Tag Archives: Graphics processor

VGA Share Memory Size from The Tech ARP BIOS Guide!

VGA Share Memory Size

Common Options for UMA : 1MB, 4MB, 8MB, 16MB, 32MB, 64MB, 128MB

Common Options for DVMT : 1MB, 8MB

 

VGA Share Memory Size : A Quick Review

The VGA Share Memory Size BIOS feature controls the amount of system memory that is allocated to the integrated graphics processor when the system boots up.

However, its effect depends on whether your motherboard supports the older Unified Memory Architecture (UMA) or the newer Dynamic Video Memory Technology (DVMT).

If you have a motherboard that supports UMA, the memory size you select determines the maximum amount of system memory that is allocated to the graphics processor. Once allocated, it can only be used as graphics memory. It is no longer accessible to the operating system or applications.

Therefore, it is recommended that you select the absolute minimum amount of system memory that the graphics processor requires for your monitor. You can calculate it by multiplying the resolution and colour depth that you are using. Of course, if you intend to play 3D games, you will need to allocate more memory.

If you have a motherboard that supports DVMT, the memory size you select determines the maximum amount of system memory that is pre-allocated to the graphics processor. Once allocated, it can only be used as graphics memory. It is no longer accessible to the operating system or applications.

However, unlike in a UMA system, this memory is only allocated for use during the boot process or with MS-DOS or legacy operating systems. Additional system memory is allocated only after the graphics driver is loaded. It is recommended that you set it to 8MB as this allows for high-resolution splash screens as well as higher resolutions in MS-DOS applications and games.

 

VGA Share Memory Size : The Full Details

Some motherboard chipsets come with an integrated graphics processor. To reduce costs, it usually makes use of UMA (Unified Memory Architecture) or DVMT (Dynamic Video Memory Technology) for its memory requirements.

Both technologies allow the integrated graphics processor to requisition some system memory for use as graphics memory. This reduces cost by obviating the need for dedicated graphics memory. Of course, it has some disadvantages :

  • Allocating system memory to the graphics processor reduces the amount of system memory available for the operating system and programs to use.
  • Sharing system memory with the graphics processor saturates the memory bus and reduces the amount of memory bandwidth for both the processor and the graphics processor.

Therefore, integrated graphics processors are usually unsuitable for high-demand 3D applications and games. They are best used for basic 2D graphics and video functions.

The VGA Share Memory Size BIOS feature controls the amount of system memory that is allocated to the integrated graphics processor when the system boots up.

However, its effect depends on whether your motherboard supports the older Unified Memory Architecture (UMA) or the newer Dynamic Video Memory Technology (DVMT).

If you have a motherboard that supports UMA, the memory size you select determines the maximum amount of system memory that is allocated to the graphics processor. Once allocated, it can only be used as graphics memory. It is no longer accessible to the operating system or applications.

Therefore, it is recommended that you select the absolute minimum amount of system memory that the graphics processor requires for your monitor. You can calculate it by multiplying the resolution and colour depth that you are using.

For example, if you use a resolution of 1600 x 1200 and a colour depth of 32-bits, the amount of graphics memory you require will be 1600 x 1200 x 32-bits = 61,440,000 bits or 7.68 MB.

After doubling that to allow for double buffering, the minimum amount of graphics memory you need would be 15.36 MB. You should set this BIOS feature to 16MB in this example.

Of course, if you intend to play 3D games, you will need to allocate more memory. But please remember that once allocated as graphics memory, it is no longer available to the operating system or applications. You need to balance the performance of your 3D games with that of your operating system and applications.

If you have a motherboard that supports DVMT, the memory size you select determines the maximum amount of system memory that is pre-allocated to the graphics processor. Once allocated, it can only be used as graphics memory. It is no longer accessible to the operating system or applications.

However, unlike in a UMA system, this memory is only allocated for use during the boot process or with MS-DOS or legacy operating systems. Additional system memory is allocated only after the graphics driver is loaded. Therefore, the amount of system memory that can be selected is small – only a choice of 1MB or 8MB.

It is recommended that you set it to 8MB as this allows for high-resolution splash screens as well as higher resolutions in MS-DOS applications and games.

 

Recommended Reading

Go Back To > Tech ARP BIOS GuideComputer | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


PCI-E Max Read Request Size – The Tech ARP BIOS Guide

PCI-E Max Read Request Size

Common Options : Automatic, Manual – User Defined

 

Quick Review of PCI-E Max Read Request Size

This BIOS feature can be used to ensure a fairer allocation of PCI Express bandwidth. It determines the largest read request any PCI Express device can generate. Reducing the maximum read request size reduces the hogging effect of any device with large reads.

When set to Automatic, the BIOS will automatically select a maximum read request size for PCI Express devices. Usually, this would be a manufacturer-preset value that’s designed with maximum “fairness“, rather than performance in mind.

When set to Manual – User Defined, you will be allowed to enter a numeric value (in bytes). Although it appears as though you can enter any value, you must only enter one of these values :

128 – This sets the maximum read request size to 128 bytes. All PCI Express devices will only be allowed to generate read requests of up to 128 bytes in size.

256 – This sets the maximum read request size to 256 bytes. All PCI Express devices will only be allowed to generate read requests of up to 256 bytes in size.

512 – This sets the maximum read request size to 512 bytes. All PCI Express devices will only be allowed to generate read requests of up to 512 bytes in size.

1024 – This sets the maximum read request size to 1024 bytes. All PCI Express devices will only be allowed to generate read requests of up to 1024 bytes in size.

2048 – This sets the maximum read request size to 2048 bytes. All PCI Express devices will only be allowed to generate read requests of up to 2048 bytes in size.

4096 – This sets the maximum read request size to 4096 bytes. This is the largest read request size currently supported by the PCI Express protocol. All PCI Express devices will be allowed to generate read requests of up to 4096 bytes in size.

It is recommended that you set this BIOS feature to 4096, as it maximizes performance by allowing all PCI Express devices to generate as large a read request as they require. However, this will be at the expense of devices that generate smaller read requests.

Even so, this is generally not a problem unless they require a certain degree of quality of service. For example, you may experience glitches with the audio output (e.g. stuttering) of a PCI Express sound card when its reads are delayed by a bandwidth-hogging graphics card.

If such problems arise, reduce the maximum read request size. This reduces the amount of bandwidth any PCI Express device can hog at the expense of the other devices.

 

Details of PCI-E Max Read Request Size

Arbitration for PCI Express bandwidth is based on the number of requests from each device. However, the size of each request is not taken into account. As such, if some devices request much larger data reads than others, the PCI Express bandwidth will be unevenly allocated between those devices.

This can cause problems for applications that have specific quality of service requirements. These application may not have timely access to the requested data simply because another PCI Express device is hogging the bandwidth by requesting for very large data reads.

This BIOS feature can be used to correct that and ensure a fairer allocation of PCI Express bandwidth. It determines the largest read request any PCI Express device can generate. Reducing the maximum read request size reduces the hogging effect of any device with large reads.

However, doing so reduces the performance of devices that generate large reads. Instead of generating large but fewer reads, they will have to generate smaller reads but in greater numbers. Because arbitration is done according to the number of requests, they will have to wait longer for the data requested.

[adrotate group=”1″]

When set to Automatic, the BIOS will automatically select a maximum read request size for PCI Express devices. Usually, this would be a manufacturer-preset value that’s designed with maximum “fairness“, rather than performance in mind.

When set to Manual – User Defined, you will be allowed to enter a numeric value (in bytes). Although it appears as though you can enter any value, you must only enter one of these values :

128 – This sets the maximum read request size to 128 bytes. All PCI Express devices will only be allowed to generate read requests of up to 128 bytes in size.

256 – This sets the maximum read request size to 256 bytes. All PCI Express devices will only be allowed to generate read requests of up to 256 bytes in size.

512 – This sets the maximum read request size to 512 bytes. All PCI Express devices will only be allowed to generate read requests of up to 512 bytes in size.

1024 – This sets the maximum read request size to 1024 bytes. All PCI Express devices will only be allowed to generate read requests of up to 1024 bytes in size.

2048 – This sets the maximum read request size to 2048 bytes. All PCI Express devices will only be allowed to generate read requests of up to 2048 bytes in size.

4096 – This sets the maximum read request size to 4096 bytes. This is the largest read request size currently supported by the PCI Express protocol. All PCI Express devices will be allowed to generate read requests of up to 4096 bytes in size.

It is recommended that you set this BIOS feature to 4096, as it maximizes performance by allowing all PCI Express devices to generate as large a read request as they require. However, this will be at the expense of devices that generate smaller read requests.

Even so, this is generally not a problem unless they require a certain degree of quality of service. For example, you may experience glitches with the audio output (e.g. stuttering) of a PCI Express sound card when its reads are delayed by a bandwidth-hogging graphics card.

If such problems arise, reduce the maximum read request size. This reduces the amount of bandwidth any PCI Express device can hog at the expense of the other devices.

Go Back To > The Tech ARP BIOS Guide | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

NVIDIA GPU Ex from The Tech ARP BIOS Guide

NVIDIA GPU Ex

Common Options : Auto, Enabled, Disabled

 

Quick Review of NVIDIA GPU Ex

NVIDIA GPU Ex is an NVIDIA chipset-specific BIOS feature. The name is purportedly a truncation of “NVIDIA GPU Extra“, hinting that it improves the performance of NVIDIA graphics cards.

According to NVIDIA, this BIOS option is a toggle for chipset-specific optimizations that improve the performance of NVIDIA graphics cards when NVIDIA GPU Ex graphics drivers are used. All NVIDIA ForceWare Release 90 or newer graphics driver support NVIDIA GPU Ex.

When set to Enabled, the BIOS will enable chipset optimizations for NVIDIA graphics cards when used with graphics drivers with NVIDIA GPU Ex support.

When set to Disabled, the BIOS will disable chipset optimizations for NVIDIA graphics cards.

When set to Auto, the BIOS will enable the chipset optimizations when it detects the presence of an NVIDIA graphics card when the computer boots up.

Benchmarking by enthusiasts have shown that enabling this BIOS feature resulted in very insignificant performance improvements (< 2%). In some cases, enabling it actually resulted in a significant drop in performance.

Therefore, we recommend that you disable this BIOS option, unless you are willing to test the effect of this BIOS option and confirm that it actually improves the performance of your NVIDIA graphics card. You should also disable this BIOS option if you are using an graphics card that is not based on an NVIDIA GPU.

[adrotate group=”1″]

 

Details of NVIDIA GPU Ex

NVIDIA GPU Ex is an NVIDIA chipset-specific BIOS feature. The name is purportedly a truncation of “NVIDIA GPU Extra“, hinting that it improves the performance of NVIDIA graphics cards.

According to NVIDIA, this BIOS option is a toggle for chipset-specific optimizations that improve the performance of NVIDIA graphics cards when NVIDIA GPU Ex graphics drivers are used. All NVIDIA ForceWare Release 90 or newer graphics driver support NVIDIA GPU Ex.

When enabled, the BIOS will enable chipset optimizations for NVIDIA graphics cards when used with graphics drivers with NVIDIA GPU Ex support.

When set to Disabled, the BIOS will disable chipset optimizations for NVIDIA graphics cards.

When set to Auto, the BIOS will enable the chipset optimizations when it detects the presence of an NVIDIA graphics card when the computer boots up.

Benchmarking by enthusiasts have shown that enabling this BIOS feature resulted in very insignificant performance improvements (< 2%). In some cases, enabling it actually resulted in a significant drop in performance.

Therefore, we recommend that you disable this BIOS option, unless you are willing to test the effect of this BIOS option and confirm that it actually improves the performance of your NVIDIA graphics card. You should also disable this BIOS option if you are using an graphics card that is not based on an NVIDIA GPU.

Go Back To > The Tech ARP BIOS Guide | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


DVMT Mode from The Tech ARP BIOS Guide

DVMT Mode

Common Options : Fixed, DVMT, Both

 

Quick Review of DVMT Mode

Unified Memory Architecture (UMA) is a concept whereby system memory is shared by both CPU and graphics processor. While this reduces cost, it also reduces the system’s performance by taking up a large portion of memory for the graphics processor.

Intel’s Dynamic Video Memory Technology (DVMT) takes that concept further by allowing the system to dynamically allocate memory resources according to the demands of the system at any point in time. The key idea in DVMT is to improve the efficiency of the memory allocated to either system or graphics processor.

The BIOS feature that controls all this is the DVMT Mode BIOS feature. It allows you to select the DVMT operating mode.

When set to Fixed, the graphics driver will reserve a fixed portion of the system memory as graphics memory. This ensures that the graphics processor has a guaranteed amount of graphics memory but the downside is once allocated, this memory cannot be used by the operating system even when it is not in use.

When set to DVMT, the graphics chip will dynamically allocate system memory as graphics memory, according to system and graphics requirements. The system memory is allocated as graphics memory when graphics-intensive applications are running but when the need for graphics memory drops, the allocated graphics memory can be released to the operating system for other uses.

When set to Both, the graphics driver will allocate a fixed amount of memory as dedicated graphics memory, as well as allow more system memory to be dynamically allocated between the graphics processor and the operating system.

It is recommended that you set this BIOS feature to DVMT for maximum performance. Setting it to DVMT ensures that system memory is dynamically allocated for optimal balance between graphics and system performance.

 

Details of DVMT Mode

Unified Memory Architecture (UMA) is a concept whereby system memory is shared by both CPU and graphics processor. While this reduces cost, it also reduces the system’s performance by taking up a large portion of memory for the graphics processor.

Intel’s Dynamic Video Memory Technology (DVMT) takes that concept further by allowing the system to dynamically allocate memory resources according to the demands of the system at any point in time. The key idea in DVMT is to improve the efficiency of the memory allocated to either system or graphics processor.

To ensure better allocation of system memory, DVMT comes with three different operating modes :

[adrotate group=”2″]
  • Fixed Memory
  • DVMT Memory
  • Fixed + DVMT Memory

But before we go into the details of each mode, it’s important to note that the system boots up with some system memory pre-allocated for graphics, irrespective of the DVMT operating mode. Depending on the motherboard manufacturer, you may or may not be allowed to select between a choice of 1MB or 8MB of pre-allocated memory.

This pre-allocated memory is dedicated to VGA/SVGA graphics and will be treated by the operating system as dedicated graphics memory. This pre-allocated memory will not be visible or accessible to the operating system. It will be used during the booting process to display the boot and splash screens, or when you run MS-DOS games and applications. It will also be used when Windows XP loads in Safe Mode.

Once an operating system with the appropriate Intel Graphics Media Accelerator Driver loads up, the graphics processor reclaims the pre-allocated memory for its use. But again, it is only available for use as graphics memory. It will never be made available to the operating system or applications. The Intel GMA driver then loads additional system memory according to the DVMT operating mode.

The Fixed Memory operating mode reserves a fixed amount of system memory as graphics memory. This is in addition to the memory already pre-allocated. Like pre-allocated memory, this fixed amount is no longer available to the operating system. But when the operating system reports the total system memory, it will include this amount, as opposed to pre-allocated memory.

The DVMT Memory operating mode allows the graphics driver to dynamically allocate system memory for use by the graphics processor. When no graphics-intensive operations are occuring, most of the DVMT memory can be reallocated to the operating system for other uses. When more graphics memory is required, the graphics driver will automatically reallocate more system memory for use as graphics memory.

The Fixed + DVMT Memory operating mode is an combination of the Fixed and DVMT operating modes. It allows you to allocate a fixed amount of reserved graphics memory (over the minimum pre-allocated amount), as well as a portion of system memory that can be dynamically allocated to both graphics processor and operating system.

This figure from Intel clearly shows the differences between the three different DVMT operating modes :

The BIOS feature that controls all this is the DVMT Mode BIOS feature. It allows you to select the DVMT operating mode.

When set to Fixed, the graphics driver will reserve a fixed portion of the system memory as graphics memory. This ensures that the graphics processor has a guaranteed amount of graphics memory.

But the downside is, once allocated, this memory cannot be used by the operating system even when it is not in use. Usually, the following configuration scheme is used :

System Memory DVMT Graphics Memory
Pre-Allocated Fixed Total
128 – 255 MB 1 MB 31 MB 32 MB
8 MB 24 MB
256 – 511 MB 1 MB 63 MB 64 MB
8 MB 56 MB
1 MB 127 MB 128 MB
8 MB 120 MB
512 MB and larger 1 MB 63 MB 64 MB
8 MB 56 MB
1 MB 127 MB 128 MB
8 MB 120 MB

When set to DVMT, the graphics chip will dynamically allocate system memory as graphics memory, according to system and graphics requirements. The system memory is allocated as graphics memory when graphics-intensive applications are running.

But when the need for graphics memory drops, the allocated graphics memory can be released to the operating system for other uses. Usually, the following configuration scheme is used :

System Memory DVMT Graphics Memory
Pre-Allocated Fixed Total
128 – 255 MB 1 MB 31 MB 32 MB
8 MB 24 MB
256 – 511 MB 1 MB 63 MB 64 MB
8 MB 56 MB
1 MB 127 MB 128 MB
8 MB 120 MB
1 MB 159 MB 160 MB
8 MB 152 MB
512 MB and larger 1 MB 63 MB 64 MB
8 MB 56 MB
1 MB 127 MB 128 MB
8 MB 120 MB
1 MB 223 MB 224 MB
8 MB 216 MB

When set to Both, the graphics driver will allocate a fixed amount of memory as dedicated graphics memory, as well as allow more system memory to be dynamically allocated between the graphics processor and the operating system. Usually, the following configuration scheme is used :

System Memory DVMT Graphics Memory
Pre-Allocated Fixed Total
128 – 255 MB NA
256 and larger 1 MB 63 MB + 64 MB 128 MB
8 MB 56 MB + 64 MB

It is recommended that you set this BIOS feature to DVMT for maximum performance. Setting it to DVMT ensures that system memory is dynamically allocated for optimal balance between graphics and system performance.

Go Back To > The Tech ARP BIOS Guide | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


Video RAM Cacheable – The Tech ARP BIOS Guide

Video RAM Cacheable

Common Options : Enabled, Disabled

 

Quick Review of Video RAM Cacheable

The Video RAM Cacheable feature aims to boost VGA graphics performance by using the processor’s Level 2 cache to cache the 64 KB VGA graphics memory area from A0000h to AFFFFh.

If this BIOS feature is enabled, the VGA graphics memory area will be cached by the processor’s Level 2 cache. This speeds up accesses to the VGA graphics memory area.

If this BIOS feature is disabled, the VGA graphics memory area will not be cached by the processor’s Level 2 cache.

From what we have discussed so far, it sounds like caching the VGA graphics memory area is logically the way to go. Caching the VGA graphics memory area will definitely speed up VGA graphics performance by caching accesses to the graphics memory area.

However, reality is far less ideal. For one thing, VGA modes are hardly used at all these days. For compatibility reason, VGA is still used in Windows XP’s Safe Mode. It is also used in real mode DOS, if you still use that. Other than that, there is no more use for VGA modes. If VGA graphics modes are not used, no benefit can possibly be realized by enabling this BIOS feature.

Even if you use DOS modes a lot, is there even a point in caching the VGA graphics memory area for better performance? Even the slowest computer today is more than capable of handling VGA graphics with ease. In short, caching the VGA graphics memory area will not bring any noticeable advantage.

On the other hand, caching this memory area will cost you some processor performance. Because some of the processor’s Level 2 cache is being diverted to cache the VGA graphics memory area, there is less to keep the processor supplied with data. Consequently, the processor’s performance suffers.

Therefore, it is highly recommended that you disable the Video RAM Cacheable feature. There is no reason to enable it even if you use real mode DOS a lot or work a lot in Windows Safe Mode.

 

Details of Video RAM Cacheable

The Upper Memory Area (UMA) is a 384KB block of memory at the top of the first megabyte of memory that is reserved for the system’s use in DOS. A portion of this Upper Memory Area is reserved as video RAM memory.

The video RAM memory area is a 128KB block from A0000h to BFFFFh. Of this 128KB, the first half (A0000h-AFFFFh) is reserved for use in VGA graphics mode. The other half is used for monochrome text mode (B0000h-B7FFFh) and colour text mode (B8000h-BFFFFh). This video RAM memory area is the only portion of the graphics card’s memory that the processor has direct access to in VGA mode.

The graphics card and the processor use this memory area to write pixel data when the computer is operating in VGA mode. This is why all VGA graphics modes take up less than 64KB of memory. The most common VGA mode is mode 0x13 which has a resolution of 320 x 200 in 256 colours. This mode uses up exactly 64,000 bytes of memory and fits nicely into the 64KB block from A0000h to AFFFFh.

The Video RAM Cacheable feature aims to boost VGA graphics performance by using the processor’s Level 2 cache to cache the 64 KB VGA graphics memory area from A0000h to AFFFFh.

If this BIOS feature is enabled, the VGA graphics memory area will be cached by the processor’s Level 2 cache. This speeds up accesses to the VGA graphics memory area.

If this BIOS feature is disabled, the VGA graphics memory area will not be cached by the processor’s Level 2 cache.

From what we have discussed so far, it sounds like caching the VGA graphics memory area is logically the way to go. Caching the VGA graphics memory area will definitely speed up VGA graphics performance by caching accesses to the graphics memory area. This is great for those old DOS games although it won’t do anything for VGA text modes.

However, reality is far less ideal. For one thing, VGA modes are hardly used at all these days. For compatibility reason, VGA is still used in Windows XP’s Safe Mode. It is also used in real mode DOS, if you still use that. Other than that, there is no more use for VGA modes. If VGA graphics modes are not used, no benefit can possibly be realized by enabling this BIOS feature.

Even if you use DOS modes a lot, is there even a point in caching the VGA graphics memory area for better performance? Even the slowest computer today is more than capable of handling VGA graphics with ease. In short, caching the VGA graphics memory area will not bring any noticeable advantage.

On the other hand, caching this memory area will cost you some processor performance. Because some of the processor’s Level 2 cache is being diverted to cache the VGA graphics memory area, there is less to keep the processor supplied with data. Consequently, the processor’s performance suffers.

[adrotate group=”2″]

If the use of the processor’s Level 2 cache can bring about significant improvement in the performance of the graphics subsystem, it would have been worth it. Unfortunately, the VGA graphics modes are rarely used at all.

Even when used, there is little or no real benefit in caching the memory area. The Video RAM Cacheable BIOS feature just wastes the processor’s Level 2 cache on something that cannot possibly improve the system’s graphics performance.

Therefore, it is highly recommended that you disable the Video RAM Cacheable feature. There is no reason to enable it even if you use real mode DOS a lot or work a lot in Windows Safe Mode.

Go Back To > The Tech ARP BIOS Guide | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


PCI Express Burn-in Mode – The BIOS Optimization Guide

PCI Express Burn-in Mode

Common Options : Default, 101.32MHz, 102.64MHz, 103.96MHz, 105.28MHz, 106.6MHz, 107.92MHz, 109.24MHz

 

Quick Review

The PCI Express Burn-in Mode BIOS feature allows you to overclock the PCI Express bus, even if Intel stamps its foot petulantly and insist that it is not meant for this purpose. While it does not give you direct control of the bus clocks, it allows some overclocking of the PCI Express bus.

When this BIOS feature is set to Default, the PCI Express bus runs at its normal speed of 33MHz.

When this BIOS feature is set to 101.32MHz, the PCI Express bus runs at a higher speed of 101.32MHz.

When this BIOS feature is set to 102.64MHz, the PCI Express bus runs at a higher speed of 102.64MHz.

When this BIOS feature is set to 103.96MHz, the PCI Express bus runs at a higher speed of 103.96MHz.

When this BIOS feature is set to 105.28MHz, the PCI Express bus runs at a higher speed of 105.28MHz.

When this BIOS feature is set to 106.6MHz, the PCI Express bus runs at a higher speed of 106.6MHz.

When this BIOS feature is set to 107.92MHz, the PCI Express bus runs at a higher speed of 107.92MHz.

When this BIOS feature is set to 109.24MHz, the PCI Express bus runs at a higher speed of 109.24MHz.

For better performance, it is recommended that you set this BIOS feature to 109.24MHz. This overclocks the PCI Express bus by about 9%, which should not cause any stability problems with most PCI Express devices. But if you encounter any stability issues, use a lower setting.

[adrotate group=”1″]

 

Details of PCI Express Burn-in Mode

While many motherboard manufacturers allow you to overclock various system clocks, Intel officially does not condone or support overclocking. Therefore, motherboards sold by Intel lack BIOS features that allow you to directly modify bus clocks.

However, some Intel motherboards come with a PCI Express Burn-in Mode BIOS feature. This ostensibly allows you to “burn-in” PCI Express devices with a slightly higher bus speed before settling back to the normal bus speed.

Of course, you can use this BIOS feature to overclock the PCI Express bus, even if Intel stamps its foot petulantly and insist that it is not meant for this purpose. While it does not give you direct control of the bus clocks, it allows some overclocking of the PCI Express bus.

When this BIOS feature is set to Default, the PCI Express bus runs at its normal speed of 33MHz.

When this BIOS feature is set to 101.32MHz, the PCI Express bus runs at a higher speed of 101.32MHz.

When this BIOS feature is set to 102.64MHz, the PCI Express bus runs at a higher speed of 102.64MHz.

When this BIOS feature is set to 103.96MHz, the PCI Express bus runs at a higher speed of 103.96MHz.

When this BIOS feature is set to 105.28MHz, the PCI Express bus runs at a higher speed of 105.28MHz.

When this BIOS feature is set to 106.6MHz, the PCI Express bus runs at a higher speed of 106.6MHz.

When this BIOS feature is set to 107.92MHz, the PCI Express bus runs at a higher speed of 107.92MHz.

When this BIOS feature is set to 109.24MHz, the PCI Express bus runs at a higher speed of 109.24MHz.

As you can see, this BIOS feature doesn’t allow much play with the clock speed. You can only adjust the clock speeds upwards by about 9%.

For better performance, it is recommended that you set this BIOS feature to 109.24MHz. This overclocks the PCI Express bus by about 9%, which should not cause any stability problems with most PCI Express devices. But if you encounter any stability issues, use a lower setting.

Go Back To > The BIOS Optimization Guide | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

PAVP Mode – The BIOS Optimization Guide

PAVP Mode

Common Options : Paranoid, Lite, Disabled

 

PAVP Mode Quick Review

PAVP (Protected Audio Video Path) controls the hardware-accelerated decoding of encrypted video streams by Intel integrated graphics processors. Intel offers two PAVP modes – Paranoid and Lite.

When set to Paranoid, the video stream is encrypted and its decoding is accelerated by the integrated graphics processor. In addition, 96 MB of system memory will be reserved exclusively for use by PAVP.

When set to Lite, the video stream is encrypted and its decoding is accelerated by the integrated graphics processor. No system memory will be reserved for use by PAVP.

When set to Disabled, the hardware-accelerated decoding of video content protected by HDCP is disabled.

If you wish to play HDCP-protected content, you should select the Lite option. It allows hardware-accelerated decoding of the video stream. The graphics core will grab system memory for use by PAVP only when it is needed and release it after use.

The allocation of PAVP stolen memory may be necessary to allow some applications to stream lossless audio formats like Dolby TrueHD or DTS-HS MA. In such cases, you will need to set the PAVP Mode BIOS option to Paranoid. However, this takes up 96 MB of system memory and also disables the Windows Aero interface.

You should only use the Disabled setting if you intend to use an external graphics card to accelerate the decoding of the video stream, or if you wish to test the ability of the CPU to handle decryption of the video stream.

 

PAVP Mode Details

PAVP (Protected Audio Video Path) is a feature available on some Intel chipsets with integrated graphics. It ensures a secure content protection path for high-definition video sources like Blu-ray discs. It also controls the hardware-accelerated decoding of encrypted video streams by the integrated graphics processor.

Intel offers two PAVP modes – Paranoid and Lite. Here is a table that summarizes the difference between the two modes :

Feature

PAVP Paranoid

PAVP Lite

Compressed video buffer is encrypted

Yes

Yes

Hardware acceleration of 128-bit AES decryption

Yes

Yes

Protected memory (96 MB reserved during boot)

Yes

No

In other words, the two modes only differ in whether 96 MB of system memory should be reserved for use by PAVP.

When set to Paranoid, the video stream is encrypted and its decoding is accelerated by the integrated graphics processor. In addition, 96 MB of system memory will be reserved exclusively for use by PAVP. This reserved memory (also known as the PAVP Stolen Memory) will not be visible to the operating system or applications.

When set to Lite, the video stream is encrypted and its decoding is accelerated by the integrated graphics processor. No system memory will be reserved for use by PAVP.

When set to Disabled, the hardware-accelerated decoding of video content protected by HDCP is disabled.

If you wish to play HDCP-protected content, you should select the Lite option. It allows hardware-accelerated decoding of the video stream. The graphics core will grab system memory for use by PAVP only when it is needed and release it after use.

The allocation of PAVP stolen memory may be necessary to allow some applications to stream lossless audio formats like Dolby TrueHD or DTS-HS MA. In such cases, you will need to set the PAVP Mode BIOS option to Paranoid. However, this takes up 96 MB of system memory and also disables the Windows Aero interface.

You should only use the Disabled setting if you intend to use an external graphics card to accelerate the decoding of the video stream, or if you wish to test the ability of the CPU to handle decryption of the video stream.

Go Back To > The BIOS Optimization Guide | Home

[adrotate group=”1″]

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

Watch AMD Vega Run DOOM On Vulkan!

AMD recently revealed architectural details of the upcoming Vega GPU, but many people were disappointed that AMD is not ready to launch it yet. Will it matter how fantastic AMD Vega is on paper if it’s nowhere close to reality? The good news is AMD appears to be close to the final silicon.

In fact, at the AMD Tech Summit in Sonoma, we were shown an AMD Vega prototype running DOOM on Vulkan. Billy Khan from id Software also came to vouch for the performance and stability of the latest Vega silicon.

 

Watch AMD Vega Run DOOM On Vulkan!

AMD showed off this Vega prototype running DOOM at the 4K resolution of 3840 x 2160, using the Ultra Quality preset. DOOM was running with the Vulkan upgrade, of course, which allows for asynchronous compute. The video shows the Vega prototype deliver frame rates of 60-70 fps, with an average of 65 fps.

We tried to take a closer look inside the chassis to catch a look of the AMD Vega graphics card, but it appears to be well-shielded from our view. All we can say is that the card appears to be rather quiet… at least it was not audible in the hubbub of the room.

 

id Software On AMD Vega And Vulkan

Earlier that day, Billy Khan (Lead Project Programmer of id Tech 6 and DOOM) spoke about AMD Vega and Vulkan. Here are the key takeaway points :

[adrotate group=”2″]
  • Vulkan brings “close to the metal” coding to the PC
  • This allows for closer development of console and PC games
  • Micro-optimisations are now very easy to port from consoles to PC
  • They have been running DOOM at 4K on super high quality at over 70 fps on an early (few weeks old) AMD Vega silicon

We came away with the feeling that Billy was very impressed with the performance and stability of the Vega GPU. Hopefully, this will assuage the frustrations of AMD fans who are eager to see the AMD Vega to take on the NVIDIA Pascal…

 

The AMD Vega Launch Date

Right now, AMD will not reveal how close they are to the final silicon. Only that they aim to launch Vega in the first half of 2017.

For more information on AMD Vega, take a look at our AMD Vega GPU Architecture Tech Report.

Go Back To > Computer Hardware + Systems | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

Video Memory Cache Mode – BIOS Optimization Guide

Video Memory Cache Mode

Common Options : USWC, UC

 

Quick Review

Video Memory Cache Mode is yet another BIOS feature with a misleading name. It does not cache the video memory or even graphics data (such data is uncacheable anyway).

This BIOS feature allows you to control the USWC (Uncached Speculative Write Combining) write combine buffers.

When set to USWC, the write combine buffers will accumulate and combine partial or smaller graphics writes from the processor and write them to the graphics card as burst writes.

When set to UC, the write combine buffers will be disabled. All graphics writes from the processor will be written to the graphics card directly.

It is highly recommended that you set the Video Memory Cache Mode option to USWC for improved graphics and processor performance.

However, if you are using an older graphics card, it may not be compatible with this feature. Enabling this feature with such graphics cards will cause a host of problems like graphics artifacts, system crashes and even the inability to boot up properly.

If you face such problems, you should set this BIOS feature to UC immediately.

 

Details

Video Memory Cache Mode is yet another BIOS feature with a misleading name. It does not cache the video memory or even graphics data (such data is uncacheable anyway). It is actually similar to the USWC Write Posting BIOS feature.

Current processors are heavily optimized for burst operations which allows for very high memory bandwidth. Unfortunately, graphics writes from the processor are mostly pixel writes which are 8 to 32-bits in nature. Because they do not fill up an entire cache line, such writes are not burstable. This results in poor graphics write performance.

To correct this deficiency, processors now come with one or more internal write combine buffers. These buffers are designed to accumulate graphics writes from the processor. These partial or smaller writes are then combined and written to the graphics card as burst writes.

The use of these internal write combine buffers provides many benefits :-

  1. Partial or smaller graphics writes from the processor are now combined into burstable writes. This greatly increases the performance of the processor and AGP (or PCI) buses.
  2. Graphics writes will require fewer transactions on the processor and AGP (or PCI) bus. This improves the bandwidth of those buses.
  3. The processor will only need write to its internal write combine buffers, instead of the processor bus. This improves its performance by allowing it to work on other tasks while the write combine buffers handle the actual write transaction.

Because the write combine buffers allow speculative reads, this feature is known as the USWC (Uncached Speculative Write Combining) feature. The older method of writing all processor writes directly to the graphics card is known as UC (UnCached).

[adrotate banner=”5″]

This BIOS feature allows you to control the USWC (Uncached Speculative Write Combining) write combine buffers.

When set to USWC, the write combine buffers will accumulate and combine partial or smaller graphics writes from the processor and write them to the graphics card as burst writes.

When set to UC, the write combine buffers will be disabled. All graphics writes from the processor will be written to the graphics card directly.

It is highly recommended that you set the Video Memory Cache Mode option to USWC for improved graphics and processor performance.

Please note that this feature must also be supported by the graphics card, the operating system and the graphics driver for it to work properly.

All Microsoft operating systems from Windows NT 4.0 onwards support USWC, so you do not need to worry if you are using a Windows NT 4.0 or newer operating system from Microsoft. As this feature has been around for some time, drivers of USWC-compatible graphics cards will fully support this feature.

However, if you are using an older graphics card, it may not be compatible with this feature. Older graphics cards make use of a FIFO (First In, First Out) I/O model which can only support the UnCached (UC) type of transaction. Enabling this feature with such graphics cards will cause a host of problems like graphics artifacts, system crashes and even the inability to boot up properly.

If you face such problems, you should set this BIOS feature to UC immediately.

[adrotate banner=”5″]

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participate in the Tech ARP Forums, or even donate to our fund. Any help you can render is greatly appreciated!

AMD Radeon RX Series Pre-Launched @ E3

Los Angeles, California, 13 June 2016 — Today at Electronic Entertainment Expo (E3) AMD CEO Lisa Su delivered a pre-launch showcase of the full line of forthcoming Radeon RX Series graphics cards set to transform PC gaming this summer by delivering enthusiast class performance and features for gamers at mainstream price points.

AMD previously showcased the Radeon RX 480 graphics card, designed for incredibly smooth AAA gaming at 1440p resolution and set to be the most affordable solution for premium VR experiences starting at just $199 SEP for the 4GB version.

Joining the Radeon RX family are the newly announced Radeon RX 470 graphics card delivering refined, power-efficient HD gaming, and the Radeon RX 460, a cool and efficient solution for the ultimate e-sports gaming experience.

 

Radeon RX Series

The Radeon RX Series of graphics processors are designed to transform the PC gaming industry across a variety of form factors, delivering on three fundamental “entitlements” for gamers and game developers:

  • Extraordinary VR experiences at price points never offered before – Previewed at Computex, the Radeon RX Series will expand the VR ecosystem by democratizing exceptional VR experiences, making them available to many form factors and millions of consumers by lowering the cost barriers to entry.
  • Great game content delivered to PC Gamers in real time – Through a combination of Radeon RX Series performance profiles and close-to-the-metal APIs that closely mirror console APIs, AMD believes that developers will be further empowered to co-develop high quality, high performing game content for both consoles and PCs, enhancing the PC gaming ecosystem.
  • Console-class GPU performance for thin and light notebooks – Gaming notebooks have traditionally been large and cumbersome or under-powered for today’s gaming needs. The Radeon RX Series addresses this with flagship technology that effectively gives mobile users GPU performance that rivals that of consoles with exceptionally low power and low-z height to drive thin, light and high-performance gaming notebooks, and 1080p 60Hz gaming experiences for both eSports and AAA titles.

“Gamers and consumers today are being left behind,” said Raja Koduri, senior vice president and chief architect, Radeon Technologies Group, AMD. “Today only the top 16 percent of PC gamers are purchasing GPUs that deliver premium VR and Gaming experiences.2 Hundreds of millions of gamers have been relegated to using outdated technology. Notebook gamers are often forced to compromise. And tens of millions more can only read about incredible PC VR experiences that they can’t enjoy for themselves. That all changes with the Radeon RX Series, placing compelling and advanced high-end gaming and VR technologies within reach of everyone.”

[adrotate banner=”5″]

Gamers in the market for a new graphics card need look no further than the forthcoming Radeon RX Series, consisting of:

  • Superior technology engineered for unprecedented performance– The Radeon RX Series features the most advanced graphics and gaming technologies ever seen in a GPU priced under $300 SEP, delivering cutting-edge engineering to everyday PC gamers and VR consumers. The Radeon RX Series harnesses the revolutionary Polaris architecture optimized for the 14nm FinFET process, the most cutting-edge process technology in the world featuring the smallest transistors ever used in a GPU, engineered to deliver unprecedented performance and power efficiency from incredibly small and thin chips.
  • Extraordinary VR experiences never widely affordable before – With models starting at $199 SEP, the Radeon RX 480 is the most affordable solution for a premium VR experience, supplying the graphics capability necessary to bring high-quality PC VR experiences from Oculus and HTC3 to anyone who wants it.
  • Future-proof technologies1 – The Radeon  RX Series continues the Radeon tradition of innovation, like being first to 14nm FinFET process technology, first in memory types and bandwidth like HBM, and first to support low overhead gaming APIs. Gamers will enjoy these products for a long time to come with a range of “future-proof” benefits including:
    •  Leading DirectX 12 and Vulkan gaming – The Polaris architecture-fueled Radeon RX Series is built to deliver phenomenal DirectX 9, DirectX 10, and DirectX 11 gaming performance, and designed to absolutely scream in DirectX 12 and Vulkan, the future of gaming. Polaris architecture uniquely supports asynchronous compute for superior experiences in games and VR applications using DirectX 12 and Vulkan. AMD brings gamers incredible DirectX 12 and Vulkan game experiences including phenomenal VR content, by collaborating with the top DirectX 12 and Vulkan developers in the world who want to develop on Radeon to bring the best games to market.
    •  Next-generation display technologies – Radeon RX Series includes support for next-generation HDR gaming and video on new HDR monitors and TVs. The Radeon RX Series also supports HDMI 2.0b and DisplayPort 1.3/1.4 supporting the new generation of high-resolution HDR and high-refresh displays. The Radeon RX Series features exceptional accelerated H.265 encoding and decoding, enabling effortless streaming or recording of 10-bit 4K video at 60 FPS4.
    •  Radeon Software designed to provide the best performance, features, stability and control – Equally as sophisticated as the Radeon RX Series graphics cards is the software that powers them. Radeon Software enables the ultimate in performance, features and stability to ensure an exceptionally smooth and fast out-of-box experience, and one that gets better with age as updates roll out.

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participate in the Tech ARP Forums, or even donate to our fund. Any help you can render is greatly appreciated!

Direct Frame Buffer – BIOS Optimization Guide

Direct Frame Buffer

Common Options : Enabled, Disabled

 

Quick Review

The Direct Frame Buffer BIOS feature controls the processor’s access to the section of system memory reserved for use by the integrated graphics processor as graphics memory. Please note that we are referring to the CPU, not the graphics processor.

When enabled, the processor is allowed to directly write to the section of system memory reserved as graphics memory. This increases the performance of applications that write directly to the frame buffer.

When disabled, the processor is not allowed to directly write to the section of system memory reserved as graphics memory. This reduces the performance of applications that write directly to the frame buffer.

It is recommended that you enable this BIOS feature for maximum performance. Of course, this only improves performance of applications that write directly to the frame buffer.

 

Details

This BIOS feature is found in VIA-based motherboards with integrated graphics processors. The integration of the graphics processor into the motherboard chipset reduces the cost of building the PC.

To further reduce cost, the integrated graphics processor does not come with dedicated graphics memory. Instead, part of system memory is cordoned off and used exclusively by the graphics processor as graphics memory.

The Direct Frame Buffer BIOS feature controls the processor’s access to the section of system memory reserved for use by the integrated graphics processor as graphics memory. Please note that we are referring to the CPU, not the graphics processor.

When enabled, the processor is allowed to directly write to the section of system memory reserved as graphics memory. This increases the performance of applications that write directly to the frame buffer.

When disabled, the processor is not allowed to directly write to the section of system memory reserved as graphics memory. This reduces the performance of applications that write directly to the frame buffer.

It is recommended that you enable this BIOS feature for maximum performance. Of course, this only improves performance of applications that write directly to the frame buffer.

Please note that this BIOS feature has no effect on the frame buffer of any discrete graphics card that you install. It only controls the processor’s access to the section of system memory reserved for use by the integrated graphics processor as graphics memory.

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participate in the Tech ARP Forums, or even donate to our fund. Any help you can render is greatly appreciated!