ATA 100 is it really that good?

[Fat|Tony]

I have no idea what my computer uses, ata66, ata100, or even if it uses ata33(god i hope not) but what's the diff??!!1

I mean is the difference that drastic?

And if anyone knows how i might be able to find out what it is then that would be very helpful.

[Iceman896]

Hi. Yeah there is a difference. The number before the ata is the theoretical transfer rate. If you can get an ata100 drive get it, because it is backwards compatible with lower speced controllers. Good luck

[AuraEdge]

33 to 66 is kinda a sizable difference, but 66 to 100 isn't at all.
To find out what speed IDE controller you have, just tell me the name of your board, the chipset on it, or if its a manufacturer made comp, Look it up and it might tell you.
Different chipsets have different speed native IDE channels.

[Yar1182]

Whatever drive you get Ultra 33,66,or 100 your substained burst rate will be closer to 33mb. This being said the faster DMA drives usually have other features like bigger buffers or faster spin rates that are worthwild. One day when serial ata comes out we'll actually be able to sustain these theoretical burst rates.

[Brangwen]

The articles regarding different ATAs are interesting. This is a recurring topic that I may one day understand. I ran SiSoft Sandra on partition E of a "5-partitioned" IBM ATA100 DMA enabled 30GB HD. Of course the drive spins at 7200 rpms. This is how it benchmarked.

<IMG SRC="http://albums.photopoint.com/j/View?u=1663681&a=12748497&p=50007308" border=0>

I'm using a T-Bird running at 1.5 GHz (Iwill non-Raid mobo) with 512 MB PC133 SDRAM CS2, if that matters. I'd like to know how to increase my HDD cache... Can someone tell me after you educate me as to the meaninglessness of these benchmark scores?

Thx!

Brangwen http://www.sysopt.com/forum/wink.gif

[Roy]

I've answered this question at least twice.

Challenge: Search for the other threads that address this topic.

[Fingers]

Here are two articles that dispute that there's any noticeable difference between ATA33/66/100.

This is the "conclusion" of an article which is admittedly nearly 2 years old, drives are certainly faster than they were in mid-1999.
The tests on ATA-33 and ATA-66 interfaces have a very clear conclusion: the use of ATA-66 interface doesn’t have any advantages, both with units at 5,400 rpm and those at 7,200 rpm. This doesn’t mean, however, that ATA-66 interface is a failure: the future possibility of using transfer rate from the disk buffer to the EIDE controller up to 66 Mbytes/sec will allow to take full advantage of even more performing mechanics, as those at 7,200 rpm and the future ones at 10,000 rpm. At the moment, the implementation of an ATA-66 controller doesn’t allow to obtain any visible increase in performance in comparison to ATA-33. <A HREF="http://www.hwupgrade.com/hd/ata66_vs_ata33/" TARGET=_blank>Ata-66 vs Ata-33 - hwupgrade.com
</A>

... but here's one from Jan2000 with basically the same conclusion:
ATA/100 - Real Performance or Marketing Hype? - realworldtech.com (http://www.realworldtech.com/page.cfm?ArticleID=RWT011701000000)


[This message has been edited by Fingers (edited 06-07-2001).]

[Axel]

One thing that I have noticed that makes a difference is the on-board cache memory on a drive.

I have been buying drives with bigger cache sizes and currently prefer the quantum 20gb 7200rpm ata66 or ata100 drives with a 2MB onboard cache.

Just do a defrag on that drive and another drive with less cache and you'll really see a huge difference in performance. I don't have numbers, but I definitely know I spend less time defraging the newer drives with exactly the same software and motherboards. Some of that is RPM - but some of it is that it doesn't have to go all the way back through the system bus to the RAM to get the data again when the bus overwhelms the write speed on the drive.