Hard Drive Speed

[Tweak155]

I just went from an ATA100 120gb 8mb HDD to a SATAII 250gb 16mb HDD and I didn't notice any difference in load times...does that sound weird to you guys?

I have run some benchmarks but i'm not really familiar with the numbers that are being shown...can someone recommend another program or maybe explain the critical numbers?

Thanks.

-Charlie

PS:
The load times I am concerned with are video games.

 

[Lord Evermore]

Not really hard to understand the numbers. They should be telling you how much data is getting transferred per second, and how quick the seek time is.

SATA hard drives are not inherently faster than IDE drives just because they have SATA. It allows higher burst speeds, but continuous reads from the drive and seek times are limited by the mechanics, not the interface. A larger cache can help, but only in instances where cache can be accessed without needing to read the drive platter (reading gigantic data files from a drive means the cache is nearly useless). It's perfectly possible for a good fast IDE drive to perform better than an average SATA drive.

The hard drive probably does perform better, in benchmark terms, because it's a newer drive with more cache. However just because the drive performs better technically doesn't mean it'll perform much better in your particular usage. If there isn't at least a 5 to 10% real, user interface measurable difference, then you aren't going to casually notice it, you'd need a stopwatch and fast reflexes to catch it.

 

[Tweak155]

Originally posted by: Lord Evermore
Not really hard to understand the numbers. They should be telling you how much data is getting transferred per second, and how quick the seek time is.

SATA hard drives are not inherently faster than IDE drives just because they have SATA. It allows higher burst speeds, but continuous reads from the drive and seek times are limited by the mechanics, not the interface. A larger cache can help, but only in instances where cache can be accessed without needing to read the drive platter (reading gigantic data files from a drive means the cache is nearly useless). It's perfectly possible for a good fast IDE drive to perform better than an average SATA drive.

The hard drive probably does perform better, in benchmark terms, because it's a newer drive with more cache. However just because the drive performs better technically doesn't mean it'll perform much better in your particular usage. If there isn't at least a 5 to 10% real, user interface measurable difference, then you aren't going to casually notice it, you'd need a stopwatch and fast reflexes to catch it.

Thats depressing, lol. Oh well, I just upgraded my CPU to 939 (I know...old still) and got a SATA II drive to match the new mobo, since it was only 80 bucks.

 

[Okasa]

yea evermore has it down, transfer speeds are leaving harddrive read/write speeds in the dust. thats why a solid state hd would really help, its not limited by mechanism read/write. ide133 is a common transfer speed for ide drives, however the actual transfer speed from the hd is more like ide66-ide100 max. it also would make sense that a larger cache would do a, prob unnoticable, but realistic difference in transfer because the cache itself is solid state memory.

 

[corkyg]

There is a lot of myth surrounding SATA speed over PATA. In most applications (except benchmarks) it is really not noticeable. The main advantage of SATA is smaller cables and better airflow because of that. It also eliminates drive jumpers.

 

[bfonnes]

Originally posted by: Lord Evermore
Not really hard to understand the numbers. They should be telling you how much data is getting transferred per second, and how quick the seek time is.

SATA hard drives are not inherently faster than IDE drives just because they have SATA. It allows higher burst speeds, but continuous reads from the drive and seek times are limited by the mechanics, not the interface. A larger cache can help, but only in instances where cache can be accessed without needing to read the drive platter (reading gigantic data files from a drive means the cache is nearly useless). It's perfectly possible for a good fast IDE drive to perform better than an average SATA drive.

The hard drive probably does perform better, in benchmark terms, because it's a newer drive with more cache. However just because the drive performs better technically doesn't mean it'll perform much better in your particular usage. If there isn't at least a 5 to 10% real, user interface measurable difference, then you aren't going to casually notice it, you'd need a stopwatch and fast reflexes to catch it.

yep, drive overhead is usually the biggest bottleneck in a system... Drive overhead involves 3 main numbers that can be different from drive to drive that must be added together to get the amount of drive overhead... for instance,
http://www.newegg.com/Product/Product.asp?Item=N82E16822136012

Look at the specs on that drive at Newegg. If newegg doesn't list all 3 numbers it doesn't mean the drive is faster than another drive. It just means it's not listed, and you would have to check the manufacturer's website for the info. A WD Raptor has about 5 ms less overhead than the newest Seagates.

BFonnes