Sign in with
Sign up | Sign in
Your question
Solved

Unstable new build failing prime95 blend test

Last response: in Systems
Share
August 16, 2011 1:55:00 PM

New Build is failing prime95 blend test. After 12 hours blend w/ 8 threads it crashes. It does not find any prime number errors and no bsod just shuts down and reboots to "windows did not shut down properly" screen.

fyi, I posted problem on my old new build thread here. Flong has been helping me troubleshoot but suggested I start a new thread. (Troubleshooting starts near end of page two.)

History/Steps taken so far:

Assembled new build (my first build) without major incident. Actually, I should mention that initially computer did failed to boot with a chassis intrusion error. I reset RTC RAM and didn't have any problems after that. So, I booted up. Installed win7 64 bit. Made no changes to bios and ran Prime95 blend test + Coretemp to monitor cpu temps. It crashed without errors after about 7 minutes--just shut down. I rebooted and confirmed that p95 results log was empty-no errors. Temperatures at full load during test were approx 53-57C.

Temperatures were a little higher than expected so I reseated cooler and re-ran blend test. This time if crashed after 5 minutes and temperatures were unchanged (53-57C).

I researched difference between blend, sml fft, and lrg fft tests here. Of interest was that fact that sml ffts test does not access memory as much as lrg ffts:

Quote:
The "Small FFTs" test uses relatively small FFTs which can fit into the CPU cache. As a result, the small FFT test is the one which accesses your main memory the least but it still makes some memory accesses. Prime95 automatically creates a FFT size range which will fit into the L2 cache of your CPU.

The "In-place large FFTs" test uses relatively large FFTs which cannot fit into the CPU cache so this test accesses main memory a lot. It only accesses a relatively small amount of main memory because it runs the FFTs in-place so it accesses the same RAM over and over.


I realize that this is not best way to test memory but decided to run the sml fft test because both times a ran the blend test, the first test it ran and crashed on was incidentally a large prime. So I ran and it ran for an hour without errors or crashing. I stopped it and concluded that it was a memory problem.

I checked bios ram settings (timing, frequency and voltage) and found that they were off by a bit. So i set explicitly to manufacture specs. Re-ran blend test. This time it ran 13 hours, produced no errors, but did still crash... again no bsod just sudden crash. Here are some temp readings during first hour or so.

time | c1 | c2 | c3 | c4
5:00 | 48c | 51c | 55c | 52c
5:13 | 52c | 54c | 58c | 55c
6:50 | 53c | 53c | 57c | 55c
.
.
.
temps were pretty consistent for duration of test. Actually I turned A/C off for bit to see how temps reacted and core 1,2 and 3 went up to ~ 60c and core 3 to 64c. But, it IS pretty hot here in beijing without A/C. I should note that crash occurred in middle of night when A/C was off so temps were probable around 60-64c before crash. btw, should i be concerned that core3 is always about 4c hotter than other cores? btw, GPU temp also around 52C.

Anyways, Stock system w/no Overclock should have no trouble burning for 24+ hours right? Not sure what to do now. Seems my system is just not quite as stable as it should be and not cooling quite as well as i would like. What do you suggest? I would like overclock a bit, but if my stock system is already marginal in terms of stability i don't know that i have any room to push it.

Btw, this is a workstation build so stability is relatively important.

Specs:

MOBO: ASUS P8P67 WS REVOLUTION
CPU: i7 2600k
GPU: NVidia Quadro 4000
RAM: G.SKILL Ripjaws X Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1600 (PC3 12800) F3-12800CL9D-8GBXL --- x2(16GB)
SSD: OCZ vertex 3 (120GB)
HDD: SAMSUNG Spinpoint F3 HD103SJ 1TB ---x2(2TB)
DVD: Sony Optiarc AD-7240S
LCD: Dell UltraSharp U2410
Case: Corsair 650D
PSU: Corsair HX750
COOLER: Corsair H60

Steps I am considering:

1. replace ram w/ same ram or different ram(perhaps try different freq/timing)
2. upgrade cooler to H80 or H100.
3. jump out the window and take my new build with me

thanks for any help
August 16, 2011 2:10:10 PM

Before you do #1, burn a memtest boot disk, boot from it and let it make a few passes testing the memory. You will probably have no issues, but will be a better test of the memory. Also, try taking the memory down to DDR3-1333 speeds and ensure you are at 1.5 volts.
http://www.memtest.org/

Let us know the results... You choose quality parts and I wouldn't expect those types of problems at stock settings.

Skip the water cooling. You will do just as good with a solid air cooler. The Hyper 212+ linked below will do a great job as-is, and even better if a second 120mm fan is added in a push / pull configuration.
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
m
0
l
August 16, 2011 4:18:39 PM

Quote:
Before you do #1, burn a memtest boot disk, boot from it and let it make a few passes testing the memory. You will probably have no issues, but will be a better test of the memory. Also, try taking the memory down to DDR3-1333 speeds and ensure you are at 1.5 volts.


running memtest now... but what do you mean by "take the memory down to DDR3-1333 speeds and ensure you are at 1.5 volts"? --to check that psu is supplying correct voltage? but why take freq down to 1333 to check that?

Quote:
Skip the water cooling. You will do just as good with a solid air cooler. The Hyper 212+ linked below will do a great job as-is, and even better if a second 120mm fan is added in a push / pull configuration.


ok, but are you just saying i paid too much for adequate cooling or do you think hyper 212+ will cool better and improve stability?

thanks
m
0
l
Related resources
August 16, 2011 5:48:00 PM

caesparktom said:
..but what do you mean by "take the memory down to DDR3-1333 speeds and ensure you are at 1.5 volts"? --to check that psu is supplying correct voltage? but why take freq down to 1333 to check that?

ok, but are you just saying i paid too much for adequate cooling or do you think hyper 212+ will cool better and improve stability?


The P8P67 motherboard SHOULD be fine with 1.6 or 1.65 volt memoy modules (or some users report), but 1.5 is what is officially supported by Sandy Bridge. In your BIOS make sure the default voltage is being set correctly at 1.5v to match the memory modules. Taking the frequency down to 1333 is to further test and possibly point at a motherboard fault. Just another troubleshooting step...

Sorry... I overlooked that you already had the H60. My point was for performance per dollar, there is still no substitute for an air cooler (..with the Hyper 212+ being a good choice.).
m
0
l
August 16, 2011 11:56:20 PM

sadams04 said:
Before you do #1, burn a memtest boot disk, boot from it and let it make a few passes testing the memory. You will probably have no issues, but will be a better test of the memory. Also, try taking the memory down to DDR3-1333 speeds and ensure you are at 1.5 volts.
http://www.memtest.org/

Let us know the results... You choose quality parts and I wouldn't expect those types of problems at stock settings.


ran memtest for 2 passes with no errors. Not sure if that means the memory is good or just that the memory has not yet been proven to be bad. Should i run few more passes? I still need to try turning ram freq down to 1333 and re-running prime95 like you suggested.

Also, Proximon suggested i run furmark to stress gpu. i ran burn test twice (xtreme burn-in+Post FX), 10 minutes each time. GPU temps leveled at 91C both times and computer didn't crash so I don't think gpu is causing instability--at least not as result of high temperatures b/c gpu temps during prime95 are in 50-60C range.

I'll probably try swapping out memory next and just see if that solves problem. Any recommendations for this build? perhaps i will have better luck w/ corsair Vengeance™ — 16GB Dual Channel DDR3 Memory Kit (CMZ16GX3M4A1600C9)


thanks
m
0
l
August 17, 2011 1:49:23 AM

caesparktom said:
Quote:
Before you do #1, burn a memtest boot disk, boot from it and let it make a few passes testing the memory. You will probably have no issues, but will be a better test of the memory. Also, try taking the memory down to DDR3-1333 speeds and ensure you are at 1.5 volts.


running memtest now... but what do you mean by "take the memory down to DDR3-1333 speeds and ensure you are at 1.5 volts"? --to check that psu is supplying correct voltage? but why take freq down to 1333 to check that?

Quote:
Skip the water cooling. You will do just as good with a solid air cooler. The Hyper 212+ linked below will do a great job as-is, and even better if a second 120mm fan is added in a push / pull configuration.


ok, but are you just saying i paid too much for adequate cooling or do you think hyper 212+ will cool better and improve stability?

thanks



The hyper 212 will NOT improve cooling over the H60. Every professional review lists the H60 as superior to the Hyper 212. Also Proximon does not think it is a temp issue with your mobo and CPU - your GPU reaching 91C seems too high though - check with others on this thread though.
m
0
l
August 17, 2011 7:41:06 PM

Well, I found ONE post from a PNY tech that said 90C at full load with a Quaddro 4000 was acceptable. The card itself just has poor cooling.

I doubt very much that the Quaddro plays any part in a Prime95 crash after 12 hours though.

Let me recruit some more help here....
m
0
l
August 17, 2011 8:03:21 PM

Proximon said:
Well, I found ONE post from a PNY tech that said 90C at full load with a Quaddro 4000 was acceptable. The card itself just has poor cooling.

I doubt very much that the Quaddro plays any part in a Prime95 crash after 12 hours though..



especially since prime95 does not fully load the gpu. gpu temps during p95 blend are about 60C not 90C. and if GPU was going to crash system, it probably would have done so during the furmark burn test.

btw, thank for recruiting help. i really appreciate it.
m
0
l
August 17, 2011 8:16:00 PM

Proximon said:
Well, I found ONE post from a PNY tech that said 90C at full load with a Quaddro 4000 was acceptable. The card itself just has poor cooling.

I doubt very much that the Quaddro plays any part in a Prime95 crash after 12 hours though.

Let me recruit some more help here....


Proximon, I had a GPU (ATI 5450) that did not cool well in a ZT Systems I-7 920 computer and it would function very well for several hours and then it would produce numerous BSODs. The symptoms of this failure look very similar to what I experienced with my GPU (in that computer I am not sure whether it is the GPU or poor case cooling). In this situation, the 650D is a top tier cooling case (I own it and my CPU temps stay under 40C most of the time, my GPU temp hovers between 50C - 60C under full load - I don't game; the mobo stays under 40C most of the time). The case is very good at cooling on high fans and yet the GPU is reaching 91C under load - this could be a defective GPU.

If the GPU is the problem, then it won't manifest until after several hours of heavy use. This is what is happening with this system. What do you think?
m
0
l
August 17, 2011 8:40:50 PM

Proximon said:
It's a Quaddro :)  Have you seen the cooling solution?

http://www.nvidia.com/object/product-quadro-4000-us.htm...

They are supposed to be good up to 110C or so, as I recall.

It would be a simple test though, just take the cover off the case and point a household fan in there for a day.



Thats an excellent idea that would not cost anything to try. It could be that as the Quadro is pushed to its limits as it nears the temperature ceiling it becomes unstable and fails sometimes but stays working at other times. 91C is pushing the 110C ceiling.
m
0
l
August 17, 2011 8:43:44 PM

Here we have a VERY good point from one of our great helpers:

Quote:
Assuming the 'Option 1' Tri-Channel set then do NOT use XMP, and instead manually set the Frequency, DRAM Voltage, and CAS Timings sometimes a QPI/VTT/VCCIO -> 1.20v helps.

Then I noticed 'F3-12800CL9D-8GBXL' I've posted here a zillion times that those will not 9/10 run Rated in 4x4GB and to use F3-12800CL8D-8GBXM IF you want DDR3 1600 MHz speed.

Solution 1 - Run the F3-12800CL9D-8GBXL at DDR3 1333 MHz and look at CPU-z 'SPD' for 667MHz CAS Timings and add Command Rate -> 2.

Solution 2 - RMA RAM and get 2 sets of F3-12800CL8D-8GBXM

Solution 3 - Get a 4x4GB Matched Set -> F3-12800CL9Q-16GBXL



I had failed to note that you bought two sets of that RAM. You should try each set separately, or just do as he suggests.
m
0
l

Best solution

August 17, 2011 8:59:25 PM

I have not read all posts, my blinders are on to the RAM. FEW people know about the 4x4GB issues so @Proximon isn't the blame - at all; 4x4GB is very rare. Personally, I'd get a 4x4GB Matched Set because they are guaranteed to work together.

You can try setting them manually: Frequency -> DDR3-1600MHz, CAS 9-9-9-24 + Command Rate -> 2, raise the DRAM Voltage -> 1.55v~1.60v, plus VCCIO -> 1.20v. Then 'see' what happens... 50/50.

Good Luck! :) 
Share
August 18, 2011 12:18:38 AM

I do actually know about the 4x4 thing, but I was lazy and clicked the link before reading that he got two 4GB kits :)  So I just saw a 2-stick G.skill 1.5V kit.
m
0
l
August 18, 2011 1:18:59 AM

caesparktom, remove one of your 2 x 4 GB RAM kits and just run 8GB. Make sure that you have them in the correct slots for this setup. Run your computer under Furmark/Prime 95 and it should tell you if this is the problem. If it is the problem, you can always get a 4 x 4 GB matched set if you want to run 16 GB.

You might try 8GB to see if it meets your needs because reviews have shown that there is only a 1-3 % performance increase by going to 16 GB.

Good luck
m
0
l
August 18, 2011 2:40:11 AM

excitedly pulling two dimms as i type. If this works, I think i will just get a 16gb matched set rather than reducing ram speed to 1333. Will corsair Vengeance™ — 16GB Dual Channel DDR3 Memory Kit (CMZ16GX3M4A1600C9) work for me? Strictly speaking, its not list on QVL

@flong I understand that performance increase is negligible, more concerned with not running out. I run very ram-intensive programs at same time. Typically have rhino, photoshop, illustrator, indesign, and 50 chrome tabs open. Have always had to tip toe around my available ram and switch to single tasking when files get too heavy.

@jaquith, great observation/solutions! thank you. setting ram timings manually almost stabilized my system so I am eager to try setting the DRAM Voltage to 1.55v~1.60v and VCCIO to 1.20v. Though will probably still get a 4x4 matched set in the end.:) 









m
0
l
August 18, 2011 8:27:16 PM

ran blend test with single 2x4 kit of ram installed. Ran 12 hours then froze.>< i'm at a loss once again. Can try the other 2x4 kit or just swap out for 4x4 corsair kit. These 12+ hour tests are killing me...
m
0
l
August 18, 2011 8:37:14 PM

Q - What exact settings? {CPU Mutiplier, vCore, DRAM, VCCIO, RAM Feq/CAS - in other words all CPU, RAM and Voltage changes}.

BTW - do not use any 'CPU Fan' settings eg run the @ 100% assuming the H60 is connected to the CPU Fan Header.
m
0
l
August 18, 2011 9:23:48 PM

jaquith said:
Q - What exact settings? {CPU Mutiplier, vCore, DRAM, VCCIO, RAM Feq/CAS - in other words all CPU, RAM and Voltage changes}.

BTW - do not use any 'CPU Fan' settings eg run the @ 100% assuming the H60 is connected to the CPU Fan Header.


Changes:
DRAM freq 1600
DRAM Timing 9 9 9 24 2
DRAM Voltage 1.5V

everything else is default or auto

H60 pump is connected to pwr fan header (~4300 rpm)
H60 fan connect to cpu_fan header (~1000 rpm)

Thats all I have done.

I do notice turbo-mode is auto enabled and that "Ai overclock Tuner" is set to auto(vs manual or xmp). Does that mean it is adjusting my manual dram settings in real-time?. At the top of bios/Ai Tweaker tab it reads

Target CPU Turbo-mode Speed: 3800MHz
Target DRAM Speed: 1600MHz


m
0
l
August 18, 2011 9:50:21 PM

Reverse:
H60 pump is connected to pwr fan header (~4300 rpm)
H60 fan connect to cpu_fan header (~1000 rpm)


H60 pump -> CPU Header
H60 fan -> PWR Fan Header

BIOS:
AI OC Tuner -> Manual
DRAM freq 1600
DRAM Timing 9 9 9 24 2
DRAM Voltage -> 1.55V
VCCIO -> 1.20v
Q-Fan -> Disabled
Phase Control -> Optimized {~4.0GHz-4.3GHz; or high loads}, Extreme {>4.3GHz}
m
0
l
August 18, 2011 10:05:48 PM

ok, i'll try and retest

Q, set phase control to optimize or extreme?

fyi, h60 manual reads:
"connect fan power to CPU_FAN header"

(btw, also have 3 unused CHA_FAN headers because chassis fans are connected to case's fan controller.)

m
0
l
August 18, 2011 10:21:19 PM

The reason for the H60 manual is most BIOS regulate the pump's speed along with the CPU temp, and the end user doesn't know to set it properly. I was looking for a BIOS setting for CPU Fan failure which is why I like the pump on the CPU Fan header -- Beeps = Pump failure. I'm hoping the BIOS considers the CPU Fan failure as a H/W failure and you still get the Beep tones.

If NO OC but high load e.g. Prime95; Phase Control -> Optimized
If OC see above.
m
0
l
August 18, 2011 11:18:13 PM

computer froze after 30 minutes of p95 blend testing.
m
0
l
August 18, 2011 11:30:37 PM

perhaps voltage bumping was premature. b/c if Ai OC tuner has been set to auto this whole time. I haven't yet run a test with the following bios settings:

AI OC Tuner -> Manual
DRAM freq 1600
DRAM Timing 9 9 9 24 2
DRAM Voltage -> 1.5V

In fact even with both 2x4 kits installed (unmatched 16gb). The system was almost stable(ran 13 hours blend) with these settings:

AI OC Tuner -> Auto
DRAM freq 1600
DRAM Timing 9 9 9 24 2
DRAM Voltage -> 1.5V

what do you think?



m
0
l
August 19, 2011 12:34:09 AM

It tells me your BIOS is unstable, not from my settings, and needs to be updated.

The latest version is 1302. Prior to BIOS Flashing please use the 'Jumper' method as outlined in your manual on page 2-15. This fully resets the BIOS to Defaults. In addition, you cannot use any USB 3.0 ports, and only use the USB 2.0 ports on the back of the MOBO {I/O}.

See for additional help -> http://www.tomshardware.com/forum/289507-30-what-flash
m
0
l
August 19, 2011 2:52:34 AM

ok, updated bios to 1302.

now, what bios settings do you recommend i start with?

For now, i have just one set (2x4gb) installed.
m
0
l
August 19, 2011 3:18:27 AM

would like a new baseline before stepping voltage back up so set ai oc tuner to XMP for first p95 blend test. running now...reasonable?
m
0
l
August 19, 2011 3:42:48 AM

Your issues may be fixed on auto now. If not the settings listed 8 posts up by Jaquith would be fine to try.
m
0
l
August 19, 2011 2:24:48 PM

Sorry, I'm EST GMT -5:00.

Prime95 and any test - If the components are both good & compatible 'Default' values under those test(s) should be able to run continuously without problems, even with 4x4GB. Further, it's very common with most any P67/Z68 that the latest BIOS is advised; the latest exception is added 22nm CPU support which seems to cause some issues.

@Proximon is correct, the settings above should work fine with the possible 4x4GB @ 1600MHz problem I first mentioned.

Don't use XMP on anything but 2x4GB, and certainly not on 4x4GB; if you ever noticed the QVL you'll see 2XDIMM and 4XDIMM this means how XMP will and won't work, and since even the QVL listed RAM is 50/50 for 4XDIMM and Manually setting the RAM is best for OC - hence my reasoning for just setting it all Manually from the get go.

Settings:

BIOS:
AI OC Tuner -> Auto
DRAM Frequency -> DDR3-1600 MHz
DRAM Timing 9-9-9-24 & 2 {CAS-TRCD-TRP-TRAS & CMD}
DRAM Voltage -> Auto ; failure 1.55v
VCCIO -> Auto ; failure 1.125v~1.20v
Q-Fan -> Disabled

As I recall last night the Default Phase Control is 'Extreme' which simply means they're all enabled.

Hopefully, Prime95 testing is going well.
m
0
l
August 20, 2011 9:36:20 PM

ok three more failed tests after reflashing bios. All test ran with 2x4gb memory.

1.

BIOS:
AI OC Tuner -> XMP

Froze after 3 hours

2.

BIOS:
AI OC Tuner -> Auto
DRAM Frequency -> DDR3-1600 MHz
DRAM Timing 9-9-9-24 & 2 {CAS-TRCD-TRP-TRAS & CMD}
DRAM Voltage -> Auto
VCCIO -> Auto
Q-Fan -> Disabled

Froze after 5 hours

3.

BIOS:
AI OC Tuner -> Auto
DRAM Frequency -> DDR3-1600 MHz
DRAM Timing 9-9-9-24 & 2 {CAS-TRCD-TRP-TRAS & CMD}
DRAM Voltage -> 1.55v
VCCIO -> 1.125v
Q-Fan -> Disabled

Froze after 3.5 hours

Perhaps its time to swap memory and remove the ram compatibility variable from the equation. Clearly either ram is not very compatible, in which case I want to replace it with something that is OR something else is the problem (motherboard??), in which case I want to replace memory so the possibility of it being incompatible can be put to rest.
m
0
l
August 20, 2011 10:42:51 PM

First, I'm certainly not going to argue against the RAM change ;)  As I mentioned 4x4GB kit or the 1600 CAS 8 set.

The failures with 3~5 hours don't necessarily prove one set of settings better than another.

What about your temps? Before you posted the following below, CPU temps typically level-off after 10~15 minutes. HWMonitor also looks at other temps like chipset, etc -> http://www.cpuid.com/softwares/hwmonitor.html

OLD:
time | c1 | c2 | c3 | c4
5:00 | 48c | 51c | 55c | 52c
5:13 | 52c | 54c | 58c | 55c
6:50 | 53c | 53c | 57c | 55c


The Corsair 650D is a good case - just make certain there more air going out than in; slight negative pressure is ideal.

If you get bored try:
4x4GB; all 4 sticks
BIOS:
AI OC Tuner -> Auto
DRAM Frequency -> DDR3-1333 MHz
DRAM Timing 8-8-8-24 & 2 {CAS-TRCD-TRP-TRAS & CMD}
DRAM Voltage -> Auto
VCCIO -> Auto
Q-Fan -> Disabled
m
0
l
August 20, 2011 11:31:52 PM

Temps have been pretty consistent with temps i posted earlier. With Q-fan off, temps are little lower at idle b/c fan is full blast but once test starts they still level off at around 53-60. I will monitor other temps with hwmonitor for all subsequent tests. thx. the fact that test were crashing at 12+ hours and are now freezing at 3-5 hours makes me think that it really is not a temperature issue at all. I could run the test the Proximon suggested a while back (case open + house fan) to put a nail in that coffin i guess.

Also, I'll try 4x4gb 1333 8-8-8-24-2 test you suggested.

a few questions about my next ram (Really don't want to botch this up a second time)

1. Is corsair Vengeance™ — 16GB Dual Channel DDR3 Memory Kit (CMZ16GX3M4A1600C9) a good choice.

2. Should i stick to ram on the QVL? Buying from http://www.360buy.com/ and not many 4x4gb kits or 1600/cas8 sets available. Should I consider 1333?

3. Any specific ram recommendations? btw, I'm not married to 1600. Will 1333 generally be more stable?

Thanks again.
m
0
l
August 21, 2011 12:26:58 AM

ANS1 - It's complex. The older Corsair Dominator were very good with plenty of 'overhead' above Rated Frequency and CAS. The Corsair got cheap with their IC's and the nice overhead 'poof' gone. {profit margins vs quality} :(  Here's a recent article illustrating what I mean -> http://www.tomshardware.com/reviews/corsair-vengeance-c... jump to the 2133 MHz. So will the Corsair run yes do 'I' like the new IC's no.

ANS2 - I'm not going to say no to QVL. DDR3 1333 CAS 7 is as fast as DDR3 1600 CAS 9. Think of Frequency as 'Speed' and CAS as 'Stopping'.
I once made this analogy - "You have to look at RAM as a whole Frequency {bandwidth} and CAS {pauses} with transmission. Think of is this way a car can instantly go 100 MPH but the Stops are longer (9 sec) vs 75 MPH but the stops are shorter (7 sec) both with a 100 stops in a mile. Further, the CPU however says you can run only 85 MPH.

How long does the car take to drive a mile?


ANS3 - RAM with 4GB density -> http://www.newegg.com/Product/Productcompare.aspx?Submi...|20-231-442^20-231-442-TS%2C20-231-456^20-231-456-TS%2C20-231-445^20-231-445-TS%2C20-231-440^20-231-440-TS%2C20-231-429^20-231-429-TS

However, only the http://search.360buy.com/Search?keyword=CMZ16GX3M4A1600... were found. Navigating/translating is difficult since I only understand English.

Your English is outstanding! I compared my Cannon D90 7,099 Yuan / 1,100 USD ; conversion was 1,236 USD.

------------------

Here's my concern now, there's no good reason that the RAM should fail when running 2x4GB. Yes, I can see and understand 1 stick being 'bad' but statistically 2 sticks is very rare.
1. Verify that you didn't mix the (2) sets. The S/N should help identify sets.
2. Run MSCONFIG -> select Diagnostic mode then test at the 2x4GB DDR3 1600 MHz & other settings.
3. Use only the Blue slots for 2x4GB

I assume Memtest ran with no errors.
m
0
l
August 21, 2011 4:22:43 AM

great article and very useful cas/freq analogy. I wonder how useful "freq/bandwidth" and "cas/pauses" are as performance indicators for workstations. I've heard people say latency is more import than speed but not sure what the context was. To be honest, i am not too interested in paying a premium for super fast memory. Just want stable ram that will remain stable at 16gb with a modestly cpu overclock. Unfortunately, @ 360.com I'm pretty much limitted to:

1. G.SKILL Ripjaws X Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1333 (PC3 10666) F3-10666CL7D-8GBXH - overpriced on 360.com -> 2x8gb kits @ 1972 cny/300 usd (240 usd on newegg)

2. corsair Vengeance™ — 16GB Dual Channel DDR3 Memory Kit CMZ16GX3M4A1600C9 - almost equally overpriced 1x16gb kit @1499 cny/234 usd (vs 114 usd on newegg)

can buy from international site, but no time to wait for it to get hear.

Quote:
Here's my concern now, there's no good reason that the RAM should fail when running 2x4GB. Yes, I can see and understand 1 stick being 'bad' but statistically 2 sticks is very rare.
1. Verify that you didn't mix the (2) sets. The S/N should help identify sets.
2. Run MSCONFIG -> select Diagnostic mode then test at the 2x4GB DDR3 1600 MHz & other settings.
3. Use only the Blue slots for 2x4GB


i did re-check the s/n numbers, and I am using blue slots (A2, B2). I will run test w/ msconfig/diagnostic tool and let you know. Also, I have been assuming that problem is with compatibility rather than bad sticks so I haven't tested the other 2x4 kit by itself which i realize is kinda dumb. so i'll do that too.

memtest passed w/ no errors with 4x4gb installed, so didn't bother testing 2x4 kits separately.



m
0
l
August 21, 2011 5:37:01 AM

caesparktom said:
great article and very useful cas/freq analogy. I wonder how useful "freq/bandwidth" and "cas/pauses" are as performance indicators for workstations. I've heard people say latency is more import than speed but not sure what the context was. To be honest, i am not too interested in paying a premium for super fast memory. Just want stable ram that will remain stable at 16gb with a modestly cpu overclock. Unfortunately, @ 360.com I'm pretty much limitted to:

1. G.SKILL Ripjaws X Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1333 (PC3 10666) F3-10666CL7D-8GBXH - overpriced on 360.com -> 2x8gb kits @ 1972 cny/300 usd (240 usd on newegg)

2. corsair Vengeance™ — 16GB Dual Channel DDR3 Memory Kit CMZ16GX3M4A1600C9 - almost equally overpriced 1x16gb kit @1499 cny/234 usd (vs 114 usd on newegg)

can buy from international site, but no time to wait for it to get hear.

Quote:
Here's my concern now, there's no good reason that the RAM should fail when running 2x4GB. Yes, I can see and understand 1 stick being 'bad' but statistically 2 sticks is very rare.
1. Verify that you didn't mix the (2) sets. The S/N should help identify sets.
2. Run MSCONFIG -> select Diagnostic mode then test at the 2x4GB DDR3 1600 MHz & other settings.
3. Use only the Blue slots for 2x4GB


i did re-check the s/n numbers, and I am using blue slots (A2, B2). I will run test w/ msconfig/diagnostic tool and let you know. Also, I have been assuming that problem is with compatibility rather than bad sticks so I haven't tested the other 2x4 kit by itself which i realize is kinda dumb. so i'll do that too.

memtest passed w/ no errors with 4x4gb installed, so didn't bother testing 2x4 kits separately.


That would be an excellent idea since you have two sets of 2 x 4GB RAM. One of the two sets may be bad. You could also try just running one 4GB stick of RAM to see if that affects the stability of your system.

The performance sweet spot for for the Intel K chips RAM is 1866 and in the USA, it is not overly expensive. Since you are a power user I would recommend that you go with 1866. My computer has DDR3 1600 and the Windows performance Index lists it at 7.6 (max is 7.9) all other components on my system are 7.9 except the CPU (I have not overclocked my 2600K yet).

jaquith, I have the 650D case and with fans on high it definitely creates a negative pressure airflow (moving more air out of the case than what is coming in). I have removed the higher HDD cage and I am getting excellent temps for the mobo, CPU and video card. Of course I have the Noctua NH-D14 cooling the the CPU and so that is a huge plus (it is probably the best air cooler on the market that does not sound like a jet engine).

Also, Jaquith and Proximon, have you considered the possibility that the intermittent BSOD could be caused by the SSD or the video card? Just a thought. While I have had no problems with my Corsair 120GB Force 3 GT, many other SSD owners of all brands including OCZ have reported problems with random BSOD from the SSD. When is is an SSD failure, the SSD has to be replaced to fix the problem. This could be tested by loading Windows 7 on another drive and booting from that drive. If the system remains stable when booting from a different HDD, then the SSD is the problem.

Not being familiar with the video card the OP has, I am not sure how to trouble-shoot that particular card in an economical way. However it could be removed and another card could be used to see if the system stabilized with a different card.

Because the intermittent failure is so persistent, it could be a major component like the SSD or the video card.
m
0
l
August 21, 2011 7:41:07 AM

Quote:
The performance sweet spot for for the Intel K chips RAM is 1866 and in the USA, it is not overly expensive. Since you are a power user I would recommend that you go with 1866. My computer has DDR3 1600 and the Windows performance Index lists it at 7.6 (max is 7.9) all other components on my system are 7.9 except the CPU (I have not overclocked my 2600K yet).


I'll consider. Do you think 1866 really outperforms 1600 in terms of noticeable difference?

G.SKILL Sniper Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1866 (PC3 14900) Desktop Memory Model F3-14900CL9D-8GBSR is an option. Price is about the same as corsair vengeance 1600 9-9-9-24. But I don't know if I would have the same problem with the unmatched 4x4gb configuration that I might be having now. Really wish i could keep my g.skill rj 1600 9-9-9-24 sticks for no other reason than they are WAAAY cheaper. paid 112 usd for 4x4gb. Ram options above are 234 & 300 bucks! and the corsair memory has the exact same timing!


m
0
l
August 21, 2011 8:14:13 AM

caesparktom said:
Quote:
The performance sweet spot for for the Intel K chips RAM is 1866 and in the USA, it is not overly expensive. Since you are a power user I would recommend that you go with 1866. My computer has DDR3 1600 and the Windows performance Index lists it at 7.6 (max is 7.9) all other components on my system are 7.9 except the CPU (I have not overclocked my 2600K yet).


I'll consider. Do you think 1866 really outperforms 1600 in terms of noticeable difference?

G.SKILL Sniper Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1866 (PC3 14900) Desktop Memory Model F3-14900CL9D-8GBSR is an option. Price is about the same as corsair vengeance 1600 9-9-9-24. But I don't know if I would have the same problem with the unmatched 4x4gb configuration that I might be having now. Really wish i could keep my g.skill rj 1600 9-9-9-24 sticks for no other reason than they are WAAAY cheaper. paid 112 usd for 4x4gb. Ram options above are 234 & 300 bucks! and the corsair memory has the exact same timing!


There is a noticeable difference. Bit-tech has an excellent article on this but I do not have the link on my new computer (it was saved on my old computer). What you have to consider is that the cost difference between the 1600 RAM and the 1866 RAM is usually less than $25 and that there is a noticeable difference performance (it is perceptible to the normal user), then it is a no-brainer to go to the 1866 RAM. CL8 or CL9 will work, you do not have to pay the extra costs for the CL7.

Conversely, there is little difference in performance between the 2133 RAM and the 1866 RAM and the 2133 RAM usually comes at a high price premium. This is only true for the Intel K chips.

Because you are a power user, this becomes all the more important. Look up the article on Bit-Tech, it is the best article on RAM I have found.

Let us know if your problem is resolved - I would really like to see you get it fixed

m
0
l
August 21, 2011 8:17:58 AM

It's benchmarks mostly, but if you are doing memory intensive tasks that take a long time, and you are losing money the longer it takes, faster RAM might be worthwhile. I have heard it can work out to perhaps 30 minutes a week for some professionals.

Here's a good write-up
http://www.anandtech.com/show/4503/sandy-bridge-memory-...
m
0
l
August 21, 2011 8:36:50 AM

Proximon said:
It's benchmarks mostly, but if you are doing memory intensive tasks that take a long time, and you are losing money the longer it takes, faster RAM might be worthwhile. I have heard it can work out to perhaps 30 minutes a week for some professionals.

Here's a good write-up
http://www.anandtech.com/show/4503/sandy-bridge-memory-...


I found the Bit-Tech article on RAM it is here:

http://www.bit-tech.net/hardware/memory/2011/01/11/the-...

It is one of the best articles on Sandy Bridge RAM performance I have found.

Read the Anandtech article - you are right it is an excellent article. The Bit-Tech article does concentrate a little more on the 1866 category RAM. It shows that it performs almost as well as the 2133 and the cost difference is significant between the two.
m
0
l
August 21, 2011 1:46:08 PM

I've been keeping an eye on the OP. There are a few possibilities, and 'I' haven't concluded anything yet. The problems can be caused from more than (1) root issue.

The MSCONFIG is a step short of 'Safe Mode' and takes 'Windows' & 'Apps' conflicts out of the equation. The 2x4GB @ 1600 should not be the problem particularly if swapped and kept in matched pairs; meaning in that state it indicates other problems. The OP needs to keep an eye on ALL temps including those of the chipset to insure Tmax isn't being reached.

One house keeping item - make certain Start = 0 for the registry values in AHCI are indeed correct; see -> http://support.microsoft.com/kb/922976 and see -> http://i1013.photobucket.com/albums/af254/Jaquith/AHCI_...

Further, it's a really good idea to look at the Event Manager to examine any Critical errors; see -> http://i1013.photobucket.com/albums/af254/Jaquith/Event...

More than likely, one common setting that helped the OP was the Command Rate -> 2T, and if you look at my memory posts almost always I give the full set of timings; XMP often does not set the 2T but instead uses 1T. The Blended/Torture tests will exploit the 1T/2T. As I said slowly were're moving in systematically. If testing in safe mode all of the drivers are dummied-down so much it's pointless to test and I wouldn't be surprised if Safe Mode passes the Blend Test, but then what...it might conclude the GPU but if Safe Mode fails then it might indicate the SSD or other Hardware issue; leave that option open.

As far as faster frequency RAM, you might want to revisit the article and look at the Cinebench R10, Image and encoding bench tests. Further, >1866 causes more problems that it solves and more than likely either the CAS Timings needs {9-10-9 -> 10-10-10 or slower} to be relaxed and/or CPU OC to keep the system stable.

Next, sure GPU drivers and SSD firmware could be the problem. The most common P67/Z68 remedy is the BIOS.
Quadro 4000 275.89 WHQL 2011.08.12 Windows 7 x64 - http://www.nvidia.com/object/quadro-tesla-win7-winvista...
OCZ Vertex 3 Firmware v2.11 - http://www.ocztechnology.com/ssd_tools/OCZ_Vertex_3,_Ve...

Let us know how the tests turned out.
m
0
l
August 21, 2011 2:21:25 PM

jaquith said:
I've been keeping an eye on the OP. There are a few possibilities, and 'I' haven't concluded anything yet. The problems can be caused from more than (1) root issue.

The MSCONFIG is a step short of 'Safe Mode' and takes 'Windows' & 'Apps' conflicts out of the equation. The 2x4GB @ 1600 should not be the problem particularly if swapped and kept in matched pairs; meaning in that state it indicates other problems. The OP needs to keep an eye on ALL temps including those of the chipset to insure Tmax isn't being reached.

One house keeping item - make certain Start = 0 for the registry values in AHCI are indeed correct; see -> http://support.microsoft.com/kb/922976 and see -> http://i1013.photobucket.com/albums/af254/Jaquith/AHCI_...

Further, it's a really good idea to look at the Event Manager to examine any Critical errors; see -> http://i1013.photobucket.com/albums/af254/Jaquith/Event...

More than likely, one common setting that helped the OP was the Command Rate -> 2T, and if you look at my memory posts almost always I give the full set of timings; XMP often does not set the 2T but instead uses 1T. The Blended/Torture tests will exploit the 1T/2T. As I said slowly were're moving in systematically. If testing in safe mode all of the drivers are dummied-down so much it's pointless to test and I wouldn't be surprised if Safe Mode passes the Blend Test, but then what...it might conclude the GPU but if Safe Mode fails then it might indicate the SSD or other Hardware issue; leave that option open.

As far as faster frequency RAM, you might want to revisit the article and look at the Cinebench R10, Image and encoding bench tests. Further, >1866 causes more problems that it solves and more than likely either the CAS Timings needs {9-10-9 -> 10-10-10 or slower} to be relaxed and/or CPU OC to keep the system stable.

Next, sure GPU drivers and SSD firmware could be the problem. The most common P67/Z68 remedy is the BIOS.
Quadro 4000 275.89 WHQL 2011.08.12 Windows 7 x64 - http://www.nvidia.com/object/quadro-tesla-win7-winvista...
OCZ Vertex 3 Firmware v2.11 - http://www.ocztechnology.com/ssd_tools/OCZ_Vertex_3,_Ve...

Let us know how the tests turned out.


I have not heard of 1866 RAM causing stability problems and there is 1866 RAM with CL timings or CL7 or CL8 and CL9. It looks like your talking about GREATER THAN, i.e. ">" than 1866, therefore 1866 is stable. I have heard that 2133 can cause problems but I had thought that was resolved.

Your point to check the SSD settings is excellent, i.e. is it in ACHI mode. I know I had to switch my computer manually in the BIOS to ACHI mode - Windows 7 did not switch it automatically even though it detected the SSD. I have the Corsair 120 GB Force 3 GT. There is an excellent guide to SSD optimization here:

http://thessdreview.com/ssd-guides/optimization-guides/...

Note that some of the items are not necessary to get full production out of your SSD.
m
0
l
August 21, 2011 4:33:33 PM

Regarding problems with DDR3 1866 MHz+ come over to my MOBO + RAM section and you will. No doubt 2133 MHz is more frequently a problem. I've seen UD7's with ultra fast frequency issues where there's no OC 'required'; specs Support for DDR3 2133/1866/1600/1333/1066 MHz memory modules - notice any (OC) in the specs?! VERY FEW people get 1866, 2000, or 2133 MHz kits, but I've handled at least a dozen problems, and in particular the solution was to both tighten and raise the CAS as I illustrated above to a normal JEDEC standard. ** Clearly, I'm not saying high Frequency is 'bad' or 'without' it's merits. ** I have a set of Dominator GT's 2000 MHz on my 980X now - you must BCLK OC to keep them fully stable in addition to CPU Multiplier OC for higher CPU frequencies.

I've lost count how many AHCI issues I've seen, and it's more 'typical' to see Start = 3 {IDE} than Start = 0 {AHCI} with ACHI in the BIOS. Everyone should check it at first boot. Gigabyte probably more than others exhibits 'unhappy' behavior, and it's a 20 second fix. Afterwards run Check Disk to remove possible corruption.
m
0
l
August 21, 2011 4:59:54 PM

jaquith said:
Regarding problems with DDR3 1866 MHz+ come over to my MOBO + RAM section and you will. No doubt 2133 MHz is more frequently a problem. I've seen UD7's with ultra fast frequency issues where there's no OC 'required'; specs Support for DDR3 2133/1866/1600/1333/1066 MHz memory modules - notice any (OC) in the specs?! VERY FEW people get 1866, 2000, or 2133 MHz kits, but I've handled at least a dozen problems, and in particular the solution was to both tighten and raise the CAS as I illustrated above to a normal JEDEC standard. ** Clearly, I'm not saying high Frequency is 'bad' or 'without' it's merits. ** I have a set of Dominator GT's 2000 MHz on my 980X now - you must BCLK OC to keep them fully stable in addition to CPU Multiplier OC for higher CPU frequencies.

I've lost count how many AHCI issues I've seen, and it's more 'typical' to see Start = 3 {IDE} than Start = 0 {AHCI} with ACHI in the BIOS. Everyone should check it at first boot. Gigabyte probably more than others exhibits 'unhappy' behavior, and it's a 20 second fix. Afterwards run Check Disk to remove possible corruption.


Can you run Check Disk with an SSD? This is my first SSD and I am still in the learning curve. I have been very fortunate because the ATTO performance for my Corsair GT Force 3 is nearly 560 mb/s read and 512 Mb/s write. It appears to be performing at its maximum level. I was careful to maximize the settings when I started up my computer. I have had no issues at all with the Corsair SSD. In some real-world benchmarks it outperforms the 240 GB OCZ SSD, which is amazing for a 120 GB drive.

I see what you are saying about RAM. People are trying to OC their RAM and causing instabilities. I think you are saying that the factory clocked 1866 RAM kits are stable but don't try to OC them to 2133 - right? I thought about overclocking my RAM from 1600 to 1866 but I didn't think it was worth the risk. It is cheaper and safer just to by a set of 1866. They regularly go on sale at Newegg for $65 (Gskill and Corsair).
m
0
l
August 21, 2011 5:43:41 PM

Yes can run Check Disk on an SSD, see -> http://i1013.photobucket.com/albums/af254/Jaquith/Scan-... I suggest both options on the Primary drive {C}, schedule and reboot...takes 20~30 minutes depending on SSD size & Speed, and a lot longer on say a 1TB or larger HDD. Yes all SSD's are nice. Though running Defrag on any SSD adds pointless writes on the SSD; to an SSD data is data wherever it's located.

No, I'm saying 'Rated' 1866 MHz RAM and faster can be an issue. Stability has all to do with the quality of the RAM's IC, CPU IMC, and BIOS and BIOS settings.
m
0
l
August 22, 2011 7:03:22 AM

success at last!! Put in the other 2x4gb set and set bios to auto 9-9-9-24-2 1.5V and ran the blend test for 24 hours without incident. So, one of the two kits that was bad. Which I suppose explains why the stress test failed in half the time with just two sticks. Little embarrassed I didn't isolate each set sooner. oh well. I learned alot fumbling around for a week when i wasn't pulling out my hair.

flong, Proximon, jaquith: Thank you very much. this is my first build and i really appreciate all your help picking parts and troubleshooting.

Bit-Tech & anandtech articles on ram both very good. I'm on the fence with 1866. Bit-Tech article is pretty convincing when it argues the benifit of higher frequency for heavy multitasking. The problem is, its a benifit-cost argument which doesn't translate in china where its not 25 dollars more but more like 125! Might actually "suffer" for a bit with my 8gb sticks (the ones that actually work) while I catch up with my work.

also, great article on SSD optimization!
m
0
l
August 22, 2011 7:58:55 AM

I'm glad it all worked out. We should have done that long ago. That will teach me to be more methodical next time :)  Be sure to award jaquith with the best answer.
m
0
l
August 22, 2011 8:21:51 AM

caesparktom said:
success at last!! Put in the other 2x4gb set and set bios to auto 9-9-9-24-2 1.5V and ran the blend test for 24 hours without incident. So, one of the two kits that was bad. Which I suppose explains why the stress test failed in half the time with just two sticks. Little embarrassed I didn't isolate each set sooner. oh well. I learned alot fumbling around for a week when i wasn't pulling out my hair.

flong, Proximon, jaquith: Thank you very much. this is my first build and i really appreciate all your help picking parts and troubleshooting.

Bit-Tech & anandtech articles on ram both very good. I'm on the fence with 1866. Bit-Tech article is pretty convincing when it argues the benifit of higher frequency for heavy multitasking. The problem is, its a benifit-cost argument which doesn't translate in china where its not 25 dollars more but more like 125! Might actually "suffer" for a bit with my 8gb sticks (the ones that actually work) while I catch up with my work.

also, great article on SSD optimization!



Yes :-)- we all share the happiness of your success. When you have four RAM modules, it is easy to not think of isolating them.

Since your RAM is guaranteed for life, you may be able to return the 2 x 4GB set for reasonable shipping (I don't know what shipping in China is). If you can, you could try the replacement set with your working set and they may work. Sometimes non-matching sets will work together. That may be the most cost efficient path for you.

Your 1600 RAM will perform very fast with the Intel K chips. You should try it first before you upgrade to 1866. You may not need to upgrade. If your 1600 RAM meets your needs, then you can upgrade to 1866 (or not) at your leisure when you have an opportunity to pick up the RAM on sale.

Props to Jaquith for pointing us to the RAM. He caught the issue when no one had been looking at it. Thanks to both Proximon and Jaquith for their help - you guys are great.

Be careful with the SSD Optimization article. Not all of those optimizations are necessary. The article is good at pointing out which ones are critical. I was really bummed to find out that System Restore had to be deactivated for an SSD. That is a very useful and practical feature. I am able to back up the data on my other drives with restore points however. I suppose if my operating system got corrupted I could do a clean install if necessary.

You really have a very good build. The 650D is one of the best cases out there (I bought it). It cools very well, is quiet and is exceptionally well laid out. It also is very up-to-date with the SATA III hot-swap drive and USB 3.0 connections. Only the Silverstone cases cool better and the the FT02 has not been updated with USB 3.0 and it does not have a a hot-swap bay. Nor do the Silverstone cases have the easy access quick release side covers. The tooless system for the case is one of the very best. The quality exceeds Cooler Master cases.

The Corsair HX 750 PSU is maybe the best out there (and has a seven year warranty) and your V3 120 GB used for your operating system will make your computer wicked fast. I have the F3 Spinpoint and I am getting nearly 150 mb/s read and write speeds with it which is remarkable for an HDD. The Seagate drive I had with my I-7 920 was about 90 -100 mb/s by comparison. The H60 cools very well with two fans - about equal to the Noctua NH-D14, which is the dominate air-cooler out there right now.

When you OC your CPU you will have a computer that is faster than the I-7 990, a $1000 CPU.

The Dell U2410 is one of the best monitors out there. It is very similar to my HP 2474 and it has a great picture. You would have to go to an NEC or similar monitor that costs twice as much to get small improvements in picture quality.

Your build and mine are very similar. I can tell you that my computer blows the doors off of my previous I-7 920, which was considered very fast in its day. I think that you will be very happy.
m
0
l
August 22, 2011 8:29:33 AM

Proximon said:
I'm glad it all worked out. We should have done that long ago. That will teach me to be more methodical next time :)  Be sure to award jaquith with the best answer.


You and Jaquith both deserve a ton of credit for helping someone in need. Thanks for all your advice.
m
0
l
August 22, 2011 12:28:47 PM

Glad to hear it's been figured-out :)  I'm sure it's a major burden off your shoulders.

However, hold the farm with SSD optimization, basically Doug's article is the same but read both the comments and the article -> http://www.tomshardware.com/reviews/ssd-performance-twe...

I think you're out of your mind to turn-off a few things: Hibernation and System Restore to name a couple. Besides the UPS most folks use Hybrid-Sleep {S3/standby + S4/hybrid} state that saves you a lot of KWH. Move the Documents, PST/OST files, etc off the SSD.
m
0
l
August 22, 2011 5:30:57 PM

jaquith said:
Glad to hear it's been figured-out :)  I'm sure it's a major burden off your shoulders.

However, hold the farm with SSD optimization, basically Doug's article is the same but read both the comments and the article -> http://www.tomshardware.com/reviews/ssd-performance-twe...

I think you're out of your mind to turn-off a few things: Hibernation and System Restore to name a couple. Besides the UPS most folks use Hybrid-Sleep {S3/standby + S4/hybrid} state that saves you a lot of KWH. Move the Documents, PST/OST files, etc off the SSD.


Wow you are the first person I have heard that has said system restore should be enabled. I will do some more research because I personally like the MS System Restore function A LOT. It is a great feature in Windows 7 and it works very well. It has saved me a couple of times. Hibernation is not very useful to me because my computer does go into hybrid sleep mode but it actually takes just about as long to put it to sleep and start it up as it does to turn it off and start it up. It boots Windows 7 in less than 25 seconds (from the point that you push the ON button).

Right now I have all data going to a 1 TB Samsung F3 and another Hitachi drive. I did not move the Outlook PST files to the F3 (they are on the SSD) but that probably is a good idea considering the size of the email files. I think I can move it without having to reinstall Outlook but I will have to look into that.

BTW what are OST files? Operating system files? Should I move them off of my SSD?

I use the Force 3 GT 120GB drive only for program files and the operating system and I don't even have it half filled. With this particular drive the performance degradation between 1/2 - 3/4 fill is insignificant and so I do not have to worry about the 1/2 fill point. My goal is to keep it near the half fill point so that I do not lose performance.

That being said, my Samsung and Hitachi ATTO scores are very close to 150 mb/s read and write speeds. So they are no slow pokes by any means. I was blown away that both of these HDDs pushed the 150 mb/s mark - the Hitachi is a 6 gb/s drive and it is connected to a SATA III port (I have extra SATA III ports). The Samsung is SATA II but it is almost as fast as the Hitachi.
m
0
l
!