Thinkpads Forum

Another data point, for what it's worth. When I got my T60p I immediately put in two 1 gb Corsair "ValueSelects" from Newegg. I've been getting random crashes, perhaps every 3-4 days. Tonight when working with VirtualPC a bit more, I went down several times.

Put my original IBM mem back in, and no problems yet.

So - I may be getting some of that fancy Crucial after all..

joe

After installing new memory, I would strongly suggest running a memory tester (memtest86 or memtestx86) to make sure that the memory is good. Just using the machine for a while will not be a sufficient test to catch some faulty memory.

It's interesting to see that for some of the memory brands the people who are complaining are the ones who actually ran a memory tester. I'm still not sure I'm going to pay the $50 premium for the Crucial memory for my T60p and X60s.

If time is not of the essence (i.e., it's not needed that day), I usually go with Crucial for clients' needs, for reasons others have brought up...that said, I couldn't pull the trigger on $150+ to bump my T60 to 1.5GB...I ended up with the GeIL 1GB PC5300 from Newegg for about $89 shipped (out of stock right now), they have a very strong reputation in the overclocker community so I felt good about the choice...so far Memtest86 running flawlessly.

http://www.newegg.com/Product/Product.a ... 6820144810

Randy

foodle wrote:After installing new memory, I would strongly suggest running a memory tester (memtest86 or memtestx86) to make sure that the memory is good. Just using the machine for a while will not be a sufficient test to catch some faulty memory.

ChiefValue is great! My UPS tracking said "2lbs". When I got the box (a rather big one for a little stick of RAM), it was filled with peanuts. The memory itself was tightly wrapped with a large roll of big air bubbles. Talk about protection and waste of packing material!

So I ran Memtest. It took forever. I was only able to run 4 tests before I gave up. Does that suffice? So far, my 1GB Patriot 667 PC5300 ran into no errors.

No one has mentioned this yet, but the CL rating (it's a timing value) can make a different (whether it is appreciable is certainly arguable). The standard memory (whatever brand that is) is rated CL5, as is most notebook memory. However newegg does stock and sell some CL3, DDR2 PC5300 SO-DIMM sticks. I'm thinking of picking up two of those.

aarong wrote:No one has mentioned this yet, but the CL rating (it's a timing value) can make a different (whether it is appreciable is certainly arguable). The standard memory (whatever brand that is) is rated CL5, as is most notebook memory. However newegg does stock and sell some CL3, DDR2 PC5300 SO-DIMM sticks. I'm thinking of picking up two of those.

Are you sure? I just looked at the site and could only find CAS latency (CL) 5 DDR2 PC5300 SO-DIMMs.

Hmm. CL lag only gets in at the beginning of reading a bulk of memory, as the probability of having appreciable difference (happens when bulk size reading time < 20 clock period) by my estimation is less than 5%, total difference memory-wise would be less than 1%. Note: usually the bottleneck is the harddrive...