Kingston ValueRAM -- worse than any other national brand?

[larry moe 'n curly]

Is Kingston ValueRAM the worst national brand cheapo memory sold?

I've had 4-5 failures in 12-13 modules tested with Gold Memory or
MemTest86 (If either diagnostic detected a fault, I considered the
module bad). I'm counting only tests done with no overclocking and
where the module failed in more than one computer (unless one machine
really underclocked the memory, i.e., PC3200 module at PC2100 speed).
This is worse than the other brands (PNY, K-Byte, Mushkin) and generics
(Fry's) I've tried.

 

[kony]

Is Kingston ValueRAM the worst national brand cheapo memory sold?

I wouldn't call it any worse than generics like Centon or
Patriot or (a bunch of other generic labels that escape me
at the moment).

Often it's cas3 though, which I try to avoid buying even if
any particular system ends up needing cas2.5 but _running_
it at cas3.

I've had 4-5 failures in 12-13 modules tested with Gold Memory or
MemTest86 (If either diagnostic detected a fault, I considered the
module bad). I'm counting only tests done with no overclocking and
where the module failed in more than one computer (unless one machine
really underclocked the memory, i.e., PC3200 module at PC2100 speed).
This is worse than the other brands (PNY, K-Byte, Mushkin) and generics
(Fry's) I've tried.

I"d try manually setting the timings, and using CPU-Z to
compare timings used with those modules contrasted with
timings used by other viable modules.

One potential I see is that if a module is spec'd as cas3 @
PC3200, but is downclocked while still using SPD timings, is
that it's SPD table may drop the CAS down to 2.5 or 2 at
lower mem bus speeds, so using the auto/SPD with
underclocked memory bus doesn't aways resolve such issues
but rather the timings need be manually set.

Beyond that, I dont' know and can only suggest avoiding KVR.
I only have (IIRC) 3 of their modules in systems around here
somewhere and nothing KVR bought in the last few months-
don't even remember which system's they're in but I believe
older, not 200MHz bus speeds... so all I really recall is
that they did pass memtest86 without recalling the
scenarios.

 

[larry moe 'n curly]

On 19 Jul 2005 00:36:41 -0700, "larry moe 'n curly"
<[email protected]> wrote:

I wouldn't call it any worse than generics like Centon or
Patriot or (a bunch of other generic labels that escape me
at the moment).

Often it's cas3 though, which I try to avoid buying even if
any particular system ends up needing cas2.5 but _running_
it at cas3.

I'd try manually setting the timings, and using CPU-Z to
compare timings used with those modules contrasted with
timings used by other viable modules.

One potential I see is that if a module is spec'd as cas3 @
PC3200, but is downclocked while still using SPD timings, is
that it's SPD table may drop the CAS down to 2.5 or 2 at
lower mem bus speeds, so using the auto/SPD with
underclocked memory bus doesn't aways resolve such issues
but rather the timings need be manually set.

When I tried one PC3200 in a mobo with integrated video, and when I
chose the safe defaults the thing would show 1/2" wide white vertical
stripes that faded in, pixel by pixel, over a couple of seconds, and
the system was frozen.

I had Kingston and K-byte PC2100 modules that would not work reliably,
no matter how the BIOS parameters were set, semi-manually
(Ultra/Turbo/Fast/Normal) or manually, unless I slowed the memory bus
speed from 266 MHz to 200 MHz. K-byte tech support was outright nasty,
while Kingston offered to replace the module but said I was using the
wrong one/right one, depending on who I talked to.

 

[BigJim]

I have and am using corsair value ram with no problems, pc3200

 

[Tim]

Is Kingston ValueRAM the worst national brand cheapo memory sold?

I've had 4-5 failures in 12-13 modules tested with Gold Memory or
MemTest86 (If either diagnostic detected a fault, I considered the
module bad). I'm counting only tests done with no overclocking and
where the module failed in more than one computer (unless one machine
really underclocked the memory, i.e., PC3200 module at PC2100 speed).
This is worse than the other brands (PNY, K-Byte, Mushkin) and generics
(Fry's) I've tried.

I use it with my Intel 865PERL (dual channel, PC3200 memory) and it runs
flawlessly. Bear in mind that Kingston ValueRAM has been tested and
approved by Intel for that particular motherboard.

 

[Ed Cregger]

I use it with my Intel 865PERL (dual channel, PC3200 memory) and it runs
flawlessly. Bear in mind that Kingston ValueRAM has been tested and
approved by Intel for that particular motherboard.

I have used Kingston RAM in several computers and have not had a problem
with it. Is it as fast as other RAM? I don't even know if that is possible,
but it has worked well for me. No, I am not an expert. Just a user.

Ed Cregger

 

[JAD]

all with the same PSU?

 

[sbb78247]

Is Kingston ValueRAM the worst national brand cheapo memory sold?

I've had 4-5 failures in 12-13 modules tested with Gold Memory or
MemTest86 (If either diagnostic detected a fault, I considered the
module bad). I'm counting only tests done with no overclocking and
where the module failed in more than one computer (unless one machine
really underclocked the memory, i.e., PC3200 module at PC2100 speed).
This is worse than the other brands (PNY, K-Byte, Mushkin) and
generics (Fry's) I've tried.

luck of the draw maybe. i have used it in the supposed doomed combination
of the msi k7n delta + value ram and it worked dual channel no problem. in
other words, your mileage may vary.

 

[larry moe 'n curly]

I use it with my Intel 865PERL (dual channel, PC3200 memory) and it runs
flawlessly. Bear in mind that Kingston ValueRAM has been tested and
approved by Intel for that particular motherboard.

Some of my Kingston ValueRAM has been flawless, too, like the PC3200
module that replaced my defective one, but how did you test yours?
This particular bad module passed MemTest86 for 5+ hours but failed
Gold Memory in 10 minutes --several times.

 

[larry moe 'n curly]

all with the same PSU?

One mobo had a 350W Fortron/Sparkle, the other a 300W Delta. How would
the PSU matter when one module was flawed while the other was fine?

 

[larry moe 'n curly]

luck of the draw maybe. i have used it in the supposed doomed combination
of the msi k7n delta + value ram and it worked dual channel no problem. in
other words, your mileage may vary.

I could believe that if it happened nearly as often with other brands
and with modules containing prime chips (i.e., the chip maker's normal
part numbers are printed on the parts), the latter being perfect for
me, so far.

 

[JAD]

voltages that are not constant, and clean will cause memory errors. IMO
anything that has 'value' or 'economy' attached to the name makes me
suspect. I have used it(value ram) few times with no problems but as a rule
I try and steer clear of 'cheap' ram, sorta like buying a cheap PSU.

 

[kony]

voltages that are not constant, and clean will cause memory errors. IMO
anything that has 'value' or 'economy' attached to the name makes me
suspect. I have used it(value ram) few times with no problems but as a rule
I try and steer clear of 'cheap' ram, sorta like buying a cheap PSU.

Generally speaking "value ram" is nothing more than chips
with slower timings. Completely different from other
commodity items (power supply, fans, etc) that may fail due
to the corner-cutting inherant in arriving at lower cost.

 

[Tim]

Some of my Kingston ValueRAM has been flawless, too, like the PC3200
module that replaced my defective one, but how did you test yours?

MemTest86 and Prime95 (blended test), each for 24 hours. This system is the
most stable I ever used so I have no cause to test the memory any further.

 

[kony]

One mobo had a 350W Fortron/Sparkle, the other a 300W Delta. How would
the PSU matter when one module was flawed while the other was fine?

It doesn't matter much at all. Memory being below the 3.3V
minimal threshold of a power supply output, means it's
always post-regulated on the board. Since I too have both
of those PSU, I can vouch for them- they're both better than
a typical 350W Antec. 350W Sparkle isn't built as well but
slightly higher current potential due to bigger transformer.
Even so, either is best suited for a box with CPU on 5V rail
rather than 12V. Come to think of it neither is even ATX12V
so *usually* that wouldn't be an issue except for that one
oddball Biostar board recently discussed.

 

[larry moe 'n curly]

news:[email protected]...

all with the same PSU?

voltages that are not constant, and clean will cause memory
errors.

Isn't that a really generic answer, straight out of an A+ certification
book rather than reality, especially when mobos have their own voltage
regulators for the CPU and DDR RAM? And how do you explain why one
DIMM always worked while the other consistently failed?

 

[Bob Adkins]

I have used Kingston RAM in several computers and have not had a problem
with it. Is it as fast as other RAM? I don't even know if that is possible,
but it has worked well for me. No, I am not an expert. Just a user.

I've used 4-5 sticks, and it worked well for me. No overclocking, mind
you... just generic systems.

Nice to see ya' BTW Ed.

 

[JAD]

Isn't that a really generic answer, straight out of an A+ certification
book rather than reality, especially when mobos have their own voltage
regulators for the CPU and DDR RAM? And how do you explain why one
DIMM always worked while the other consistently failed?

maybe...they don't print that stuff for nothing...although, I have not that
cert. Reality..software memtesters are not reliable, because for one thing
they do not isolate everything in between(controller etc). Although it has
been a rare occurrence that I have a ram memory problem, a few have been
solved by replacing the PSU(if there is not enough or erratic voltages,
regulators won't help in some cases.) AFA One working rather than 4, I would
say the PSU was marginal possibly. If the software test makes me suspect or
the machine is getting squirrelly during ram hungry apps, I would pay the
5'r and have it hard tested, before I start tossing ram out.

Here's a question...do you ask questions about things you know everything
about? I asked a simple question about an obvious scenario. Had the multiple
PSU info been in your first post, I wouldn't have asked. Good luck
determining your answer, stats say it seems to be only your problem. l8tr

l

 

[larry moe 'n curly]

news:[email protected]...

voltages that are not constant, and clean will cause memory
errors.

Here's a question...do you ask questions about things you know
everything about?

That would make as little sense as replying to an unsolicited question
without having a good idea of what the answer is.

 

[jonkx]

Memory is problematic.

In General, memory is somewhat motherboard dependent. There are small
electrical compatibility/interface differences among motherboards as
well as among memory brands and product lines. (It may be that more
expensive memory is tested among a wider variety of motherboard brands.
I am assuming that you have the memory type correct for the
motherboard - with respect to ECC/non-ECC, buffered/registered or
unbuffered/unregistered.)

Most motherboard manufacturers will list on their web site the sticks
that the motherboard was tested with, including the brand of stick or
the brand of the memory chips.

Also, some motherboards function poorly or not at all with all the
memory slots filled; others function even better. Another factor which
varies success is how much memory
is on each stick.

For best results, use those or buy from a dealer who guarantees
compatibility with your motherboard such as Crucial. Use the same
brand and same size of stick in each slot. Another approach, perhaps
less safe, is to use the same motherboard/memory combination as someone
else who has been successful, such as in a magazine project.

Just my 3 cents worth,
Jon