Overclock.net › Forums › AMD › AMD Motherboards › [Official] ASUS M2N-SLI and M2N32-SLI Club!
New Posts  All Forums:Forum Nav:

[Official] ASUS M2N-SLI and M2N32-SLI Club! - Page 899

post #8981 of 12360
I am using my SSD connected the Nforce ports and where my SSD is located in my case, I don't have a cord long enough to reach the Silicon img. controller to test that.

I'm using the latest Nforce drivers in windows as well. (the may 2011 ones)

I finally got around to doing ATTO on my drive. I hear that this benchmark uses compressed data and SandFroce drives tend to score much higher.

I also posted the S.M.A.R.T. information on my drive using Crystal Disk Info so that you can see it's running at SATA2 speeds. The damn power on hours got reset with the recent firmware update. My drive has been running close to 3 months now.

The speeds on your drive look very low compared to mine and they should be in the 200 range unless the SF drives make a bigger deal about AHCI which I highly doubt.

I would have a look at the SSD forums here on OCN, they have been a great help to me when I was looking for my SSD. Check out the AS-SSD thread table as well which is located here

http://www.overclock.net/ssd/754763-...rk-thread.html

http://imageshack.us/photo/my-images/822/attonforce.jpg

http://imageshack.us/photo/my-images/40/smartinfo.jpg
Edited by Agent_kenshin - 8/23/11 at 4:10pm
post #8982 of 12360
The OCZ boards haven't given me the help I was looking for. They got a bit snippy when I suggested the firmware update may somehow have something to do with slowing down the drive.

The one suggestion they've given me but which I haven't tried yet is to clear the CMOS. They seem very insistent that this is essential when updating the firmware on the SSD, but the stated reasons usually revolve around AHCI which I don't have. I will to the CMOS clear just to get that off my plate possible over the next weekend or two.

For those that have SATA-3 capable boards with Intel chipsets, there seems to be conjecture that the power state throttling on the SATA link causes the drive to either stutter or cause the BSODs in Windows. Consequently OCZ if fighting a major customer sat issue and most are blaming the Sandforce controller.

Have you updated the firmware on your drives?
post #8983 of 12360
Quote:
Originally Posted by Agent_kenshin View Post
I am using my SSD connected the Nforce ports and where my SSD is located in my case, I don't have a cord long enough to reach the Silicon img. controller to test that.

I'm using the latest Nforce drivers in windows as well. (the may 2011 ones)

I finally got around to doing ATTO on my drive. I hear that this benchmark uses compressed data and SandFroce drives tend to score much higher.

I also posted the S.M.A.R.T. information on my drive using Crystal Disk Info so that you can see it's running at SATA2 speeds. The damn power on hours got reset with the recent firmware update. My drive has been running close to 3 months now.

The speeds on your drive look very low compared to mine and they should be in the 200 range unless the SF drives make a bigger deal about AHCI which I highly doubt.

I would have a look at the SSD forums here on OCN, they have been a great help to me when I was looking for my SSD. Check out the AS-SSD thread table as well which is located here

http://www.overclock.net/ssd/754763-...rk-thread.html

http://imageshack.us/photo/my-images/822/attonforce.jpg

http://imageshack.us/photo/my-images/40/smartinfo.jpg
I just looked at your graphs. Very nice indeed. What is the WEI score if I might ask, please?
post #8984 of 12360
Resetting the BIOS properly by loading defaults/removing the battery and moving that jumper might be a good idea. I bet that you have already tried a number of ports on your board.

My WEI was at 7.4 when my drive was only 15% full but about 2 weeks ago it went down to 7.3 all of a sudden. I have put more stuff on my drive since then and it's about 55% full. I decided to run my benchmarks again and my speeds really did not change too much. I hear that this is quite normal for a SSD for the WEI to drop a bit and see very little changes in benchmarks.

As for the firmware update for my drive, it was only to address the 8MB bug issue on this drive which basically if the drive was not powered down properly for some reason, (power outage, crash, etc) the drive would only show the capacity of only 8MB and you could not see anymore of the drive which meant that your data was lost. The only way you can get the full capacity back was to secure erase it.

I have heard a case that came up that the firmware did not fix this but I'm gonna wait for an official word from Intel on this. Until then, I will keep backing up my drive at least 3-4 times a week.

I know that the SF drives do have some odd throttling system with their drives. Extreme Systems encountered this issue on their experiment with 34nm versus 25nm NAND write endurance. Very intresting stuff they are doing. You can find out more about it here http://www.xtremesystems.org/forums/...e-25nm-Vs-34nm
post #8985 of 12360
Quote:
Originally Posted by Agent_kenshin View Post
Resetting the BIOS properly by loading defaults/removing the battery and moving that jumper might be a good idea. I bet that you have already tried a number of ports on your board.

My WEI was at 7.4 when my drive was only 15% full but about 2 weeks ago it went down to 7.3 all of a sudden. I have put more stuff on my drive since then and it's about 55% full. I decided to run my benchmarks again and my speeds really did not change too much. I hear that this is quite normal for a SSD for the WEI to drop a bit and see very little changes in benchmarks.

As for the firmware update for my drive, it was only to address the 8MB bug issue on this drive which basically if the drive was not powered down properly for some reason, (power outage, crash, etc) the drive would only show the capacity of only 8MB and you could not see anymore of the drive which meant that your data was lost. The only way you can get the full capacity back was to secure erase it.

I have heard a case that came up that the firmware did not fix this but I'm gonna wait for an official word from Intel on this. Until then, I will keep backing up my drive at least 3-4 times a week.

I know that the SF drives do have some odd throttling system with their drives. Extreme Systems encountered this issue on their experiment with 34nm versus 25nm NAND write endurance. Very intresting stuff they are doing. You can find out more about it here http://www.xtremesystems.org/forums/...e-25nm-Vs-34nm
Yes, I tried different 590 ports and cables before someone that doesn't even own our board reminded me of the other SATA controller.

My knowledge of SSDs is still sparse, but I understand that one way of overcoming degraded performance over time was to implement TRIM in the device. Windows 7 supports this natively. I know my SSD has it but it's almost impossible to know if TRIM is actually running. Not enough experience yet to know if that will allow WEI to hold it's value over time.
post #8986 of 12360
Quote:
Originally Posted by Vogel View Post
I got the ssd working on the Sil32 controller.


It looks to be faster in small read/writes than the nvidia controller, but no 200MB/s.


Switching to the Sil32 Controller unlinked all my libraries etc. from my data hdd, so I changed it back for now (thankfully that restored all the links). Maybe I'll do some more experimenting later.
I bought this board with upgrading in mind but I only started doing that 6 months ago. Looks like I'm going to have buy that new board anyway.


Cheers
See yours has similar issues to mine. If I can trust Windows, your drive is only running in UDMA mode 5. I don't understand why that's happening, but if it's accurate, the interface is running much slower than it should be.
post #8987 of 12360
Clearing the CMOS made no difference as I expected.

Booting into another OS HDD and testing the SSD specs produced no different results.

Here is an empirical analysis comparing the OCZ Solid3 and the Crucial C300.

These results won't surprise you and made me feel partially better because it exonerates the M2N32, but still leaves me with a problem on the OCZ SSD.

This is something I hope OCZ takes the time to investigate.

I was fortunate enough to be allowed to borrow my son's Crucial C300. This was one of the first SATA-3 SSDs on the market I'm aware of and as such doesn't have the same raw specs as the Solid3, but at least it gives me a basis of comparison. The premise made by a few is that the M2N32 is too retarded (because of faulty Nvidia chipset) or too old to properly run an SSD at SATA-2 speeds.

So as things stand, there is no problem with my motherboard's ability to run a SATA-2 SSD. The Solid3 however is not able to be properly detected.

Dunno where this leaves me, but I wish I could downgrade my Solid3 back to 2.08 to validate my premise.

Here are side by side comparisons of the two drives using ATTO.

post #8988 of 12360
In some of my more recent posts fighting SSD gremlins, I was asked to clear CMOS to reset ACPI data. While this proved fruitless for the SSD situation, it showed me some interesting datapoints.

The first curious one is that Phenommsrtweaker is able to manipulate the CPU multiplier and Vcore independently of what the BIOS is set to. That was a wonderful revelation that means there are few less parameters to have to play with for my current OC setup. I still have to set the board's OC frequency manually although I may be able to do that from the OS using Nvidia System Tools, but that is too quirky to use. With the current PMT settings, I can set the CPU multis and to OC the system on demand.

The second interesting observation is that I can run the M2N32 in stock trim without too much of a perceived loss of performance leaving the stock 200MHz frequency, and the CPU core temperatures are a lot lower, 35C vs 43C. I found that to be astonishing actually. I attribute the performance gains mostly to the lower latency of the SSD vs my SATA-2 HDDs.

I'm guessing now that the temperature difference is mostly due to OCing the board and running the NB voltage at 1.3V. I still wish I could effectively undervolt that setting from within the OS, but that doesn't appear to be possible for me.

Cheers.
post #8989 of 12360
I had made the mistake previously of setting Vcore (1.36V) and CPU to NB voltage (1.3V) in BIOS when trying to run PMT, not realizing the two would be fighting each other in the process. Setting these values in BIOS was necessary back at the time I was trying to establish a stable overclocking setup.

So when I got PMT sort of working, I could not understand why the power states of the CPU wouldn't change when the frequency throttles. Nowhere have I ever seen the point of leaving the two voltages in BIOS set to Auto mentioned when trying to use PMT, although it's probably obvious to most, but not to me.
Consequently with this discovery, I can now run the 965BE with both voltages set to 1.05V and the resulting core temps have come down to 37C. Keep in mind it is quite warm where I live.

I have the mainboard frequency set to 254MHz, so the idle CPU freq is 1024MHz.

If anyone else is running this way, I'm interested to know your experience in undervolting these K10 processors. How low can you go?
post #8990 of 12360
Quote:
Originally Posted by gr8sho View Post
Clearing the CMOS made no difference as I expected.

Booting into another OS HDD and testing the SSD specs produced no different results.

Here is an empirical analysis comparing the OCZ Solid3 and the Crucial C300.

These results won't surprise you and made me feel partially better because it exonerates the M2N32, but still leaves me with a problem on the OCZ SSD.

This is something I hope OCZ takes the time to investigate.

I was fortunate enough to be allowed to borrow my son's Crucial C300. This was one of the first SATA-3 SSDs on the market I'm aware of and as such doesn't have the same raw specs as the Solid3, but at least it gives me a basis of comparison. The premise made by a few is that the M2N32 is too retarded (because of faulty Nvidia chipset) or too old to properly run an SSD at SATA-2 speeds.

So as things stand, there is no problem with my motherboard's ability to run a SATA-2 SSD. The Solid3 however is not able to be properly detected.

Dunno where this leaves me, but I wish I could downgrade my Solid3 back to 2.08 to validate my premise.

Here are side by side comparisons of the two drives using ATTO.
Well if you can get the Crucial C300 to work like that it offers new hope for my M4 as it has basically the same controller, it just uses different firmware.

I also noticed something weird. In devicemanager I noticed the 590 controllers (3 of them?) were placed under the "IDE ATA/ATAPI" section. I always thought those belonged there but when I asked windows to search for new drivers they were installed and the 590 controllers were suddenly moved to the (I believe that translates to) "storagecontrollers" section alongside the Sil32 controller which was always filed there. It didn't do anything to speed up the drive unfortunately.
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Ph II x4 945 C3 Asus M2N32 SLI Deluxe MSI R6870 Hawk 6 Gb Corsair 6400C4DHX 
Hard DriveOptical DriveOSMonitor
Crucial M4 64 Gb + 1 Tb Samsung W7 64 Samsung P2370 
PowerCase
Antec EarthWatts 500 Antec Sonata II 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Ph II x4 945 C3 Asus M2N32 SLI Deluxe MSI R6870 Hawk 6 Gb Corsair 6400C4DHX 
Hard DriveOptical DriveOSMonitor
Crucial M4 64 Gb + 1 Tb Samsung W7 64 Samsung P2370 
PowerCase
Antec EarthWatts 500 Antec Sonata II 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD Motherboards
Overclock.net › Forums › AMD › AMD Motherboards › [Official] ASUS M2N-SLI and M2N32-SLI Club!