Windows 7 on a Netbook 1 gig vs 2gig

A couple weeks ago I compared Windows Xp that comes on the Acer Aspire One D250 to the new Windows 7 to basically find out how they compared to each other in terms of performance. The D250 I have comes with 1gig of ram, so As sort of a part two to that article I thought I’d toss in 2 gigs of ram and re-run all of my tests to see what, if any difference there might be between running 1 gig or 2 gigs. The results of the testing very much surprised me in that 2gigs doesn’t seem to be better at all really, and I’m not sure why that is. Maybe someone else can shed some light on this for me…

Testing:

The ram that comes with the Acer Aspire One D250 is a single 1gig stick of Samsung DDR2 5300 running at 4-4-4-12 timings. I threw a 2gig stick of OCZ Ram with the same specs into the D250, you’d think the performance would get better, but it didn’t really. I did nothing to the system, no changes, nothing at all, I just put the ram in and that was it. I’m sure I could have went in and tweaked some things a bit, but I wanted the results to be just what the average person might see if they decided to toss a new stick of ram in their netbook.

Here’s the screenshots from CPU-Z telling us about the ram, looks the same to me, the only difference is that the Samsung is running at 266.0 MHZ and the OCZ is running at 265.4MHZ, but that’s not going to skew the results.

When running the tests I made sure the wi-fi was turned off to make sure it wasn’t trying to update, and I made sure there wasn’t anything else running in the background either that might affect the testing.

The first test is using CrystalMark 2004R3. This program works well at testing basically all aspects of a computer, in the graph I’ve got the overall results or marks, and I’ve included screen shots of the individual tests as well so you can see the exact results.

Starting with Crystalmark we can see a small difference running the 2gigs of ram over the stock 1gig, it seems better really, but not much though. Where you’d think thinks might be the same they aren’t, and where you think they’d improve, they don’t.

==========

The next test I’ve got in Cinebench 10. If you don’t know what Cinebench is, then here’s a quote from the site:

CINEBENCH is a real-world test suite that assesses your computer’s performance capabilities. MAXON CINEBENCH runs several tests on your computer to measure the performance of the main processor and the graphics card under real-world circumstances.

The test procedure consists of two main components: The first test sequence is dedicated to the computer’s main processor. A 3D scene file is used to render a photoreaslistic image. The scene makes use of various CPU-intensive features such as reflection, ambient occlusion, area lights and procedural shaders. During the first run the benchmark only uses one CPU (or CPU core) to ascertain a reference value. On computers that have multiple CPUs or CPU cores and on those that simulate multiple CPUs (via HyperThreading or similar technolgies), MAXON CINEBENCH will run a second test using all available CPU power.

The second test measures graphics card performance and is run inside the 3D editor window. The project file used can test all graphics cards that support the OpenGL standard. In this scene, only the camera was animated. This scene places medium to low demands on graphics cards and tests the maximum speed with which the scene can be properly displayed.

Overall we see a major drop in performance when running 2gigs of ram, but why? What is going on here exactly is what I’d like to find out.

==========

Super Pi would be next and time to calculate Pi to 1 million places.

Apparently more ram can shave a few seconds off of Pi calculations.. ok

==========

SiSoftware, makers of SiSoft Sandra, recently released service pack 4 for their 2009 Sandra program, which made it more compatible with Windows 7. I like using Sandra as it gives me a good comparison, and it includes several tests all in one program.

The first test is the Cache and Memory Test.

Processor Cache and Memory

Benchmark the processors’ caches and memory access (transfer speed).

Results Interpretation

Cache/Memory Bandwidth (MB/s) – higher results are better, i.e. faster memory bandwidth.

Speed Factor (MB/s) – lower results are better, i.e. less difference between processor cache speed and memory speed.

Combined Index: is a composite figure representing the overall performance rating of the entire Cache-Memory performance in terms of MB/s. The value is the logarithmic average of all the results for the entire address space. (Higher is better, i.e. better performance)

Speed Factor: is a figure representing the speed differential between the CPU’s cache and memory. The value is the ratio of the fastest cache (i.e. L1) bandwidth to the main memory bandwidth. (Lower is better, i.e. the memory is not very much slower than CPU’s cache)

With this test we see a very slight variation between the two sticks of ram, a very slight drop in the actual test but a slight jump in the Speed Factor.

==========

The next test is CPU Arithmetic:

Processor Arithmetic

Benchmarks the ALU and FPU processor units. Shows how your processors handle arithmetic and floating point instructions in comparison to other typical processors.

Results Interpretation

Dhrystone (MIPS) – higher results are better, i.e. better integer performance.

Whetstone (MFLOPS) – higher results are better, i.e. better floating-point performance.

With this test you might think that running two gigs of ram would be beneficial, but it’s the same. Though this is a CPU specific test, so the ram isn’t much of a factor.

==========

Next up is CPU Multi-Media:

Processor Multi-Media

Benchmark the (W)MMX(2), SSE(2/3/4), AVX processor units. Shows how your processors handle multi-media instructions and data in comparison to other typical processors.

Such operations are used by more specialised software, e.g. image manipulation, video decoders/encoders, games.

Results Interpretation

Multi-Media Integer (Pixels/s) – higher results are better, i.e. better integer performance.

Multi-Media Single/Double Float (Pixels/s) – higher results are better, i.e. better floating-point performance.

Here we see a very slight, insignificant really, performance increase when running two gigs of ram. Which is could be attributed to a number of factors, like some sneaky service starting up in the background.

==========

The next test would be the File Systems test:

File Systems

Benchmark mounted file systems (i.e. volumes). Shows how your file systems connected to storage adapters and storage hosts compare to other devices in a typical computer.

This is not the raw disk performance that other benchmarks test – but the speed of the volume itself that depends on many more factors like file system, operating system cache, position on disk, etc. Thus this is the performance you get at the file system level.

Drive Index: is a composite figure representing an overall performance rating based on the average of the read, write, and seek tests, and file and cache size. The Drive Index is intended to represent drive performance under typical use in a PC. A larger number means better performance. The weighting of the results is not equal it represents the distribution of different files sizes as used on these devices (obtained through field research).

With this test, one might think it would be the same considering that it’s the same hard drive, but no, here we see an odd jump in Random Access time, and an actual drop in the Drive Index.

==========

Next up we have the Memory latency test:

Memory Latency

Benchmark the latency (response time) of processors’ caches and memory

The latency of caches is measured in processor clocks (i.e. how many clocks it takes for the data to be ready) as it is dependent on the processor clock speed.

The latency of memory is measured in nanoseconds as it is typically independent on processor clock speed.

Here, with this test, we’re focusing on the ram and I’m not seeing what I expected to see. The two sticks of ram are, far all intents and purposes, the same. So why is there such a big difference between them?

==========

and the next test would be Memory Bandwidth:

Memory Bandwidth

Benchmark the memory bandwidth of your computer.

Results Interpretation

Integer Memory Bandwidth (MB/s) – higher results are better, i.e. faster memory bandwidth.

Float Memory Bandwidth (MB/s) – higher results are better, i.e. faster memory bandwidth

With this test we see very close results. These differences are insignificant really, at least you’d never ‘feel’ the differences in real life, but that’s why we run these tests to find things down at the system or chip level that you can’t notice.

==========

And the final test is the Power Management Test:

Power Management Efficiency

Benchmark the power management efficiency of the processors. Shows how efficient the power management of your processors is in comparison to other typical processors.

The ability of the processors to step-down in frequency and voltage at different workloads is measured. The more a processor steps down in both frequency and voltage the better the score at the specific workload. The test stops when the workload is too great the processor even at 100% efficiency.

The ALU/FPU score is a geometric mean based on the whole range of workloads; thus the power of the processors does matter in obtaining a higher score.

The Power Efficiency score is a geometric mean based on the supported workloads only. Thus the power of the processors does not matter.

Explanations

MP Dhrystone (MIPS) ALU (Arithmetic) Benchmark

MP Whetstone (MFLOPS) FPU (Floating Point) Benchmark

Higher scores are better.

I ran this test without really thinking about it, it really has no bearing on the ram, but since I ran it I tossed it in here. No need to waste right?

Conclusion:

So, you tell me what’s going on here.

Seriously, I’d like to get your opinion on these results.

I don’t want to be insulted thanks, I can get that anywhere. I am a bit smarter than the average bear, and I can figure it out myself, but I want to get a discussion going and maybe just learn something that I didn’t know from you.

I’d like to know what you think is happening here and why. Give me your hypotheses and give me some suggestions to fix it and I’ll take your advice and try it out and let you know the outcome.