• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

I think I've traced the source of the Mushkin Blue NF7 etc... Compatibillity Problems

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

fldrice

Member
Joined
Sep 25, 2003
Location
UCI/Newport Coast
I think I've traced the source of the Mushkin Blue NF7 etc... Compatibillity Problems

Background:

After going through many sets of Mushkin Blues w/ Winbond UTT I am ready to report back my findings... It seems that Fry's/Outpost received two batches of these modules. The first, with the date code 030705GN used Winbond CH UTT ICs exclusively. Out of the second batch, I have consistantly received modules with the date code of 031805VN, and with the info provider by Dounger, we know that these modules were produced with both BH and CH quality Winbond UTT ICs.

Average Memtest86 Results:

030705GN (single modules and pairs)
2.5vdimm
200mhz @ 2-3-3-6 -PASSED-
210mhz @ 2-3-3-6 -PASSED-
2.6vdimm
215mhz @ 2-3-3-6 -PASSED-
220mhz @ 2-3-2-6 -FAILED- (~50% modules passed)
3.2vdimm
200mhz @ 2-2-2-10 -PASSED-
210mhz @ 2-2-2-10 -PASSED-
220mhz @ 2-2-2-10 -FAILED- (~80% modules passed)
230mhz @ 2-2-2-10 -FAILED- (25~33% modules passed)
3.3vdimm
230mhz @ 2-2-2-10 -FAILED- (~60% modules passed)
3.4vdimm
220mhz @ 2-2-2-10 -PASSED-
230mhz @ 2-2-2-10 -FAILED- (~80% modules passed)
240mhz @ 2-2-2-10 ~60% pass rate
250mhz @ 2-2-2-10 ~40% pass rate

Note: I did find 2 modules that only needed 3.3v to reach DDR500 w/ 2-2-2 timings.

031805VN (single modules)
2.5vdimm
200mhz @ 2-3-3-6 -PASSED-
210mhz @ 2-3-3-6 -PASSED-
2.6vdimm
215mhz @ 2-3-3-6 -FAILED- (~30% modules passed)
220mhz @ 2-3-2-6 -FAILED- (~10% modules passed)
3.2vdimm
200mhz @ 2-2-2-10 -PASSED-
210mhz @ 2-2-2-10 -PASSED-
220mhz @ 2-2-2-10 -FAILED- (~70% modules passed)
230mhz @ 2-2-2-10 -FAILED- (~25% modules passed)
3.3vdimm
230mhz @ 2-2-2-10 -FAILED- (~50% modules passed)
3.4vdimm
220mhz @ 2-2-2-10 -PASSED-
230mhz @ 2-2-2-10 -FAILED- (~75% modules passed)
240mhz @ 2-2-2-10 ~45% pass rate
250mhz @ 2-2-2-10 ~40% pass rate

031805VN (Pairs)
2.5vdimm
200mhz @ 2-3-3-6 -PASSED-
210mhz @ 2-3-3-6 -PASSED-
2.6vdimm
215mhz @ 2-3-3-6 -FAILED- (~30% modules passed)
220mhz @ 2-3-2-6 -FAILED- (~10% modules passed)
2.8vdimm
200mhz @ 2-3-3-6 -FAILED-
210mhz @ 2-3-3-6 -FAILED-
220mhz @ 2-3-2-6 -FAILED-

:confused: This can't be right, so I popped the VNs into my brother's NF7 just to make sure. The results were even worst then before. 100% failure rate with both paired and single module configurations.

Conclusion:
Upon further inspection, the GNs have ICs that have a high resemblance to my the CH-5 ICs on my LVL Ones. This should be expected, as we know these are CH quality ICs. However, the ICs on the VN are different. While the VN's ICs have the two staples on their sides like all Winbond ICs, these are much thicker. This physical difference has lead me to conclude that the 031805VN modules were built using the BH quality ICs. Remembering that the dual-packs of Blues were all built using BH quality ICs, and that the RMA rate for these packs was unusually high, my results began to make a lot more sense. I believe that Mushkin received a Bad/Strange batch of BH UTT. These modules on the BH quality IC pass testing in single module configurations, and thus are able to pass factory screening. However, when one more modules are used they began to error. The nforce2 memory controller, on the other hand, is not as forgiving as the A64's or the factory testing equipment. Consequently, memory errors occur right away, as experienced by many nf7 users. Therefore, I do not believe it is the BP PCB that is causing this incompatibility issue, but I think the problem lies within the actual IC.

Dounger, feel free to comment as much as you like. From my conversations with you, you are aware that the Blues had some strange reported incompatiblity issues. If my finding are wrong please correct me and report back with your findings.

EDIT: Just realized the paired VN results @ 2.5v and 2.6v are off. They are now corrected
 
Last edited:
Interesting. Looks like you tested a bunch of modules- how many?

Thanks,
CJ
 
I remember some time ago Duonger stated that the blues had problems with the NF7 and I believe the rmas sent to people used Mosel chips...... I haven't had much success using UTT and my NF7..... :-/
 
Back