Jump to content

Skyrim Project Optimization (by rgabriel15)


Besidilo

Recommended Posts

It helps the most with the populated interiors like taverns especially at night. Try it at the Bannered Mare and get about 5-10 FPS depending on time of day. If you already are at 60 FPS in most interiors like castles it will not help a ton.

 

From what I understand about Skyrim is that it already uses Occlusion Culling, but the author, rgabriel took it a bit further than Beth did. He did leave a list a of places that are optimized and a list of places that cannot be optimized here. His list is pretty extensive so testing all the interiors is going to take a while, but if we start with castles and taverns (night and day) it would probably give us a good idea about the performance impact. Looks like spreadsheet would work for this vs a ton of screenshots.

Damn that! :lol: LOL! I'm not going to be testing every location that has been edited...don't have the time or willingness for that. So far on my list, I have 11 locations where testing will be done. I'll being benching vanilla FPS, STEP FPS (baseline), and then STEP with this mod FPS. That should give plenty of results to see if the mod actually helps in terms of performance.
Link to comment
Share on other sites

  • Replies 48
  • Created
  • Last Reply

Top Posters In This Topic

Okay... testing is done, numbers are in, results are calculated. I ended up with only 8 of the original 11 locations as some location's numbers remained unchanged throughout the testing so I simply excluded them from the final results. Below in the spoiler is the raw data for those interested. But for those that just want a summary this is for you:

 

Skyrim Project Optimization (SPO) is a fair candidate for STEP inclusion. Testing has shown that the mod does in fact provide a performance increase, although it is nothing dramatic as shown in some the screenshots on the mod's nexus page. The numbers show that with the mod enabled the average FPS and minimum FPS are higher than without it; though this increase was only 2-5 FPS on average. Although this might not seem like much, any time you can get the minimum and maximum FPS closer together the result will be smoother gameplay, and that's just what I noticed. As, Ithian mentioned above, there was a noticeable smoother gameplay with the mod enabled in the larger locations that I tested (Bards College, Dragonsreach, Jorrvaskr Living Quarters).

 

Since I was testing, I decided not to stop with just the usual "STEP + Mod" testing. SPO's original design was for Realistic Lighting Overhaul running on lower-end systems. I didn't have a lower-end system laying around, but I did have the latest RLO release (which is a work of art, btw!). I also had my preferred ENB laying around so I tested both and the results were the same. An increase of 2-5 FPS with the mod enabled.

 

So you get a small FPS boost and some added smoothness as well. What's not to like about SPO? Nothing in my book!

 

Interestingly, if you'd like to look at the numbers below you'll see how RLO causes, on average, a 4 FPS drop. ...and we won't even talk about the performance hit from the ENB (average of 18 FPS drop).

 

 

 

Test 1  - Vanilla

Test 2a - STEP only

Test 2b - STEP + SPO

Test 3a - STEP + RLO

Test 3b - STEP + RLO + SPO

Test 4a - STEP + RLO + ENB

Test 4b - STEP + RLO + ENB + SPO

 

For those looking at these numbers, keep in mind that my max FPS is normally limited to 60 on my system.

 

Bards College - Solitude

Test 1  - Avg: 59.505 - Min: 58 - Max: 60

Test 2a - Avg: 59.433 - Min: 54 - Max: 62

Test 2b - Avg: 59.500 - Min: 57 - Max: 62

Test 3a - Avg: 59.267 - Min: 53 - Max: 62

Test 3b - Avg: 59.500 - Min: 58 - Max: 61

Test 4a - Avg: 40.500 - Min: 33 - Max: 43

Test 4b - Avg: 43.533 - Min: 38 - Max: 45

 

Castle Dour - Solitude

Test 1  - Avg: 59.527 - Min: 58 - Max: 61

Test 2a - Avg: 59.500 - Min: 56 - Max: 63

Test 2b - Avg: 59.500 - Min: 58 - Max: 61

Test 3a - Avg: 59.500 - Min: 57 - Max: 62

Test 3b - Avg: 59.500 - Min: 58 - Max: 61

Test 4a - Avg: 42.567 - Min: 38 - Max: 45

Test 4b - Avg: 47.633 - Min: 46 - Max: 49

 

Angeline's Aromatics - Solitude

Test 1  - Avg: 59.515 - Min: 59 - Max: 60

Test 2a - Avg: 59.500 - Min: 59 - Max: 60

Test 2b - Avg: 59.500 - Min: 59 - Max: 60

Test 3a - Avg: 55.333 - Min: 54 - Max: 56

Test 3b - Avg: 59.033 - Min: 57 - Max: 61

Test 4a - Avg: 37.367 - Min: 37 - Max: 38

Test 4b - Avg: 38.967 - Min: 38 - Max: 40

 

Bits & Pieces - Solitude

Test 1  - Avg: 59.502 - Min: 58 - Max: 61

Test 2a - Avg: 59.500 - Min: 59 - Max: 60

Test 2b - Avg: 59.500 - Min: 59 - Max: 60

Test 3a - Avg: 59.500 - Min: 58 - Max: 60

Test 3b - Avg: 59.500 - Min: 59 - Max: 61

Test 4a - Avg: 44.500 - Min: 43 - Max: 46

Test 4b - Avg: 44.733 - Min: 42 - Max: 47

 

Bannered Mare - Whiterun

Test 1  - Avg: 59.496 - Min: 57 - Max: 62

Test 2a - Avg: 59.500 - Min: 57 - Max: 62

Test 2b - Avg: 59.500 - Min: 58 - Max: 62

Test 3a - Avg: 56.567 - Min: 51 - Max: 58

Test 3b - Avg: 54.833 - Min: 51 - Max: 56

Test 4a - Avg: 38.400 - Min: 33 - Max: 43

Test 4b - Avg: 38.833 - Min: 33 - Max: 41

 

Belethor's General Store - Whiterun

Test 1  - Avg: 59.492 - Min: 58 - Max: 61

Test 2a - Avg: 59.500 - Min: 58 - Max: 61

Test 2b - Avg: 59.500 - Min: 58 - Max: 61

Test 3a - Avg: 55.900 - Min: 54 - Max: 58

Test 3b - Avg: 57.533 - Min: 57 - Max: 58

Test 4a - Avg: 36.733 - Min: 36 - Max: 38

Test 4b - Avg: 37.333 - Min: 36 - Max: 38

 

Dragonsreach - Whiterun

Test 1  - Avg: 59.511 - Min: 58 - Max: 61

Test 2a - Avg: 59.500 - Min: 58 - Max: 61

Test 2b - Avg: 59.500 - Min: 58 - Max: 61

Test 3a - Avg: 56.200 - Min: 55 - Max: 58

Test 3b - Avg: 56.133 - Min: 53 - Max: 58

Test 4a - Avg: 39.733 - Min: 39 - Max: 41

Test 4b - Avg: 40.033 - Min: 39 - Max: 41

 

Honningbrew Meadery - Whiterun

Test 1  - Avg: 59.508 - Min: 59 - Max: 60

Test 2a - Avg: 59.500 - Min: 58 - Max: 61

Test 2b - Avg: 59.500 - Min: 59 - Max: 60

Test 3a - Avg: 59.500 - Min: 59 - Max: 60

Test 3b - Avg: 59.500 - Min: 58 - Max: 61

Test 4a - Avg: 47.233 - Min: 46 - Max: 48

Test 4b - Avg: 47.600 - Min: 47 - Max: 48

 

 

Link to comment
Share on other sites

Agree that the trends indicate a slight performance improvement (too bad only for interiors). However, since it is only slight on tech's laptop, it will not be really even noticeable on more powerful systems (but likely quite noticeable on low-end systems).

 

Therefore, I would say that this is an ideal candidate for Performance STEP, but probably optional for Baseline and not worth the plugin slot for Extreme.

 

I guess this means that it gets into STEP right now with modification when we get the other guide levels up and running.

Link to comment
Share on other sites

I agree with your analysis, Z. It was originally created with low-end systems in mind and they will benefit the most from this mod. As such, most Baseline users will only see a slight performance increase; however, may notice the increased smoothness as I did. Every little bit helps though. The real question is where to put this mod. The only place in my mind, without adding a "performance" category is to put it in "Fixes", since it adds to the occlusion culling that is already in the game. Seems like Bethesda could have done the same thing SPO is doing to improve performance but chose not to for whatever reason.

 

I don't know if it's even possible, but I wouldn't mind seeing so type of "outdoor" version of this mod.

Link to comment
Share on other sites

Yes, I agree that this is a "fix", since it elaborates on an insufficiently-implemented optimization.

 

It should be totally applicable to non-LOD outdoor environment using the same methods. It may be much more challenging though, since there are many more possible view angles.

Link to comment
Share on other sites

Yes, I agree that this is a "fix", since it elaborates on an insufficiently-implemented optimization.

 

It should be totally applicable to non-LOD outdoor environment using the same methods. It may be much more challenging though, since there are many more possible view angles.

I was googling and reading up on it and it's completely doable in exterior environments. Especially within cities and forts where the buildings and walls provide a line-of-site "barrier" to only render what is in view.

 

This site is for a 3D program, but provide a nice before and after image to give you idea of how it works:

https://docs.unity3d.com/Documentation/Manual/OcclusionCulling.html

 

And this site explains how the occlusion planes work:

https://geck.bethsoft.com/index.php/Occlusion_Culling

 

Very interesting stuff. It would take a lot of work, but it's completely doable in more the just interiors. I asked the author if he has and plans on expanding to exteriors and he has stated that the exteriors are already optimized by default, but he would look at the cities to see if any improvements could be made.

Link to comment
Share on other sites

Candlehearth Hall in Windhelm was the only tavern or castle that I got any jump in FPS between 6-13 FPS, and Jorvaskr Basement seemed to be a better at the entrance.

 

If you have 60 FPS indoors already then it might not help, but if any place drops under 50 FPS, you will see a nice gain. Other than Fixes it could be in Animations & Effects, since it adds the culling effect.

 

The outdoor culling effect may already be as good as it can get, and even many of the indoor places are as good as they can get.

Link to comment
Share on other sites

  • 2 weeks later...

Since there are multiple versions of this and very little information on which to install and why, we'll need some feedback on the recommended installation asap.

 

I am assuming full version and that with homes includes Hearthfire support ... ?

 

 

Can anyone verify?

Link to comment
Share on other sites

From what I can tell, the Full version with Homes doesn't contain anything to do with Hearthfire. When I was looking through the files in TES5Edit, though, I did notice that all the Dragonborn cells got a "DUPLICATE001" added onto the end of them, which I think stops them from working?

 

Also, the Full version that I downloaded has both an ESM and an ESP in it. From what I can see, they contain exactly the same stuff, the only differences being that one's an ESM and one's an ESP.

Link to comment
Share on other sites

  • 3 weeks later...

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

By using this site, you agree to our Guidelines, Privacy Policy, and Terms of Use.