- Dec 1, 2002
- 542
- 0
- 0
Hello,
I have set up my SCSI machine and I am currently trying to benchmark it against my current ATA100 drive. OK, I'll describe my current set up and methodology for you to have a clear view on what it going on.
Ok, here is my setup:
AMD Athlon XP 2500+ @ 2.083GHz
Thermalright SLK-800 & Panaflo FBA08A12U1A
EPoX EP-8K3A+ V1.2
512mb Mushkin High Performance LEVEL 2 PC2700
Antec SX840 cooled by Panaflo
ATI Radeon 9700 Pro 128mb
Creative Labs SoundBlaster Audigy Gamer
3com 3C905-TX
Adaptec ASC-29160N SCSI Ultra160
Seagate Cheetah X15 36LP 36.7gb SCSI Ultra160
IBM Deskstar 180GXP 120gb
Well, I decided to have a better look on "how much" my HDD has increased. I don't know numerous benchmarks for HDD so I used HD Tach 2.61 to begin with. This benchmark gives me horrible results! I cannot link a copy clip of the results so I will write them down.
Here they are:
Seagate Cheetah X15 36LP 36.7gb SCSI Ultra160
Read speed
Maximum: 34961.0 kps
minimum: 14305.0 kps
average: 34680.1 kps
Access time: 5.9ms
Read burst speed : 34.1 mbps
CPU utilization: 52.9%
If I look the graph, the layout is maxed at 34000 kps...
IBM Deskstar 180GXP 120gb
Maximum: 60492.0 kps
minimum: 15425.0 kps
average: 45182.8 kps
Access time : 12.8ms
Read burst speed : maxed out at 80 mbps
CPU utilization: 16.1%
As you can see by the results, there is something that goes wrong! When my IBM was my windows boot drive, the results were the same except the CPU utilization which was about 60%. At that time, I was thinking that maybe a VIA based motherboard could the problem. Now, this drive is only used for plain storage so that may explain the difference in CPU utilization. What I don't understand is WHY the SCSI drive seems to top at 34 mbps?!? My controller is a U160 based so it may not be the problem.
Is this benchmark accurate or is my system needs more tweaks? I am using WinXP SP1 actually and I know that there is still a debate regarding the "XP cache filter thingie"... The answer isn't clear as some people says that there is no problem with XP while other argue that XP sucks with SCSI drives. Anyway, I don't understand why there is a so large gap in performance between those 2 drives.
If someone has expenrience with such setups, please, give me a hint because it sucks badly when you see the performance "loss" while having so much invested in CA$H. Frustrating..
Thanks in advance,
Parabellum
I have set up my SCSI machine and I am currently trying to benchmark it against my current ATA100 drive. OK, I'll describe my current set up and methodology for you to have a clear view on what it going on.
Ok, here is my setup:
AMD Athlon XP 2500+ @ 2.083GHz
Thermalright SLK-800 & Panaflo FBA08A12U1A
EPoX EP-8K3A+ V1.2
512mb Mushkin High Performance LEVEL 2 PC2700
Antec SX840 cooled by Panaflo
ATI Radeon 9700 Pro 128mb
Creative Labs SoundBlaster Audigy Gamer
3com 3C905-TX
Adaptec ASC-29160N SCSI Ultra160
Seagate Cheetah X15 36LP 36.7gb SCSI Ultra160
IBM Deskstar 180GXP 120gb
Well, I decided to have a better look on "how much" my HDD has increased. I don't know numerous benchmarks for HDD so I used HD Tach 2.61 to begin with. This benchmark gives me horrible results! I cannot link a copy clip of the results so I will write them down.
Here they are:
Seagate Cheetah X15 36LP 36.7gb SCSI Ultra160
Read speed
Maximum: 34961.0 kps
minimum: 14305.0 kps
average: 34680.1 kps
Access time: 5.9ms
Read burst speed : 34.1 mbps
CPU utilization: 52.9%
If I look the graph, the layout is maxed at 34000 kps...
IBM Deskstar 180GXP 120gb
Maximum: 60492.0 kps
minimum: 15425.0 kps
average: 45182.8 kps
Access time : 12.8ms
Read burst speed : maxed out at 80 mbps
CPU utilization: 16.1%
As you can see by the results, there is something that goes wrong! When my IBM was my windows boot drive, the results were the same except the CPU utilization which was about 60%. At that time, I was thinking that maybe a VIA based motherboard could the problem. Now, this drive is only used for plain storage so that may explain the difference in CPU utilization. What I don't understand is WHY the SCSI drive seems to top at 34 mbps?!? My controller is a U160 based so it may not be the problem.
Is this benchmark accurate or is my system needs more tweaks? I am using WinXP SP1 actually and I know that there is still a debate regarding the "XP cache filter thingie"... The answer isn't clear as some people says that there is no problem with XP while other argue that XP sucks with SCSI drives. Anyway, I don't understand why there is a so large gap in performance between those 2 drives.
If someone has expenrience with such setups, please, give me a hint because it sucks badly when you see the performance "loss" while having so much invested in CA$H. Frustrating..
Thanks in advance,
Parabellum