ATI Radeon Xpress 200: Performance, PCI Express & DX9 for Athlon 64
by Wesley Fink on November 8, 2004 6:00 AM EST- Posted in
- CPUs
SidePort: On-Board GPU Memory
Just before the launch of the Athlon 64, we found that many chipset manufacturers were a bit worried about the performance of integrated graphics solutions and AMD's new CPU. The worries stemmed from the fact that in previous CPU/chipset architectures, the integrated graphics cores resided on the North Bridge and shared access to the system memory controller - also located on the North Bridge. With the Athlon 64, however, the memory controller resides in the CPU, increasing memory access latencies from the perspective of the integrated graphics core.The Radeon Xpress 200 supports a local frame buffer attached to what ATI refers to as their "SidePort". The SidePort is a 32-bit DDR memory interface that the integrated graphics can use either instead of or alongside the Athlon 64's memory controller.
While we assumed that SidePort was included to hide some of the latencies of using the Athlon 64's memory controller, that ended up not being true as performance in UMA mode (using the Athlon 64's memory controller) was quite respectable. It turns out that most games don't benefit too much from lower latency memory accesses (through SidePort). So, why would ATI include support for a local frame buffer with the Radeon Xpress 200? Although performance is improved with SidePort enabled, the biggest reason for supporting the feature is to reduce power consumption in mobile environments. Without SidePort enabled, the CPU needs to be awake to fetch data for refreshing the display, but with SidePort enabled, all memory accesses can occur via the Radeon Xpress 200 and the CPU can remain asleep in power saving modes.
Because of the added cost of supporting SidePort, it isn't a requirement - the Radeon Xpress 200 has four memory operating modes:
- SidePort only - In this mode, the integrated graphics core treats the SidePort memory as its local memory. If more memory is needed, it is allocated dynamically through system memory by the driver, which is significantly higher latency than the local SidePort memory.
- UMA only - In UMA mode, the only memory to which the integrated graphics has access is a dynamically allocated partition of system memory. The size of the parition is selectable from within the BIOS (ATI's reference board allows for 16 - 128MB sizes). If more memory is needed, it is allocated dynamically through system memory by the driver.
- UMA + SidePort (Interleaving Disabled) - In this mode, the total amount of "local" graphics memory is the size of the UMA partition and the amount of memory connected to the Radeon Xpress' SidePort. The integrated graphics core will first use SidePort memory until it runs out, then using system memory. If more memory is needed, it is allocated dynamically through system memory by the driver.
- UMA + SidePort (Interleaving Enabled) - By enabling Interleaving and setting the UMA frame buffer size to the same size as the memory connected to the Radeon Xpress' SidePort, a special Interleaving mode is enabled. In this mode, the integrated graphics cores will request data from both the UMA space and SidePort memory. The benefit of Interleaving is that now two reads or writes can occur at the same time, whereas with just SidePort only a single 32-bit read/write can happen at any given time. Despite the fact that UMA accesses will be higher latency, the dual ported nature of this setup improves overall performance. There are situations when a SidePort only configuration will offer greater performance if the application depends on lower latency memory accesses. If more memory is needed it is dynamically allocated through system memory by the driver.
ATI's reference board features 16MB of DDR memory attached to the Radeon Xpress' SidePort. The memory can either run synchronously with the system memory clock (200MHz for DDR400) or asynchronously, where the speed is bound by the type of memory used. In our case, the 2.5ns Samsung DDR located on the board was capable of running at the maximum frequency the BIOS allowed - 350MHz.
As we mentioned before, the SidePort memory interface is a single 32-bit channel, which at 350MHz provides 1.4GB/s of bandwidth to the integrated graphics core. At 200MHz SidePort can only provide 800MB/s of bandwidth, so the additional latency incurred by running the SidePort asynchronously with main memory is well worth the additional bandwidth.
What's truly interesting is the pretty impressive performance of running in SidePort-only mode. Granted you are limited to low resolutions, but as you will soon see, the integrated graphics core isn't really designed to run at very high resolutions. In fact, running in SidePort-only mode is faster than running in UMA only mode with a single-channel Socket-754 Athlon 64.
The charts below do a good job of showing off the performance advantages to the various operating modes of the Radeon Xpress 200.
The first thing we see is that there's a huge performance advantage to the dual channel memory controller of the Socket-939 Athlon 64, - 33% in Doom 3 and 27% in UT2004. This is far from unexpected given that the more system memory bandwidth you have, the more graphics memory bandwidth you have.
The performance advantage to using the SidePort + UMA configuration isn't insignificant either - 8.5% in Doom 3 and 7.7% in UT2004, however with the added cost we would say that the SidePort isn't absolutely necessary for desktops (but we understand its usefulness in notebooks).
We compared the graphics performance of the Radeon Xpress 200 to ATI's lowest end discrete PCI Express graphics card: the Radeon X300 SE. The X300 SE is a four-pipe version of the Radeon Xpress 200 but with only a 64-bit DDR memory interface, so the Radeon Xpress 200 actually holds a memory bandwidth advantage over the X300 SE while it is at a fill rate deficit.
We also compared the Radeon Xpress 200 to Intel's Graphics Media Accelerator 900. While the GMA 900 is obviously only available on the Intel-only 915G and the Radeon Xpress 200 is an AMD-only solution, the two offerings are slow enough that most games end up being completely GPU limited and thus the CPU differences become negligible.
You'll notice that not all of the benchmarks have scores for Intel's GMA 900; those that don't have GMA 900 scores are ones where the GMA 900 was not able to either run the game or complete the benchmark without crashing.
Facebook
Twitter
RSS
45 Comments
View All Comments
ImJacksAmygdala - Monday, November 8, 2004 - link
Good job Wesley and Anandtech staff, very nice article! Thanks for answering all the questions Wesley...Things are really heating up for AMD64 chipsets... I might just have to wait for Q1 2005, but then dual core CPUs will be the rage of rumor and HL2 is right around the corner ARGH!!!!!!
I might just bite the bullet and pull the trigger on an AMD64 3500+ with a Nforce4 or ATI chipset this holiday season because I really can't run HL2.... I cannot continue to wait for the next best thing. I have been waiting to upgrade since dual DDR was just a rumor... LOL!
xsilver - Monday, November 8, 2004 - link
Are these going to be shipping mainly for s754 or s939? I think determining factor will be price, cheaper than nforce 4 ultra? isnt that going to be $150+? I think they need to get closer to $100 for any serious challenges..Penty - Monday, November 8, 2004 - link
I don't know, I still want Tyan's new dual Opteron with dual SLI board.http://my.so-net.net.tw/brandon640410/NF41/004.JPG
Zebo - Monday, November 8, 2004 - link
Nice article. But I'm afraid, like the out-standing SiS ref board, these may never see light of day but in budget off brand products like ECS, FOXCONN, MATSONIC etc nVidia just seems to dominate the enthusiast sector.:(Wesley Fink - Monday, November 8, 2004 - link
#29 - A new Sounthbridge, called SB450, is due to debut the first of the year. It will feature High-Definition audio and other updates.RAID - Here is the reply from ATI regarding RAID on the Rx480/SB400:
"Your Raid answers are as follows:
Our Raid implementation will support up to 8 SATA devices. This is limited by ODM implementing appropriate amount of SATA channels on pcb.
1) Currently SB400 supports RAID on SATA, not on IDE (PATA).
2) IDE cannot be combined with SATA in RAID. Limited to SATA only.
3) We only support RAID 0 and RAID 1.
4) We do support hot swap for RAID 1. We can replace the failed hard disk and rebuild a new driver in RAID 1"
The review will be updated.
keitaro - Monday, November 8, 2004 - link
Here's a hardware question for you folks. Given that SLI is becoming more and more of an option, I'd like to know if the scenerio I thought up can be possible. Also, a general question is "What can a HyperTransport link be used? Is it specifically a link from CPU to north/south-bridge or is it a general link for general access?"The above question will basically determine if the scenerio I thought of is possible. Basically have a HT link big enough for an additional chip where it supports additional PCI Express lanes for 1x, 2x, or 4x slots. I see current chipsets designed with 20 lanes, or in ATi's case 22 lanes, and it got me to wondering if such a scenerio is truly possible that a full dual 16x instead of dual 8x can be a reality. If you can explain or answer this, that'll be great.
mctmcpoop - Monday, November 8, 2004 - link
The main problem of the ATI chipset will be the compatibility of south bridge ... Their south bridge sucks and that's why some of the RS350 motherboard use ULI south bridge instead of their solution ...http://techreport.com/reviews/2004q4/radeon-xpress...
You can check with the USB performance ... That can be called the most worst that I have ever seen ... And we still do not know if there is any compatibility issue of the USB interface unless there is a lot of ATI chipset based retail board hit the market …
And a preview board of chipset vendor is nothing important ... Where is the highly phrased SiS 755/755FX chipset based and good quality motherboard we can buy ? Sometimes there is something inside the chipset that we can reveal from the demo board if all the motherboard maker does not have the solution …
Anyway , time will tell …
FinalFantasy - Monday, November 8, 2004 - link
#27 I doubt ATI is doing that. If the mobo's that ship are noticealbe different (performance wise) from the one's ATI are sending out for review, I know that there will be a huge backlash from the hobbyist community and ATI knows this.blckgrffn - Monday, November 8, 2004 - link
Is no one else suspicious that ATI shipped out hand picked, thouroughly tested MOBO's that may not be representative of shipping boards? I will wait until I see a review of boards using this chipset in the future before I make up my mind.FinalFantasy - Monday, November 8, 2004 - link
Wow...ATI really stepped it up with their new chipset and integrated graphics solutions. This chipset should really give nVidia's NF4 a run for it's money. Also I agree w/#22...let's see some HDD performance benchies and what about an integrated software firewall or on-board RAID controller (article says similar to NF4's "Any-Drive")...how do these components compare to NF4's or does ATi's chipset even have them.I know I was skeptical about ATI coming out with an A64 chipset, but they've really gone above and beyond what I've expected with their "first" board. I am VERY impressed with their OC'ing performance.
I don't want to say that this new chipset is the best thing since sliced bread (I'm sure nVidia has a response to ATI's chipset in the works), but when it comes time to buy my 90nm A64 rev E0, ATI looks like they might have my business/money/whatever the hell you want to call it ;D.