Windows Experience Index says my disk transfer rate sucks

[fulle]

Thanks Rako. I had the same issue, and assumed it was just a bug. Nice to know it was just a little bug when write caching is enabled.

 

[rako77]

Thanks for the link, and the tip. I as well wondered about the low disk scores in win7, but now it makes good sense. Basically, caching a cache is never a good thing.

I think I'll run HDTune with and without that setting before I commit myself, but I suspect that it will confirm my thoughts.

I think you'll find better performance with cache enabled its just you have to disable it to use wei because somehow the cache can throw it off.

 

[NewbieTechGodII]

My 3 WDYS2500's have Write Caching and Advanced Performance enabled and my WEI score is still 5.5 on both Vista 32 and Windows 7 x64. It has thrown me that MS has upped the max score for W7 but not for Vista...yet.


EDIT: I turned off the above-mentioned "Performance enhancers", restarted my system, and re-ran WEI. Same score. So much for that doo-doo.

 

[rako77]

its only a fix if your score is significantly lower there was nothing to fix for your hard drive.

 

[Hurricane Andrew]

Seems to be an issue with SATA drives. From another post of mine:

Same thing here. Athlon 64 X2 5200+, 8GB RAM, 2 SATA hard drives, nForce 4 Ultra chipset (Abit KN9-Ultra), Radeon HD3650 512MB PCIe video card. Disk score in Vista was 5.7, now it's 3.0. Processor score actually went up from 5.2 to 5.7?

That disk score though is somewhat disconcerting, but I fear all too real. In a dual boot setup with Vista x86 and Windows 7 64-bit, I used Nero vision to create a DVD from an .avi file. The video was 1:22 (hr:min). In Windows 7 it was well over an hour, and I was just over half done. Switched to the Vista install and it completed in just over 45 minutes. Clearly there are other variables here, one obvious one is that Win7 was using Nero 8 and Vista using Nero 9. Also, the 64-bit Win 7 vs. 32-bit Vista (both Ultimate). I could clearly see the video in the window as it was being encoded was somewhat "jerky" in Windows 7. Given the size of the files and the disk usage during encoding, it leads me to believe that the lower disk score reflects a very real underlying issue.

Graphics scores were also down, but only by .6 and .5, and I attribute that to driver issues. The ATI Beta driver that Windows Update recommended left me with an ugly situation where my Samsung LCD was not recognized and the display adapter in Device Manager had the lovely yellow exclamation point. I had to uninstall and manually install the Catalyst 8.12 drivers to restore things to normal.

UPDATE: It seems to be an issue with SATA disks. I've seen a few rumblings around with folks having similar issues with SATA drives. Strangely, my test work PC, which is only a single core Athlon 64 3400+, with 3GB RAM, and an older single 80GB IDE disk, which scored a 5.4 on the WEI for the Primary Disk.

 

[Scotteq]

I think I stumbled on the answer why we are seeing this:

Microsoft changed the way WEI is testing the discs, because they were finding that the way some were handling random I/O was less than optimal. This also seems to explain why enabling/disabling write caching has the effect on the score that it appears to. As explained below:


From: http://blogs.msdn.com/e7/archive/2009/01/19/engineering-the-windows-7-windows-experience-index.aspx


Key quote:

With respect to disk scores, as discussed in our recent post on Windows Performance, we’ve been developing a comprehensive performance feedback loop for quite some time. With that loop, we’ve been able to capture thousands of detailed traces covering periods of time where the computer’s current user indicated an application, or Windows, was experiencing severe responsiveness problems. In analyzing these traces we saw a connection to disk I/O and we often found typical 4KB disk reads to take longer than expected, much, much longer in fact (10x to 30x). Instead of taking 10s of milliseconds to complete, we’d often find sequences where individual disk reads took many hundreds of milliseconds to finish. When sequences of these accumulate, higher level application responsiveness can suffer dramatically.

With the problem recognized, we synthesized many of the I/O sequences and undertook a large study on many, many disk drives, including solid state drives. While we did find a good number of drives to be excellent, we unfortunately also found many to have significant challenges under this type of load, which based on telemetry is rather common. In particular, we found the first generation of solid state drives to be broadly challenged when confronted with these commonly seen client I/O sequences.

An example problematic sequence consists of a series of sequential and random I/Os intermixed with one or more flushes. During these sequences, many of the random writes complete in unrealistically short periods of time (say 500 microseconds). Very short I/O completion times indicate caching; the actual work of moving the bits to spinning media, or to flash cells, is postponed. After a period of returning success very quickly, a backlog of deferred work is built up. What happens next is different from drive to drive. Some drives continue to consistently respond to reads as expected, no matter the earlier issued and postponed writes/flushes, which yields good performance and no perceived problems for the person using the PC. Some drives, however, reads are often held off for very lengthy periods as the drives apparently attempt to clear their backlog of work and this results in a perceived “blocking” state or almost a “locked system”. To validate this, on some systems, we replaced poor performing disks with known good disks and observed dramatically improved performance. In a few cases, updating the drive’s firmware was sufficient to very noticeably improve responsiveness.

To reflect this real world learning, in the Windows 7 Beta code, we have capped scores for drives which appear to exhibit the problematic behavior (during the scoring) and are using our feedback system to send back information to us to further evaluate these results. Scores of 1.9, 2.0, 2.9 and 3.0 for the system disk are possible because of our current capping rules. Internally, we feel confident in the beta disk assessment and these caps based on the data we have observed so far. Of course, we expect to learn from data coming from the broader beta population and from feedback and conversations we have with drive manufacturers.

For those obtaining low disk scores but are otherwise satisfied with the performance, we aren’t recommending any action (Of course the WEI is not a tool to recommend hardware changes of any kind). It is entirely possible that the sequence of I/Os being issued for your common workload and applications isn’t encountering the issues we are noting. As we’ve said, the WEI is a metric but only you can apply that metric to your computing needs.

 

[charles48864]

disable disk caching and it will go up. Bug in windows 7 for sata drives

 

[garygs22]

My hard disk score is much lower in w7 too. 5.3 in vista and 2.0 with W7.
I am running w7 on a dual boot partition. w7 partition is only 30GB and may be the cause of my low score.
maybe I am wrong, please comment.

I too am scoring a low 2.9 with a SATA 500GB Seagate, with only Windows 7 installed on it, and this is on my desktop Quad 6600 with 4gb of ram; so I don't think his hd is the issue. I scored a 5.9 on Vista 32 bit with this drive.

 

[charles48864]

I also seen updating the intel driver update from the intel site will help. Know bug in windows 7 for sata drives with the intel sata controller. Also seen turning off Command Queuing


http://www.sevenforums.com/general-discussion/342-show-us-your-wei-16.html#post10171

 

[garygs22]

I also seen updating the intel driver update from the intel site will help. Know bug in windows 7 for sata drives with the intel sata controller. Also seen turning off Command Queuing


http://www.sevenforums.com/general-discussion/342-show-us-your-wei-16.html#post10171

You were correct about the Disc Caching, that cured the problem !
Now running back up to a 5.9, thanks.

 

[mesh1]

Anyone has:

Windows 7
SSD or VelociRaptor (one or in RAID-0)

This would give us an idea of what could be the maximum?

I have 2 VelociRaptors in RAid0 and the data transfer rate is 6.2
P6T Deluxe
Core i7 920 @ 3.8
6gb dominator
ati 4870 (1gb)

Im running the Latest windows 7 build (the rumored gold RTM build)

 

[RoboG]

Core i7 2.66 (Score 7.4)
6GB Corsair (Score 7.6)
Ati 4870 512GB (Graphics : Score 7.4 - Gaming Graph : Score 7.4 )
HD Data Transfer : (Score 5.9) :-(((( What the f***

The HD is new 2 weeks Barracuda 7,200 rpm with cache 16MB

Why that low..??

 

[Zoron]

Did anyone read Scott's post? There is a good explanation of what's happening there... there is no "bug". MS did this intentionally.

 

[sminlal]

HD Data Transfer : (Score 5.9) :-(((( What the f***

The HD is new 2 weeks Barracuda 7,200 rpm with cache 16MB

Why that low..??

That's not low. It's actually the highest possible score under Vista. Under Win7 they extended the range to 7.9 to account for faster solid state disks.

 

[NarbutM]

I can agree with Mesh1 results as mine were similar:

Anyone has:

Windows 7
SSD or VelociRaptor (one or in RAID-0)

This would give us an idea of what could be the maximum?

My rating is 6.1 (Much lower than expected given all the effort and expense that went into building the RAID 0 array with the fastest SATA hard drives on the planet.

 

[Scotteq]

...just means Microsoft set the score high enough that we'll probably have to raid array the next generation drives in order to maxx it out.

Though if it bothers people *that* much, the scores are kept in a text format which can be edited.

 

[sminlal]

My rating is 6.1 (Much lower than expected given all the effort and expense that went into building the RAID 0 array with the fastest SATA hard drives on the planet.

...and what this should be telling you is that if you really want the best possible performance, stop messing around with hard drives.

 

[Bullheaded67]

My processor is 7.6 with an i7 920 OC'd to 3.6Ghz.
My memory is 7.7 with 6GB DDR3 1600 but its slightly underclocked to 14xx due to the multiplier
My graphics is 7.9 with 2x GTX285 in SLI
My Primary hard disk is 5.9 with 2 WD Black 640GB hard disks in RAID 0 - go figure.

 

[IQ11110002]

I can agree with Mesh1 results as mine were similar:

Anyone has:

Windows 7
SSD or VelociRaptor (one or in RAID-0)

This would give us an idea of what could be the maximum?

My rating is 6.1 (Much lower than expected given all the effort and expense that went into building the RAID 0 array with the fastest SATA hard drives on the planet.

Here is my SSD rig for comparison

Processor 7.5 Intel i7 950
Memory 7.5 Corsair 6gb DDR 3 1800C8
Graphics 6.4 Nvidia 280GTX
Gaming Graphics 6.4 ^ same ^
Primary Hard disk 6.6 OCZ Vertex Turbo 120gb SSD

 

[IQ11110002]

I read on another site that 2X Intel X-25 SSD's in Raid 0 =7.9
Hmmm curious if anyone has Vertex or Vertex turbo in a raid 0 array, Just to see if it scores 7.9 also. Would have to be close I'm guessing.

 

[IQ11110002]

Sorry my bad, Ram is 1600 C8 not 1800 C8 rating not that it matters, half asleep lol!

 

[Zoron]

If you don't notice the performance (or lack thereof) of your hard drives anywhere other than WEI, then I would give exactly two sh*ts about you WEI hard drive score. All you're going to do in trying to obtain that elusive 7.9 score is waste a lot of money for absolutely no reason. WEI is not a good indication of real world performance.

 

[IQ11110002]

WEI may not be perfect but it does give a general idea about performance!
Going from a raptor which gave me a score of 5.4 to SSD which gives me 6.6 and I notice the difference in real world apps. I bet moving from 1x SSD to SSD Raid 0 will not only give a higher WEI score but depending on what you do with pc you will also notice the performance jump.
WEI is not meant to be the be all and end all benchmark index it is just an indicator designed for mainstream or should I say noobs to use to give them a general indication wether their pc components are decent enough to run games/apps.
Otherwise Joe Bloggs down the road would have no idea what part to upgrade!
In that regard it does a good enough job,He/she buys a game for instance, looks at the rating and sees their graphics card has a lower score than game recommended settings,At least they know what to expect and do in future to make it better!

"If you don't notice the performance (or lack thereof) of your hard drives anywhere other than WEI, then I would give exactly two sh*ts about you WEI hard drive score. All you're going to do in trying to obtain that elusive 7.9 score is waste a lot of money for absolutely no reason. WEI is not a good indication of real world performance."
LOL maybe,But then again some people have more money to waste than others,While others enjoy bragging rights even though SLI is just as much a waste compared to just holding out for 3-6 months, Until next gen single card or overclocked current card gives just as much or more performance at a cheaper price!
All part of the game we play

 

[Zoron]

Bears repeating, since most people seem to be skipping over it:

With respect to disk scores, as discussed in our recent post on Windows Performance, we’ve been developing a comprehensive performance feedback loop for quite some time. With that loop, we’ve been able to capture thousands of detailed traces covering periods of time where the computer’s current user indicated an application, or Windows, was experiencing severe responsiveness problems. In analyzing these traces we saw a connection to disk I/O and we often found typical 4KB disk reads to take longer than expected, much, much longer in fact (10x to 30x). Instead of taking 10s of milliseconds to complete, we’d often find sequences where individual disk reads took many hundreds of milliseconds to finish. When sequences of these accumulate, higher level application responsiveness can suffer dramatically.

With the problem recognized, we synthesized many of the I/O sequences and undertook a large study on many, many disk drives, including solid state drives. While we did find a good number of drives to be excellent, we unfortunately also found many to have significant challenges under this type of load, which based on telemetry is rather common. In particular, we found the first generation of solid state drives to be broadly challenged when confronted with these commonly seen client I/O sequences.

An example problematic sequence consists of a series of sequential and random I/Os intermixed with one or more flushes. During these sequences, many of the random writes complete in unrealistically short periods of time (say 500 microseconds). Very short I/O completion times indicate caching; the actual work of moving the bits to spinning media, or to flash cells, is postponed. After a period of returning success very quickly, a backlog of deferred work is built up. What happens next is different from drive to drive. Some drives continue to consistently respond to reads as expected, no matter the earlier issued and postponed writes/flushes, which yields good performance and no perceived problems for the person using the PC. Some drives, however, reads are often held off for very lengthy periods as the drives apparently attempt to clear their backlog of work and this results in a perceived “blocking” state or almost a “locked system”. To validate this, on some systems, we replaced poor performing disks with known good disks and observed dramatically improved performance. In a few cases, updating the drive’s firmware was sufficient to very noticeably improve responsiveness.

To reflect this real world learning, in the Windows 7 Beta code, we have capped scores for drives which appear to exhibit the problematic behavior (during the scoring) and are using our feedback system to send back information to us to further evaluate these results. Scores of 1.9, 2.0, 2.9 and 3.0 for the system disk are possible because of our current capping rules. Internally, we feel confident in the beta disk assessment and these caps based on the data we have observed so far. Of course, we expect to learn from data coming from the broader beta population and from feedback and conversations we have with drive manufacturers.

For those obtaining low disk scores but are otherwise satisfied with the performance, we aren’t recommending any action (Of course the WEI is not a tool to recommend hardware changes of any kind). It is entirely possible that the sequence of I/Os being issued for your common workload and applications isn’t encountering the issues we are noting. As we’ve said, the WEI is a metric but only you can apply that metric to your computing needs.

 

[IQ11110002]

WEI is a very basic assessment intended for the masses.
I fail to understand what your getting at in above post? That is way to complicated for average joe to sink in.
Regardless of whatever hd benchmarking program they used, SSD drives still come out on top in nearly all tests,That equates to a noticeable feel in real world use.
All the person wanted to know was what score an SSD drive either single or RAID 0 would give in WIN 7, Wether it's intended for just bragging rights or a general guide to slowest component in WEI.
Once again I say X-25m SSD in raid 0 will give a score of 7.9 the highest possible WEI score and believe me you will notice the difference, As I already do just with 1 OCZ SSD!
My old Raptor booted WIN 7 in around 40 seconds,This single OCZ SSD boots in 19 seconds flat from when I switch on power button,Half of that time is Bios post so it's more like 10 seconds once windows logo pops up.
I hope that helps any newbies out there trying to decide on an SSD or not for next upgrade, 6.6 WEI score seems about right. 7.9 score for X25 SSD raid 0 would be right also and it is amongst the best drives for 4kb reads so the I/O issue is meaningless in real world performance.
The real issues that need to be resolved for SSD's are of course price but more importantly native TRIM support to keep the drives running at their best over time,Hopefully the finished Windows 7 will do a reasonable job of this.

Wolfenstein Recommended WEI is 5.0
Required is 4.0
Current system 6.4
Pointless to tech heads like you and I and most readers on these sites, But just may give average joe out there an understanding of why their pc sucks and won't run game or application to it's fullest potential. WEI may not be perfect but it's a step in the right direction!