While the RAM is not named the "list" from Intel (No mushkin RAM is), it fully complies with all the specifications laid out in the compatibility list; Mushkin assures me that the RAM is fully XMP profile compliant for the DX58SO.
The bad RAM was tried at all standard speeds (1066, 1333, 1600), no post.
The RAM was set according to Mushkin's specifications for this board and cpu (as per a conversation with Mushkin support moderators, according to Mushkin it works)
The RAM was memtested when it was initially received at the recommded settings/speed and passed multiple passes without error.
The RAM sticks that continue to function all passed memtest, multiple passes at Mushkin recommended settings, individually and together, in different slots. (I spent multiple days doing this)
Mushkin called me today and offered to swap my Redline Kit for Ridgebacks. They are a little slower, 6-8-6-24. The Redlines were 6-7-6-19, but, I have heard good things about the Ridgebacks, so I will not moan too much about it. I am curious if Mushkin recognises that there is a problem with the Redline DDR3 and the Intel DX58SO and is attempting to limit the RMAs. They told me that they are discontinuing Redlines and are not able to get those chips anymore: that was my 3rd RMA for that RAM. Obviously if there is a problem again, I will have to chalk it up to the board entirely. Hopefully, this will be the last time I have to RMA RAM and the quirks of this motherboard end.
This would otherwise be a great board, except for a few things that are strange/con about it, including: electrical noise when QPI power management enabled; lack of memory voltage adjustment in the 1.63-1.65v range (SERIOUSLY BAD); strange placement of sata ports in the way of the upper PCI-E X16 with long video card; peculiar RAM channel handicap on the 4th slot affecting all slots when active; 4th slot too close to HSF for Coolermaster V8 and other large units; no USB 3.0; only 1X pci 2.0; not all Japanese capacitors; tendency to be slow at POST.... Enough already!
!!!!electrical noise when QPI power management enabled !!!!
I dont think our cpu supports this feature, this was added to the bios with the new core i7 800 line , and is not intended for the older i7 920 line last time i looked it up on the net. It is also disabled in optimal bios settings so i reccomend disabling it. Last time i enabled it i could not post.
I can't argue that the CPU's written specs say 800/1066; but referring to the specifications sheet, as far as I am concerned, is a recourse of last resort as there are thousands of users utilizing DDR3 1333+ & above without this problem. Before I accept the necessity of strictly adhering to spec sheets an explanation as to why the DX58SO even supports 1333+ (as advertised on the package) and the consequences of using 1333+ and/or some kind of semi-official notice citing a general event in the neighborhood of malfunction would be required to convince me. Short of that, I would accept wide agreement of experienced users that the dx58so with i7-920 CAN NOT run in the configuration I have propoed. and that the problem cited would result if done. Curiously, Mushkin is aware of my configuration and has not issued a warning that the RAM would malfunction. In fact, I have not read any RAM manufacturer or reseller literature stating that the i7-920 on the DX58SO will malfunction if DDR3 1333+ is used at its rated speed.
Given that this board and CPU have been on the market since 2008, is seems highly unlikely that that this is a matter of simple malfunction or "doesn't work with your RAM".
Thanks for the reply!
Since you are soliciting ideas/opinions as to why you have to keep on changing your ram (because the board destroys the ram?), I have presented information (not my own but from the maker of the board and the processor) that might be the possible reason. The decision whether the information cited applies/is acceptable/applicable to your situation is solely yours.
Added: Let me expound on what I have cited earlier. Intel says about their DX58SO board that:
a.) the board supports the I7 processor in an LGA1366 socket
b.) the board has four 240 pin DDR3 SDRAM DIMM sockets
c.) the board support DDR3 1333, 1066, and 800 DIMMs
Intel also says about their I7-920 processor that:
a.) It can operate up to 24 GB of memory (dependent on the type of memory)
b.) the memory type is DDR3 800/1066
With regards to specification I always interpret this as a guarantee from the maker that their product will operate satisfactorily within the provided specification. If the user choose to operate it beyond the given specs, then there is no guarantee from the maker that it will operate satisfactorily. In other words, the risk is yours only for whatever consequences that may occur.
As to your statement that the specs will only be considered as a last resort, technically and logically, I simply disagree. But this is beyond the issue at hand.
Message was edited by: Boyet
To make an analogy,
Often, road speed limits are set at 55 MPH, and this is the "supported" speed. Yet, my gas pedal pedal doesn't restrict me from accelerating perfectly well. While there might be an increased risk of damage in the event of error, to site the "rules-of-the-road" state that speed "55 MPH is the speed limit" isn't helpful if I inquire as to the reason my engine is breaking timing belts. To say, "It's because you are doing 60 MPH!" does little to solve my problem when everyone else's car is passing me. I am sorry if that is not clear.
Incidentally, I was not asking for a guarantee, I was merely asking for help to get it running. I am quite certain that if I were to "underclock" the said RAM, it would have still malfunctioned; due, it seems to either a problem with the RAM itself or with this motherboard
I am merely trying to help by pointing the possibility of the the wrong use of ram with reason why by providing the spec of the processor and explaining that spec given by the manufacturer about their product are assured by them to work at the indicated values. Of course you can always do whatever you want with their product when you buy them.
I understand that you are not asking for a guarantee. The guarantee that I'm talking about only pertain to my explanation with regards to the product spec and nothing to do with granting you the guarantee (I have no right to give you one). The help that I have provided is the information of the possibility of using an incorrect ram and the support why it is so. As of the moment, I can't find any other reasons/suggestions to help you with your problem. However, if you feel that the info provided does not apply to your problem you can always disregard it (forget about it) and go on with whatever solution that you find aggreeable. It is your problem, and the suggestions provided are given as is, and the decision whether to accept or reject it is solely yours.
Thanks for the reply.
Regrettably, there is no enthusiast forum on the Intel msg. boards where I can post regarding what ultimately is an enthusiast question. Strictly speaking, Intel's specifications are outside of my setup scope; however, given the number of other similar setups I have seen, it strikes me that this issue is a case of bad luck regarding either the board and/or the memory. Hopefully someone who has successfully troubleshot a similar setup will have some advice, or, alternatively, an Intel representative will suggest an RMA of the board.
The memory manufacturer is confident that their product will run in my rig as outlined, and their msg. board is no stranger to this setup. However, the fact that I RMA-ed the RAM 3 times makes me suspicious that the motherboard is affecting a product that is supposed to run under the conditions I have outlined. The CPU may not be guaranteed to run with RAM at this speed, but the RAM itself is - at it is the RAM that needs recurrent replacing. It is possible that I had lousy RAM thrice, but, as you can imagine, three instances of three different RAM kits' failure makes me suspicious. I have no reason to suspect that the CPU is being stressed or is otherwise malfunctioning as the problem disappears (for a time) after the RAM has been replaced - had the CPU or motherboard been damaged, the problem sh/would have persisted despite changing the RAM
To boot, I have attempted to run the RAM at the Intel i7 920 spec. speeds: it did not remedy the problem.
Given that this performance RAM has been paid for with no possibility of refund, I am resistant to replacing the current solution with some DD3-800/1066 and would much rather get it working as specified. Hopefully this makes the issue clear.