Sign in to follow this  
Guest Eugene

Two Highly Anticipated Drives

Recommended Posts

Guest Eugene

Lots of folks have written in asking about an ETA on a Deskstar 180GXP review and to a lesser extent the SCSI crowd has been asking about the Atlas 10k IV.

As an enterprising reader noted in another thread, complete Deskstar 180GXP results may be found in our database.

We also have just about all the results for the Atlas 10k IV in there, excepting only operating temperature.

I was hoping to have a DiamondMax 16 and +9 in my hands by now, but no sight yet unfortunately :/. I'll keep hoping ;)

Share this post


Link to post
Share on other sites
Guest Eugene
DO YOUR DARNEDEST TO GET THEM REVIEWED.  wE CANNOT WAIT EUGENE!!!...

Hence the practice of releasing figures immediately to the database.

Share this post


Link to post
Share on other sites

eugene: as you've hopefully noticed the posts about command queuing on the 180GXP, this may make a real performance difference, but driver support is lacking. Perhaps you could try the 180GXP with an adaptor that has driver support for this feature so we can see if it's worthwhile.

I'm sure you've seen the posts about the relative performance differences of the different Dmax+ models, specifically the difference between 60GB per platter and 80GB per platter, and between the +8 and +9 models. It would be good if you could cover this when you get them eventually :)

I still haven't got my 80GB Dmax+9 I ordered in September :(

Share this post


Link to post
Share on other sites
Guest Eugene
eugene: as you've hopefully noticed the posts about command queuing on the 180GXP, this may make a real performance difference, but driver support is lacking. Perhaps you could try the 180GXP with an adaptor that has driver support for this feature so we can see if it's worthwhile.

Win2k and XP have provisions to disable tagged queuing on SCSI drives through the device manager. Informal tests that we've run in the past with it toggled on and off haven't revealed major performance differences... of course, we haven't pursued it much (for all I know, it may not actually be toggling anything) so we don't really know yet. But... I believe most folks overestimate its value in single-user situations.

Regards,

Eugene

Share this post


Link to post
Share on other sites

It may have more effect on the server tests perhaps. Anyway, it would be instructive to have some tests which actually show that command queuing is not very important (if that's the case), rather than subjective speculation. Educated speculation perhaps :)

Since IBM make a big point about this, then it's possible that their implementation is better. It would be good to put their claims to the test, since many people could be impressed by the hype.

Reading through the Testbed3 stuff, I see that you used a Promise UDMA66 controller. Now these apparently support command queuing, though the UDMA66 may limit the performance a bit. Worth an experiment maybe?

Share this post


Link to post
Share on other sites

What are you talking about Euegene? You honestly believe that Command tagged queing does little for the desktop environment, wake up BROTHER! You know, since it is assumed that you know hdds, there is a change that I have noticed in recent months. More people are using SCSI for desktops! And since SCSI is much more advanced than ATA, some of its features really make scsi more powerful than ATA in really high-end desktop workstations. One of those SCSI features is command tagged queing!!! This feature does benefit SCSI; and since there is only ONE hdd manufacturer that is smart enough to use it in their ATA lines (IBM), then I think that they know what they are doing. IBM, as any hdd manufacturer has the ability, to thoroughly evaluate its products better than you and I ever could! Point is, since IBM is marketing tag 'n seek with their ATA drives (for the desktop market), then trust me, they know (and I know too, so do the rest of us) that it does do something SIGNIFICANT for the drives performance. It is like saying that WD does not tweak their fireware to get better benchmarks. In short, IBM's command tagged queing does do wonders; it is a shame that no one else has added it to their product lines!!!...

God God has blessed this message for your enjoyment and protection!!!...

LATES!!!...

Share this post


Link to post
Share on other sites
What are you talking about Euegene? You honestly believe that Command tagged queing does little for the desktop environment, wake up BROTHER! You know, since it is assumed that you know hdds, there is a change that I have noticed in recent months.
The queue depth of a desktop IDE drive is rarely greater than one, but why am I even replying to you? Go away.

Somebody ban this person. Why? Read the history of posts (just don't read mine!)

Share this post


Link to post
Share on other sites

Sivar get the hell off this site. Intelligence is wanted not stupidity (hence you must leave cuz I said so)..

Sivar, if you had a brain, you would know something about hdds. I know more than YOU, so go smoke a joint in your Santa Barbara California, and do something else than wasting SR's hdd space by writing your constant babble.

Share this post


Link to post
Share on other sites

I'm sorry, but I couldn't let that slide. Godgod, if you want people to listen to what you have to say, I would suggest not YELLING at people. Eugene has done extensive work benchmarking HDD's and points out he is basing his opinion on tagged queuing from informal tests he has run.

If you disagree, please show data demonstrating otherwise.

Just because a HDD firm markets something as "the next big thing" doesn't actually mean it really is. ;) Until I see data on tagged queuing from an impartial third party (i.e. not just from a HDD maker), I will reserve judgment on how effective and important it really is for the typical desktop user.

Share this post


Link to post
Share on other sites

From a post by H-a-n-d (another of godgod's handles) dated september 4, 2002;

Under the last two months You have known me As Jonathan Davis? It seems a number of you so called members Do not like my presence here that is unfortunate, In an international forum you must except different points of view and different ways of expressing Your point, that is the difference between A dictatorship And a free society, Otherwise it will grow inside itself and destroy itself like any cancer.  

In any society there is always a difference of opinion Which comes in many forms that is how a society grows the Good with the Bad, Intolerance with Tolerance, Hatred with Love, If this forum is to survive And grow it must Except In tolerant opinions which are confrontational As well and not only the opinions that conform To certain people’s Way Of thinking and the way they would like the world To be seen.  

If you don’t like certain posts don’t read them?  

Two all you bleeding heart liberals my advice to you is to get a life and stop complaining about the world giving you such a hard deal?  

Have A Nice Day  

PS I will be watching you knew from time to time If need be I will be back  

Former Nicks  

Its.fubar  

Monica Lewinski

Note the date "godgod" registered. He's been banned once already. He sent malicious email to Eugene and several other members. His act is really getting old.

Share this post


Link to post
Share on other sites
I'm sorry, but I couldn't let that slide. Godgod, if you want people to listen to what you have to say, I would suggest not YELLING at people. Eugene has done extensive work benchmarking HDD's and points out he is basing his opinion on tagged queuing from informal tests he has run.

If you disagree, please show data demonstrating otherwise.  

Just because a HDD firm markets something as "the next big thing" doesn't actually mean it really is. ;) Until I see data on tagged queuing from an impartial third party (i.e. not just from a HDD maker), I will reserve judgment on how effective and important it really is for the typical desktop user.

Just ignore him Mickey. He's a troll who Eugene and Davin banned a while back and he still holds a grudge against them. He's only about 1/2 as clever as he thinks he is and always manages to slip up and get caught.

Share this post


Link to post
Share on other sites

There can't be that many readers on his subnet. A single line in an IPchains script can do wonders.

Share this post


Link to post
Share on other sites
hey flagreen, you must be related to sivar, stupidity runs rampant here.
At least, it has since about September 18, 2002. :roll:

Share this post


Link to post
Share on other sites
I think that the leaderboard should have the newest additions to its list even though they may not be available yet? 

Hey S_I_V_A_R I'M back!!!...

ahahahaha

Hmm...This proves he was here before :evil:. He has been trying to start trouble with other members including me("#$%ed up pic (all shot up), why dont you do that to yourself" "WHEN IT DOES GO SUPERNOVA, MAKE SURE THAT "YOU" ARE IN THE WAY SO YOU WILL DIE.") since he first registered as H_A_N_D and its.fubar.

Share this post


Link to post
Share on other sites
Guest russofris

Gentlemen,

I am fairly certain that Eugene is aware of the problem and has likely taken steps to correct it.

Thank you for your time,

Frank Russo

Share this post


Link to post
Share on other sites

He's definitely ignorant. Hasn't come here to help or learn but to

show what he *doesn't* know.

Anyway, those stats are great! I can't wait for the Atlas IV.

I'm supposing the increased retrieved data aacounts for the

higher CPU usage of the Seagate 15K.3 and Atlas IV (compared

to the previous models). The only thing missing was the temp

for the Atlas IV. Is it still in the ballpark of the Atlas III -in other

words not excessively hot?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this