Seagate 7200 -> WD VRap 300GB

[Guest Nick_N]

Nick, tell me what to make of this . . .I have a 150Gb Vrap now. I'd like to clone my existing 250Gb 7200rpm SATA II drive to the new drive. The problem is, I use Seagate's DiskWizard for cloning, and it will require me to delete the existing partition on the Vrap. In doing so, will that kill the 64K clusters? Seems like it would, but I don't know. If it does, do I just copy the old drive (instead of cloning) to the Vrap in Windows after formatting and setting the cluster size in the Vrap? What do you know?Also, what the heck kind of controller are you using to get those benchmarks? This thing isn't even close--it's as Word Not Allowed's is. Is this effect mostly from the controller card you have, or is it the larger drive? If it's the latter, what % of the improved perf is attributable to the drive size versus the controller card.Again, what controller card? They range hugely in price and I'm not sure what to look for.Thank you,Noel

Using 3rd party disk management software such as Acronis Disk Director the user can convert the cluster size (of any drive) after such a clone to 64K without formatting (Acronis: Right click drive, select ADVANCED - CHANGE CLUSTER SIZE) however the user will need to properly run defrag maintenance as such cloning around :( will leave the install fragmented. NOTE: The OS dirve can ONLY be 4K.. only data drives (primary or logical) may use >4K cluster formats and any drive which has a Windows page file on it must be 4K as Windows VM operates in 4K allocation.Card: http://www.newegg.com/Product/Product.aspx...N82E16816116042WARNING: If you use this card for RAID0, -or- for the OS and data drives, for 100% secure data write protection in the event of a power loss or BSOD REGARDLESS of a UPS in operation you MUST install this BBU for the card: http://www.newegg.com/Product/Product.aspx...N82E16816116051If you use that card strictly for FSX or other game drive in which a UPS is in use the need for the BBU is not critical, however, it is still possible under the right circumstances data corruption may occur in that situation during a power failure. It would only be an issue in the event the user is installing FSX or a addon in a write operation and a power glitch may force a write error to the card buffer. In that event the drive would need to be completely wiped and restoredCorrect Performance Settings (single OR RAID array operation): http://www.simforums.com/forums/forum_post...D=161336#161336Make SURE to obtain the latest 3Ware Code set (Driver, firmware and interface) from the 3ware website and make sure the card driver/firmware is updated to the latest versionNOTE: The card requires a 2nd PCIe x16 slot which will deliver at minimum PCIe x8. Some motherboards may have issues with this card in the 2nd PCIe slot if the user is planning on booting the OS off the card. In that event the card must be in the 1st PCIe slot and the video adapter in the 2nd. Not all motherboards display this issue as most will boot the OS fine from the 2nd slot. The Asus P5E3-D does display the problem: http://www.simforums.com/forums/forum_post...D=165540#165540That should provide the answers you were looking for.

[Noel 4,507]

Using 3rd party disk management software such as Acronis Disk Director the user can convert the cluster size (of any drive) after such a clone to 64K without formatting (Acronis: Right click drive, select ADVANCED - CHANGE CLUSTER SIZE) however the user will need to properly run defrag maintenance as such cloning around :( will leave the install fragmented. NOTE: The OS dirve can ONLY be 4K.. only data drives (primary or logical) may use >4K cluster formats and any drive which has a Windows page file on it must be 4K as Windows VM operates in 4K allocation.Card: http://www.newegg.com/Product/Product.aspx...N82E16816116042WARNING: If you use this card for RAID0, -or- for the OS and data drives, for 100% secure data write protection in the event of a power loss or BSOD REGARDLESS of a UPS in operation you MUST install this BBU for the card: http://www.newegg.com/Product/Product.aspx...N82E16816116051If you use that card strictly for FSX or other game drive in which a UPS is in use the need for the BBU is not critical, however, it is still possible under the right circumstances data corruption may occur in that situation during a power failure. It would only be an issue in the event the user is installing FSX or a addon in a write operation and a power glitch may force a write error to the card buffer. In that event the drive would need to be completely wiped and restoredCorrect Performance Settings (single OR RAID array operation): http://www.simforums.com/forums/forum_post...D=161336#161336Make SURE to obtain the latest 3Ware Code set (Driver, firmware and interface) from the 3ware website and make sure the card driver/firmware is updated to the latest versionNOTE: The card requires a 2nd PCIe x16 slot which will deliver at minimum PCIe x8. Some motherboards may have issues with this card in the 2nd PCIe slot if the user is planning on booting the OS off the card. In that event the card must be in the 1st PCIe slot and the video adapter in the 2nd. Not all motherboards display this issue as most will boot the OS fine from the 2nd slot. The Asus P5E3-D does display the problem: http://www.simforums.com/forums/forum_post...D=165540#165540That should provide the answers you were looking for.

Nice, I'll read them thru. Does the card add much additional boot time, like my SCSI card did? I have a 2nd x16 slot so I guess I'm ok there. Thank you Nick for the detailed reply.

[Guest Nick_N]

Nice, I'll read them thru. Does the card add much additional boot time, like my SCSI card did? I have a 2nd x16 slot so I guess I'm ok there. Thank you Nick for the detailed reply.

As with any high end controller card and using them for OS launch there are 2 boots that take place with them.. the initialization of the card itself to the system BIOS for the slot it is located just after POST and then if the card is set up for OS operation, the boot from the card itself.There are always pauses involved with those operations.I do not find them excessive but they are there none the less. I only have one tower I use for OS boot off the card. The others I simply use the card for drives which hold FSX, A/V and or other engineering applications as the target for the perf is in the app in which constant storage access of some type is required unlike the OS which is essentially loaded at boot although there will be some access to that OS drive as the system is used. The goal in optimizing the OS is to lean it for initial load into physical memory at boot, and, the defrag solution does the rest in keeping the resources for the small amount of access taking place to the OS at a minimum. As for your question about file read performance from another thread.. the card and its user interface setup is what increases the efficiency of those read/write operations to the levels shown, not the cluster size.The freeware versions of Disk Management software will more than likely not offer their advanced features for free. I do not use those versions so I have no frame of reference to look and see if the options are available but I would say since the 'change cluster size' feature is located under the ADVANCED menu, it is not.

[Noel 4,507]

As for your question about file read performance from another thread.. the card and its user interface setup is what increases the efficiency of those read/write operations to the levels shown, not the cluster size.

Ahh, sounds like the card is the big issue. And a neat idea using it for FSX primarily. It IS nice tho to not have the SCSI boot delay.I'm getting pretty darn nice performance currently. Does the new drive show ANY perceivable change in FSX performance--ie, my subjective perception? I'd have to say a resounding . . . maybe. But as we ALL know hopefully by now, one can only compare EXACT situation comparisons. Perhaps this is because of the missing SATA card, which now makes me wonder how well the low end Seagate SATA II's would have performed ON A CARD. Benchwise, the VRap definitely outperforms the budget SATA II, but only about 20% I/O tops, and half the access time, but I will tell you CPU use has doubled with the Vrap over the budget SATA II drive--up to about 3.2% which is a new record high for me. My Cheetah's never went that far up. I don't know what to make of that. I do think FSX is performing close to perfectly, but truthfully, this was the case before the Vrap. About the only thing that stresses it now is too much autogen in the wrong place, and also certain cloud situations using ASX can be troublesome. I build machines for friends, not FSX'ers per se, so it's pretty easy to offload a part here or there so the cost doesn't bother me.Thank you again for your detailed, targeted and useful replies NickN.I do use a nice APC UPS, so I guess I'm ok on the issue mentioned re data loss.

[Guest Nick_N]

Ahh, sounds like the card is the big issue. And a neat idea using it for FSX primarily. It IS nice tho to not have the SCSI boot delay.I'm getting pretty darn nice performance currently. Does the new drive show ANY perceivable change in FSX performance--ie, my subjective perception? I'd have to say a resounding . . . maybe. But as we ALL know hopefully by now, one can only compare EXACT situation comparisons. Perhaps this is because of the missing SATA card, which now makes me wonder how well the low end Seagate SATA II's would have performed ON A CARD. Benchwise, the VRap definitely outperforms the budget SATA II, but only about 20% I/O tops, and half the access time, but I will tell you CPU use has doubled with the Vrap over the budget SATA II drive--up to about 3.2% which is a new record high for me. My Cheetah's never went that far up. I don't know what to make of that. I do think FSX is performing close to perfectly, but truthfully, this was the case before the Vrap. About the only thing that stresses it now is too much autogen in the wrong place, and also certain cloud situations using ASX can be troublesome. I build machines for friends, not FSX'ers per se, so it's pretty easy to offload a part here or there so the cost doesn't bother me.Thank you again for your detailed, targeted and useful replies NickN.I do use a nice APC UPS, so I guess I'm ok on the issue mentioned re data loss.

I have provided the answers to all the questions you are asking however I will sum it up here.First of all, this word

[Noel 4,507]

The random access benchmark shows alot of difference between my two drives. Seems that would be a decent bench for FSX, no?

[Guest Nick_N]

The random access benchmark shows alot of difference between my two drives. Seems that would be a decent bench for FSX, no?

No.. it is not a good test for any 'application' in any way shape or form NoelIt gives the engineer a baseline to project if the device is functioning normally as it is designed by the manufactureJust like the disk manufacture does not tell you to move your data to the outer edge of the platter, or, that as you fill their drive the performance drops, or, that a hardware controller will take much better advantage of the buffer on a drive serving data to the system... the engineer setting up the system uses such tests to confirm what they installed is working as it should be and if it is not, they look for the issue and correct it be it a driver change or other problem... or a defective unit or motherboard controllerReal-time, real application random access performance is based on actual application use and recording data in real time. How the application is being used, how the data is set up on the disk based on how the application best calls that data, and, how the controller device makes use of the drives internal design and communicates with the buss makes the call on performance for the application in questionA syth benchmark is useless for thatHowever.. it does give a user a good look at where the 'baseline' differences are in the storage devices they are using.The controller card may not show much of any difference in that random access benchmark.. but the file system reads/writes are getting their boost from how the card accesses the drive buffer and communicates with the buss... something motherboard SATA will not do.

[Guest Nick_N]

I did forget to add... from the benchmark you displayed the Vrap is performing correctly and in spec

[Noel 4,507]

I did forget to add... from the benchmark you displayed the Vrap is performing correctly and in spec

Many thanks again Nick for your insights and thoughts.