Post your Roadkill HDD Results!

[Madwand1]

Roadkil's a very simple synthetic benchmark. It's not an application benchmark. It measures three things, and gets at least one of them -- STR -- quite wrong in some cases.

STR is not complicated. The Raptor's one of the most benchmarked drives around. It doesn't do 177 MB/s STR.

 

[Minerva]

Originally posted by: EricMartello


Caching is an integral part of disk subsystem performance, and ignoring it and ONLY focusing on the raw performance of the drive is interesting but ultimately pointless.

As I mentioned before, the logic chipset that controls the HD itself plays a big role in how "fast" it is in real world situations, much moreso than peak transfer rates. Did you know that the peak transfer rate of the drive varies drastically with the type of access pattern? For example, when booting WinXP even a drive like the RaptorX will peak around 8-9 MB/s! That's a far cry from the raw benchmark results that show 75-80 MB/s...

Benchmarks are supposed to measure performance, and I think that saying cache invalidates a given benchmark is off the mark. If the cache and burst transfers can provide the data at near interface speeds, that means the disk cache is doing its job. It's supposed to be that fast...furthermore, how are people saying that a benchmark is supposed to give some idea of "real world" performance if they choose to ignore the effects of caching and bursting? Do you all run your OSes with caching disabled?

Hard drives are still the main bottleneck in a modern computer, but they've come a long way since the days of PATA. I don't think that looking at transfer rates really says much about the drive's actual performance. You need to evaluate the package, which includes factors like caching effectiveness, on-drive logic and processing as well as bursting.

It may be pointless but when a benchmark is skewed as the result of the cache one needs to know this! As pointed out a single Raptor does NOT have a STR of 175+ MB/S.

Burst speed is completely meaningless but again it's measured to ensure the interface is operating correctly.

It does not matter what drive one has under high queue depths encountered with the latest NT based operating systems when starting up the throughput is going to be several MB/S at best on average. There is no argument about the Raptor 150's firmware tuning for single user applications. The damn thing is FAST!


If the STR is WAY off on RK benchmark, then something must be skewing the results whether it's hardware or software cache. Years ago in the DOS 3.3 days one could purchase a program called PC Kwik and get outrageous scores in Norton Utilities and Check IT benchmarks. :Q (Norton had ncache.exe TSR as well but I preferred smartdrv.sys especially when Win 3.1 was getting popular.)

HD Tach will show the STR usually. It does on the FIRST time here. Subsequent times it runs in the cache and shows 0.0 ms access and > 1GB/S STR across the disk! The disk activity lights are quiet on the second run. It's funny to show it to people and tell them you have a 1TB SS system though.

 

[EricMartello]

Originally posted by: Madwand1
Roadkil's a very simple synthetic benchmark. It's not an application benchmark. It measures three things, and gets at least one of them -- STR -- quite wrong in some cases.

STR is not complicated. The Raptor's one of the most benchmarked drives around. It doesn't do 177 MB/s STR.

Which is why the ATTO results I posted show more realistic figures for my drive.

 

[EricMartello]

Originally posted by: Minerva
It may be pointless but when a benchmark is skewed as the result of the cache one needs to know this! As pointed out a single Raptor does NOT have a STR of 175+ MB/S.

I never said that it did! I simply posted the results I got.

Burst speed is completely meaningless but again it's measured to ensure the interface is operating correctly.

Not really. Most drives have onboard cache...burst transfers usually occur between that cache and the system. Your system would be very slow without this.

If the STR is WAY off on RK benchmark, then something must be skewing the results whether it's hardware or software cache. Years ago in the DOS 3.3 days one could purchase a program called PC Kwik and get outrageous scores in Norton Utilities and Check IT benchmarks. :Q (Norton had ncache.exe TSR as well but I preferred smartdrv.sys especially when Win 3.1 was getting popular.)

HD Tach will show the STR usually. It does on the FIRST time here. Subsequent times it runs in the cache and shows 0.0 ms access and > 1GB/S STR across the disk! The disk activity lights are quiet on the second run. It's funny to show it to people and tell them you have a 1TB SS system though.

Smartdrive was da bomb. I loaded it 2MB of cache...back in those days, 2MB was plenty.

 

[soydios]

9469 with a single Seagate 7200.10 320GB SATAII

 

[bob4432]

Originally posted by: EricMartello

Originally posted by: bob4432
it still strikes me as interesting that other with raptors don't get what you get, that is what i am saying. why is yours higher? i don't care about benches but just find your results way out of the ordinary

and according to wiki:

First-generation SATA interfaces, also known as SATA/150 or SATA 1, run at 1.5 gigabits per second (Gbit/s). Serial ATA uses 8B/10B encoding at the physical layer. This encoding scheme has an efficiency of 80%, resulting in an actual data transfer rate of 1.2 Gbit/s, or 150 megabytes per second (MB/s) (or 143.05 MiB/s). The relative simplicity of a serial link and the use of LVDS allow both the use of longer drive cables and an easier transition path to higher speeds.

is this information incorrect?

Wiki info is hit-or-miss. Not wrong, but not always 100% correct. I think the fact that this bench shows 177 MB/s contradicts the 80% efficiency generalization. At 1.5 Gbps, 20% overhead is very high.

Why are my results higher? Computers I build always run faster than what other people build. Seriously tho, my MB is the Asus A8N32 Deluxe (latest bios), and the drives are connected to the nvidia SATA with raid disabled in the bios. Running XP Pro with MS drivers for the drives. Maybe someone else with a similar setup will chime in and post their results.

here is the actual WD documentation stating 150MB/s

Acutally no...it's stating:

SATA Hard Drives
150 GB, 1.5 Gb/s, 16 MB Cache, 10,000 RPM

Maybe you misread 150 GB as 150 MB/s?

no, i am talking interface-

Interface SATA 150 MB/s

as per the wd page

 

[Minerva]

Originally posted by: EricMartello

Originally posted by: Minerva
It may be pointless but when a benchmark is skewed as the result of the cache one needs to know this! As pointed out a single Raptor does NOT have a STR of 175+ MB/S.

I never said that it did! I simply posted the results I got.

Burst speed is completely meaningless but again it's measured to ensure the interface is operating correctly.

Not really. Most drives have onboard cache...burst transfers usually occur between that cache and the system. Your system would be very slow without this.

If the STR is WAY off on RK benchmark, then something must be skewing the results whether it's hardware or software cache. Years ago in the DOS 3.3 days one could purchase a program called PC Kwik and get outrageous scores in Norton Utilities and Check IT benchmarks. :Q (Norton had ncache.exe TSR as well but I preferred smartdrv.sys especially when Win 3.1 was getting popular.)

HD Tach will show the STR usually. It does on the FIRST time here. Subsequent times it runs in the cache and shows 0.0 ms access and > 1GB/S STR across the disk! The disk activity lights are quiet on the second run. It's funny to show it to people and tell them you have a 1TB SS system though.

Smartdrive was da bomb. I loaded it 2MB of cache...back in those days, 2MB was plenty.

If the cache (buffer) on the hard disk was completely disabled it probably would be quite slow. However if you consider a SATA2 drive will show a much higher burst in HD Tach when running in SATA2 mode. Now jumper it so it's forced to run in SATA1 mode. Everything looks the same in HD Tach except the burst speed. Windows performance is very similar and most definitely indistinguishable to the end user. Eventually as areal density increases with perp recording we shall see outer track STR's approaching SATA1's limit thus the need for SATA2 will be mandated. But until now it's like the days of UDMA-66. If I recall the very first U66 drive was a WDC 5400 RPM Caviar (AC313000) that was nowhere near that limit. Sounds familiar.

Found it!

That drive was a perfect example of firmware for benchmark tuning and this is discussed in the article.

 

[EricMartello]

Originally posted by: bob4432
no, i am talking interface-

Interface SATA 150 MB/s

as per the wd page

It's not 150 MB/s, the SATA interface is either 1.5 or 3.0 Gbps...I assure you. If it was only 150 MB/s it would barely be an improvement over UDMA133 ATA.

Originally posted by: Minerva
If the cache (buffer) on the hard disk was completely disabled it probably would be quite slow. However if you consider a SATA2 drive will show a much higher burst in HD Tach when running in SATA2 mode. Now jumper it so it's forced to run in SATA1 mode. Everything looks the same in HD Tach except the burst speed. Windows performance is very similar and most definitely indistinguishable to the end user. Eventually as areal density increases with perp recording we shall see outer track STR's approaching SATA1's limit thus the need for SATA2 will be mandated. But until now it's like the days of UDMA-66. If I recall the very first U66 drive was a WDC 5400 RPM Caviar (AC313000) that was nowhere near that limit. Sounds familiar.

Found it!

That drive was a perfect example of firmware for benchmark tuning and this is discussed in the article.

If you are talking about everday use such as browsing the web, checking email, word processing, etc then I would say it is hard to notice a big difference between a Pentium 3 and a Athlon 64. None of these apps really push either system to any limits, so despite the A64 scoring much higher benchmarks, you wouldn't notice that until you ran an app that demanded the extra performance.

With your statement about burst transfers not showing any difference even if you swap interfaces - I think the fact that HD Tach shows a much higher rate is indicative of better potential performance in a given situation...possibly when a lot of simultaneous IO operations are occurring. But saying that it doesn't matter because you couldn't notice it using "Windows performance" as a gauge doesn't invalidate the fact that it is faster.

One good reason to switch over to the SATA 3.0 setup if you are upgrading, is because it doesn't cost any more for a drive that offers SATA 3.0 - its necessity can be debated, but I'm not about upgrading to yesterday's technology. I do wish the RaptorX had an SATA 3.0 interface.

 

[stickybytes]

Seagate 7200.10 320GB SATAII single drive 9686.

Btw, should i exit all programs before i run this or it doesn't matter? Im getting readings varying by 200-300 everytime i run it.