A couple of years ago the newest greatest IDE standard was first introduced. The ATA-33. Although we were promised great gains in performance preliminary performance... 82% Rating:
Chipset: HPT366 Ultra DMA66 IDE Controller Bus: PCI Drive Mode Support: Ultra ATA 4/3/2/1/0, PIO 4/3/2/1/0, DMA 2/1/0 Specifications:
Ultra DMA 66MB/sec data transfer rates, Two independent ATA channels, 256 byte FIFO per ATA channel, Concurrent PIO and bus master access, Compliant with Plug & Play, Up to 4 IDE devices supported BIOS Support: Auto identifies and configures drive type, Auto detects and supports Ultra Mode, Recognizes drives up to 128GB Advanced Features: Supports new CRC enhanced data protection for Ultra ATA drives, Supports dual data channels to allow separate device timings for Ultra ATA and EIDE devices, Supports ACPI functions Operating Systems Supported: MS DOS 5.x and higher, MS Windows 95/98, MS Windows NT 4
Performance
The benchmarks below were run a minimum of five times, and the best 3 scores were averaged. I used SiSoft Sandra 99 Professional and Winbench 99 to measure subsystem performance..
Test System Configuration
Processor(s) Intel Celeron 300A @ 504Mhz Memory 64 Megs PC-100 SDRAM Hard Disk Drive WD Caviar 13.1 Gig ATA-66 5400 rpm Video Card 3dfx Voodoo 3 2000 AGP Operating System Windows 98
Sandra 99 Pro Performance
The way the above chart is presented you would think that there is a huge difference between the two results. But looking at the numbers you can see from the performance differences between ATA-33 and ATA-66 modes are very minimal. Clearly in ATA-66 mode the performance is slightly higher, but it's so minimal that real world performance would not be affected. In case you are wondering how I managed to run the hard drive in ATA-33 mode using the Hot Rod controller, I simply used a normal 40 conductor cable instead of the 80 conductor ATA-66 cable. You must remember that ALL requirements must be met in order for ATA-66 mode to work.
Let's compare to the ABIT BE6 mainboard
Same
story here. Not much difference between the ABIT BE6 ATA-66 controller and the
Hot Rod 66. And there shouldn't be. They utilize the same chip. Furthermore,
comparing ATA-33 to ATA-66 yields the same comparable results. So far we haven't
seen a particular advantage in using an ATA-66 controller as far as real world
perfomance is concerned. Synthetic benchmarks such as Sandra 99 Pro are good for
concentrating testing in very specific areas, but we must not forget real world
performance especially when results are so close to each other. Let's continue.