ULi M1575 for ATI Crossfire – SATA2 and Fast USB for the ATI Chipset
by Wesley Fink on October 13, 2005 4:21 PM EST- Posted in
- CPUs
Disk Controller Performance
With so many chipsets and brands of storage controllers on current Athlon 64 boards, we standardized on Anand’s storage benchmark, first described in Q2 2004 Desktop Hard Drive Comparison: WD Raptor vs the World, as a standard means of measuring disk controller performance. To refresh your memory, the iPeak test was designed to measure "pure" hard disk performance, and in this case we kept the hard drive as consistent as possible while varying the hard drive controller. The idea is to measure the performance of a hard drive controller with a consistent hard drive. We played back Anand’s raw files that recorded I/O operations when running a real world benchmark - the entire Winstone 2004 suite. Intel's IPEAK utility was then used to play back the trace of all the IO operations that take place during a single run of Business Winstone 2004 and MCC Winstone 2004. To try to isolate performance difference to the controllers we were testing, we used Seagate 7200.7 model SATA and IDE hard drives for all tests.
iPeak gives a mean service time in milliseconds; in other words, the average time that each drive took to fulfill each IO operation. In order to make the data more understandable, we report the scores as an average number of IO operations per second so that higher scores translate into better performance. This number should not be used to report hard disk performance as it is just the number of IO operations completed in a second. However, the scores are useful for comparing “pure” performance of the storage controllers in this case.
In past benchmarking, IDE has provided the slowest IO performance in this roundup. However, ULi and ATI IDE break that trend, with IDE performance being the best that we have measured since we have been testing with iPeak.
There are no additional SATA/SATA2 controllers on the ULi M1575 Reference Board, but with excellent IDE and SATA2 disk performance, there is really no need for additional discrete controllers unless you need more than four SATA2 drives.
With so many chipsets and brands of storage controllers on current Athlon 64 boards, we standardized on Anand’s storage benchmark, first described in Q2 2004 Desktop Hard Drive Comparison: WD Raptor vs the World, as a standard means of measuring disk controller performance. To refresh your memory, the iPeak test was designed to measure "pure" hard disk performance, and in this case we kept the hard drive as consistent as possible while varying the hard drive controller. The idea is to measure the performance of a hard drive controller with a consistent hard drive. We played back Anand’s raw files that recorded I/O operations when running a real world benchmark - the entire Winstone 2004 suite. Intel's IPEAK utility was then used to play back the trace of all the IO operations that take place during a single run of Business Winstone 2004 and MCC Winstone 2004. To try to isolate performance difference to the controllers we were testing, we used Seagate 7200.7 model SATA and IDE hard drives for all tests.
iPeak gives a mean service time in milliseconds; in other words, the average time that each drive took to fulfill each IO operation. In order to make the data more understandable, we report the scores as an average number of IO operations per second so that higher scores translate into better performance. This number should not be used to report hard disk performance as it is just the number of IO operations completed in a second. However, the scores are useful for comparing “pure” performance of the storage controllers in this case.
In past benchmarking, IDE has provided the slowest IO performance in this roundup. However, ULi and ATI IDE break that trend, with IDE performance being the best that we have measured since we have been testing with iPeak.
There are no additional SATA/SATA2 controllers on the ULi M1575 Reference Board, but with excellent IDE and SATA2 disk performance, there is really no need for additional discrete controllers unless you need more than four SATA2 drives.
Facebook
Twitter
RSS
15 Comments
View All Comments
nlr_2000 - Saturday, October 15, 2005 - link
nlr_2000 - Saturday, October 15, 2005 - link
It's on page 5.DRavisher - Friday, October 14, 2005 - link
Just you wait. In a couple of years ATi will finaly figure out how USB works and release a self made southbridge under the slogan "USB done right!", where they will tout their superiority in USB over nvidia, and how important good USB performance from the same company that designed the northbridge is for a company.sprockkets - Friday, October 14, 2005 - link
why bother with another southbridge? Keep it simple for linux driver support.But since video with nvidia is better supported anyhow, might as well just use the 6100 vs the exp 200
LoneWolf15 - Friday, October 14, 2005 - link
Good review. I guess for me it all comes down to the same thing though, economics.SLI and Crossfire have a high cost. Two video cards, a motherboard that supports them, a power supply that supports them, plus additional cooling. And then within 6-10 months, a new high-end card comes out that's a single-slot solution that can beat previous SLI/Crossfire solutions before it, and since it's single slot, doesn't need the special mainboard.
I think we'll see in a year, maybe two tops, that vendors will begin implementing either dual-core GPU chips, or two GPU chips on a single PCIe card much more often, making SLI and Crossfire a moot point for all but people who need a high-performance multi-monitor setup. I think SLI and Crossfire are just stopgap solutions until this arrives, and a lot of money to spend when a mid-to-high-end single card will give you what you need for less.
notposting - Thursday, October 13, 2005 - link
Don't suppose there's been any word on more motherboard's based on ULi's 1695 North bridge?The Asrock looks like a pretty good board, but I when I toss my chips into the 64bit pile I'll be trying to get a top line mobo featurewise, Firewire, SATA2, gigabit Lan, great sound.
Good to see some competition for Nvidia (and Intel ironically by Nvidia) in the chipset department. Better for us and pushes them (all the manufacturers) harder.
tuteja1986 - Thursday, October 13, 2005 - link
My main reason for wanting to buy a ATI chipset at the moment is because of Sata 2 performance and overclocking abilities. I couldn’t care less for USB since I only use a USB mouse, and Joystick that computer. If i really need a fast USB port than i would just buy a cheap $10 PCI USB/Firewire card from my local computer store.Madellga - Thursday, October 13, 2005 - link
We got an important piece of information provided by Wesley: The Crossfire motherboards are not a paperlaunch due to ATI delays, they were delayed by the OEMs due a last minute change to the ULI1575 chipset.It is worth in my opinion the delay, as we can get a truly competitive solution, even if one doesn't need the Crossfire feature.
I still have a question for Wesley: some say that the ATI/ULI combo chipsets run cooler than Nvidia's. Is that true?
Wesley Fink - Thursday, October 13, 2005 - link
I've had the bare ULi SB running for hours. It's very warm to the touch but not hot - without any heatsink at all. A passive heatsink should be plenty. The ATI RS480 north bridge has a passive heatsink which is completely cool to the touch. The final Crossfir RD482 uses the .11 process compared to the .13 on theis RS480.The ATI solution is 2 chips compared to the single-chip nVidia, which may partially account for the extra heat many complain about on the nVidia nF4.
The ATI chipset delays are a result of several factors - the ULI M1575 being just one. I have been told the chipset has been ready for months, but the video component was a moving target. Final release needed a coordinated Crossfire video and chipset solution, and Crossfire has only recently been solidified to the point of production. The video delays made the chipset late since no manufacturer wanted to launch a new chipset only to face potentail board revisions in a month or two.
Cygni - Thursday, October 13, 2005 - link
Not entirly true... the boards NOT using ULi southbridges (using the ATI SB540) arent exactly flooding the market, ya know.No, the last minute change simply kept a few major players like ASUS from (not really) launching with the rest of the boards that are using the ATI southbridge.