Game Skyrim Forum

Original file http://www.tesnexus.com/downloads/file.php?id=13092

SI fixed file http://www.tesnexus.com/downloads/file.php?id=31757

does it really do anything or mess up things?

Don't bother with it, it won't help you all that much. The author once mentioned that all you'll see is something like a .1 FPS improvement. Not worth bothering with and it used to have a number of bugs. I don't know if the SI version fixes them, but even still it's not worth the trouble.

Disagree.

If you are willing to do some testing (what mod doesn't need any testing at all?), to get a free performance boost (completely variable, depends on your CPU, and to be honest how do you even track a 0.1 FPS improvement?) why wouldn't you try it?

If you find any problems you can always disable it, and revert to a previous save file, but if you don't, congratulations you just got some performance boost for nothing.

I personally use the SI fixed version, and haven't found any problems so far.

Agree. The consensus from those in the know say it's not worth having, and even introduces a few bugs. Scripting is pretty efficient, you've gotta have a pretty powerful GPU to be throttled by the CPU, which will be held back by other CPU tasks long before scripting performance becomes an issue...

Because even the author has mentioned that the performance increase is practically non-existent. Why would you even bother if the person who made it can't even recommend it? If you want to waste time trying to track a .1 FPS boost be my guest, but I'd rather worry about other things in my game. It's just filling up one of your .esp slots for no real benefit.

First time i downloaded it, many many months ago, such claim from the author was not present, plus i like to hear the opinion of many people around and not focus on a single one.

Anyway it doesn't matter because i was talking about my experience, you say you didn't even try it, so you have little to say but what you have heard from others, you may as well link or quote instead of making such claims as yours.

Scripts run entirely on the CPU, if you have an old CPU and a game running lots of scripts (and some very heavy scripts in my case) and get into combat with many NPCs around, even if you have a nVidia GeForce GTX 590 Ultra, your PC will bleed due to the CPU being a bottleneck.

On a Pentium 4 PC this helps a lot by increasing performance by around 2-4 frames depending place and situation currently running in game, and this is by experience not by theory or calculation, and as i said before, the gain is variable depending on your CPU.

So worthless? no, bugless? doubt it, no bugs in my experience though, like most of the mods out there, it's up to you.

Check the TESNexus page for more comments, there are many people that claim performance gains as well.

EDIT.

This is not to bash you Belanos obviously, ITPalading asked "does it really do anything or mess up things?" and i felt your answer wasn't complete.

Yes for some it does something, and yes for some it mess up things, is the trade off it's worth it? It's for me on my old P4 PC, but i don't use it on my C2D laptop.

Your experience is purely placebo effect. What you describe cannot be the result of what his mod changed.

Stating a fact without an argument, in somebody else's custom modded Oblivion, saying a 2-4 fps increase with HrmnsOblivionScriptOptimizationv1.0 enabled using TDT and an external program it's a placebo effect, EVEN when the fps are gone with the esp disabled (and due the already low frame rate this PC handles, 2-4fps makes a visual difference) has to be a very wise observation... or not.

In fact I did try it, I used it for quite awhile at first. But when I started hearing about the bugs it can cause I removed it. And I didn't notice any difference at all in my performance by doing so. So I'm not just going by what other people are saying, but from personal experience as well. The mod simply isn't worth the effort of downloading and installing it, and basically just takes up space.

This.

Any performance gain perceived while using this mod is entirely placebo. Many of us HAVE tried it on heavily modded games and its impact was zero. While it may not contain any real bugs, what it claims to do even under the most ideal of conditions simply doesn't happen. The author himself has said that the gains would be no greater than 0.5fps, and if you're not going to believe the author, that's fine.

Oblivion's real CPU eaters are AI heavy mods. The AI system will bring you down far quicker than any script unless that script is badly written. None of the vanilla scripts are badly written.

Show me a statistical anolysis of your FPS with and without the mod, and I'll consider what you have to say. Until such a time as you produce such an anolysis, everything I know about Oblivion, it's scripting engine, and what affects its performance, as well as my own tests with the mod, tell me that you are absolutely wrong, and in fact the effect you describe is impossible.


This. There is no script written for Oblivion that is a tenth as complicated as a single NPC's AI, barring actual bugs. I'd be surprised if there are any even a hundredth as complicated. While I'd argue that some of Bethesda's scripts are "badly written" in a number of cases (a particular script in SI's chapel thing springs to mind immediately), none of them have such a glaring, performance-destroying bug.

If you are having performance issues, the first thing to check is graphics settings, graphics mods, and your computer's graphics capabilities. If you've assured yourself that everything in that area is running efficiently, you might consider AI; large crowds of NPCs will cause problems for even the greatest of CPUs. Since Bethesda did a reasonably good job limiting the number of NPCs in one spot, and most mod authors try to as well, for most people the graphical processes are the major bottleneck, and improving CPU issues does nothing so long as its waiting on the graphics card to finish everything - which is the situation that most Oblivion installs find themselves in. Only if you're in the highly unusual position of having a stellar graphics card and a terrible CPU would I even worry much about AI (barring a mod that adds ridiculous levels of crowding), since graphics are almost certainly the culprit. Scripts don't even come into play.

Note that this is very different from Morrowind. In Morrowind, Scripts could absolutely cripple the game; they were not nearly as efficient as Oblivion's. Modders had to be a lot more cautious about their use and substantial gains could be found by optimizing them. While it is always best practice to optimize your script as much as you can (after all, no point in wasting CPU time, that's just sloppy), bug-free scripts are basically never going to affect your game's performance.

(barring the brick wall effect, which is basically not going to come into play under any actual play scenario; you have to be trying to hit that)

I have some questions for Arthmoor and DragoonWraith.
Why do you both have two oo... no sorry that was not one of the questions
1. Arthmoor knows I wanted to turn off all NPC's low level processing to asses how much CPU drain that entails - I haven't found any automated way to do that yet and I wont click through hundreds, possible thousands of NPC's in all my mods just to test.
So here is one question - With no processing done, that effectively shut down all AI-packages that are not in the same cell; how much improvement in CPU-throughput would the gain be - circa?

2. Now we are back in a normal game. Are the actors in the same cell allot heavier CPU wise than those running their AI-packages in other cells (that is, the rest of the game world)? Se also q4. I am wondering about how much heavier - I know they are heavier.

3. As a script is so much more efficient, maybe when characters have a set schedule just to be different places at times, use a script with moveto instead of having an AI-package that does this with allot more cpu overhead? What I mean is, when you enter a cell the script checks for NPC's schedules and if they should be in the same cell as you, and move them if that is the case.
Please don't tell me it is a silly idea because of the amount of work it would entail to write this script and check all the NPC's packages - I am asking a question and wondering if someone has an answer.

4. DragoonWraith mentioned AI programming. There is a difference between NPC's AI in the same cell and those NPC's running there AI-packeges in other cells right. With the NPC's in the same cell they have to account for collision, decision making, conversation and what not.
Do you got any good suggestions how to best set up NPC's for better throughput CPU-wise?

I am spiking my cores and since the only thing that is truly evil in that area is the NPC's according to many here, I am looking to learn more. Also please note I am looking for information, not to be belittled.

Cheers!

I knew there was a reason I had made a note of this: http://www.uesp.net/wiki/Tes4Mod:Script_Functions

Specifically, you might try these commands:

ToggleLowProcess
ToggleMiddleHighProcess
ToggleMiddleLowProcess

According to that page, these should shut off the low and middle range processing, leaving only the high level AI still on.

The TAI command, as you already pointed out, turns off ALL AI processing. But if you're looking to assess the impact of middle and low level processing, you need to turn those off selectively.

Also, if you want to assess the specific impact of scripts, use the TSCR command to turn off all script processing. Keeping in mind this will kill *ALL* scripts, not just the ones running in GameMode - only the GameMode scripts are relevant to the issue of background processing.

Thanks Arthmoor, you are a champ :foodndrink: