does memory brand really matter?

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View

I'd like to upgrade the memory in my laptop (old Gateway 7330GZ) I was
originally going to just go with Kingston, but it's about 2x the price
of many of the other options for a 1GB DDR.  Is there a significant
difference in quality/performance/reliability or should I go for one of
the cheaper options?

here's what I'm looking at



replace "roosters" with "cox" to reply.
** Posted from **

Re: does memory brand really matter?

Nate Nagel wrote:

Quoted text here. Click to load it

You want memory modules made with prime quality chips, that is, chips
that passed all of the chip manufacturer's tests and are guaranteed to
meet all its published specifications.  These chips will have their
full manufacturer part numbers printed on them, and if you Google such
a number you'll get dozens of pages of technical information about the

Unfortunately most retail memory modules aren't made with prime
quality chips but instead with either UTT -- UnTesTed -- chips, which
weren't fully tested at the factory or didn't pass all tests.  Module
makers buy these and test the chips themselves, or with whole wafers
(can't be fully tested) that the module maker slices and dices
itself.  Unfortunately module makers don't seem to be as strict or
sophisticated as the chip makers are, and many module companies test
with nothing but ordinary PCs.

AFAIK, the only sure way to get retail modules with prime quality
chips on them is by choosing Crucial/Micron modules that do NOT have
heatsinks on them and that are rated to run reliably at standard
voltage (3.3V for SDR, 2.5V for DDR, 1.8V for DDR2).  The ones I've
bought were made with Micron, Elpida, or Samsung chips, and, oddly,
the later modules were even made by Samsung.  OTOH Crucial's Ballistix
DDR2 memory modules, which have been subject of a lot of consumer
complaints, have heatsinks stuck to their chips, their chips aren't
marked with standard Micron part numbers, and they're specified to run
at 2.0V or higher.   Avoid.

Whatever modules you buy, test them overnight with at least two
unrelated memory diagnostics, including MemTest86 ver. 3.x and Gold
Memory ver. 5.x, and don't accept even a single defective bit when the
modules are run at their default SPD settings.

Re: does memory brand really matter?

On Sat, 06 Sep 2008 10:27:36 -0400, Nate Nagel thoughfully wrote:

Quoted text here. Click to load it
Quoted text here. Click to load it

Any of the generic name brands shown should work, actual performance not
a factor.  Just check the memory module specifications against the
motherboards specified requirements.  

If the memory module doesn't work you'll have to add in shipping costs
for exchanging the memory.  

Re: does memory brand really matter?

Quoted text here. Click to load it

Although the "generic" memory should probably be OK...
I have definately had problems with that on the very low end of the

If it's guaranteed to work, then the risk isn't too great however.

By all means avoid the cheap stuff, that cannot be returned!!!

Re: does memory brand really matter?

On Sat, 06 Sep 2008 14:04:15 -0500, philo thoughfully wrote:

Quoted text here. Click to load it

I tried to exclude the cheap stuff by stating those shown in his link.  
However, I've never used A-Data and I've had problems with Crucial,
Kingston and Corsair before with Kingston being the most costly

Re: does memory brand really matter?

On Sat, 06 Sep 2008 10:27:36 -0400, Nate Nagel

Quoted text here. Click to load it

Kingston memory is often no better and sometimes worse than
other brands, there is no justification at all for it to
cost more.

I'd get this one, haven't had any problems with A-Data
memory and it's priced a little better than some because of
the free shipping.  Plus, on the off chance that the memory
doesn't work (similar chance with the Kingston or any
other), having free shipping means you aren't out that cost
when you get a refund.

Site Timeline