1 |
On Thursday 29 Jan 2015 22:13:28 Volker Armin Hemmann wrote: |
2 |
> Am 28.01.2015 um 00:28 schrieb walt: |
3 |
> > Yesterday I installed 4GB more of RAM in this machine for a total of 8GB, |
4 |
> > and the machine soon began random segfaulting and even a kernel crash or |
5 |
> > two, so obviously I suspected the new RAM was faulty. |
6 |
> > |
7 |
> > I let memtest86+ run overnight and it found zero memory errors. Today I |
8 |
> > exchanged the new RAM anyway and got a different brand this time, and |
9 |
> > that fixed the problem. |
10 |
> > |
11 |
> > My question is why didn't memtest86+ find any errors? Could it be that |
12 |
> > the first RAM I bought was actually okay but this machine didn't like it |
13 |
> > for some reason? Both were DDR3/1333MHz, just from different |
14 |
> > manufacturers. |
15 |
> |
16 |
> Since this was not mentioned yet: |
17 |
> |
18 |
> Maybe because the ram was not faulty at all. |
19 |
> |
20 |
> Maybe it really operated in the range of allowed tolerances - and those |
21 |
> were never crossed with memtest as a very light system load. |
22 |
> |
23 |
> But with an OS booted, the CPU, graphics solution, harddisks all sucking |
24 |
> power like mad, your mainboard or PSU might not be able to deliver as |
25 |
> stable currents as the specifications demand. Some memory is more |
26 |
> tolerant than other. |
27 |
|
28 |
Yes, I've witnessed this too after adding 2 new memory modules of a different |
29 |
size to the originals and from a different manufacturer, in a box with a |
30 |
suspect PSU. Memtest+86 was not erroring out, but the system was crashing |
31 |
when put under pressure. Typically I would get errors when more than the size |
32 |
of the old memory started being used. This got worse over time, as the PSU |
33 |
components were ageing. Eventually I replaced a capacitor in the PSU and the |
34 |
memory problems disappeared. |
35 |
|
36 |
It has been already mentioned, but it is worth noting that some BIOS/MoBos are |
37 |
more sensitive to different brands of memory. In those cases I found that |
38 |
using the same make and size modules resolves the problems. |
39 |
-- |
40 |
Regards, |
41 |
Mick |