A recent Digital Foundry podcast addressed a long-running debate over the Unreal Engine. There is an argument over whether Unreal Engine ini mods actually improve game performance, especially the stuttering problems in the Oblivion Remastered edition.
The team carefully tested two popular mods and has ended this debate, at least for Oblivion. These widely used mods deliver almost no real performance benefits. One of the mods tested was the “Oblivion BSA Uncompressor.” This is meant to reduce stuttering by decompressing game files ahead of time instead of during gameplay. The idea was that real-time decompression puts extra strain on the CPU, which might cause stuttering.
Multiple five-minute benchmark tests in The Imperial City, both with and without the mod, showed almost no difference. After carefully comparing frame times, there was a tiny improvement of about 1.9% in average frame rate with the mod, but this was so small that it could easily be due to normal variations in CPU-heavy benchmarks.
The second mod tested was “Ultimate Engine Tweaks,” a common type of ini file modification that claims to optimize various game settings, including graphics, loading behavior, and how the engine uses multiple CPU cores. The results were very similar to the first test.
Running the game with and without the “Ultimate Engine Tweaks” mod showed almost no difference in frame times or stuttering. Some settings in the ini file were unclear in their purpose, but the ones that could be understood either didn’t matter in the final version of the game or only affected graphical details unrelated to performance.
Even a small increase in average frame rate doesn’t matter if the problem, inconsistent frame times causing stuttering, isn’t fixed. CPU-bound benchmarks naturally have some variability, making tiny percentage improvements meaningless. Basically, the best way to judge these mods is by whether they reduce stuttering, not by a minor boost in average frame rate.
Neither mod made a noticeable difference in gameplay, so there is a huge gap between what the mods claim to do and the actual lack of improvement in testing. Even still, these mods are popular.
Players might test the game in different areas or situations where performance seems better after installing a mod, even if the mod isn’t the real cause. Another reason is shader compilation stutters, which are often confused with general performance problems.
If a player installs a mod after experiencing bad shader stutters, the next play session might seem smoother just because those stutters were temporary and not because the mod helped. This effect is especially common in early DirectX 11 games, which helped create the false belief that these mods work.
This is just Oblivion, so there is always the chance that they do work in other games, but the results make me believe they don’t work at all. If there was some benefit or noticeable improvement, then I’d believe it’d work elsewhere, but there’s none, and this game has a lot of processing in the background. So your mods may just be taking up space.
Source: Digital Foundry
