SuperMicro X10SL7-F

[Sir.Robin]

Looks like most of you with this board are using Xeons.
What kind of Heatsink/Fan are you using?

I'm using a Noctua tower cooler. 120mm fan. Higly recommended :)

 

[Z300M]

Looks like most of you with this board are using Xeons.
What kind of Heatsink/Fan are you using?

I'm using the heatsink and fan that came with the (boxed) CPU -- work fine.

 

[indy]

The Intel cooler works absolutely fine and quiet for the cpu.
I am not as happy regarding the sas-controller though - it gets hot to the touch idling with 1 disk attached.
Honestly I would stay away from tower coolers since they direct even less airflow towards it than the stock cooler.
I have a top-blow-cooler in the mail (Noctua NH-L12), I will post the results.

 

[DJABE]

Tower coolers are designed with a different approach to solve thermal silicon issues... it does not matter that they direct no airflow to the CPU socket! They're keeping cold heat pipes who are in a direct contact with the CPU surface / die...

 

[Z300M]

Nah... didn't get any sencible answer or explanation. Even told about other guys in the forum. Just bollocks answer. Sorry sorry etc.
They even claimed first that theyve NEVER had 8GB sticks for it. Ffs... Kingston: You've really lost a bucket of stars in my book. :(

I'll sell them and get something else i guess.... but not yet. When i'm ready for 4x8GB.

Maybe that's best for me too: keep the ones I have until I need to go to 32GB (if ever).

 

[cyberjock]

To be honest, I'm somewhat in shock at this whole Kingston RAM versus Supermicro thing. And when I say shocked I mean shocked at the comments here by people that apparently haven't seen this before. This is fairly common practice and isn't too shocking in my book.

I can't tell you how many times I've seen HCLs updated and see stuff removed that was previously there. And when you call and ask they try to deny it all. This is nothing new.

Last time I saw this a RAID controller had a HCL. A friend bought 12 drives that were on the HCL along with the controller. Sure enough, over the course of 2 months the drives would randomly drop from the array forcing rebuilds. Twice he had to recover from backups. He started talking to the hard drive and RAID controller manufacturer to get answers. Each one pointed the finger at the other. His drives matched the HCL perfectly.. model, firmware version, everything. Over the course of 6 months they made a couple of revisions to the HCL, and the last revision removed his drives from the list. He called to ask for more info and no anwer was provided.

 

[crumbz]

Looks like most of you with this board are using Xeons.
What kind of Heatsink/Fan are you using?

Stock Intel fan.

 

[crumbz]

To be honest, I'm somewhat in shock at this whole Kingston RAM versus Supermicro thing. And when I say shocked I mean shocked at the comments here by people that apparently haven't seen this before. This is fairly common practice and isn't too shocking in my book.

Even if this happened before, as you claim, people can still be disappointed when it happens again - and to them.

I think it's good to discuss this, as it will hopefully deter someone from buying Supermicro/Kingston products, or at least they will know the limitations of the products beforehand.

 

[Sir.Robin]

To be honest, I'm somewhat in shock at this whole Kingston RAM versus Supermicro thing. And when I say shocked I mean shocked at the comments here by people that apparently haven't seen this before. This is fairly common practice and isn't too shocking in my book.

I have seen this before. I have also seen other companies take a dump, but not hide from it.
Intel for one. Theyve crapped many a times, but they at least inform about it.

Cisco has several product (switches) out there with faulty memory chips, leading to a dead box in some surcomstances.
What they do? They put out information on the subject.
Who is affected and why and what to do.

Shit happens, but when it does we need to know why and what so we can take appropriate measures.

Can i trust my Kingston memory as of now?
Since i have nothing from Kingston other than that they chose to remove it from their list, i would have to say no.

And yes, i for one is a bit surprised because Kingston is pretty large. And still acts like a smaller turd company in this case.

I am dissapointed (actually i'm a bit pissed off). And i will probably never buy Kingston again, because theyve just shown they cannot be trusted.

 

[Z300M]

I have seen this before. I have also seen other companies take a dump, but not hide from it.
Intel for one. Theyve crapped many a times, but they at least inform about it.

Cisco has several product (switches) out there with faulty memory chips, leading to a dead box in some surcomstances.
What they do? They put out information on the subject.
Who is affected and why and what to do.

Shit happens, but when it does we need to know why and what so we can take appropriate measures.

Can i trust my Kingston memory as of now?
Since i have nothing from Kingston other than that they chose to remove it from their list, i would have to say no.

I am taking refuge in what the Kingston Tech Support told me in a chat session (of which I saved the transcript): that the ECC errors reported were only with four 8GB modules. And someone reported here (don't recall whether it was this thread or another one) that he has two X10SL7-F boards running with four KVR16E11/8EF modules with no problems.

Edit: That's four modules on each motherboard.

Is this a problem of some board components and some RAM modules being only just within tolerance and not close enough to the design parameters to work together?

And yes, i for one is a bit surprised because Kingston is pretty large. And still acts like a smaller turd company in this case.

I am dissapointed (actually i'm a bit pissed off). And i will probably never buy Kingston again, because theyve just shown they cannot be trusted.

Agreed.

 

[cyberjock]

I'm not saying it doesn't suck. But, if you are seriously going to stop buying a product because they change their HQL list without giving you details, you should stop buying from the following companies immediately:

Kingston
Samsung
Western Digital
Seagate
Highpoint Technology
LSI
Hitachi
Toshiba
Dell
HP

I think that's all the ones I've dealt with personally...

 

[crumbz]

If you want the manufacturers to change their behavior, a good place to start is to stop buying their wares.

The fact that many others do it also, doesn't make it any more right.

 

[cyberjock]

If you want the manufacturers to change their behavior, a good place to start is to stop buying their wares.

The fact that many others do it also, doesn't make it any more right.

No, I'm just saying that choosing not to buy a particular brand because of these behaviors, there's a lot more brands you'd better avoid. And I can pretty much assure you that my list is not all-inclusive either. My guess would be that anyone in IT, if they followed their moral values(which I'm not arguing is incorrect) would probably be unable to build or buy a computer because of the number of computing companies that would be excluded.

I applaud the behavior, as that's your option and responsibility to do. I agree that companies that do sneaky things deserve to be punished. I'm just saying that the problem is widespread in the industry. Also, there's no clear smoking gun as to whether the problem is from Supermicro's side or Kingston's.

Think about this: Was this poor engineering on Supermicro's part? Whas this poor engineering in Kingston's part? Did they both engage in poor engineering behaviors leading to this problem? Who changed their design and why(the why only being for curiosity)? Was this that testing wasn't thorough enough from the start? Who is responsible for that testing? Who should be responsible when a 3rd party changes their design? If they may be legally responsible for NOT updating the HCL matrix, would you have preferred that? If they do not want to discuss the issues because they don't want to be legally responsible, do you think the laws should be changed? If so, why and how?

Right now, as an outsider with no interest in either company "saving face" and without me owning any of the hardware we are discussing, I am:

1. Glad I don't own the hardware.
2. Be asking for refunds if I bought hardware from HCLs that contained now-incompatible hardware.
3. Asking questions just as many here are asking.
4. Want answers to the questions in #3.
5. Be wanting to hold the company responsible. (Be that legal ramifications if significant money is on the line, or simply avoiding that brand in the future).

For all we know, there may be a manufacturing reason that has forced them to redesign their DIMMs and the new DIMMs don't function the same. Remember, these businesses have to abide by many laws. Many companies had significant problems when non-lead based solder was outlawed. Many companies didn't survive the transition. This *could* be a case where neither company is directly responsible for the problem. We just don't know.

And I'd urge you to get these answers before you start punishing a company for what appears to be wrong-doing. They may not have had a choice or had any control of the situation.

 

[DJABE]

I have seen this before. I have also seen other companies take a dump, but not hide from it.
Intel for one. Theyve crapped many a times, but they at least inform about it.

Cisco has several product (switches) out there with faulty memory chips, leading to a dead box in some surcomstances.
What they do? They put out information on the subject.
Who is affected and why and what to do.

Shit happens, but when it does we need to know why and what so we can take appropriate measures.

Can i trust my Kingston memory as of now?
Since i have nothing from Kingston other than that they chose to remove it from their list, i would have to say no.

And yes, i for one is a bit surprised because Kingston is pretty large. And still acts like a smaller turd company in this case.

I am dissapointed (actually i'm a bit pissed off). And i will probably never buy Kingston again, because theyve just shown they cannot be trusted.

Well, that's why I don't like Intel since they screwed me with VT support on my brand new notebook back in 2009 (C2D Penryn core). At first they stated on their official ARK and other papers this series of CPU have VT support enabled. But after a couple of months, after I (and many other folks) have bought notebooks from various vendors which we believed has VT, Intel simply erased a VT support from ARK and other pages. Even some people claimed to sue them, I'm not sure what happened eventually...

Same thing with ECC and MB here. From my experience, I have never took much care with HCL's (when building Desktops) and never ever had any issue with compatibility.

What about Supermicro MBD-X10SLM-F-O and Kingston ValueRAM 2x8 kit KVR1333D3E9SK2/16G compatibility?

 

[jgreco]

As a datapoint here, the ASRock Avoton C2750 board here fails spectacularly on 4x Kingston 1600 modules.

Kingston has never been on the Supermicro HCL AFAICT, so I am hesitant to point fingers at Supermicro.

 

[Z300M]

As a datapoint here, the ASRock Avoton C2750 board here fails spectacularly on 4x Kingston 1600 modules.

Kingston has never been on the Supermicro HCL AFAICT, so I am hesitant to point fingers at Supermicro.

No Kingston RAM was on the Supermicro list for the X10SL7-F either, but I'm almost certain that Kingston was claiming that the RAM they were recommending had been certified compatible by some outside body.

 

[jgreco]

I am reasonably certain it and the 1333's had been listed on the compat table for the X10's.

The KVR16E11/8 is still listed for X9S.

 

[jgreco]

For all we know, there may be a manufacturing reason that has forced them to redesign their DIMMs and the new DIMMs don't function the same. Remember, these businesses have to abide by many laws. Many companies had significant problems when non-lead based solder was outlawed. Many companies didn't survive the transition. This *could* be a case where neither company is directly responsible for the problem. We just don't know.

And I'd urge you to get these answers before you start punishing a company for what appears to be wrong-doing. They may not have had a choice or had any control of the situation.

In the early 2000's, when the BX was still popular, Kingston was selling double sided 256MB parts. Then one day they became single sided, and the BX boards would only see 128MB. There was no part number difference and if you shipped them a defective dual sided module you had a very significant chance of getting a single in return.

Stuff happens. If you're DIY'ing stuff, you have to be prepared for some possible troubles.

 

[panz]

As a datapoint here, the ASRock Avoton C2750 board here fails spectacularly on 4x Kingston 1600 modules.

Kingston has never been on the Supermicro HCL AFAICT, so I am hesitant to point fingers at Supermicro.

So, I'm an unlucky owner of 32GB of Kingston RAM, just purchased for my system? (see signature for details) :(

 

[Z300M]

So, I'm an unlucky owner of 32GB of Kingston RAM, just purchased for my system? (see signature for details) :(

Not necessarily: check for yourself, but at least for the KVR16E11/8EF (the "Server Premier" line) many X9-series Supermicro motherboards were still listed when I checked a few days ago. Search for the module number on the Kingston site, then click the link for compatible motherboards, etc.