It's simple.
Computers run many tasks all the time, but a single CPU can only run ONE at a time. Normally there are many tasks to perform, more than CPU's or CPU cores, so tasks are queued and are given CPU time in one or another processing scheme. This means that tasks that take too long to finish will have to be constantly queued again and this means that the more tasks there are, the longer standing in line will take and therefore the longer each task will have to wait to be serviced again.
All of these tasks require resources, such as CPU time, memory, disk access, network access, etc. So in a way they also compete for resources. The more tasks there are using the network or the disk, the longer it will take for the whole set to finish.
Operating systems run a lot of background tasks that control the whole computer and use up resources. Add to that user space applications such as text editors, web browsers, or games. Some applications, such as games, require much more dedication than others to function properly. However if there are a lot of applications to run, resources will be split accordingly. Some applications have priority over others, so while we run a game our web browser in the background may not interfere much. However some others remain in the foreground, and demand constant CPU time, memory, disk access, etc.
Most types of modern DRM are like that last type. They require computing resources to launch the game and to keep it running, and essentially take a chunk of computing power away that could be used in making the game run better. In extreme cases you will notice the difference, for instance in framerate. Imagine that while you run your game, another program needs CPU time, dedicated memory, and may even request access to game files which in turn slows down their proper utilization within the game.
If you wanted to measure the effect of a game with and without DRM, you'd have to create a controlled scenario with a computer running a certain amount of background and system processes and then run both versions and measure their performance. But this measure would be dependent on several things, such as the type of DRM used, the type of game, and your own computer. For instance, perhaps a 2D adventure game isn't as heavily affected by a background process as a computing-intensive game like The Witcher 2. In the old days games used simple copy protection such as a CD-key check during installation, so this scheme doesn't interfere with actual gameplay, while others such as SecuROM and Denuvo evidently do. The computer might affect measured performance too, it wouldn't be the same if you run a set of processes that fill up your RAM and required virtual memory (ugh!) vs. a set of processes that barely use up 50% of your memory. Also not the same having more L1 cache with a multicore processor.
Summarizing, the "impact of DRM in game performance" is a function of several variables, and I don't think it can be generalized across games and/or DRM systems. One thing is for sure though, any DRM-related performance loss greater than 0 is already too much, since it is unnecessary and unethical.
Computers run many tasks all the time, but a single CPU can only run ONE at a time. Normally there are many tasks to perform, more than CPU's or CPU cores, so tasks are queued and are given CPU time in one or another processing scheme. This means that tasks that take too long to finish will have to be constantly queued again and this means that the more tasks there are, the longer standing in line will take and therefore the longer each task will have to wait to be serviced again.
All of these tasks require resources, such as CPU time, memory, disk access, network access, etc. So in a way they also compete for resources. The more tasks there are using the network or the disk, the longer it will take for the whole set to finish.
Operating systems run a lot of background tasks that control the whole computer and use up resources. Add to that user space applications such as text editors, web browsers, or games. Some applications, such as games, require much more dedication than others to function properly. However if there are a lot of applications to run, resources will be split accordingly. Some applications have priority over others, so while we run a game our web browser in the background may not interfere much. However some others remain in the foreground, and demand constant CPU time, memory, disk access, etc.
Most types of modern DRM are like that last type. They require computing resources to launch the game and to keep it running, and essentially take a chunk of computing power away that could be used in making the game run better. In extreme cases you will notice the difference, for instance in framerate. Imagine that while you run your game, another program needs CPU time, dedicated memory, and may even request access to game files which in turn slows down their proper utilization within the game.
If you wanted to measure the effect of a game with and without DRM, you'd have to create a controlled scenario with a computer running a certain amount of background and system processes and then run both versions and measure their performance. But this measure would be dependent on several things, such as the type of DRM used, the type of game, and your own computer. For instance, perhaps a 2D adventure game isn't as heavily affected by a background process as a computing-intensive game like The Witcher 2. In the old days games used simple copy protection such as a CD-key check during installation, so this scheme doesn't interfere with actual gameplay, while others such as SecuROM and Denuvo evidently do. The computer might affect measured performance too, it wouldn't be the same if you run a set of processes that fill up your RAM and required virtual memory (ugh!) vs. a set of processes that barely use up 50% of your memory. Also not the same having more L1 cache with a multicore processor.
Summarizing, the "impact of DRM in game performance" is a function of several variables, and I don't think it can be generalized across games and/or DRM systems. One thing is for sure though, any DRM-related performance loss greater than 0 is already too much, since it is unnecessary and unethical.
Last edited:
