Today we have another sample of Crucial’s awesome DDR-2 memory up on the review slab. This time it comes in the PC-5300 flavor. Crucial has been very good with the review samples lately, which tells me they want the public to be informed about the memory they offer you. This in itself is a good indicator of a good company. Isn’t it nice to see a PC hardware company that’s interested in your choice rather than how much money they can squeeze out of you? Memory Specs

• -->Part Number: BL2KIT12864AA663 -->
• -->Module Size: 2GB kit (1GBx2) -->
• -->Package: Ballistix 240-pin DIMM -->
• -->Feature: DDR2 PC2-5300 -->
• -->Configuration: 128Meg x 64 -->
• -->DIMM Type: UNBUFFERED -->
• -->Error Checking: NON-ECC -->
• -->Speed: DDR2-667 -->
• -->Voltage: 2.2V -->
• -->Memory Timings: 3-3-3-12 -->
• -->Specs: DDR2 PC2-5300 • 3-3-3-12 • UNBUFFERED • NON-ECC • DDR2-667 • 2.2V • 128Meg x 64-->

Awesome numbers there in that list eehh? Notice the low timings of these modules? These 3-3-3-12 specs aren’t your average DDR-2 timings, lower latency means the information is being processed faster inside of the memory IC modules. In the world of memory timings lower is better. I wish I could tell you what kind of memory IC modules is on this memory. But, I cannot because I can’t’ find out for myself about them. I called crucial up to ask their PR rep what they were and he said due to the memory business being so competitive he couldn’t share that information. So the IC chips on this memory are secret. System Specs

• Asus P5WD2-E Premium
• Intel Pentium 4 630 Retail
• Stock Intel HSF
• Crucial Balistics pc-5300 2x1gig
• ATI X800-XL (Cat 6.2)
• WD 80g IDE 8mb
• NEC 3500 optical drive
• Win XP Pro (SP 1&2)

Memory Pics Crucial shipped me the memory kit directly from their warehouse and it came in the standard box that they ship all of their memory in. Each stick is enclosed in an anti-static sleeve to protect the memory sticks from any electrical charges that may develop through shipping or static buildup. Next we see another angle of the memory kit pre-installed for your viewing pleasure. The last pic shows a close-up of the memory labels and its ID codes. Here it is installed in the motherboard. Those gold heatspreaders really stand out don’t they? Performance This is what you have all been waiting for, the performance. Let’s see how these stick really do. We will start out with the factory timings and speed first. These sticks are factory clocked at 667Mhz as indicated by the mother board bios. Below we see a CPU-z shot of the stock speed and timings. Everything looks perfect, including the motherboard memory divider that’s needed to run this memory at 667Mhz with a stock 800fsb P4. The testing programs I will be using are Sandra v.1069, Everest 2.50, ScienceMark 2.0. These programs create rigorous memory intensive tests in order to measure memory performance in mega-bytes per second. For stock I decided to incorporate all three tests in one graph due to the very small overclock I was able to achieve at stock. That was a paltry 15 more Mhz than 200fsb with stock timings and that yielded only 300mbs on an average in the testing programs. So the stock overclock is actually not even worth mentioning. The scores below aren’t to shabby for some stock action eehh? Overclock Performance (All following FSB speeds obtained by using 14x multiplier) This is where we find out how far and fast this memory can really go! In the previous section we discovered that the tight timings of this memory really kept the speed to a minimum yet beyond spec for this SPD type. SPD (factory speed & timing) is what matters to most computer users. Overclocking on the other hand appeals to less yet is growing in popularity among computer users. The less popular segment of computer users which is what is commonly known as the enthusiast will definitely appreciate the next set of findings on this awesome memory. Next we will take this memory out of its rated SPD and unlock its hidden potential. I will start out with the most common middle of the road timings of 4-4-4-12 and then progress to the looser but higher clocked timings of 5-5-5-15 which is what we are seeing in the newer faster DDR2 kits that are out. The best stable FSB I could get with 4-4-4-12 was 255 on the CPU. This speed allowed me use of all of the memory dividers with the exception of the 1:2. That divider would give me black screens with reboots and was unusable. But all of the other dividers were stable. The chart below shows the very nice performance boost with the higher FSB and lower timings. Now we will slack up the timings and bump up the FSB. Loosening the timings will enable us to take the memory higher in speed. The memory will be less efficient with looser timings but the higher clock speed will makeup for the loss and then some. :>) For this next graph I have selected a lower speed of 250 FSB to get a good comparison with the previous tighter timings. You can see how the tighter more efficient timings really make a difference here. Even though this graph is 5fsb slower than the previous one you get a good idea. We can now move on to the next increase in speed. This next round of tests will be done at 260fsb. The scores from this speed match the closest to the lower FSB scores of the tighter timed 4-4-4-12 testing done earlier. Some scores are just below while others surpass those of the previous tighter timed test. Lastly we have our highest stable speed at the 5 series of memory timings. 270 FSB was the most I could achieve from his set of memory which is excellent for the rated SPD that this kit is marketed at. I tried for DDR-1000 (pc-8500) and all I could get was a suicide shot of CPU-z before it crashed and rebooted. But don’t let that place doubt in your mind about this memory. Just about any thing under that speed was stable through any other memory divider. It took more elevations in the FSB but our scores here beat out the highest score of the tighter timed memory speed of our 4 series of memory timing test. Gaming Performance Our next round of testing will be done with the ever popular Future Mark 2005 & 2006 benchmarking suites. I have kept the same timing and speed groups to show how the various bandwidths translate into 3D applications. Each benchmark will be run at the 3, 4, & 5, series memory timings as well as the respective elevated FSB as in the previous testing methods. Conclusion It looks like Crucial has another awesome kit of memory here folks. Is it as fast as their top notch pc-6400? No, its not but, almost! As you can see in the gaming graph overclocking didn’t’ help the score “that much” but it did improve. My suggestion would be to save your hard earned money and buy this memory instead of the pc-6400 series. It does almost as much as its big brother yet is considerably cheaper. I liked this ram very much and have yet to buy or receive any thing bad from Crucial. I liked the flexibility of the memory timings and the killer overclock that I was able to get from it as well. If you are in the market for some cheaper 2gig set of ram then this is the set for you. The Good: Cheaper than pc-6400, very flexible timings, great overclock, 2 gigs density The Bad: none-:>) 9 out of 10