Issue with 2 Kinects on Dell XPS 1530

TheMightyThor · **Joined:** Sun Dec 04, 2011 5:11 pm **Posts:** 7

atmuc wrote:

http://crystalmark.info/software/CrystalDiskMark/index-e.html

use this program to test if your disk can record 53 MB/s at random write. my disk can write 87 MB/s seq and 38 MB/s random.

use usb 3.0 empty disk. with empty disk your sequential write chance will be higher.

increment your ram to 8 gb.

for external disk set removal policy to better performance. bu don't disconnect it without "Safely Remove Hardware"

thanks for the Reply, at the moment the hard drive is being defragged, will test speeds later tonight, i appreciate your help

atmuc · **Joined:** Thu Sep 22, 2011 2:31 am **Posts:** 70

TheMightyThor wrote:

thanks for the Reply, at the moment the hard drive is being defragged, will test speeds later tonight, i appreciate your help

defragmentation is useful for reading disk. it is bad for large empty space. defragmentation try to join separated parts of files. so it uses all empty spaces to make files as one part. for fast writing we need one part large empty space. it does not matter how space do you have on your disk. if all parts of files placed at the beginning of the disk, it means you have empty space at the end of disk. so operating system writes all parts sequentially. disk head doesnt move part to part. so sequential read and write is faster than random read/write. what i mention occurs on magnetic disks. ssd has random access. so both sequential and random read/write speed is the same.

another option is to have a minimum sized ssd (20 gb enough) on usb 3.0 port.

in fact 4 mb empty memory should be enough for buffering.

TheMightyThor · **Joined:** Sun Dec 04, 2011 5:11 pm **Posts:** 7

atmuc wrote:

TheMightyThor wrote:

thanks for the Reply, at the moment the hard drive is being defragged, will test speeds later tonight, i appreciate your help

defragmentation is useful for reading disk. it is bad for large empty space. defragmentation try to join separated parts of files. so it uses all empty spaces to make files as one part. for fast writing we need one part large empty space. it does not matter how space do you have on your disk. if all parts of files placed at the beginning of the disk, it means you have empty space at the end of disk. so operating system writes all parts sequentially. disk head doesnt move part to part. so sequential read and write is faster than random read/write. what i mention occurs on magnetic disks. ssd has random access. so both sequential and random read/write speed is the same.

another option is to have a minimum sized ssd (20 gb enough) on usb 3.0 port.

in fact 4 mb empty memory should be enough for buffering.

this system has 4 Gig but it is also a windows XP. unfortunately, out of all my systems (2), this system is the only one that appears to have separate controllers, so i am trying everything out on this. i have a new system on its way, its a Alienware m14x with windows ultimate 64 bit, 8 gig of ram, two usb 3.0 ports and one 2.0 port, 500 gig hard drive, so i am hoping that is the solution.

for now i am just testing out the software on the HP 8710w until the alienware gets here. Hopefully that solves the problem. again i do appreciate your help! thanks again

atmuc · **Joined:** Thu Sep 22, 2011 2:31 am **Posts:** 70

now i have the same problem with my hp pavillion dv6. i thought it was because it has 3 gb memory. i changed it to 4. i have the same problem. i have no problem with hp elite core 2 duo 4 gb ram. for few secs it records. its buffer indicator is full. then buffer starts to down. then i get buffer error. during this period i have 2 gb free memory. so problem is not related memory. ipisoft recorder uses 80 MB. so it has no buffer. i think it uses system buffer. there must be a way to increase this buffer while we have free memory.

what can i record is around 40 secs and 2GB.

Issue with 2 Kinects on Dell XPS 1530

Who is online