Will ATA 133 be much faster than ATA 66 or ATA 100? What doe

[imtim83 0]

I am wondering if ATA 133 HD will be worth going to when you already have a ATA100 HD ? Also with these new hard drivers will seek time go lower with these hard drives or just tranfer rates ? And will the speed difference be noticeable by feel ?

 

Thanks

[Brian Frank 0]

Well, the spindle speed will have a lot to do with that as the normal data throughput is around 40MB/s even though they can hit 100MB/s.

Also, the thing is that the IDE controllers are on the PCI bus which is limited to 133MB/s. I can assure you it will not be using all that bandwith up or there's going to be problems unless they bump up the PCI bus bandwith. There probably won't be that much performance increase either, unless the spindle speed is increased along with it.

[Preacher 0]

I think that the whole specifications above ATA33 are pure marketing strategies. No harddrive I know of (except SCSI server drives) can do more than 20 megs/sec random.

 

So, why random and not sequentially? Easy, if you put files onto your harddrive, they will become fragmented thus cannot be read in sequence.

Defrag utilities? You really believe in them? The last good defragger I had was the one in PC-Tools 9.0. If todays defraggers were any good, how comes that a completely defragged disk in the first util shows up as heavily fragmented in all the other utils?

 

ATA66 and above ist just the same as DDR-RAM: althou they sound great they´re absolutely unnecessary. The only reason to buy DDR-RAM is if you do lots of rendering, but that would be another thread.

[Deviant 0]

You're right Preacher !

 

Better idea is to buy 7.200 or 10.000 RPM HDD instead of upgrading to ATA133 motherboard & HDD.

 

D.

[CrazyKillerMan 0]

I wouldnt upgrade to ata133 at all. I have dual ata100's in raid.

[BladeRunner 0]

Even if ATA133 was the best thing since bread came sliced, there is no way I'm suddenly going to get rid of my 4 60GB IBM 60GXP HD's for what will be a minimal performance increase, if any.

[ValkyerieK7 0]

the ata jargon is a marketing scheme nonetheless mainly due to fact that sustained transfer rates have yet to surpass or even meet the old 66MB /sec standard

[pmistry 0]

Spindle speed is what you need:

 

I had an ATA100 board [the KA7-100] and my Maxtor 20GB has identical performance now on my ATA66 Asus K7V. Also an ATA66 7200 RPM drive will be faster than ATA100 5400 RPM and the difference will be REALLY noticeable. I just want a new ATA card for the sole purpose of getting more ports to plug things in and the fact that my HDD is ATA100 ready as for ATA133 Maxtor can't get anything out of it because of PCI's 133 limit already and you ain't gonna get up that high.

[ValkyerieK7 0]

the only way i would see an increase in pci performance is overclocking the processor bus

[Mua_Dib 0]

I've noticed slight improvements since the days of UDMA33 drives.

 

But the increased performance is negligable since the previous incarnation.

The ATA design seems to be getting old and tired and arent moving in larger steps inline with what other hardware manufacturers are doing. For example; Processor giants arent increasing the speeds of their flagship chips by 33mhz increments anymore. Adaptec have also stopped increasing their interface bandwidth by 20/40 mhz upgrades and are now doubling in each new design.

 

As soon as i get some dosh together i'm moving to SCSI160/320 with a Seagate Cheeta 15k drive.

 

-Mua

[Brian Frank 0]

Of course, it would be nice to see 10K rpm IDE drives ya know.

[ValkyerieK7 0]

I'm with u on that one brian. I've been patiently waiting for those to come out. I guess they're more into marketing gimmicks than actually improving product line

[BladeRunner 0]

I believe it's a heat issue.

I've been reading various white papers from IBM's web site and one of our new hardware testers here at work used to work for 'big blue'.

Apparently due to many factors and the technology involved etc 10,000rpm IDE drives generate too much heat to be reliable, but it is being worked on and will happen.

[Brian Frank 0]

If it can be done with SCSI, surely it can be done with IDE, as you don't hear of a massive amount of SCSI hard drives frying all of a sudden.

Spindle speed is really going to give you the bang for your buck here not so much the pipeline it's getting pumped through.

And yeah, the PCI bus needs to be upped, and I've heard of the so called PCI-X, which is faster, but not much else.

[FadedTimes 0]

ATA 133 is only Maxtor. I believe IBM and Intel have already rejected it, since most drives never get over the 66mb spec. Most of all it is because of the PCI bus itself and spindle speed, even 64bit PCI can't sustain IDE 100mb. I have a 5400rpm ATA 100, and my 7200 rpm ATA 66 beats it, barely. I do a have a 7200rpm ATA 100 drive that flies, and can get over 66mb mark on my Promise controller, but when running apps, doing large disk transfers, playing games, compiling, encoding, I don't notice the difference. I did notice going from ATA 33 to ATA 66, but not much since. My next machine after the 2nd gen P4's come out will just have SCSI.