ATA133 drive on ATA100 controller

[3Suns]

I just want to make sure before I buy...

If I put a new Maxtor ATA133 drive on my ATA100 onboard controller, it should work but at ATA100 speeds, correct?

Thanks,
3Suns

 

[thorin]

The best IDE/ATA drive on the market the WD 1000BB-JB (SE) series only pushes ~38MB/Sec Average Sustained Transfer Rate (49 Max). This is far far far below even ATA66 speed. Putting a ATA133 drive (which doesn't even reach ATA66 max throughput) on a ATA100 controller won't cause any performance issues at all.

Yes it will work, and at the best speed it possibly can.

Thorin

 

[Whitedog]

Yes, the drive will work.

Thorin, I believe the ATA100 interface is superior to the ATA66, thus yeilding better performance.

I don't recall seeing any benches where they took the same drive and on an ATA66 interface and ATA100 and compared scores, but I think the scores are supposed to be a tad faster on the ATA100...

I'd be interested in seeing some benches to support this if anyone knows of any.

Thanks

 

[mboy]

I have 1 on my 8KHA+ and even with the so called IDE problems with VIA chipsets, I get 26000 on the HD bench in SANDRA on my primary OS partition.

 

[rbV5]

The only real problem would be if you put a new maxtor>128 GB capacity on that controller, it will only recognise 128 GB max.

 

[Mavrick007]

I believe that would be any drive larger than 137 Gigs since that is where ATA133 should help. I'm not sure if this refers to "formatted" size which would be smaller since companies say a meg = 1000x1000 bytes instead of 1024x1024 bytes which is an actual meg.

 

[DN]

<< Yes, the drive will work.

Thorin, I believe the ATA100 interface is superior to the ATA66, thus yeilding better performance.

I don't recall seeing any benches where they took the same drive and on an ATA66 interface and ATA100 and compared scores, but I think the scores are supposed to be a tad faster on the ATA100...

I'd be interested in seeing some benches to support this if anyone knows of any.

Thanks >>



Actually, Thorin is correct.. I've seen benches to support what he's said, but I don't recall where I saw them.. It also makes sense that if the hard drive is the bottleneck (as is the case at the moment), no matter how much you improve the controller, it won't make a difference..

 

[rbV5]

<< I believe that would be any drive larger than 137 Gigs >>



You're probably correct, I was going by the maximum supported drive on my promise controller, which is 128GB. OOPS😱

 

[CQuinn]

And I've seen benches that agree with Whitedog, so YMMV. It would seem the newer drives may gain some
benefit in command execution thru a faster bus, but overall the difference in long term performance is negligible for
single drive to drive usage.

 

[Whitedog]

I think we've made a whole lot more out of the original question...

But the answer to the Original question is: YES 😉

 

[majewski9]

It will work fine! I have an ATA 133 drive on an ATA 100 controller. I want to move to ATA 133 but Ill get a new board first.

 

[Pariah]

The limit is 137GB decimal, 128GB binary. 2^28 sectors of 512bytes each.

 

[Whitedog]

hehehe, I can't help it but reply to Pariah's post...

HUH?? WTF are you talking about?

hehe, I think you got the wrong thread. 😎

 

[ViRGE]

Whitedog, no he's in the right place. As he said, the previous ATA limit was 137GB if it was measured by manufacturer's standards(1000 bytes in a KB, 1000KB in a MB, ect). However, in terms of real storage, which is based off binary(2^10, or 1024 bytes in a KB, 1024KB in a MB, ect), the limit is really 128GB.

 

[Pariah]

You guys were talking about the IDE size limit for a drive. You were both right. It's 137GB decimal value, 128GB binary value. The way that number is calculated is that ATA uses a 28bit number to address sectors, each sector containing 512bytes of storage space, resulting in 2^28 * 512 being the maximum capacity.

 

[jacksco]

The only time you'll see transfer rates anywhere near 100 or 133MB/s is when the data you are trying to access is sitting in the cache. The burst from the cache is limited by the maximum transfer rate of the protocol. Otherwise you're still limited by the media-to-buffer transfer rate (platter to buffer) like everyone else is saying (usually 40MB/s ->55MB/s max).