pixels are square, you need to blur bigger range at least to make smooth image. Partially possible to implement bicubic filtering in postpass shader for low resolution textures, huge quality boost, but still with artifacts noticable when rotating camera on small bright objects, so bigger blurring range must be used anyway. I suggest to disable depth in cycles of two texture passes, then modify scaling and watch how it looks for SamplerBloom4 only output (link tempF1.x variable for simpler scaling). Another problem is separate blurring for horizontal and vertical lines, skipping any pixels leads to kinda moire lines. They starts to appear after certain scaling, so when bloom was 768 it was visible for resolutions like 3k*1.5k. I don't understand how amount of samples computed for cycles:
Code: Select all
for(int i = 1; i < LOOPCOUNT && SigmaSum < Q; ++i)
_________________
i9-9900k, 64Gb RAM, RTX 3060 12Gb, Win7
i9-9900k, 64Gb RAM, RTX 3060 12Gb, Win7