+67% Performance for Maxtor HDD

[thorin]

Originally posted by: thegreypilgrim
BTW I have to appologize for the YELLING thing... Sorry, I'm just a french guy...

Ok you're forgiven

Thorin

Edit: Fixed Smiley.......

 

[mooojojojo]

Originally posted by: thorin

Originally posted by: thegreypilgrim
Try it cause IT IS NOT A CACHE PROGRAM ! I can't believe it is at least. Cause this thing decrease the time of transfert on a 687Mb DivX file from one HDD to another.

Am I wrong or this file doesn't fit in 2Mo (in my old but still usefull 10G drive)

OMG I can't believe how poor education is around the globe these days, I swear 75% of the population would fail basic reading comprehension tests. Not to mention the fact that people start YELLING even when they have to facts to back up what they're attempting to comprehend.

Here's a quote right off the page linked originally.

The MaxBoost driver intelligently caches data in the host system RAM before it is written to and read from the Maxtor disk drive, optimizing the effective storage speed of your system in a variety of applications.

Thorin

I'm sure thegreypiligrim is better in some areas which are not related to reading a language that is not native for him. You though, are the king of elitist posts. I'm sorry for coming out like an ass but you always reply with a tone like if you are somehow smarter than everyone around you.

On topic - now if only Seagate would come up with an app like that. Anyone tried yet to run it with a Barracuda? It doesn't explicitly say that it won't work with drives other than Maxtor/Quantum.

 

[MDE]

What are the minimum system requirements to use MaxBoost?
Maxtor or Quantum ATA or SATA hard drive (RAID not supported)

But I'm sure someone will figure a way to get Seagates\WDs to work with it, or just find another app that does the same thing.

 

[thorin]

Originally posted by: mooojojojo
I'm sure thegreypiligrim is better in some areas which are not related to reading a language that is not native for him. You though, are the king of elitist posts. I'm sorry for coming out like an ass but you always reply with a tone like if you are somehow smarter than everyone around you.

No yer not coming out like an ass at all, I know I'm mean sometimes. It just pisses me off when people try to speak authoritatively on things they don't know about. And if you've read alot of my replies (as you seem to have) then you know I'm the first to admit it when proven wrong (and likely even thank the person for pointing me to new info and/or adding to what I know).

On a side note you've replied 5 minutes after he appologized for yelling and I replied to him. Also I don't think language barriers are any valid excuse, I wouldn't go to a french/asian/whatever board and start blabing away about things I couldn't read and fully comprehend (which would be true of the majority I suspect).

Moral of the story is ppl shouldn't talk out of their butts (in a native or foreign language) unless they're willing to be corrected (sometimes harshly) by someone else who may know more then them. (Including myself of course.....as already mentioned above).

Thorin

 

[arcenite]

I want to try it on my WD... but... I want someone else to try it first... lol

Bill

 

[BD2003]

Originally posted by: MonkeyDriveExpress
From the download confirmation page

Thank you for downloading the MaxBoost Beta. You should receive your 10% off MaxStore.com coupon* via e-mail within the next 72 hours. Please note that the MaxBoost Beta software will expire and automatically disable itself 60 days after installation.

Looks like they expect this to generate a little cash for them.

Youre making a major jump in logic there. It says nowhere that youre going to have to pay for it, only that youll get 10% off stuff in their store. Perhaps thats their way of saying thanks for beta testing? They dont want you using it more than 60 days cause its a freaking beta.

Of course I'm not saying theres a possibility theyll charge, but it doesnt imply that at all.

 

[Adul]

This is a news worthy thread.
http://www.anandtech.com/news/shownews.html?i=20594

 

[MDE]

Originally posted by: BD2003

Originally posted by: MonkeyDriveExpress
From the download confirmation page

Thank you for downloading the MaxBoost Beta. You should receive your 10% off MaxStore.com coupon* via e-mail within the next 72 hours. Please note that the MaxBoost Beta software will expire and automatically disable itself 60 days after installation.

Looks like they expect this to generate a little cash for them.

Youre making a major jump in logic there. It says nowhere that youre going to have to pay for it, only that youll get 10% off stuff in their store. Perhaps thats their way of saying thanks for beta testing? They dont want you using it more than 60 days cause its a freaking beta.

Of course I'm not saying theres a possibility theyll charge, but it doesnt imply that at all.

My point is that they want you to pay for it. IMO, they should release it as a free goodie (kinda like MS does with the PowerToys for Windows) to thank people for buying their drives.

 

[Boonesmi]

seems like you would run a much higher risk of data loss and corruption with a sudden power failure when using this

 

[InlineFive]

A caching program, no wonder it sucks up memory. Luckily, I have 1GB! However the thing that you have to buy it stinks. Some programmer will probably buy it and release their own version (basically the same) under a different name, such as "M.A.X." or Maxtor Accelerator eXtreme! Like Asus and Abit did with PAT.

-Por

 

[RagingGuardian]

Has anybody tried this yet? My nforce board has horrible IDE performance so this might help.

 

[WobbleWobble]

Originally posted by: MonkeyDriveExpress

My point is that they want you to pay for it. IMO, they should release it as a free goodie (kinda like MS does with the PowerToys for Windows) to thank people for buying their drives.

I still don't know why you think they want you to pay for it. As for the 10% off at MaxStore.com coupon, read the main page. It is a thank you coupon.

Joining is easy: Just click on the Download Now link below, provide your user name and e-mail address, agree to the MaxBoost License Agreement, and download the MaxBoost software. As a thank you for joining the beta test, we'll send an e-mail to you with a 10% off coupon* good at MaxStore.com.

 

[JustAnAverageGuy]

uh... Am I doing something wrong... the download isn't working :'(

Entered my spammed email, a user name... clicked accept.

"Thank you for downloading the MaxBoost Beta. You should receive your 10% off MaxStore.com coupon* via e-mail within the next 72 hours. Please note that the MaxBoost Beta software will expire and automatically disable itself 60 days after installation."

Got that page. No download dialog box :/

* Maxtor only allows one coupon per beta user, regardless of the number of submissions.

I'm assuming they send the download link in the email?

*wants a linkified for the lazy link*

 

[JohnnyCNote]

I installed it but it wouldn't detect either of my Maxtor HDD's. I guess there's no way to find out what the problem is since they said there's no tech support....

 

[Auric]

It stops working so you don't keep using beta software. Common practice. They request feedback at the end of the trial. I doubt the final version will cost anything.

adendum:
user name and email are not required... enter anything and the download is immediate. The first time I did it though the link similarly did not appear, until I disabled my filter... don't know if it was pop-up (didn't appear to be one), auto refresh (also didn't seem to be), or wha... maybe some tracking thing.

 

[Pariah]

Windows already does the exact same thing this Maxtor driver does. If you have windows memory management setup properly it shouldn't make much of a difference unless Maxtor has done a much better job of writing the algorithm. Using Winbench as a barometer of performance is not a good idea. It is an excellent lowlevel benchmark, but the highlevel parts (WB Business, WB Highend) are completely inaccurate and should not be used for any sort of performance analysis.

This driver appears to be free. The 10% coupon is a gift for joining the beta test team.

 

[Budman]

Originally posted by: thegreypilgrim
No dual CPU ? No HyperThreading ?
I have an TBird so if anyone want to try it on a P4....

It will not install for me,the stupid program thinks my hyperthreading P4 cpu is a dual cpu system. :disgust:

 

[Walleye]

i've never installed disk drivers. was i supposed to?

 

[dszd0g]

Originally posted by: Boonesmi
seems like you would run a much higher risk of data loss and corruption with a sudden power failure when using this

Agreed. Windows does its own disk caching, but it takes steps in case power goes out or the system crashes to minimize damage. If this program does not take those steps that would explain the performance increase, but is it worth the risk of corrupting the system or one's data?

Why won't MaxBoost work on my older computer with less RAM and a slower processor?

MaxBoost requires a minimum set of memory and processor resources to effectively boost disk performance. The minimum specifications necessary to ensure a consistent performance boost are a 700 MHz processor and 256MB of RAM.

Those of us with 1GB of memory won't necessarily mind giving up 16-32MB for increased performance. But has anyone taken numbers for CPU usage with heavy I/O with and without this utility? I believe Winbench provides CPU usage numbers. IOmeter would probably give better numbers, but it's kind of a pain to setup.

 

[thegreypilgrim]

Any tests results with IOMeter ?

 

[DerekWilson]

Originally posted by: Pariah
Windows already does the exact same thing this Maxtor driver does. If you have windows memory management setup properly it shouldn't make much of a difference unless Maxtor has done a much better job of writing the algorithm. Using Winbench as a barometer of performance is not a good idea. It is an excellent lowlevel benchmark, but the highlevel parts (WB Business, WB Highend) are completely inaccurate and should not be used for any sort of performance analysis.

This driver appears to be free. The 10% coupon is a gift for joining the beta test team.

windows may do something like this, but it would have to be on a *much* smaller (Kb maybe?) scale. I know that windows does prefetches from the drive and has a bit of a buffer to store such data in, but I'm not sure about implimenting what seems like a full writeback cache in ram. One of the things that strikes me as fishy about this is that windows uses pagefiles for virutal memory on the disk to handle situations where memory beyond the capacity of physical ram is needed (or in certain other application specific kinds of things). It seems counter intuitive to think that these very writes to a disk would be cached in the ram that windows is trying to extend onto the disk... But, I don't have any cold hard facts that say whether or not (and to what extent) windows does hd cacheing in ram, so if anyone knows of a definitive source on this, I would really appreciate linkage.

Originally posted by: Boonesmi
seems like you would run a much higher risk of data loss and corruption with a sudden power failure when using this

You definitly do run these risks with such a system in place. With those 8MB buffers on some drives there is similar potential for loss, though sticking a writeback cache in system ram is much more risky in all regaurds.

Of course, for years with linux, people could create a huge ramdrive in system memory and use that like a harddrive, then just save everything to a physical disk when done with it ;-) ... teh uberrisk. but very fast

I do also take the view that any software which increases the performance of my hardware deserves some intense scepticism before I'll be comforatble. But perhaps this really will turn out to be a valid way to use the insane amount of RAM everyone will have in their 64bit systems o_0 ;-)

 

[zephyrprime]

windows may do something like this, but it would have to be on a *much* smaller (Kb maybe?) scale. I know that windows does prefetches from the drive and has a bit of a buffer to store such data in, but I'm not sure about implimenting what seems like a full writeback cache in ram.

Windows definitely has a write back disk cache. It's huge, not small. Windows disk cache is called VCache because it is integrated into the virtual memory system. It will use all available ram for it's disk cache. In the windows 3.1 day, the windows disk cache was proven to be king among x86 disk caches. But that was before VCache.

One of the things that strikes me as fishy about this is that windows uses pagefiles for virutal memory on the disk to handle situations where memory beyond the capacity of physical ram is needed (or in certain other application specific kinds of things). It seems counter intuitive to think that these very writes to a disk would be cached in the ram that windows is trying to extend onto the disk.

All windows disk caches ever made have been smart enough to avoid this problem so there's no need to worry.

 

[sharkeeper]

This is triones all over again.

Forget this folks and move on.

Cheers!

 

[zephyrprime]

Agreed. Windows does its own disk caching, but it takes steps in case power goes out or the system crashes to minimize damage. If this program does not take those steps that would explain the performance increase, but is it worth the risk of corrupting the system or one's data?

The windows XP write back cache is like 40MB which is bigger than the entire cache used by this program. The size of the write back cache will become unlimited if the system is set to Optimize for Services.

 

[zephyrprime]

Could someone try this on a Windows Server system and see if produces the same results?

I've noticed that non-server versions of Window XP seem to have an intentionally crippled disk subsystem that will not allow more than 6 pending IO interactions to occur. I can't be sure of this however. I'm wondering if this driver works by bypassing this limitation somehow.