What is the difference between iGPU and dGPU connections for external monitors?

Introduction

On laptops with integrated and dedicated graphics, external display ports may be connected either to the integrated GPU (iGPU) or directly to the dedicated GPU (dGPU).

Both configurations support external monitors, including for demanding games. However, they differ in gaming performance, latency, power consumption and support for certain display technologies.

Summary

Please allow us to explain this in more detail in this article.

What does it mean for a monitor to be connected to the iGPU or dGPU?

Many modern laptops contain two different graphics processors.

The integrated GPU (iGPU) is part of the processor or system-on-chip (SoC). It uses comparatively little power and is particularly well suited to everyday tasks such as office applications, web browsing and video playback.

The dedicated GPU (dGPU), such as an NVIDIA GeForce RTX Laptop GPU, has its own processing units and video memory. It is designed for demanding 3D applications and games but generally consumes significantly more power.

Almost all laptops with a dGPU also have an iGPU. This setup is known as hybrid graphics, in which the two GPUs operate in parallel at certain levels. Two fundamental tasks can be distinguished:

These two tasks do not necessarily have to be handled by the same GPU.

External monitor connected to the iGPU

If a display port is connected to the integrated GPU, the iGPU handles display output.

However, this does not mean that games have to be rendered exclusively by the comparatively less powerful iGPU.

Instead, the dedicated GPU can handle rendering. The completed frames are then transferred to the iGPU via an internal data connection, and the iGPU outputs them to the monitor.

On laptops with NVIDIA graphics, this approach is commonly referred to as NVIDIA Optimus. The general term MSHybrid (Microsoft Hybrid Graphics) is also widely used.

The advantage: When the dedicated GPU is not needed, it can enter a particularly low-power state or go to sleep completely. Meanwhile, the monitor continues to be driven by the iGPU.

The disadvantage: When games are rendered by the dGPU, the completed frames also have to be transferred to the iGPU. This extra step introduces some technical overhead and can affect performance and latency. The performance penalty is most noticeable at high FPS; the latency difference is on the order of a few milliseconds - measurable, but not necessarily noticeable to everyone. These effects are explained in more detail later in this article.

External monitor connected to the dGPU

When a display port is connected directly to the dedicated GPU, that GPU handles both rendering and display output. The rendered frames therefore do not first have to pass through the integrated GPU.

This eliminates the additional transfer step involved in hybrid graphics, potentially resulting in higher frame rates (FPS) and lower latency.

However, the dedicated GPU must remain active for as long as it is driving the external monitor. As a result, power consumption is higher than with an iGPU connection even during simple tasks such as writing text or playing videos.

Technical details

How does GPU connectivity affect gaming performance?

Users who primarily use an external monitor for gaming often benefit from a direct dGPU connection.

This is not because connecting a monitor to the iGPU reduces the dGPU’s available processing power. In principle, the dGPU can still render the game using its full 3D processing capability.

The difference lies in how the rendered frames are transferred and displayed.

FPS and latency

With an iGPU connection, every frame rendered by the dGPU must first be made available to the integrated GPU for display output.

This additional processing can affect both the maximum achievable FPS and latency. Depending on the system, it can also affect frame-time consistency (frametimes).

These differences are particularly relevant in situations where very high frame rates can be achieved:

• Esports titles and other competitive games
• Gaming at low resolutions or reduced graphics settings
• Games in which the GPU is not fully utilised and can render a particularly high number of frames per second

For graphically demanding games at maximum quality settings, the difference is often smaller. Here, the dGPU’s 3D processing power is often the limiting factor for the achievable frame rate.

For this reason, a blanket claim such as “a direct dGPU connection delivers 10% more FPS” would be unjustified. The actual benefit can vary considerably depending on the game and system.

Modern optimisations reduce the drawbacks of hybrid graphics

Hybrid graphics operation has been further optimised over the years.

One example is Cross Adapter Scan-Out (CASO), a Microsoft technology that makes data transfers between the two GPUs more efficient on supported Windows 11 systems.

Earlier implementations sometimes had to copy rendered frames multiple times between different areas of memory. CASO can reduce this overhead by allowing display output to access a shared memory area directly.

This can improve performance and latency in hybrid graphics mode.

However, the fundamental distinction remains: With a direct dGPU connection, frames do not first have to be made available for output through another GPU.

For the highest possible FPS and lowest latency, a display port connected directly to the dGPU therefore remains the recommended option in general. Those who mainly use office applications or only play games occasionally, however, should not regard an iGPU connection as a disadvantage.

How does GPU connectivity affect power consumption?

Power consumption is particularly important in laptops: Higher power draw can not only reduce battery life but also increase heat output and, consequently, fan noise.

When an external monitor is driven by the iGPU, the dedicated GPU can enter sleep mode completely under light loads, provided no other programs or system functions require it to remain active.

This reduces the laptop’s power consumption.

This is particularly beneficial when only simple applications such as web browsers, office software or video playback are used for extended periods.

With a direct dGPU connection, by contrast, the dedicated GPU must remain active to drive the monitor. Although it can still reduce its clock speeds and power consumption, it cannot enter the same deep sleep state as it can when the display is driven solely by the iGPU.

Under completely idle conditions, the difference can sometimes be 20-30 W, depending on the CPU and GPU - a significant amount. There are two reasons why the difference can be so large despite dynamic GPU clock speeds:

• A dGPU is relatively complex and requires a certain amount of power even at idle, whether or not a monitor is connected. This is why it is generally desirable for the dGPU to enter a full sleep state when idle (at which point it consumes zero power).
• An active dGPU also prevents the processor from entering particularly deep package C-states, as the CPU must maintain the PCIe connection to the dGPU. This increases power consumption on the CPU side as well.

The difference is only relevant at idle or under very light load. When a demanding game is running on the dGPU, that GPU is active regardless of how the monitor is connected.

How does GPU connectivity affect G-SYNC, FreeSync and Adaptive Sync?

Variable Refresh Rate (VRR) refers to technologies that allow a monitor to adjust its refresh rate dynamically to match the frames actually being displayed.

This helps prevent distracting screen tearing and can make motion appear smoother.

Well-known implementations include NVIDIA G-SYNC, AMD FreeSync and Intel Adaptive Sync.

Which of these technologies can be used depends, among other factors, on which GPU actually drives the monitor.

In general, all modern iGPUs support these technologies. An exception involves older monitors with NVIDIA’s proprietary first-generation G-SYNC module, which only works with a direct connection to an NVIDIA GPU. We have compiled more information in the following articles:

• Which XMG laptops are compatible with G-SYNC and FreeSync displays?
• What are the differences between NVIDIA G-SYNC, AMD FreeSync and Intel Adaptive Sync?

Why do VR headsets generally require a direct dGPU connection?

A direct connection to the dedicated GPU is particularly important for virtual reality applications.

Wired PC VR headsets receiving their display signal via HDMI or DisplayPort expect a direct connection to the GPU rendering the VR content.

This is partly because VR headsets have different requirements for display output, synchronisation and latency, and often use specialised GPU interfaces. Rather than simply acting as external monitors, they operate as integrated VR systems. This specialised operating mode is also known as “Direct Mode”.

Hybrid graphics output through the iGPU is therefore generally unsuitable for VR headsets. For more information: Which XMG laptops are compatible with VR headsets?

Can you switch between an iGPU and dGPU connection?

The physical connections of the various external display ports are generally determined by the motherboard design. While almost all modern gaming laptops feature a switch (known as a “MUX switch”) for changing how the laptop’s built-in display is connected, this is generally not the case for external ports. The GPU assignment of external display ports therefore cannot be changed through software or firmware.

More information about the MUX switch for the laptop’s built-in display:

• In which XMG laptops can NVIDIA Optimus be disabled?

Difference between GPU selection and monitor connectivity

In Windows, you can specify which GPU individual programs and games should preferably use for rendering. See also: How can I choose whether a program or game runs on the iGPU or dGPU?

However, this setting does not automatically change the physical GPU connection of the display port.

For example, if a program is set to use the NVIDIA GPU while the external monitor is connected to the iGPU, the NVIDIA GPU can still handle rendering. The completed frames are then output via the iGPU.

Changing the GPU selection setting therefore does not turn an iGPU-connected port into a dGPU-connected port.

How can you find out which GPU a monitor is connected to?

For our laptops, the GPU assignment of each display port is listed in the spec sheet. In addition, Windows provides several ways to check which GPU is currently driving a monitor.

Windows display settings

In Windows 11, you can check the monitor’s current GPU connection directly in the system settings:

1. Open Windows Settings.
2. Navigate to System > Display > Advanced display.
3. If multiple displays are connected, select the external monitor you want to check.
4. Under the display information, check which GPU is shown as connected.

If an Intel or AMD GPU is listed, for example, the monitor is currently driven by that iGPU. If an NVIDIA GeForce RTX Laptop GPU is listed instead, display output is handled by the dedicated GPU. This indicates which GPU is currently driving the display, not which GPU a particular game uses for rendering.

NVIDIA App

On laptops with NVIDIA graphics, the NVIDIA App provides another way to check how external monitors are connected.

1. Open the NVIDIA App.
2. Navigate to System > Displays.
3. Check which monitors are detected and driven by the NVIDIA GPU.

The NVIDIA App lists monitors that are connected directly to the NVIDIA GPU in this section. Alongside the monitor names, it also shows information such as the connection type and supported display features.

If the “Displays” section is completely missing from the NVIDIA App, it is likely that no monitor is connected to the dGPU. To confirm this, we recommend checking the Windows display settings as described above.

Conclusion

For office work, web browsing, multimedia and energy-efficient operation, an iGPU connection is generally a very good choice. The dedicated GPU can enter sleep mode under light loads. For gaming with the highest possible FPS and lowest latency, on the other hand, a direct dGPU connection is recommended. This is also usually the preferred option for NVIDIA G-SYNC. Ultimately, the better choice depends on whether maximum graphics performance or minimum power consumption is the priority.