Yellowbeard

No there isn't. And, that is not what the numbers indicate.

There is no way to determine that other than to look at the ICs themselves.

1. No, there is no way to reliably verify this with any utility that I am aware of at this time. 2. EVERY device that uses NAND flash is going to show bad sectors. It's the nature of NAND flash to have sectors that do not work. This is true with SSDs, USB FDDs, cell phones, iPods, etc etc etc. So, if you do have bad sectors and this count is accurate, it's about 40KB of bad sectors from 120GB (unformatted value of course). If your drive malfunctions, we'll gladly replace it. However, I would not suggest an RMA based on what you have posted here.

Currently, the suggested method to manually assess SSD health and performance is to run ATTO. However, repeated benchmarking fills up the drive so it is not desirable to bench excessively as this in itself will lead to performance degradation. I think what he means here is that it would be great if you could interpret this type of information correctly with an SSD tool as opposed to having to bench it. However, health status and performance are 2 very different things so even a properly functioning SMART utility will not tell you everything you may want to know.

Did you actually read the first post? :confused:

It's not the SMART technology itself. It's the lack of standardization in how it's reported combined with the utilities that are inaccurate. It's sort a chicken vs egg argument.

No, it is in no way reliable.

According to our Engineering people, the F40 and F80 drives do not have temperature sensors. So, any temp readings using those drives should be ignored. Also, readings on the other Force drives may still be incorrect with some utilities even after the 2.0 firmware update.

In the last few weeks, I have spent quite a bit of time looking at the SMART data from a variety of Corsair SSDs. This thread is here to address the specific things I have observed about a few popular SMART reporting utilities, the SMART attributes and data, and the Corsair Force series SSDs. I discovered some very interesting things about SMART monitoring and reporting. SMART as defined by the ATA standard is standardized. However, the structures or methods used by device manufacturers to report SMART attributes and status are not. Although there are some common standards observed across device manufacturers, there are many that are unique to each. This problem is exacerbated by the fact that many SMART values have what is known as a raw value which is wholly determined by the device manufacturer. This raw value may or may not correctly correspond to a physical unit or normalized value, numbered 1-253. A great example of this is the recent issue of some SMART monitoring utilities reporting SSD temperatures of 156c or more! :flamethro Additionally, there are a set of SMART attributes that only apply to SSDs. And, even within these SSD specific attributes, there are variations of the terminology used for these attributes. Finally, to completely complicate things, not all SSD controllers report the same SMART attributes. Many SMART tools were originally only designed to monitor HDDs. Some current tools are adaptations of older SMART utilities that have been modified or updated to include SSD attributes. Unfortunately, due to the variations of SMART attribute reporting across the many different SSD controllers, it is almost impossible to find a tool that is accurate across all brands of controllers AND that also correctly deals with all the SSD specific SMART attributes. We are working dilligently with SandForce and with the developers of some of the more popular SMART monitoring utilities to improve the accuracy when using these utilities with Corsair SandForce based SSDs. In the mean time, here are some screenshots of some of these issues that you may want to be aware of. In short, if your drive is performing within specification, then you should be skeptical of any SMART utility that is indicating issues. Keep in mind that this is not a condemnation of each respective software developer or criticism of their efforts. However, we feel it is important to show our customers some examples of what is going on as we work our way through this situation. An important item of note is the method of determining Health Status or SSD Life Left. Some SSD controllers report an attribute not seen on SandForce attributes, 205 (hex-D0). This attribute is the MAX Program/Erase cycles for the SSD. This attribute is used on some controllers combined with atttribute 5 (Reallocated Sector Count aka Retired Block Count) to determine SSD Life Left. Since this attribute is not reported by the SandForce controller, it is impossible to determine how the monitoring utilities are determining SSD Life Left or Health Status reports. Crystaldisk 3.7.0 In this example, we can see 2 different temperature attributes reported. Obviously a working drive is not at 0C and there is no raw or normalized value given. This report is also problematic as the second attribute, E7 (aka 231) is still reported by some tools as a drive temperature. Here it is showing SSD Life Left which is an SSD attribute. However, we have also verified that the SandForce controllers are not currently reporting the E7 attribute correctly. So, we believe that it is not a reliable reporting of SSD Life Left. As you can see, this drive is new having been cycled only 12 times and been powered on for 3 total hours. I took it out of the box, installed Windows 7, and took these screenshots. It has never been benchmarked. With so little usage, and these reporting anomolies, we don't believe that 96% is an accurate estimate of Health Status for this drive. DiskCheckup This utility, with a release date of August 19, 2010, shows several Unknown Attributes, some of which are circled in red. These attributes are reported by other tools including CrystalDisk. Therefore, it is unclear why this tool would not show known SMART attributes shown by other utilities. We again see the E7/231 attribute reported incorrectly as a Temperature. HD Tune The Health Warning indication for attribute 05 is an erroneous reading. We are looking into this and have not as of yet determined why this occurs. This same drive checked with the SandForce utility does not show any status indicating an issue. Again, we also see the SSD Life Left attribute E7/231 reported as a temperature. As previously stated, this attribute has been determined to not be reported correctly by the SandForce controller. So what does all of this mean to end users? As stated above, we don't believe that the current SMART tools are completely accurate in their reporting. Our suggestion is to use performance as read by ATTO as a tool to determine if your drive is functioning properly. Remember, no SSD is going to report 100% life left, even fresh out of the box.

Read the sticky and try the method I posted here.

Have you tried the method given in this sticky? Also, what OS are you using?

Memtest Forums FAQ It's a very good idea to read over this before your testing if you are not already familiar with Memtest.

Check at http://www.bootdisk.com Also, most burning programs will make a bootable CD for you. You have to supply the image though so you'd need to find an ISO of the WIN98 program since you don't have access to a floppy. Also, if you have access to another computer with a floppy drive, you could use another system to do this.

Reserved .... to be continued.

I recently had a Flash Voyager 8mb drive get to a state in which I could not use it or format it. The entire file directory was corrupted and the file names all were intermingled with what I can only describe as hieroglyphics. Sorry, I did not get a screen shot of it but, if you see it, you will know what I mean. Also, I could not format the drive using any of the regular formatting methods. This made me think that not only was the file table corrupted, the Master Boot Record was corrupted and that the partition (and thus the entire drive) was unusable and unformattable. I tried a Windows based partition utility, a command line format, and Partition Magic in DOS. None of these worked. As a last resort, I dug out the trusty old Windows98 boot floppy. This floppy utility has the Fdisk program and with it, I was able to revive my drive. For those of you unfamiliar with how to use Fdisk, here is a step by step tutorial. I hope this helps some people. I used the WIN98 SE OEM image which I found at http://www.bootdisk.com Lots of good utilities to be had there. You can also use a disk burning program to make a bootable CD with the Windows98 utility on it. If you do this, obviously, you'll need to set the system to boot from the CD first. However, I did mine with a floppy so we'll go with that. Once you have extracted the image and made your floppy disk, you will need to enter your bios and make certain that your system will boot from the floppy disk. If your floppy is already first in your boot order, you can skip this step. http://www.houseofhelp.com/upload/Images/USBFdisk/1.jpg Save the settings, and reboot: http://www.houseofhelp.com/upload/Images/USBFdisk/2.jpg When WIN98 boots, you will have 3 options. In this case, select option 2 as you will not need CD-ROM support. If you are using a CD-ROM copy, use option # 1. http://www.houseofhelp.com/upload/Images/USBFdisk/3.jpg Once it boots, you will get an A: prompt. Type Fdisk and hit enter: http://www.houseofhelp.com/upload/Images/USBFdisk/4.jpg Your next 2 prompts are inquiries about disk size, enter (Y) for both: http://www.houseofhelp.com/upload/Images/USBFdisk/5.jpg http://www.houseofhelp.com/upload/Images/USBFdisk/6.jpg Here, you will want to check and make sure the current fixed disk drive is your USB drive as most likely Fdisk will not detect your USB drive as # 1. So, select option # 5: http://www.houseofhelp.com/upload/Images/USBFdisk/7.jpg In my example, my 8gb Flash Voyager is disk #2, so I selected (2) and hit enter: http://www.houseofhelp.com/upload/Images/USBFdisk/8.jpg Your partition should show up, even if it is damaged. You'll want to select #3 to delete the existing partition: http://www.houseofhelp.com/upload/Images/USBFdisk/9.jpg You'll now want to select option # 1 to delete the primary DOS partition: http://www.houseofhelp.com/upload/Images/USBFdisk/10.jpg The next step is a verification of the proper partition. As I only had 1 partition, I selected (1): http://www.houseofhelp.com/upload/Images/USBFdisk/11.jpg Enter volume label is another verification step. Since mine has no label, I left this blank and hit enter: http://www.houseofhelp.com/upload/Images/USBFdisk/12.jpg And, the point of no return, ARE YOU SURE? I chose (Y): http://www.houseofhelp.com/upload/Images/USBFdisk/13.jpg Now that your drive has no partition information at all, you'll need to create a new DOS partition, option # 1: http://www.houseofhelp.com/upload/Images/USBFdisk/14.jpg And, option #1 again as you must have a primary partition first: http://www.houseofhelp.com/upload/Images/USBFdisk/15.jpg Once you make this choice, the program will verify the entire drive integrity and take you to the next menu: http://www.houseofhelp.com/upload/Images/USBFdisk/16.jpg The next option will ask if you want to use the entire drive capacity for the primary partition. I chose (Y) and the program again verifies the drive integrity: http://www.houseofhelp.com/upload/Images/USBFdisk/17.jpg http://www.houseofhelp.com/upload/Images/USBFdisk/18.jpg Then, it again verifies the choice to use the maximum drive size and I chose (Y): http://www.houseofhelp.com/upload/Images/USBFdisk/19.jpg Once this is complete, if your drive is functioning properly, you will see a verification that your drive has a primary DOS partition and that it is using the entire drive space. You can now format your drive and use it again. http://www.houseofhelp.com/upload/Images/USBFdisk/20.jpg

Very cool stuff Eva2000. I recently adapted my Hydrocool to use on my DFI+AM2 FX-62 setup. The Hydrocool block design and the newer Nautilus 500 block design are very similiar so, I used the Nautilus 500 hold down on the Hydrocool. The Nautilus block is slightly thicker than the Hydrocool so I used some slightly thicker foam to adapt. My CPU is OEM and I have no air cooling to compare but, the Hydrocool does a wonderful job. LONG LIVE THE HYDROCOOL!