Hey guys, I have a toshiba satellite a130 st1312. I installed some new ram and am not noticeing any differnece. let me break down what I have, then I'll tell you guys what is going on.
According to CPUZ, I have a the following components::
under CPU tab:::
Processor::: intel mobile core 2 duo T7200
Core speed::: goes from 997mhz at 6x multiplier up to 2000MHz at 12 x multiplier. They advertised as a 2.0 GHz cores so that seems okay?
Bus speed::: 166.3 MHz
Rated FSB::: 665.0 MHz
Cache::
L1 data is 2x32Kbytes 8 way
L1 Inst. is 2x32Kbytes 8 way
Level 2 is 4096Kbytes 16 way
Under Caches tab:::
L1 D-Cache :: size 32 Kbytes x 2, 8 way set associative, 64 byte line size.
L1 l-Cache:: same as L1 D-Cache.
L2 Cache:: 4096 Kbytes, 16 way set associative , 64 byte line size.
Under Mainboard tab:::
Motherboard::: Toshiba model- IAKAA 1.00
chipset::: intel i945PM, rev. 03
southbridge::: intel 82801GHM (ICH7-M/U)
Bios::: Toshiba version V1.10, date 12/08/2006
Graphic Interface::: Version PCI- Express, link width x16, max. supported x16.
Under Memory tab:::
General:: DDR2 dual channel, 4096 Mbytes
Dram Frequency:: 332.5 MHz
FSB:RAM:: 1:2
timings are 4-4-4-12 and a (20 clocks bank cycle time)
Under SPD:::
DDR2 PC2-5300 (333Mhz) kingston
part number:: KHX5300s2ll/2G it is the hyper x model.
Okay guys, take an advil after all that and now I can tell you what is going on with my computer.
I was running 2x1gb sticks of ram in dual channel which were also 667Mhz sticks. I ran pcwizard and got a 23,000 benchmark score. I decided to upgrade to 2x2gb of kingston hyper x 667MHz ram and run it in dual channel. Well, I did a few benchmarks on pcwizard 09 and I am getting very low scores for what I have. 22,940. I took out one of my 2gb sticks and ran 1 of the 2gb sticks alone and got around the same score as with both of them in???? What could be the problem here. is the 1:2 fsb:dram thing screwing me up here?? I know the dual channel pc2-5300 ram is allowing me up to around 10GB/s transfer speeds, is my fsb bottlenecking my ram? Please let me know what I can do to fix this situation guys. I just spent a lot of money and it has hindered me, not helped the slightest bit!
Please help me guys, you guys are geniuses and I'm just smart, I'm an engineering student who doesn't know much more than what I've read over the past week or so about computers.
-
Don't worry about it. There is nothing wrong with your computer. I just installed pcwizard and ran the tests. All the test measures is the latency of your memory, including the L1, L2, and in some cases L3 cache on the processor itself. That first benchmark only tests your memory's latency, nothing else. Maybe the new sticks have slightly higher latency or that is just a random difference. Both are equally likely since it is never going to get the same scores twice due to other things running on the computer. Run the second memory test, the global one, and look at the numbers that are more than 4MB (your L2 cache) and look at the bandwidth scores between 1 stick and 2. There will be a big difference there between single and dual channel. The amount of memory on your computer will not affect any of these tests. So please don't put much into pcwizard's memory benchmarks.
-
moral hazard Notebook Nobel Laureate
can you try some other benchmarking software?
I don't think it has anything to do with the fsb:dram ratio. -
Okay, I didn't know that it was checking things which didn't have much to do with my memory that I just installed. But, wouldn't it be optimal to have my fsb match my ram. It seems as if I'm bottlenecked by the fsb of 667. Since I have dual channel ram pc2-5300 , I figured I'd be able to do about 5333x2(dual channel) gb/s because 667mhz x 128bits(dual channel)/8bits..... and my fsb will only allow 667x64/8(8bits in a byte)= 5336gb/s??? I'm not sure if I'm doing this correctly. so it seems as if my 1:2 ratio is correct. Would I be able to squeeze out some more performance if I could at least raise up the front size bus??? What should my transfer rates be in GB/s on benchmarking considering my set up? What do you guys think I could do to optimize what I have? Thanks for all of your help and time cause time is valuable and help is what I need.
-
moral hazard Notebook Nobel Laureate
If you increase the FSB then your CPU and ram will be overclocked. it will result in a big performence increase.
You can use software called setfsb for that.
If setfsb doesn't work (or you don't know how to use it), let me know and I will do my best to help you out.
I overclock my FSB a lot. recently I got it from 200mhz to 270mhz. -
Bandwidth 4 MB : 16120.90 Marks
Bandwidth 8 MB : 5846.40 Marks
Bandwidth 16 MB : 5373.88 Marks
Bandwidth 32 MB : 5383.48 Marks
Bandwidth 64 MB : 5413.19 Marks
Bandwidth 128 MB : 5392.88 Marks
Bandwidth 256 MB : 5390.90 Marks
Bandwidth 512 MB : 5405.75 Marks
With 1 stick:
Bandwidth 4 MB : 17964.22 Marks
Bandwidth 8 MB : 5992.27 Marks
Bandwidth 16 MB : 4686.25 Marks
Bandwidth 32 MB : 4765.17 Marks
Bandwidth 64 MB : 5335.13 Marks
Bandwidth 128 MB : 5582.57 Marks
Bandwidth 256 MB : 5713.84 Marks
Bandwidth 512 MB : 5542.52 Marks
It actually did better with 1 stick than with 2? I already tested both sticks, they both give the same score when I put them in alone, I've tested the memory slots, they both work. What could be happening? -
And now that I took one stick out, should'nt the fsb match the ram because it is in a single setup now??? Somebody please tell me where my reasoning is wrong becasue cpuz still shows a 1:2 ratio. Thanks all
-
moral hazard Notebook Nobel Laureate
when you have a dual channel ram setup you are not chaging your ratio.
yes you double your ram bandwidth, but you're not doubeling your ram frequency. -
But please, oh please just stop it. Your computer is fine and there is nothing to worry about. If you computer tells you that it is running dual channel, it is running dual channel. If it tells you that it has 4GB, it has 4GB.
Memory problems
Discussion in 'Hardware Components and Aftermarket Upgrades' started by 400awhp, Sep 10, 2009.