Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

otheronenorehto

macrumors newbie
Original poster
Oct 12, 2006
5
0
I posted this on the apple support discussions but have found no solace! Below it the thread from apple discussions, please I would like to know if anyone has any ideas or is experiencing similar problems. Please know that I have spent hours on google trying to research this so don't just post a link to the bare feats article or to xceler8yourmac:) Anything else is much appreciated!



Topic: Benchmark test results seagate 320gb 7200.10 dissapointing... help!

otheronenorehto
Posts: 27
From: New Jersey
Registered: Oct 12, 2006
Benchmark test results seagate 320gb 7200.10 dissapointing... help!
Posted: Dec 12, 2006 12:24 PM Reply Email


My previous post was deleted... Thanks admins keep up the good work. So I am posting this in a more constructive manner. I am going to post the dissapointing bench mark results for the seagate 320gb drives model #ST3320620AS.

I have been walked around in circles by both apple and seagate on this so far... Here are the things that I have done that have not fixed the problem at all:

1. Install 10.4.8 and all EFI updates
2. remove the 1.5gb/s jumpers from the back of the drives
3. update the firmware on all the drives to 3aae (3 of my 4 drive I RMA'ed and got brand spanking new with the 3aae firmware)
5. Disabled spotlight indexing which may have been interfering with the tests.

So I started running tests on every possible configuration of drives. All tests done on raid configurations were built with a 32k block size.

I will notate it like this:
discs will be labeled d1 d2 d3 and d4 the number also coresponds to the bay they are in. r0 means raid 0 (the only raid configuration I have tested so far). R0 will always be followed by what discs were being tested (example r0d1d2 would be a test on discs 1 and 2 in a striped configuration). Sometimes I tested a configuration more than once (labeled test1, test2 etc.)Here are the results:

d1
Results 54.47
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type ST3320620AS
Disk Test 54.47
Sequential 109.65
Uncached Write 84.47 51.86 MB/sec [4K blocks]
Uncached Write 139.95 79.18 MB/sec [256K blocks]
Uncached Read 90.56 26.50 MB/sec [4K blocks]
Uncached Read 154.94 77.87 MB/sec [256K blocks]
Random 36.23
Uncached Write 12.05 1.28 MB/sec [4K blocks]
Uncached Write 98.60 31.57 MB/sec [256K blocks]
Uncached Read 93.52 0.66 MB/sec [4K blocks]
Uncached Read 151.87 28.18 MB/sec [256K blocks]

d2
Results 51.08
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type ST3320620AS
Disk Test 51.08
Sequential 107.24
Uncached Write 81.96 50.33 MB/sec [4K blocks]
Uncached Write 133.91 75.77 MB/sec [256K blocks]
Uncached Read 91.42 26.76 MB/sec [4K blocks]
Uncached Read 149.36 75.07 MB/sec [256K blocks]
Random 33.52
Uncached Write 11.01 1.17 MB/sec [4K blocks]
Uncached Write 95.10 30.44 MB/sec [256K blocks]
Uncached Read 90.06 0.64 MB/sec [4K blocks]
Uncached Read 145.54 27.01 MB/sec [256K blocks]

d3
Results 53.99
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type ST3320620AS
Disk Test 53.99
Sequential 108.15
Uncached Write 82.18 50.46 MB/sec [4K blocks]
Uncached Write 134.25 75.96 MB/sec [256K blocks]
Uncached Read 94.87 27.76 MB/sec [4K blocks]
Uncached Read 146.47 73.61 MB/sec [256K blocks]
Random 35.97
Uncached Write 12.15 1.29 MB/sec [4K blocks]
Uncached Write 85.98 27.53 MB/sec [256K blocks]
Uncached Read 94.92 0.67 MB/sec [4K blocks]
Uncached Read 148.31 27.52 MB/sec [256K blocks]

d4
Results 54.58
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type ST3320620AS
Disk Test 54.58
Sequential 105.11
Uncached Write 81.88 50.27 MB/sec [4K blocks]
Uncached Write 132.51 74.97 MB/sec [256K blocks]
Uncached Read 88.64 25.94 MB/sec [4K blocks]
Uncached Read 142.60 71.67 MB/sec [256K blocks]
Random 36.86
Uncached Write 12.40 1.31 MB/sec [4K blocks]
Uncached Write 94.12 30.13 MB/sec [256K blocks]
Uncached Read 95.41 0.68 MB/sec [4K blocks]
Uncached Read 148.51 27.56 MB/sec [256K blocks]

r0d1d2
Results 101.36
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type raid0
Disk Test 101.36
Sequential 181.48
Uncached Write 321.97 197.69 MB/sec [4K blocks]
Uncached Write 272.48 154.17 MB/sec [256K blocks]
Uncached Read 84.61 24.76 MB/sec [4K blocks]
Uncached Read 290.18 145.84 MB/sec [256K blocks]
Random 70.31
Uncached Write 29.02 3.07 MB/sec [4K blocks]
Uncached Write 154.56 49.48 MB/sec [256K blocks]
Uncached Read 103.53 0.73 MB/sec [4K blocks]
Uncached Read 158.72 29.45 MB/sec [256K blocks]

r0d1d3
Results 72.48
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 6
Disk Test 72.48
Sequential 92.49
Uncached Write 39.51 24.26 MB/sec [4K blocks]
Uncached Write 287.36 162.59 MB/sec [256K blocks]
Uncached Read 89.81 26.28 MB/sec [4K blocks]
Uncached Read 300.95 151.26 MB/sec [256K blocks]
Random 59.59
Uncached Write 22.29 2.36 MB/sec [4K blocks]
Uncached Write 157.55 50.44 MB/sec [256K blocks]
Uncached Read 103.89 0.74 MB/sec [4K blocks]
Uncached Read 159.26 29.55 MB/sec [256K blocks]

r0d1d4
Results 103.08
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 10
Disk Test 103.08
Sequential 185.57
Uncached Write 333.56 204.80 MB/sec [4K blocks]
Uncached Write 289.58 163.85 MB/sec [256K blocks]
Uncached Read 84.42 24.70 MB/sec [4K blocks]
Uncached Read 306.98 154.28 MB/sec [256K blocks]
Random 71.36
Uncached Write 29.84 3.16 MB/sec [4K blocks]
Uncached Write 148.53 47.55 MB/sec [256K blocks]
Uncached Read 102.42 0.73 MB/sec [4K blocks]
Uncached Read 165.63 30.73 MB/sec [256K blocks]

r0d2d3test1
Results 45.62
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 15
Disk Test 45.62
Sequential 184.22
Uncached Write 322.72 198.14 MB/sec [4K blocks]
Uncached Write 265.55 150.25 MB/sec [256K blocks]
Uncached Read 87.23 25.53 MB/sec [4K blocks]
Uncached Read 295.40 148.46 MB/sec [256K blocks]
Random 26.03
Uncached Write 28.61 3.03 MB/sec [4K blocks]
Uncached Write 9.73 3.11 MB/sec [256K blocks]
Uncached Read 102.95 0.73 MB/sec [4K blocks]
Uncached Read 160.72 29.82 MB/sec [256K blocks]

r0d2d3test2
Results 100.29
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 15
Disk Test 100.29
Sequential 174.55
Uncached Write 232.76 142.91 MB/sec [4K blocks]
Uncached Write 265.61 150.28 MB/sec [256K blocks]
Uncached Read 87.29 25.54 MB/sec [4K blocks]
Uncached Read 294.31 147.92 MB/sec [256K blocks]
Random 70.36
Uncached Write 28.71 3.04 MB/sec [4K blocks]
Uncached Write 158.48 50.73 MB/sec [256K blocks]
Uncached Read 101.79 0.72 MB/sec [4K blocks]
Uncached Read 170.02 31.55 MB/sec [256K blocks]

r0d2d4
Results 69.27
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 21
Disk Test 69.27
Sequential 85.77
Uncached Write 36.35 22.32 MB/sec [4K blocks]
Uncached Write 248.70 140.72 MB/sec [256K blocks]
Uncached Read 86.05 25.18 MB/sec [4K blocks]
Uncached Read 287.16 144.33 MB/sec [256K blocks]
Random 58.10
Uncached Write 21.67 2.29 MB/sec [4K blocks]
Uncached Write 156.41 50.07 MB/sec [256K blocks]
Uncached Read 102.18 0.72 MB/sec [4K blocks]
Uncached Read 153.06 28.40 MB/sec [256K blocks]
r0d3d4
Results 99.46
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 28
Disk Test 99.46
Sequential 177.96
Uncached Write 330.90 203.17 MB/sec [4K blocks]
Uncached Write 266.56 150.82 MB/sec [256K blocks]
Uncached Read 81.09 23.73 MB/sec [4K blocks]
Uncached Read 296.62 149.08 MB/sec [256K blocks]
Random 69.01
Uncached Write 28.26 2.99 MB/sec [4K blocks]
Uncached Write 154.15 49.35 MB/sec [256K blocks]
Uncached Read 101.56 0.72 MB/sec [4K blocks]
Uncached Read 160.32 29.75 MB/sec [256K blocks]

r0d1d2d3test1
Results 17.96
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 36
Disk Test 17.96
Sequential 19.08
Uncached Write 5.28 3.24 MB/sec [4K blocks]
Uncached Write 162.52 91.95 MB/sec [256K blocks]
Uncached Read 86.14 25.21 MB/sec [4K blocks]
Uncached Read 436.83 219.55 MB/sec [256K blocks]
Random 16.96
Uncached Write 45.16 4.78 MB/sec [4K blocks]
Uncached Write 5.04 1.61 MB/sec [256K blocks]
Uncached Read 112.17 0.79 MB/sec [4K blocks]
Uncached Read 158.29 29.37 MB/sec [256K blocks]

r0d1d2d3test2
Results 17.60
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 36
Disk Test 17.60
Sequential 35.16
Uncached Write 241.12 148.04 MB/sec [4K blocks]
Uncached Write 150.11 84.93 MB/sec [256K blocks]
Uncached Read 9.93 2.91 MB/sec [4K blocks]
Uncached Read 438.42 220.35 MB/sec [256K blocks]
Random 11.74
Uncached Write 32.47 3.44 MB/sec [4K blocks]
Uncached Write 3.40 1.09 MB/sec [256K blocks]
Uncached Read 111.56 0.79 MB/sec [4K blocks]
Uncached Read 157.14 29.16 MB/sec [256K blocks]

r0d1d2d3test3
Results 8.72
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 36
Disk Test 8.72
Sequential 4.57
Uncached Write 234.32 143.87 MB/sec [4K blocks]
Uncached Write 1.17 0.66 MB/sec [256K blocks]
Uncached Read 85.08 24.90 MB/sec [4K blocks]
Uncached Read 435.78 219.02 MB/sec [256K blocks]
Random 95.68
Uncached Write 45.65 4.83 MB/sec [4K blocks]
Uncached Write 213.35 68.30 MB/sec [256K blocks]
Uncached Read 111.28 0.79 MB/sec [4K blocks]
Uncached Read 160.65 29.81 MB/sec [256K blocks]

r0d1d2d4test1
Results 8.11
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 1
Disk Test 8.11
Sequential 4.50
Uncached Write 5.28 3.24 MB/sec [4K blocks]
Uncached Write 1.46 0.82 MB/sec [256K blocks]
Uncached Read 83.30 24.38 MB/sec [4K blocks]
Uncached Read 446.70 224.51 MB/sec [256K blocks]
Random 41.36
Uncached Write 47.83 5.06 MB/sec [4K blocks]
Uncached Write 197.79 63.32 MB/sec [256K blocks]
Uncached Read 114.39 0.81 MB/sec [4K blocks]
Uncached Read 16.13 2.99 MB/sec [256K blocks]

r0d1d2d4test2
Results 4.96
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 1
Disk Test 4.96
Sequential 2.55
Uncached Write 5.26 3.23 MB/sec [4K blocks]
Uncached Write 0.73 0.41 MB/sec [256K blocks]
Uncached Read 82.61 24.18 MB/sec [4K blocks]
Uncached Read 426.18 214.19 MB/sec [256K blocks]
Random 95.72
Uncached Write 46.39 4.91 MB/sec [4K blocks]
Uncached Write 201.90 64.64 MB/sec [256K blocks]
Uncached Read 112.03 0.79 MB/sec [4K blocks]
Uncached Read 157.40 29.21 MB/sec [256K blocks]

r0d1d2d4test3
Results 13.89
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 1
Disk Test 13.89
Sequential 7.48
Uncached Write 259.49 159.32 MB/sec [4K blocks]
Uncached Write 1.94 1.10 MB/sec [256K blocks]
Uncached Read 83.53 24.45 MB/sec [4K blocks]
Uncached Read 426.89 214.55 MB/sec [256K blocks]
Random 96.49
Uncached Write 47.13 4.99 MB/sec [4K blocks]
Uncached Write 194.30 62.20 MB/sec [256K blocks]
Uncached Read 113.59 0.80 MB/sec [4K blocks]
Uncached Read 159.07 29.52 MB/sec [256K blocks]



r0d1d3d4
Results 134.02
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 55
Disk Test 134.02
Sequential 216.98
Uncached Write 503.13 308.92 MB/sec [4K blocks]
Uncached Write 420.89 238.14 MB/sec [256K blocks]
Uncached Read 85.13 24.91 MB/sec [4K blocks]
Uncached Read 430.18 216.21 MB/sec [256K blocks]
Random 96.95
Uncached Write 46.86 4.96 MB/sec [4K blocks]
Uncached Write 204.20 65.37 MB/sec [256K blocks]
Uncached Read 112.17 0.79 MB/sec [4K blocks]
Uncached Read 163.84 30.40 MB/sec [256K blocks]

r0d2d3d4
Results 22.34
System Info
Xbench Version 1.3
System Version 10.4.8 (8L2127)
Physical RAM 2048 MB
Model MacPro1,1
Drive Type Untitled RAID Set 66
Disk Test 22.34
Sequential 157.29
Uncached Write 232.38 142.68 MB/sec [4K blocks]
Uncached Write 139.35 78.85 MB/sec [256K blocks]
Uncached Read 85.38 24.99 MB/sec [4K blocks]
Uncached Read 446.47 224.39 MB/sec [256K blocks]
Random 12.02
Uncached Write 45.94 4.86 MB/sec [4K blocks]
Uncached Write 3.38 1.08 MB/sec [256K blocks]
Uncached Read 113.25 0.80 MB/sec [4K blocks]
Uncached Read 157.21 29.17 MB/sec [256K blocks]


Mac pro 2.66 Mac OS X (10.4.8)





Phil U.
Posts: 13
From: Virginia
Registered: Sep 16, 2006
Re: Benchmark test results seagate 320gb 7200.10 dissapointing... help!
Posted: Dec 12, 2006 12:47 PM in response to: otheronenorehto Reply Email


Interesting, I'd love to find a solution as I also have four of these babies in my Mac Pro.

Let me ask you this: How'd you go about updating the firmware on them? Where'd you get the firmware image?

Mac Pro Mac OS X (10.4.8)





otheronenorehto
Posts: 27
From: New Jersey
Registered: Oct 12, 2006
Re: Benchmark test results seagate 320gb 7200.10 dissapointing... help!
Posted: Dec 12, 2006 1:12 PM in response to: otheronenorehto Reply Email


For those of you who don't want to digest all this data I will sum things up. The Seagate SATAII 320gb 7200.10 drive preform very poorly with small block sizes especially with random writes. The other thing I noticed is that often the test results would vary widely without an apparent reason (as you can see in r0d1d2d3 tests 1 2 and 3).

In the above referenced test write times in test one for 4k blocks were an abysmal 3 mb/s (this is for a 3 disc raid!) but the increased to a more reasonable 140 mb/s in the second test. However, the read time in the second test went from 20 mb/s(already very low) to 3 mb/s.

Also I just wanted to explain something that may be confusing. The first number in each line is the Xbench score which is a comparison between my score and a dual g5 running a single 7200.9 drive. so for instance in this line:

Uncached Read 9.93 2.91 MB/sec [4K blocks]

9.93 is my score which means my test result was 9.93% that of a dual 2.0 g5 (or about 90% slower). The second number is the actual real life speed.

Some of you may wonder if these results translate to real world performance issues. I have been quoted copy times of over a week on a 100gb file I tried to transfer.

I would like to take this chance to point out that part of the problem is most likely Apples OS X raid implimentation. I can not confirm for sure but I hear that even simple raid 1 and 0 functions are only available through software.

I was aware that Apple had dissabled Raid 5 on the Intel ESB2 chip used in the Mac Pro's. This intel chip is capable of handling raid 0/1/10/5/JBOD via the hardware. Many linux and PC users have been able to set up these types of raid using this same chipset. If Apple doesn't want to support raid 5 I will reluctantly live with that.

However, that they would bypass onboard hardware capabilities for raid 1/0/10 in favor of the software raid available through disk utility boggles my mind. I think that at least in part the erratic test results are a function of shoving the raid processes on the CPU forcing DATA to run all over the motherboard so to speak. I would love to hear from anyone else trying to run these drives in raid on thier mac pro.

Mac pro 2.66 Mac OS X (10.4.8)



otheronenorehto
Posts: 27
From: New Jersey
Registered: Oct 12, 2006
Re: Benchmark test results seagate 320gb 7200.10 dissapointing... help!
Posted: Dec 12, 2006 1:17 PM in response to: Phil U. Reply Email


Updating the firmware was a huge pain in the butt! I got the update tool directly from seagate disc support. Eventually I sent 3 of the discs back to newegg, who were kind enough to send me 3 discs with the latest firmware installed at the factory. The 4th disc I was eventually able to update from DOS.

I actually figured out how to boot my mac pro to DOS but apparently the Mac Pro controller is not recognised by the firmware, so I connected the drive to a dell I had and was able to update it from there. Long story short All the drives have the same firmware version.

Mac pro 2.66 Mac OS X (10.4.8) running boot camp


otheronenorehto
Posts: 27
From: New Jersey
Registered: Oct 12, 2006
Re: Benchmark test results seagate 320gb 7200.10 dissapointing... help!
Posted: Dec 13, 2006 9:48 AM in response to: otheronenorehto Reply Email


Just to let you all know I recieved this email from a seagate tech this morning:

Hello,

The attachment didn't make it through, or wasn't attached with the Email.

Since the drives have been updated to the latest version of firmware, any issues with the drives at this point should be resolved.

Have you verified with the company who RAID hardware/software you are using, if they might have any updates for the controller/software?

Bryce S.
Seagate Technical Support

========== Customer's original message ========== The disk was formatted properly with a call I made to support yesterday.
However the reason for updating my drives has not been resolved. I am about to send you a .tar archive containing plain text files with methodical x-bench test results. It may be delivered to your junk mail folder.

My computer is a Mac Pro 2.66 with 2gb of ram. I am running os x 10.4.8 with all the apple firmware revisions. All discs are running your latest firmware (version e). All 1.5gb/s jumpers have been removed from the drives. All raid configurations tested were built with 32k size blocks.

The file name of the text file represents which disc(s) and in what configuration the test was run. For instance:

D1=disc one (1 also represents what bay the disc is in) R0D1D2=raid 0 using discs 1 and 2

You can see that sometimes I ran the test multiple times on the same
configuration: r0d1d2test1 and r0d1d2test2

I made sure no other processes were running during these tests. I only did all possible raid 0 configurations except for with all four discs.
You can see that there is definitely something very funky going on!

Jeremiah

Mac pro 2.66 Mac OS X (10.4.8) running boot camp



otheronenorehto
Posts: 27
From: New Jersey
Registered: Oct 12, 2006
Re: Benchmark test results seagate 320gb 7200.10 dissapointing... help!
Posted: Dec 21, 2006 7:53 AM in response to: otheronenorehto Reply Email


OK here is an update for those of you who have been following this debacle! At this point I have spent upwards of 40 hours working on this issue over the last 3 months and it is just plain disappointing.

I continued my tests on mirrored raid drives which followed a pattern similar to the raid 0 test results above.

Notable similarities were:

1. lack of performance in small read/write block sizes
2. results would sometimes vary greatly on the same test on the same drive!

Everything breaks down as soon as I try to write to three or more drives at the same time! As with the three drive raid results above raid 10 completely breaks down. I can barely build raid 10 as the I/O to a seemingly random drive will slow to a crawl. After I managed to build a raid 10 set tests were too slow to run and copying a 10gb file took about 20 minutes. After the 10gb file copied inspecting the raid set in disk utility yielded two failed drives. So I deleted the raid set and booted up to windows. I ran the seagate diagnostic utility and there were no S.M.A.R.T. errors and no bad sectors.

In addition to all the steps I have taken to make this work so far at this point I decided to zero out all the drives! This took me about 3 hours per drive. I then tested the drives one at a time.

I would mount a drive in a bay format it, test it, disable spotlight indexing then shutdown and unmount it and mount the next drive. All drives behaved properly. Once all drives were properly formatted and tested I shut down the computer and mounted all the drives. after powering on I tested all the drive individually again when everything was mounted the results were within an acceptable margin of the previous tests.

So I set up a stripe of two mirrored raids in disk utility, added the disks and built it, I used 256k blocks on all raid sets and on the mirrored sets I checked the automatically rebuild option. It took the raid set about 30 minutes to build the majority of which was spent building the final raid set. It seems like things fail when the computer tries to write across 3-4 disks. Just an interesting side note:

I am a video editor so I kind of get a kick out of having a running all time record for the longest estimated time to complete On video project I often get times of about 24 hours. My current record was "about a week". This does not include when I render fractal animations which took my dual 2.0 g5 upwards of 3 weeks to render. I don't count them because there is no little bar and time estimate. Anyway to get to the point while building this raid, and I am not kidding you here, my Mac estimated the time to complete as 1,193,046 hours. Now it didn't take that long obviously, but it said that for a good 7-8 minutes, which means that the I/O must have been phenomenally low!

After the raid set was built I disabled spotlight indexing on it and ran a test. The test froze up which just means that I/O was so slow that it was apparently doing nothing in the GUI but was really doing what it is supposed to at an incredibly slow pace. This was confirmed by grabbing a sample of Xbench from the activity monitor.

So I booted up Kona which allows you to view text results for each block. It seemingly froze up as well. In the past when this has happened leaving the test to run over night will usually be enough for it to complete showing speeds a fraction of a mb per second!

So I tried a real test copying the same 10gb file to the array. Again the transfer took around 20 minutes and this time only one of the disks failed. I am currently in the process of rebuilding the failed drive. I will try and test it again but I plan on calling Applecare today to have them replace the entire Mac. At this point I have to assume some part of my SATA controller or hard drive interface is faulty.

Mac pro 2.66 Mac OS X (10.4.8) running boot camp



otheronenorehto
Posts: 27
From: New Jersey
Registered: Oct 12, 2006
Re: Benchmark test results seagate 320gb 7200.10 dissapointing... help!
Posted: Dec 21, 2006 7:56 AM in response to: otheronenorehto Reply Email


I emplore anyone reading this thread to please post if you can think of any other possibilities that might cause such a problem. So far seagate tech's have been either unwilling or incompetent to do anything other than say it's not our problem. We will see how well apple responds.

Mac pro 2.66 Mac OS X (10.4.8) running boot camp
 

ero87

macrumors 65816
Jan 17, 2006
1,196
1
New York City
i don't know much about this topic, so i unfortunately can't help you, but you will get many more replies with a much more concise post. EDIT it, you don't to post all that stuff. :cool:
 

otheronenorehto

macrumors newbie
Original poster
Oct 12, 2006
5
0
i don't know much about this topic, so i unfortunately can't help you, but you will get many more replies with a much more concise post. EDIT it, you don't to post all that stuff. :cool:

Unfortuneately If I edit things out people will only ask me the same questions and offer the same solutions that have already been given. I am willing to accept that I will get fewer replies this way. Hopefully they will be of high value to me though.

Most of the information above especially the test results can be skimmed.

Thanks though for the reply may you be one of many and may one actually someday solve my problem.
 

davepk

macrumors member
Nov 20, 2006
45
0
Otherone,

I too ran into the problem you describe with the seagate 320 7200.10 not working properly. I had intended to use a couple of them in raid0 for a boot drive but that turned out to be impossible due to their performance.

My solution was to purchase a couple WD Raptor 10k drives and use them as my primary drives. I now use one 7200.10 as a backup drive and the other as a bootcamp drive. The raptors tho expensive are in my case easily worth it due to there performance.

I noticed your other post over at discussions.apple.com and thought i'd point you to "Fan control for Mac Pro" as it was mentioned to you as a possible solution. https://forums.macrumors.com/threads/253965/ This version was modified to work properly for a Mac Pro.

I doubt it will solve your Seagate problem but i guess it might be worth a try.
 

cynerjist

macrumors regular
Nov 8, 2006
170
0
I want to understand the problem a little bit better as I have a mixture of 7200.9 and 7200.10 500gb drives in my mac pro.

How common is it to randomly read/write a 4K block? What is the significance of this benchmark as it relates to everyday use?
 

otheronenorehto

macrumors newbie
Original poster
Oct 12, 2006
5
0
It is more common that you think due to file fragmentation to write smaller blocks. Also file fragmentation contributes to the frequency of random or non-consecutive read/write. all told it is not that common which is why you have to back it up with real world tests. My drives failed the real world test as well. However with KONA you can test block sizes up to 4gb and let me assure you that the preformance issue carries over.
 

MaaseyRacer

macrumors regular
Oct 30, 2005
160
0
San Francisco, Ca.
try disabling Speed Spectrum Clocking (SSC), and see if that works. There might be a jumper setting or you may have to plug it into a PC to do so. I think seagate has a windows utility to do it.
 

lenn0x

macrumors newbie
May 22, 2005
7
0
I am also experiencing such results myself, just enabled SATAII and my score is 33.94, with a RAID 0 of 2 500GB 7200.10's. I verifed the firmware is set to 3.AAE. Model is ST3500630AS
 

lenn0x

macrumors newbie
May 22, 2005
7
0
one interesting note i found was this:

http://www.barefeats.com/quad07.html

The improved results for four Seagate 750GB test units are due to the firmware update. Our drives came direct from the Seagate test engineers and were labeled "3.AEE + 2F00."

This firmware also fixed the slow sustained write speeds for single 7200.10 inside the Mac Pro.

I can not find anywhere the firmware for '3.AEE + 2F00.'
 

trainguy77

macrumors 68040
Nov 13, 2003
3,567
1
This is not just a .10 problem. I have both a .9 which came standard with the machine as its the 160 gb drive. But I also own a .10 320 gb drive. The .10 got a overall score of 51.08. With similar numbers to yours. Now the main drive which is a .9 160gb I got a total score of 49.9 with all numbers a little worse! I would like to see what type of scores people get with other stock drives such as the Western digitals? If others get higher performance with stock drives that are not .9s or .10s then I think I will take this straight to apple and demand a different 160 gb drive. :D And if they can't give me a different brand name in 160 gb then I want a 250 at no charge!!!! :D
But anyway what performance does other peoples stock drives get?
 

OldCorpse

macrumors 68000
Dec 7, 2005
1,758
347
compost heap
Why are we assuming that the problem is at Seagate's end... might just as well be an Apple problem that's responsible... have the drives been tested in other configurations?
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.