Beware! MaxBoost

The_Neon_Cowboy · Dec 20, 2003, 02:56 PM

Quote:

[URL=http://www.hardwareheaven.com/showthread.php?s=&postid=263176]
Originally posted by johnybravo
Maxtor Corporation – a well-known maker of hard disk drives – has announced its new software driver that may boost performance of HDDs produced by the company by stunning 60%, according to some reports

MaxBoost is performance-boosting driver software which operates under Microsoft Windows 2000 and XP and which is designed to complement your Parallel ATA or Serial ATA hard disk drive by Maxtor. The MaxBoost driver intelligently caches data in the host system RAM before it is written to and read from the Maxtor disk drive, enhancing the effective storage speed of your system under a variety of system conditions and applications.

“Maxtor has released a new version of its MaxBoost Driver that is supposed to boost performances up to 60% on their hard disk drives,” a person, who tested the latest beta version of MaxBoost said.

You can grab the software to increase performance of your http://www.maxtor.com/en/support/dow...t/download.cfm

two drives tested as proof a 60gb 2mb cache maxtor and a 120gb 8mb cache maxtor

60gb

test with out maxtor utill

test with maxtor utill

120gb

test with out maxtor utill

test with maxtor utill

so the hdd cpu use is doubled!
the read time delayed
and the throughput reduced

this does exactly the opposite. it supposed to do

**MIG-31** · Dec 20, 2003, 03:14 PM

good of you to point that out. still keeping up the good work i see.

The_Neon_Cowboy · Dec 20, 2003, 04:21 PM

Quote:

Originally posted by johnybravo
good of you to point that out. still keeping up the good work i see.

thanks,

now it's possable that it could help in some circumstances were many small files are being created and then read back real fast but i cant think of anything thing would do this... but as tested it doesnt look good

even if it worked as advertised 58+% cpu useage is way to high...

the bench gives you a idea of the perfromance your gonna get. becouse you system reads alot more then it writes.... and if you have set your io pagelock to a larger size you'll see a much better gain then this program can provide.

before you guys ask

HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\Se ssion Manager\Memory Management\IoPageLockLimit

for example to set you set your pagelock to 256 mb

so hex value would be 10000000 (dec 268435456)

windows defult is set to 512k

No_Style · Dec 20, 2003, 07:59 PM

Looks like they don't support RAID setup.

Could you post some test system specs?

This is still TrialWare though.

hohum · Dec 20, 2003, 08:21 PM

Sounds a lot like the Intel Application Accelerator (which I use), except does the exact opposite of what it's supposed to do. Which is nice. :P

DarkFoss · Dec 20, 2003, 08:47 PM

Nice tip Neon_Cowboy, I haven't seen that one.
Unfortunately I don't have the key IoPageLockLimit under Memory Managment or it's sub-group pre-fetch parameters. How do I go about making it ? I know right clicking will create a new key but I don't know which choice to select ; string value, binary value, DWORD value, multi-string value or expandable string value. Afterwhich I just enter your values posted above and save correct ?

I agree with your observations 100%.
I also registered and loaded maxtors utility to try and it's pure junk no increases what so ever. Just increased cpu usage.

DonnoD · Dec 20, 2003, 08:58 PM

Why are you using HD tach? It's completely useless in this case. Of course software can NOT increase the performance of the actuator nor the Sustained Transfer rate. These performance aspects are limited almost exclusively by the hardware. To measure application performance, you must use a benchmark that simulates application performance. Such as Winbench (popular but outdated, not accurate), IPEAK's WinTrace32-RankDisk (good benchmark for single users), and IOMeter (good server/multi-user bench).

The_Neon_Cowboy, you even quoted in your own post that this software increases performance by CACHING. Caching will not improve low-level hardware performance (which is all that HD tach is capable of measuring). Caching will improve performance by intelligently pre-fetching and storing data into fast system ram instead of reading it off the slower Hard drive.

Keep in mind that windows already does this, so I assume maxtor hopes to increase performance even more by better pre-fetch algorithms and better understanding of the internals/firmware of their harddrives.

The_Neon_Cowboy · Dec 20, 2003, 09:04 PM

Quote:

Originally posted by No_Style
Looks like they don't support RAID setup.

Could you post some test system specs?

This is still TrialWare though.

2500+
9800pro
windows xp
1 gb geil ram
a7n8x (nforce2)

the two drives by them selfs primary master / secondary master
using rounded/shielded ata 133 cables

Quote:

Originally posted by DarkFoss
Nice tip Neon_Cowboy, I haven't seen that one.
Unfortunately I don't have the key IoPageLockLimit under Memory Managment or it's sub-group pre-fetch parameters. How do I go about making it ? I know right clicking will create a new key but I don't know which choice to select ; string value, binary value, DWORD value, multi-string value or expandable string value. Afterwhich I just enter your values posted above and save correct ?

I agree with your observations 100%.
I also registered and loaded maxtors utility to try and it's pure junk no increases what so ever. Just increased cpu usage.

choose DWORD value

and 256 works good for me... u can use more but it's not a good idea unless you got alot of ram... i've got a gig but if you running 512 i'd set it @ 128mb

heres a utill for you to run... it'll create the keys for you and tweak a buch of other settings for u

http://content-g.kontiki.com/fserve/...3/cachm550.exe
just go in a run the wizzerd from the menu and keep click'n next/ok till it finshs then you need to set it not to auto load (u less you really want it to) and then choos save and then reboot

the tweak works well and can be seen expecailly when installing software... it's like releseing a brake on your installer sometimes... sure makes office xp install faster

as for the no increse i found it funny my bechmark and diskeeper 8 pro tell me my acess times
while maxtors utill kept trying to sell to me i was getting like .58 ns write and 3 ns read ....
me thinks the maxtor utill is lieing

DarkFoss · Dec 20, 2003, 09:07 PM

Yeah it does state on maxtor's site that most benching programs like Sandra, HD Tach won't show any improvement and recommends using Winbench99 to test for performance gains. They even provide a link for downloading it. I haven't configured everything to be able to complete full tests with it yet, frelling bench whines about everything and shuts down before completion.

Thanks for the info and wizzy link Neon

can't wait to give it a whirl.

The_Neon_Cowboy · Dec 21, 2003, 02:28 AM

Quote:

Originally posted by DarkFoss
Yeah it does state on maxtor's site that most benching programs like Sandra, HD Tach won't show any improvement and recommends using Winbench99 to test for performance gains. They even provide a link for downloading it. I haven't configured everything to be able to complete full tests with it yet, frelling bench whines about everything and shuts down before completion.

Thanks for the info and wizzy link Neon can't wait to give it a whirl.

Winbench99 ? rofl its a several year old program! were not benching "windows 98" here.... of couse it gonna run good rofl it out of date... hardware has moved beyond that lets take a look..............

so intersiong info about how inacure winbench is

"the results scored in this benchmark did not show any conformity with frame rates scored in any real world 3D game."

http://www6.tomshardware.com/graphic...214/index.html

edit OMFG....

ok the memory cache that maxtor program offers is 32mb ...
the total program of winbench is 32 mb meaing the entire program fits in thier cache.....

also heres the system equirements
640x480 video
16 mb ram

edit again....

it writes and times 16 mb then times reading 16 mb.. as the defult test i'm setting it to rwea real test value

fin · Dec 21, 2003, 03:55 PM

Check this

note decreased acces time

and my benchmarks in PCMark 2002 with
Celeron 1100@1667 / ABIT SA6 (i815EP) / 512 MB SDRAM / Quantum CR 8,4 GB / Windows 2000 SP4

note File copy score (MB/s)

P.S. HDTach 2.70 is far less reliable than 2.61, in Russian you can check this here or use babelfish

FluffyChicken · Dec 21, 2003, 04:07 PM

It's a caching program and HDTach (which is a poor program v 2.7 being worse than 2.61 btw) should not show any difference.

There is no way it can !!!!

As stated above,
Use Application Benchmark/ Database benchmark or similar that will make use of the cache. It is designed to improve general drive performance (not play magic with physics

)

WinBench99 .... use the STR graphing part of it instead of HDTach

As far as I can see it just a caching program optimised to run in conjunction with Maxtor's ondrive Cache instead of using generic .....

anyway what has Framerates (from the second most unreliable review site

) got to do with it?

The_Neon_Cowboy · Dec 21, 2003, 04:34 PM

Quote:

Originally posted by FluffyChicken
It's a caching program and HDTach (which is a poor program v 2.7 being worse than 2.61 btw) should not show any difference.

There is no way it can !!!!

As stated above,
Use Application Benchmark/ Database benchmark or similar that will make use of the cache. It is designed to improve general drive performance (not play magic with physics )

WinBench99 .... use the STR graphing part of it instead of HDTach

As far as I can see it just a caching program optimised to run in conjunction with Maxtor's ondrive Cache instead of using generic .....

anyway what has Framerates (from the second most unreliable review site ) got to do with it?

the defult and the max settingas allowed by winbench99 fit nicely into a 32 mb cache
winbean99 is a inappreate/usless test unless your back in 1998

i will user other tests ... but not one that are to old and olny seek to defeate 2mb the defult cache on the
hdd to ensure proper testing....

do you realize how much slower your system is with 60% of you cpu time locked to the hdd

that means a 2 ghz system would have only have 800 mhz free to process programs!!!!!
the ultra high cpu usage is unacceptable in any test....

by useing max boost my hard drive was useing and extra 600 mhz to process hdd data!!!!

no mater the test such cpu usage would slow the system down emencly

FluffyChicken · Dec 21, 2003, 05:56 PM

I would not believe the CPU usage fugures myself.

Get a Process Monitor, find the drives thread and graph it usage yourself.

I said the STR test !!!!!
i.e. plots out the Sequential Transfer Rates
it is more accurate than the HDTach you run and also works fine when testing RAID configuations.
That was for a direct an alternative to HDTach.

ie.
Maxtor DM+9 (8Mb cache, 80GB/platter)

EDIT: btw, that THG link was to 3DWinBench -- nothing related to anything here, it's even the wrong program

FluffyChicken · Dec 21, 2003, 06:01 PM

One Benchmark you may want to try is Passmark5, it got quite a comprehensive disc benchmark now BUT I've not tried it yet myself (not sure if you need the Pro version or not)

Mosquito · Dec 21, 2003, 06:04 PM

I found this through a Google search:http://www.techspot.com/vb/showthread.php?threadid=509

Quote:

A few FACTS, not speculation...

First off, it's most definitely in bytes, not kilobytes. (The first thing the code does with it is shift it right 12 bits, to turn the number into number of pages. If you have a debugger hooked up you can see the result in the global MmLockPagesLimit...)

Second, if you set it to something less than 524,288 (512K), the system will just set it to 524,288 before converting it to pages and storing the result. Also, if you set it too large for the available memory, the system will, you guessed it, set it back down to 512K. Since lots of "tweak" sites are saying that the value is expressed in KB, this results in very small numbers in the registry value, resulting in the default value being used internally.

Third, very few setups are going to even use a fraction of the 512K default. Setting it larger will not make anything run faster unless you're already running into the limit. This value sets a quota on how much memory can be temporarily page-locked for the duration of currently-pending DMA IO requests. You need to understand that most of the time you have only one IO request pending per disk device, and most of those requests are 64Kbytes or less in size. (Use PerfMon and see.) Most systems never bump into this limit and if they do, the result is FAILURE of I/O requests, not poor performance.

Fourth, IoPageLockLimit is ignored as of Win2K SP1, SP2, and Windows XP. Fergetaboutit, unless you're running a very old version of the OS. You can prove this by running Mark Russinovich's "strings" on ntoskrnl.exe, and noting that the string "IoPageLockLimit" is gone as of SP1, or by using his "regmon" and noticing that the value is no longer being read.

All indications to the contrary are placebo effect, i.e. "I tweaked this and the result was so and so," but that doesn't prove that the tweak CAUSED so-and-so. Basic post hoc fallacy.

The_Neon_Cowboy · Dec 21, 2003, 06:13 PM

iopagelock

thats funny becouse it makes my office xp, and game intalls take about 1/2 the time
also it reduces the time of transfer of files frm hdd to hdd

max boost
ok max bost is giveing you 32 mb hdd cacheing @ a micrese of cpu usage

windows can be set to cache it via iopagelock 32 mb? i have mine set at 256 mb meaning thier is no need for maxboost to even exist!

as for bechmarking it ... it does no goos to uses any benchmark thats useing a file or amount of data to read write .. thats less then 32 ...

pleasw tell me ho manty program you run that the program and all its dependaceis that are a total of 32mb?

my adverage files size is about 128 mb!
most of my game folder 2+ gb
my videos etc....

that utillity blows nuff said

well have to do this all over agian when we get a new release

Dec 20, 2003, 03:14 PM	#2
MIG-31 HardwareHeaven News Mod Join Date: Dec 2002 Location: Nottingham,UK Posts: 55,770 Rep Power: 444 System Specs	good of you to point that out. still keeping up the good work i see. __________________ Please Post Your News Submissions Here Last edited by MIG-31; Dec 20, 2003 at 03:32 PM.

Dec 20, 2003, 07:59 PM	#4
No_Style Styleless Wonder Join Date: Jun 2002 Location: Ottawa, Ontario Posts: 6,034 Rep Power: 0	Good Work, Neon Looks like they don't support RAID setup. Could you post some test system specs? This is still TrialWare though. __________________ "The Best Style Is No Style" Specifications Here

Dec 20, 2003, 08:47 PM	#6
DarkFoss HardwareHeaven Senior Member Join Date: May 2002 Location: Downers Grove IL Posts: 688 Rep Power: 0	Nice tip Neon_Cowboy, I haven't seen that one. Unfortunately I don't have the key IoPageLockLimit under Memory Managment or it's sub-group pre-fetch parameters. How do I go about making it ? I know right clicking will create a new key but I don't know which choice to select ; string value, binary value, DWORD value, multi-string value or expandable string value. Afterwhich I just enter your values posted above and save correct ? I agree with your observations 100%. I also registered and loaded maxtors utility to try and it's pure junk no increases what so ever. Just increased cpu usage. __________________ Those who would give up Essential Liberty to purchase a little Temporary Safety,deserve neither Liberty nor Safety. Ben Franklin, 1755

Dec 20, 2003, 09:07 PM	#9
DarkFoss HardwareHeaven Senior Member Join Date: May 2002 Location: Downers Grove IL Posts: 688 Rep Power: 0	Yeah it does state on maxtor's site that most benching programs like Sandra, HD Tach won't show any improvement and recommends using Winbench99 to test for performance gains. They even provide a link for downloading it. I haven't configured everything to be able to complete full tests with it yet, frelling bench whines about everything and shuts down before completion. Thanks for the info and wizzy link Neon can't wait to give it a whirl. __________________ Those who would give up Essential Liberty to purchase a little Temporary Safety,deserve neither Liberty nor Safety. Ben Franklin, 1755

Dec 21, 2003, 03:55 PM	#11
fin HardwareHeaven Newbie Join Date: Aug 2003 Location: Odessa, Ukraine Posts: 11 Rep Power: 0	Check this note decreased acces time and my benchmarks in PCMark 2002 with Celeron 1100@1667 / ABIT SA6 (i815EP) / 512 MB SDRAM / Quantum CR 8,4 GB / Windows 2000 SP4 note File copy score (MB/s) P.S. HDTach 2.70 is far less reliable than 2.61, in Russian you can check this here or use babelfish Last edited by fin; Dec 21, 2003 at 04:05 PM.

Dec 20, 2003, 08:21 PM	#5
hohum HardwareHeaven Addict Join Date: Apr 2003 Location: UK Posts: 298 Rep Power: 54	Sounds a lot like the Intel Application Accelerator (which I use), except does the exact opposite of what it's supposed to do. Which is nice. :P

Dec 20, 2003, 08:58 PM	#7
DonnoD DriverHeaven Junior Member Join Date: Jun 2002 Posts: 41 Rep Power: 0	Why are you using HD tach? It's completely useless in this case. Of course software can NOT increase the performance of the actuator nor the Sustained Transfer rate. These performance aspects are limited almost exclusively by the hardware. To measure application performance, you must use a benchmark that simulates application performance. Such as Winbench (popular but outdated, not accurate), IPEAK's WinTrace32-RankDisk (good benchmark for single users), and IOMeter (good server/multi-user bench). The_Neon_Cowboy, you even quoted in your own post that this software increases performance by CACHING. Caching will not improve low-level hardware performance (which is all that HD tach is capable of measuring). Caching will improve performance by intelligently pre-fetching and storing data into fast system ram instead of reading it off the slower Hard drive. Keep in mind that windows already does this, so I assume maxtor hopes to increase performance even more by better pre-fetch algorithms and better understanding of the internals/firmware of their harddrives.

Dec 21, 2003, 04:07 PM	#12
FluffyChicken DriverHeaven Addict Join Date: Nov 2002 Location: England Posts: 320 Rep Power: 0	It's a caching program and HDTach (which is a poor program v 2.7 being worse than 2.61 btw) should not show any difference. There is no way it can !!!! As stated above, Use Application Benchmark/ Database benchmark or similar that will make use of the cache. It is designed to improve general drive performance (not play magic with physics ) WinBench99 .... use the STR graphing part of it instead of HDTach As far as I can see it just a caching program optimised to run in conjunction with Maxtor's ondrive Cache instead of using generic ..... anyway what has Framerates (from the second most unreliable review site ) got to do with it? __________________ [COLOR=#455b64]Team mauisun.org [/COLOR]

Dec 21, 2003, 05:56 PM	#14
FluffyChicken DriverHeaven Addict Join Date: Nov 2002 Location: England Posts: 320 Rep Power: 0	I would not believe the CPU usage fugures myself. Get a Process Monitor, find the drives thread and graph it usage yourself. I said the STR test !!!!! i.e. plots out the Sequential Transfer Rates it is more accurate than the HDTach you run and also works fine when testing RAID configuations. That was for a direct an alternative to HDTach. ie. Maxtor DM+9 (8Mb cache, 80GB/platter) EDIT: btw, that THG link was to 3DWinBench -- nothing related to anything here, it's even the wrong program __________________ [COLOR=#455b64]Team mauisun.org [/COLOR]

Dec 21, 2003, 06:01 PM	#15
FluffyChicken DriverHeaven Addict Join Date: Nov 2002 Location: England Posts: 320 Rep Power: 0	One Benchmark you may want to try is Passmark5, it got quite a comprehensive disc benchmark now BUT I've not tried it yet myself (not sure if you need the Pro version or not) __________________ [COLOR=#455b64]Team mauisun.org [/COLOR]

Dec 21, 2003, 06:13 PM	Thread Starter #17
The_Neon_Cowboy HardwareHeaven Extreme Member Join Date: Dec 2002 Location: U.S.A. Posts: 16,009 Rep Power: 90 System Specs	iopagelock thats funny becouse it makes my office xp, and game intalls take about 1/2 the time also it reduces the time of transfer of files frm hdd to hdd max boost ok max bost is giveing you 32 mb hdd cacheing @ a micrese of cpu usage windows can be set to cache it via iopagelock 32 mb? i have mine set at 256 mb meaning thier is no need for maxboost to even exist! as for bechmarking it ... it does no goos to uses any benchmark thats useing a file or amount of data to read write .. thats less then 32 ... pleasw tell me ho manty program you run that the program and all its dependaceis that are a total of 32mb? my adverage files size is about 128 mb! most of my game folder 2+ gb my videos etc.... that utillity blows nuff said well have to do this all over agian when we get a new release __________________ Last edited by The_Neon_Cowboy; Dec 24, 2003 at 08:41 AM.