LITTLEBLACKDOG.COM

Posted: Sun Nov 11, 2001 7:04 pm Post subject:

Feel free to post any questions, comments, or feedback regarding this article. We will do our best to answer any questions you may have, and look forward to your feedback on this article!

Original Article - http://www.littlewhitedog.com/reviews_hardware_00034.asp

Moderator Joined: 05 Dec 2000 Posts: 5941 Location: USA

This is the best review yet ... w Big Smile

f - w

f !!!

September 11, 2001

Stray Dog Joined: 12 Nov 2001 Posts: 10 Location: Norway

Here's my 2 cents :
I appriciate all the hard work put into the article, good work but :

First off I assume you had the 40GB ver. of D740X for the test? I assume since it's listed as a 1 platter disk.
If so, then comparing it to a WD200BB or any other drive "like" it will give you a big difference. Nomatter if they got ATA100 or ATA133.. The main reason for making this statment is the following :
WD200BB -> 10GB pr. platter.(Assumed, since it's listed with 2 platter's)
D740X -> 40GB pr. platter.
This fact in itselves will give you some performance difference, if not a substantial difference!
Why? Because when you got a higher data-desity on the platters this will give you : higher max transfer rate/read from the platters themselves, lower rotational latency(neglible difference), lower search time/latency to name some..

I think it would have made a more fair comparison using a ATA100 drive with the "same" specs as the Maxtor D740X..

This might have given a result more "fair"! Atleast it might have shown the difference between ATA100 and ATA133 better, or more accurate! And not the disks themselves.. As I think this article does!

Dont get me wrong, I do appreciate all the hard work you put into it! And you probably havent got "every" hardware thing! at your disposal!

Best regards
Zyztem off

Moderator Joined: 05 Dec 2000 Posts: 5941 Location: USA

Welcome to the LWD Forums Zyztem ... w Big Smile

f !!!

September 11, 2001

Stray Dog Joined: 12 Nov 2001 Posts: 1

quick questions regarding the maxtor 160G drive (probably stupid and not possible)

1. if you create 2 partitions for the drive would a normal old style motherboard detect 2 80G drives and be able to see them?

2. Can you attach the maxtor drive to an older motherboard that only supports ultra100 or ultra66? i.e. is it backwards compatible or no?

is there going to be no way to get it working with an older (or current) mb without the promise style adapter?

Big Dog Joined: 20 Oct 2000 Posts: 3689 Location: USA

Quote:

1. if you create 2 partitions for the drive would a normal old style motherboard detect 2 80G drives and be able to see them?

nope . . you'd just have one physical drive with 2 partitions

Quote:

2. Can you attach the maxtor drive to an older motherboard that only supports ultra100 or ultra66? i.e. is it backwards compatible or no?

probably not without running some sort of "overlay" software because most of those controllers have a 128gig upper limit

Quote:

is there going to be no way to get it working with an older (or current) mb without the promise style adapter?

again, probably not without running some crappy overlay software

temporary replacement signature . . . please stand by

Stray Dog Joined: 12 Nov 2001 Posts: 10 Location: Norway

Thx for the welcome squito Big Smile

See you've been around here for a while Wink Smile

Moderator Joined: 05 Dec 2000 Posts: 5941 Location: USA

Quote:
Thx for the welcome squito
See you've been around here for a while

Your welcome ... seems like I just got here yesterday ...

September 11, 2001

Big Dog Joined: 18 Oct 2000 Posts: 4419

Quote:

Here's my 2 cents :
I appriciate all the hard work put into the article, good work but :

First off I assume you had the 40GB ver. of D740X for the test? I assume since it's listed as a 1 platter disk.
If so, then comparing it to a WD200BB or any other drive "like" it will give you a big difference. Nomatter if they got ATA100 or ATA133.. The main reason for making this statment is the following :
WD200BB -> 10GB pr. platter.(Assumed, since it's listed with 2 platter's)
D740X -> 40GB pr. platter.
This fact in itselves will give you some performance difference, if not a substantial difference!
Why? Because when you got a higher data-desity on the platters this will give you : higher max transfer rate/read from the platters themselves, lower rotational latency(neglible difference), lower search time/latency to name some..

I think it would have made a more fair comparison using a ATA100 drive with the "same" specs as the Maxtor D740X..

This might have given a result more "fair"! Atleast it might have shown the difference between ATA100 and ATA133 better, or more accurate! And not the disks themselves.. As I think this article does!

Dont get me wrong, I do appreciate all the hard work you put into it! And you probably havent got "every" hardware thing! at your disposal!

Best regards
Zyztem off

Welcome Zyztem!

You make several very valid points.

That's why I tested each drive, on each controller. The 2 other drives used were to compare the differences between the drives (granted they weren't very equal) and the testing of each drive on a different controller was to test the controller speed. I'm sure you get the point... so I'll stop rambling. Again, welcome to The Pound!

Stray Dog Joined: 15 Nov 2001 Posts: 1

hi umm. i was wondering where did you guys obtain a promise ata 133 tx2 card? i am shopping around for a controller card, and i couldnt find this one anywhere.

Big Dog Joined: 18 Oct 2000 Posts: 4419

The card came directly from Promise (in a round about way). Try a Pricewatch search for ATA133, and remember that most of the places that sell the D740X will have them since Maxtor is offering their version of the promise card (it's the same card) for free:

http://www.maxtordirect.com/product.asp?sku=1856736

Stray Dog Joined: 05 Jan 2002 Posts: 2

Cable Support - Includes single Ultra ATA 80-wire/40-pin cable; cable required for Ultra ATA/133 drives to reduce signal noise (“crosstalk”) during transfers

Anybody care to explain how extra wires reduce "crosstalk"

Original Article - http://www.littlewhitedog.com/reviews_hardware_00034.asp

[/quote]

Big Dog Joined: 20 Oct 2000 Posts: 3689 Location: USA

I'll give you a quick example

[40 pin cable]

_______ wire A

_______ wire B

_______ wire C

_______ wire D

_______ wire E
[/40 pin cable]
Versus
[80 pin cable]
_______
_______ wire A
_______
_______ wire B
_______
_______ wire C
_______
_______ wire D
_______
_______ wire E
_______
[80 pin cable]

this is a technically incorrect example, so don't take it too seriously but here's my explanation

the red wires above represent "hot" wires (with signal activity)

when you have red wires right next to each other, and the signal strength in one is strong at the same time as the signal strength in the wire on either side of it there will be bleed-over

this is because shooting the electrical activity up and down the wire produces a magnetic feild

this magnetic field will be strong enough to exert an electrical pull on the copper wire next to it (the pull will be there whether the wire next to it is hot ot not)

OK . . let me refine the example:

if wire B in the 40 pin diagram is hot, then it is exerting and electricl pull on wires A and C.

this isn't as much of a problem at ata33 speeds because the signal lengths are "longer" and there is more time for the electronics to determine whether a pulse is really a one or a zero.

the problem comes when you try to cram more data through in the same amount of time. the only ways to do this are to add more wires to carry the data, or to make the signal lengths shorter. once you make the signal shorter, there's less time for the electronics to determint the value of each individual bit, and the minor variance that wire B causes on wires A and C can be enough to change what the electronics interpret a single bit to be because the duration is shorter

the easiest way to over come this problem is to add "insulator" wires between each of your "hot" wires. and keep the same original amount of carriers (wires with data on them)
the insulator wires now pick up and ground out the "interference" generated by each hot wire, and keep that interference from reaching the "hot" wires on either side

if you put one "insulator" between each of the "hot wires" you almost double the amount or wires (which gets you from 40 to 80) and this is why ata66, ata100, and ata133 all require the 80 pin cables.

I don't know how good my explanation was, so if anyone else can draw it out better feel free to jump right in [ Very Happy

]

temporary replacement signature . . . please stand by

Stray Dog Joined: 05 Jan 2002 Posts: 2

Thanks Brain!.

I get the idea. Thanks for taking time in replying..sweet of you.

Thanks Again

Spectrum

Edited by - Spectrum on 01/07/2002 11:48:55

Stray Dog Joined: 03 Feb 2002 Posts: 1 Location: USA

I am building my first system, & I wanted the latest info on ATA133, GREAT REVIEW!!! Big Smile

Now for another unrelated question on ATA133 supported hardware: How will PC2700 (DDR333) figure into the future of building a fast PC? Thanks! BTW, the crosstalk info was pretty much on the money.