TES Skyrim 0.144

[jonwd7]

Oyama
Same as for Oyama, i need only precise F0 variable value, it can't change more than 10% each frame when camera is not moving.

I gave you the variable. It changes far too rapidly for me to tell you any exact number. My tests were *extremely* objective. That is a typical view in game, with tons of up-close detail.

And it definitely changes more than 10%. I never moved the camera once in my tests. (Paused the game with ~ , and toggled the menus)

[Oyama]

Boris
OK, back to testing, will disable SLI.
But true, values are changing pretty quick, even if not moving the cam.

[jonwd7]

And also, I don't understand why FPS doesn't matter. My tests were all done in the exact same place, loaded from the same save, with the same camera position. I immediately enter the console to pause the game, and the FPS stays constant for each test. The Temp variable however changes wildly each frame.

Anyway, all three of my tests have the exact same camera vantage point, as seen in the gallery. So FPS certainly tells you exactly how much slower/faster v144 vs Flow vs NoFlow is.

[ENBSeries]

Fps doens't matter, because at the measure begin all rendering is flushed and mod waiting till drawing finished, only then it starting to compute ssao. After that it waiting again till the end of rendering confirmed by videodriver, so only difference in time matter for me something, fps is always very low in this mode.
Value can't change much, because videocard performance is constant all the time, it's like cpu computations of something, time is always the same. Also value interpolated in 10 frames, so smooth enough even to reduce delays of driver job and interruptions, for me value changes in about 1% only.
That camera angle meanless, what the sense to argue with me if i said not? Should i describe every step what, why, how? Okay, code have optimization when quality changes over distance, range of sampling changes, flow control optimization happen less frequently (and this i need to test).

[jonwd7]

That camera angle meanless, what the sense to argue with me if i said not?

Because you're not my boss and I decide what I give to you. You described "no sky", and I gave you "almost no sky". The image is roughly 1-2% sky... I calculated this. You described exterior, with lots of objects. No moving the camera. Gave you everything you needed, yet you still act unhappy.

Furthermore, here: http://www.youtube.com/watch?v=wDKNCqtT6Oo

Proof that the variable does not do what you are claiming. You'll have to watch in 1080p to see it clearly. The value cycles rapidly like this no matter whether I pause the game with the console key, whether I'm moving the camera, whether I'm looking at 0.0% sky or not. So arguing with me is in fact useless, I can agree with you on this.

[Oyama]

Boris
Been noting some values, will finish testing in few hours and report.
Would have been better making some video, but I suck with it
That's quite a hit when using one GPU only
Try taking some time to breathe

One question : did you make anything on latest binaries ( since you changed the detector ) that could have impacr on bloom passes's .xyz ? Like kind of internal limit or threshold-trigger ?
Seems that bloom is bleeding / spreading quite differently with post .14x versions...

[Wolfstryder]

ENB 144 looks very nice but I'm having a major fps drop. Going from ENB 119 to 144 I'm losing at least 15 fps. Way more at times. I've tested many times. I've turned every feature off, SSAO, AA etc. and still the same result. I hate to say it but the major framerate dips are actually making the game unplayable by being too jerky and way too slow at times.

I've even tried several driver versions for my video card and get the same result.

[ENBSeries]

Wolfstryder
Only for budget videocards or supersampling at huge resolution versions after 0.119 are slower like you said (don't know base fps to compare with). Even changes in almost all shaders can't affect performance so much. Turn off detailed shadows, shadowqualityfix, bilateral filter and restart the game.

Oyama
Don't need testing, nobody care, so why should i? Not playing the game, better will do something else instead of damn ssao.
Bloom code is not changed (and anything else color related).

[Oyama]

I DO care !
Well,.... you decide, it's your work of course.
I'm pretty sure you don't need anyone to lick your ass (so I won't), but hell, pity some don't realize your AO algorythm is WAY better than any out there.... That's a fact.

Be well.

[chidosity]

Don't need testing, nobody care, so why should i? Not playing the game, better will do something else instead of damn ssao.
Bloom code is not changed (and anything else color related).

Bullshit.

I just spent an ungodly amount of money on a new rig. I was able to play Skyrim fine on the old rig. But I couldn't play Skyrim fine while using your damn ssao. I simply can't live without it. Hell, your forum is one of 3 websites that I check every single day, religiously, just to see what you are up to.

I'm useless as a tester, sorry about that, so I know that I add nothing of value here, other than my bad attitude.

This is your project, and you do it for your own reasons, with your own motivations, but don't delude yourself into thinking nobody cares.

Seems to me you've found yourself at the edge of a black-hole of tedium. You've taken your ssao very far, but now you are confronted with a mountain of work for very little gain. Maybe your perfectionism is getting the better of you. The SSAO isn't perfect, but it's good enough for right now, isn't it? Maybe shelf this aspect of the project for a while and find something a little more exciting to delve into. You have been on a rather manic code-writing frenzy lately. All this time spent modding games. Do you ever take the time to actually play them?