So this is where my system finally sits... Just being jacked by having a Raptor drive & not SSD So ... I was gonna say "gay" here but that would be insensitive to any of our gay members, so instead of using "gay" as a slur, I'll say "so lame" or something equally non-offensive
The memory running at 1600 rather than 1200 made a big difference. Also pushing the CPU at 4.0GHz also gave me the 7.9 I needed/wanted
So this is where my system finally sits... Just being jacked by having a Raptor drive & not SSD So Gay!
Hawt.
Unfortunately for spindle-disk users, the 16k random seek score is what kills you even on a mega-fast drive (my WD Black scores a 5.9 as well). SSD Seek times are just simply so much faster than spindles.
I can't wait to get my hands on Nvidia's new GPU. I have been hearing good things about it.
Another thing I noticed not only in this test but in Super PI is that voltage seems to make a big difference. i was running 4.0GHz at 1.25V and my super PI score was 11.37sec I thank cranked it to 1.35 and got in the sub 10sec range...
I would love to hear someone thoughts on that situation.
Considering the number of systems hitting the Max of 7.9 I think this rating system is meaningless for top performing systems and hardware. it has more meaning for lower end systems and can point out bottlenecks on a system that should be upgraded
The i7 platform is a bit different than Core 2, though... from what I understand, the Core 2 platform doesn't really gain much benefit above 1:1 with the FSB.
I have no idea how this thing calculates HD score...
Here is what Winsat does to produce the performance benchmarks
for Hard drive it's a combined score that measures
Sequential I/O read performance
Random I/O write performance
flush performance
For CPU
encryption and decryption for AES 256bit and SHA1 hash
compression and decompression for Lempel-Zev algorithm and Windows internal algorithms
For memory
memory bandwidth in a manner reflective of large memory to memory buffer copies
For Graphics
graphics memory throughput (bandwidth) metric
D3D assessment
all the scores are then placed on an arbitrary scale that microsoft controls and claims to update with 1 point per month. (the win 7 version of winsat has a different scale then vista used and some changes to how ratings are measured)
I get a 6.0 if I have the November ATI drivers installed for my 4870x2, I get a 7.0 if I don't have them installed...
CPU is 7.5
Memory is 7.9
Graphics is 6.0
Gaming Graphics is 7.9
Primary Hard Disk is 7.6
I get a 6.0 if I have the November ATI drivers installed for my 4870x2, I get a 7.0 if I don't have them installed...
CPU is 7.5
Memory is 7.9
Graphics is 6.0
Gaming Graphics is 7.9
Primary Hard Disk is 7.6
A 6.0 on a 4870x2 is definitely a weird score. My Studio 1735's mobility 3650 got a 6.0...
While I don't completely understand the math used to determine the hard disk subscore, I'm starting to suspect that random seek times must play into it. Even when I had my entire set of 5 drives in a RAID5 array, I couldn't get anywhere near a 7.5 - but I was able to confirm with a separate testing tool that the array could sustain-read at 410MB/s...
Random seek time would explain the discrepancy. No matter how many drives there are in a RAID array, they're still limited by the physics of read/write head movement.
While I don't completely understand the math used to determine the hard disk subscore, I'm starting to suspect that random seek times must play into it. Even when I had my entire set of 5 drives in a RAID5 array, I couldn't get anywhere near a 7.5 - but I was able to confirm with a separate testing tool that the array could sustain-read at 410MB/s...
Random seek time would explain the discrepancy. No matter how many drives there are in a RAID array, they're still limited by the physics of read/write head movement.
Random I/O write performance is one of the 3 parts used to calculate the score (if you had read my other post I already had mentioned that)
Photodude, I didn't happen to equate "random I/O writes" with "random seek times" in my mind. It was an honest mistake - was it really worth getting sarcastic with me over?
Also, your post is not entirely correct. In fact, the testing does random and sequential tests on both read and write performance.
Comments
If you see no more than 150MB/s then you are on SATA1
The memory running at 1600 rather than 1200 made a big difference. Also pushing the CPU at 4.0GHz also gave me the 7.9 I needed/wanted
Hawt.
Unfortunately for spindle-disk users, the 16k random seek score is what kills you even on a mega-fast drive (my WD Black scores a 5.9 as well). SSD Seek times are just simply so much faster than spindles.
I can't wait to get my hands on Nvidia's new GPU. I have been hearing good things about it.
Another thing I noticed not only in this test but in Super PI is that voltage seems to make a big difference. i was running 4.0GHz at 1.25V and my super PI score was 11.37sec I thank cranked it to 1.35 and got in the sub 10sec range...
I would love to hear someone thoughts on that situation.
Considering the number of systems hitting the Max of 7.9 I think this rating system is meaningless for top performing systems and hardware. it has more meaning for lower end systems and can point out bottlenecks on a system that should be upgraded
I see your 6.2, and raise you:
5 platter drives in RAID5.
I'd love me some SSDs though.
I thought my score would suck but I guess not.
I'm starting too look at the Patriot Warp ssd...hmm....
And of course Sledge's rig kicks all of our's asses'.
[shatner]
beeeeeeeeaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaan!
[/shatner]
Still think these numbers are some what meaning less, other then I need a better GPU
Now to save up for a Quadrofx 1800 or Firepro v8700 and an SSD boot drive
cpu: Intel i5 - 750 @ 3.4ghz
MLB: Msi gd-65
Memory: patriot ddr3 cas7
Video: Galaxy gtx 260+
HDD: WD 1 tb dual processor
My Asus 1000HE Netbook (at best it's a 5400RPM drive) got 5.7 in Windows 7 performance index. My RAID5 array gets 6.1-6.3, depending how it feels.
:banghead:
Here is what Winsat does to produce the performance benchmarks
for Hard drive it's a combined score that measures
For CPU
For memory
For Graphics
all the scores are then placed on an arbitrary scale that microsoft controls and claims to update with 1 point per month. (the win 7 version of winsat has a different scale then vista used and some changes to how ratings are measured)
CPU is 7.5
Memory is 7.9
Graphics is 6.0
Gaming Graphics is 7.9
Primary Hard Disk is 7.6
A 6.0 on a 4870x2 is definitely a weird score. My Studio 1735's mobility 3650 got a 6.0...
Random seek time would explain the discrepancy. No matter how many drives there are in a RAID array, they're still limited by the physics of read/write head movement.
Random I/O write performance is one of the 3 parts used to calculate the score (if you had read my other post I already had mentioned that)
Also, your post is not entirely correct. In fact, the testing does random and sequential tests on both read and write performance.