Memory modules?

A friend had a problem with his laptop and we suspected it was the ram, since we both have HP’s, his a T9300 64bits mine a T7500 32bits (4GB 2x2 PC2-5300 667MHz DDR2 SODIMM), we took mine out of this and put it in his.
We were right and it fixed the problem, since I have the use of another T7500 32bits I let him use my ram until he bought some.
After we re-installed the ram back into this my laptop’s FPS has dropped by half. I know this because I used the same stress test on both this laptop and the other T7500 32bits.
Question: Is this a software issue or have we damaged some of the memory modules?
Any suggestions would be greatly appreciated.
Cheers Heresy

 

Sounds like you probably don't have the RAM seated right - try reseating the RAM.

Or, his laptop could be putting out too much voltage and screwed your RAM over

 

How are you reading the memory speed?

If you are using CPU-z then it reports the clock frequency. DDR2 means that the data speed is twice the clock frequency.

If you in doubt about the performance then download SiSoftware Sandra and run the memory bandwidth benchmark and compare your score with the results for the same chipset in Sandra's database.

John

 

I’ve taken the ram back out twice, tried reversing them top on bottom. Tried a recovery, plus a reset default in bios, I don’t see how the ram could of been fried by the T9300, because that laptop was running fine when we replace this ram for the new? Can’t see how they could have been damaged, because they’re not difficult to install. I run a Microsoft Windows Memory Diagnostic and it said no errors.This is what I use to test the FPS http://www.daionet.gr.jp/~masa/rthdribl

 

Is it possible that when I took the ram out of this it could of changed or lost setting that was stored in the ram? I know I’m grasping at straws here but I really don’t have anything else.

 

Make sure all the conditions are the same as the last time you benchmarked your machine. It is far more likely that your laptop is in power saving mode or something like that than your ram being damaged.

 

When the ram was taken out of this it was running off the battery, as I had taken this laptop to his place and could very well been in sleep mode. So if I’m hearing you right I should wack the battery back in, let it go into sleep mode and then what?

 

IF you removed the ram while it was in sleep that could cause not only damage to the ram but to the sockets them selves. You need to remove power AND the battery before changing ram modules.

 

I wasn’t there when the ram was taken out or replaced, I was just told the battery was out when he took it out but in when he replaced it.
I’m guessing it would of been turned off but that wasn't good either? So I’m up for new ram on the bottom and new sockets as well at worst and of course murder chargers?

 

Isolate the problem. Run memory bandwidth bench. I don't at this point think you have a RAM problem. You should be around 4000MB/s.

 

My point was to make sure you have the AC adapter plugged in when you re-run the benchmark if it was when you initially benchmarked your computer, or to set your laptop for high performance instead of power saving mode if it was set to that when you first ran the benchmark. You need to make sure that your processor and everything else are running at the same frequencies they were when you first ran the benchmark.

If you really think your RAM is damaged, I'd suggest making a memtest86 boot disk and run memtest86 until you either get an error or ~12-24 hours have gone by without any errors.

 

You think I should run both sticks for 12 hours each?

 

I don't think there is anything wrong with your ram, so I don't think you should even bother running more memory tests unless you have nothing better to do with your time.

There is no set amount of time to run memtest86- you run it until you either get errors or it has been running for a "long time" without errors- long enough for you to feel confident that your ram never makes a single mistake.

If you decide to run memtest, you can run it with both sticks in or one stick at a time. Testing both at the same time would take less time, but if you do get error(s), you will have to test sticks individually to determine which one is bad.

I don't know anything about the benchmark that you've been using, but I would imagine that your processor speed matters WAY more than your RAM. Are you sure your processor is running at 2.2ghz while running this benchmark? I'd suggest running using a few different benchmark programs instead of running memtest86.

 

http://forum.notebookreview.com/showthread.php?t=142746 1 pass each stick this guy reckons, yea I doubt my ram is dead but better safer. By this post it should take my system just over 3 hours both sticks, can do I still have use of other computers.
Thanks for everyone's help it's been invaluable

gotta say that I might need more ;0)

 

This is not conclusive, it is a different notebook. You mention same CPU T7500 but is it even the same model notebook? Even if it is your methodology is sketchy.

memtest is fine but potentially time consuming. For the sake of expedience I do not think you are at the point were using memtest is worthwhile.

Your speculations and conclusions are on a shaky foundation. Please step back and consider a more systematic diagnostic approach. It will likely help you to arrive at the correct answer. Which will result in a fix to your problem.

I do not know your computer knowledge level so if I am off base please forgive and ignore. I assume it is at least in this area entry level?

Is the GPU an IGP? If it is not then jumping to system RAM on a DX bench seems like checking your radiator when your engine oil light goes on.

If IGP still troubled by your conclusions based on a DX bench. The reason is you can isolate the RAM's function more directly. John mentioned first (SiSoft Sandra) memory bandwidth.

Your post is confusing I think John was thinking when you said "FPS" you meant "FSB" as it clearly made more sense. So he mentioned CPU-Z. You have sense cleared up and well still confusing. Back to your conclusions.

JPZ made a very good observation as far as power saving features. CPU is now downclocked? That would certainly explain all you describe.

Being aware of circumstance (RAM being removed and reinstalled) is important. Using coincidence to draw conclusions is dangerous (well at least in the sense of trying to resolve).

So testing on different systems with the same CPU is not a valid for comparison in most cases other than CPU specific testing and still open to much interpretation.

Just do a memory bandwidth bench already.

 

look dude I barely understood a word, I can tell you this, A. My playing environment was good before I took the ram out, now it's not. B. I have 2 computer here with the same spec's (identical, got them off the same seller both dv9653cl) and the guy I'm here with said I'm paranoid, there's nothing wrong with my rig. C. So I downloaded a stress test and ran them on both rigs with identical spec's. D. The other rig's score is twice my. E. What did I do for this to happen? D. Took out the ram, now the ram has been damaged or some setting has changed and that's a fact, no matter how much I know or don't!

 

First stick completed with no errors, took 1hr 17minutes

 

Dude I am on your side. Other than your being paranoid I would add "freaking out". You don't have to go there. I believe you. Something is going on. Let's fix it.

Please, please do the bandwidth test. I made it simple. Download here, link. Then run the memory bandwidth benchmark. That will help us to help you.

And just so you know if it helps, I think we all think your RAM might be fine. And with the correct info we can get you up and running very shortly. Is that not good news.

 

Thanks I do appreciate your help and I will run this other test with pleasure. BTW I'm not freaking or paranoid, just a freak!
And just which link is this download? There's more links there than the German railway?

 

Yes many links. Pick one in your country or one you know. All are the same file.

 

I downloaded to both laptops and the one with a problem was 2,06/7GB and the one without was 4.05/6GB Since I memtested 1 of the sticks 100% and 70% of the other and both of the stick were sitting in the bottom slot. Could it be safe to say there might be a problem with the top slot if not a software issue?

 

It should like this.

SiSoftware Sandra

Benchmark Results
Int Buff'd iSSE2 Memory Bandwidth : 3980 MB/s
Float Buff'd iSSE2 Memory Bandwidth : 4015 MB/sResults Interpretation : Higher index values are better.

Int Buff'd iSSE2 Memory Bandwidth
Assignment : 3951 MB/s
Scaling : 3914 MB/s
Addition : 4030 MB/s
Triad : 4027 MB/s
Data Item Size : 16 byte(s)
Buffering Used : Yes
Offset Displacement Used : Yes
Bandwidth Efficiency : 62% (estimated)

Float Buff'd iSSE2 Memory Bandwidth
Assignment : 3941 MB/s
Scaling : 3959 MB/s
Addition : 4102 MB/s
Triad : 4058 MB/s
Data Item Size : 16 byte(s)
Buffering Used : Yes
Offset Displacement Used : Yes
Bandwidth Efficiency : 63% (estimated)

Performance Test Status
Run ID : ZEPTO-6024W on 22 August 2007 at 19:33:08
Platform Compliance : Win32 x86
Memory Used by Test : 512MB
NUMA Support : No
SMP (Multi-Processor) Benchmark : No
Total Test Threads : 1
Multi-Core Test : Yes
SMT (Multi-Threaded) Benchmark : No
Processor Affinity : P0C0T0
System Timer : 14MHz
Page Size : 4kB
Use Large Memory Pages : No

Features
SSE Technology : Yes
SSE2 Technology : Yes
SSE3 Technology : Yes
Supplemental SSE3 Technology : Yes
SSE4 Technology : No
EMMX - Extended MMX Technology : No
SSE4A Technology : No
HTT - Hyper-Threading Technology : No

Chipset 1
Model : Inventec Corp Mobile PM965/GM965/GL960 Express Processor to DRAM Controller
Revision : A4
Front Side Bus Speed : 4x 200MHz (800MHz data rate)
Width : 64-bit
Maximum Bus Bandwidth : 6400MB/s (estimated)

Logical/Chipset 1 Memory Banks
Bank 0 : 512MB DDR2-SDRAM 5.0-5-5-15 (tCL-tRCD-tRP-tRAS) CR1
Bank 1 : 512MB DDR2-SDRAM 5.0-5-5-15 (tCL-tRCD-tRP-tRAS) CR1
Bank 4 : 512MB DDR2-SDRAM 5.0-5-5-15 (tCL-tRCD-tRP-tRAS) CR1
Bank 5 : 512MB DDR2-SDRAM 5.0-5-5-15 (tCL-tRCD-tRP-tRAS) CR1
Shared Memory : 8MB
Channels : 2
Memory Bus Speed : 4x 333MHz (1332MHz data rate)
Width : 64-bitPerformance Acceleration Technology : No
Memory Controller in Processor : No
Maximum Memory Bus Bandwidth : 21312MB/s (estimated)

Performance Tips
Notice 5405 : System bandwidth appears FSB limited. Attempt to increase FSB.
Notice 5008 : To change benchmarks, click Options.
Notice 5004 : Synthetic benchmark. May not tally with 'real-life' performance.
Notice 5006 : Only compare the results with ones obtained using the same version!
Warning 5010 : Cannot use Large Memory Pages due to lack of privileges.
Warning 2544 : System/Video shared memory greatly reduces performance. Use external video card.
Tip 2 : Double-click tip or press Enter while a tip is selected for more information about the tip.

The high lighted is what I want to know.

I do not understand your numbers at this point?

Do not do the Cache test, bandwidth is what I want.

 

Sorry dude lost connection for near on an hour.
SiSoftware Sandra

Benchmark Results
Int Buff'd iSSE2 Memory Bandwidth : 2.04GB/s
Float Buff'd iSSE2 Memory Bandwidth : 2.07GB/s
Results Interpretation : Higher index values are better.

Windows Experience Index
Current Chipset/Memory : 4.8
Results Interpretation : Higher index values are better.

Performance vs. Speed
Int Buff'd iSSE2 Memory Bandwidth : 1.14MB/s/MHz
Float Buff'd iSSE2 Memory Bandwidth : 1.16MB/s/MHz
Results Interpretation : Higher index values are better.

Performance vs. Power
Chipset(s)/Memory Power : 13.39W
Int Buff'd iSSE2 Memory Bandwidth : 156.09MB/s/W
Float Buff'd iSSE2 Memory Bandwidth : 158.63MB/s/W
Results Interpretation : Higher index values are better.

Capacity vs Power
Memory Capacity : 306MB/W
Results Interpretation : Higher index values are better.

Int Buff'd iSSE2 Memory Bandwidth
Assignment : 1.98GB/s
Scaling : 1.90GB/s
Addition : 2.12GB/s
Triad : 2.17GB/s
Data Item Size : 16bytes
Buffering Used : Yes
Offset Displacement Used : Yes
Bandwidth Efficiency : 11.90%

Float Buff'd iSSE2 Memory Bandwidth
Assignment : 2.00GB/s
Scaling : 2.05GB/s
Addition : 2.11GB/s
Triad : 2.13GB/s
Data Item Size : 16bytes
Buffering Used : Yes
Offset Displacement Used : Yes
Bandwidth Efficiency : 12.09%

Performance Test Status
Run ID : Intel Mobile PM965/GM965/GL960 Express Processor to DRAM Controller; 2x 2GB DDR2 SO-DIMM PC2-14600 (5-5-5-15 3-20-5-3)
Platform Compliance : x86
Total Memory : 3GB
Memory Used by Test : 256.03MB
NUMA Support : No
SMP (Multi-Processor) Benchmark : No
Total Test Threads : 1
Multi-Core Test : Yes
SMT (Multi-Threaded) Benchmark : No
Processor Affinity : P0C0T0
System Timer : 14.32MHz
Page Size : 4kB
Use Large Memory Pages : No

Features
SSE Technology : Yes
SSE2 Technology : Yes
SSE3 Technology : Yes
Supplemental SSE3 Technology : Yes
SSE4.1 Technology : No
SSE4.2 Technology : No
AVX - Advanced Vector eXtensions : No
FMA - Fused Multiply Add eXtensions : No
SSE4A Technology : No
SSE5 Technology : No
HTT - Hyper-Threading Technology : No

Chipset
Model : HP Mobile PM965/GM965/GL960 Express Processor to DRAM Controller
Revision : A13
Front Side Bus Speed : 4x 549MHz (2.20GHz)
Width : 64-bit
Maximum Bus Bandwidth : 17.16GB/s

Logical/Chipset Memory Banks
Bank 0 : 1GB DDR2 SO-DIMM 5-5-5-15 3-20-5-3 1T
Bank 1 : 1GB DDR2 SO-DIMM 5-5-5-15 3-20-5-3 1T
Bank 4 : 1GB DDR2 SO-DIMM 5-5-5-15 3-20-5-3 1T
Bank 5 : 1GB DDR2 SO-DIMM 5-5-5-15 3-20-5-3 1T
Channels : 2
Memory Bus Speed : 2x 915MHz (1.83GHz)
Multiplier : 5/3x
Width : 64-bit
FMA - Fast Memory Access Support : No
Memory Controller in Processor : No
Maximum Memory Bus Bandwidth : 28.59GB/s

Memory Module(s)
Memory Module : Super Talent MALABS 2GB DDR2 SO-DIMM PC2-5300U DDR2-666 5.0-5-5-15 3-20-5-3
Memory Module : Super Talent MALABS 2GB DDR2 SO-DIMM PC2-5300U DDR2-666 5.0-5-5-15 3-20-5-3

Performance Tips
Notice 5008 : To change benchmarks, click Options.
Notice 5004 : Synthetic benchmark. May not tally with 'real-life' performance.
Notice 5006 : Only compare the results with ones obtained using the same version!
Notice 5405 : System bandwidth appears FSB limited. Attempt to increase FSB.
Warning 5400 : Low bandwidth efficiency (advanced). Check memory timings and settings.
Warning 5010 : Cannot use Large Memory Pages due to lack of privileges.
Tip 2 : Double-click tip or press Enter while a tip is selected for more information about the tip.

 

HerEsY JPZ had it correct when he suggested your power saving options were set to max battery. I ran the benchmark with mine set that way and got the same bandwidth as you, 2GB vs 4GB.

I do not see RAM "failing" in this way more of an on and off.

I control my power options with Notebook Hardware Control, link. Install and then second tab, CPU speed set at Max power or Dynamic switching (dynamic switching is the best). Your system should get back to where it should be.

Why this happened I have no clue but am certain that will solve your problem. There are other ways to do the same thing but this is the one I use and know.

I can see the benchmark sees both DIMMs and is running in dual channel. Your RAM is fine.

 

With battery @ 62% charged I got the benchmark below, with the plugged in power I got a bit better, 4.07GB/s & 3.01GB/s respectively. These numbers came about from just installing the software and not changing anything.
Because I’m getting an error message when ever I try to change the CPU speed settings in Processor speed management (Error on changing the CPU speed) to either the Max. Performance or Dynamic switching, but it allows me to change to Max. battery from Battery optimized, this error message occurs in both fields plugged in and battery drop down lists. I checked and unchecked the Only NHC is allowed to change the power plane box to see if that made a difference with no result.
I still feel the laptop is not running as it was before the problem started for 2 reasons, it’s not as responsive both physically and visionally and I’m getting a little click from the speakers whenever I open or close a program or I get a dialog box opening or closing.

I’m going to let it go into sleep mode and run down the battery (while writing this bit it visionally it started looking as before the problem started, battery @ 33%) to see if that will change anything.

Int Buff'd iSSE2 Memory Bandwidth : 3.87GB/s
Float Buff'd iSSE2 Memory Bandwidth : 2.53GB/s
Results Interpretation : Higher index values are better.
Channels : 2
Memory Bus Speed : 2x 333MHz (666MHz)
Multiplier : 5/3x
Width : 64-bit
FMA - Fast Memory Access Support : No
Memory Controller in Processor : No
Maximum Memory Bus Bandwidth : 10.41GB/s

If I don’t hear from you again, thanks you help was much appreciated.

 

When you get 4GB/s bandwidth you are fine. You could still have a problem just not memory?

I am confused but understand the application issues. Wish others could enter more info for you.

Looks like RAM not the issue?

 

No ram was not the issue, but no doubt removing the ram caused the issue!

I can see by the shear amount of posts you have, that you knowledge base far surpasses what I know about computers. Saying that, I don’t think looking at the ram as the possible problem as jumping to conclusions.
The only reason I run a stress test was to prove to someone there was a problem with my playing environment, not because I believed there was a problem with my GPU.
BTW After running the battery flat, on reboot with power I got a large crackle from the speakers and the clicking issue has gone.

Cheers and thanks again for all your and everyone’s help, it was invaluable!

 

And I also have a question regarding a RAM upgrade. Did not want to create a new thread to prevent traffic

I currently got a Acer 9523 with 2*1Gb DDR PC2-5300 and Intel T5600. From what I have understood, upgrading this to 2+1 will not result in a net +50% gain but something less. I couldn't exactly undertstood the formula in the sticky.

Does it mean that the net gain will be 2/3 *1Gb ram = 667Mb upgrade?
As I'm running a Vista 32 bit and don't have much money, rather then buying 2*2, I want to get a 1+2 system. Will I really benefit the upgrade?
Because currently, the new game I purchased uses 1.2GB ram alone, thus leaving the computer very busy all the time.

Thanks for all the help!

 

Basically just getting more ram won't help your gaming environment that much if your graphics card isn't up to the job, saying that buy more than you need is the answer many here would give you, because you never know what you'll need tomorrow.
And if this helps your budget any, you have to use scientific tests to tell the difference between brand name ram and good generic ram, in short good generic ram works good enough for the average user.

 

tersagun I think more RAM would be a good idea as very cheap. May not improve game play much. GPU is more critical for games. And unless you where actually running out of memory or hitting page file gameplay likely to be the same.

HerEsy can you install CPU-Z, link. Click the Memory Tab, what is the frequency?

 

Thanks for the replies guys.. My GPU is 7900GS with 512m Ram. So while not the coolest piece, for a Directx 9 game it should me more then enough..

I can "feel" the lack of ram playing the game. Alt-tabbing, for example, takes like 5 secs and then another 15-20 secs to become smooth. Alt tabbing from WoW take 1 secs.. The hard drive is also on the run all the time.

 

tersagun the HDD activity does indicate more RAM might help. What game are you having issues with again?

 

The new Warhammer Online.

 

With no windows open 332.1

 

Well that is what I wanted to see. Is it still acting up on you?

 

No, in fact I'm noticing a better game environment,at least it seem as if I got more speed. I installed the CPU-Z on the other dv9653cl to see if the frequency was identical, and everything is a mirror image of the spec's on this rig.

I spoke with the dude who I borrowed the ram to while the store had his on order, to ask if he could of taken the ram out of this rig while it was in sleep mode and with the battery in because it was strange that it was stuck in battery power saver mode. He said no way, but he did say he had the battery power saver mode selected before he took the ram out of this rig.

Stupid question from a stupid man, could this issue have been resolved by just checking to see what power setting this rig was set to? :cry:

 

Yes that was JPZ's first thought on page one maybe. But Oh well we all know so much more now. Good luck.

 

That hurts big time!

 

Not meant to hurt! Nothing but love at NBR!

Most important thing is now it is working for you!

 

That would be OK if I came here looking for love, lucky the only thing I love nowadays just made me feel inept; kind of like the ex did, it's all starting to feel a bit weird.

What do you buy a laptop to show you care?

 

A bigger flash card?

 

A cooling pad!