Xeon Owners Club

[Edwired]

Sorry about the last message as I was annoyed with the my setup thinking it was damaged I had one last try to see if I can fix it but the good news the board is back up and running. Had funky issues like the motherboard bios threw a new CPU was installed along with chassis intrusion error even so there's no jumper on the board for that to enable or disable. So cleaned the socket and lightly brushed the pins in case it might be out of alignment or dust. So as for the evga gtx 1060 6gb sc there's a possibility the memory or the gpu die could be degraded have to test which one

 

[1freedude]

The joys and misery of x58. It's a fickle beast that doesn't accept change very well.

You have nothing to apologize for.

Glad you got it figured out.

 

[Edwired]

The joys and misery of x58. It's a fickle beast that doesn't accept change very well.

You have nothing to apologize for.

Glad you got it figured out.

Thanks for the support bud I was worried at first thinking that the motherboard gone down the drain after spending €110 but got lucky this time thank god. Now I might figured something out about the evga gtx 1060 6gb sc when it in stock state with the power limit at 100% the card keeps crashing regardless even the temps hovering 50c to 54c unigine heaven with all the Tess level maxed out with graphic in high and anti align x8 so I did the same test over and over with core clock underclocked and memory underclocked same result with crashing. Then I already replaced the thermal pad that covered the vrm it was skinny strip not enough to cover it so I decided to sacrifice my broken Asus p6x58d-e as it have two thick thermal pads maybe 2mm in height but squishable to fit in between the base plate and heatsink to give it more surface to spread heat. Same result crashing. So I'm thinking what causing it to crash then I said to myself it must be something to do with the power limit maybe it broken in the vbios so I turned it down abit to see what will happen but will have to do more testing.

Possible case the power supply could be degrading but have to wait and see what comes first

 

[Sora]

?!? Got a screen shot? That can't work, at least as far as I know.

It's called Flex mode, the lower region of memory is run in triple channel and the upper region that is only present on one channel is acccessed as if it were single channel.

in a motherboard with 4 slots like this, you would want 2 channels with single 8GB dimms and the last channel that has two slots populated by 2x4GB to achieve consistent triple channel performance.

With your i7 working fine but the Xeon glitching, you may have a stability issue.

I checked the CPU support chart on ASUS.com;

RAMPAGE IV EXTREME - Support

www.asus.com

The 1680V2 is not in that list. I wonder if perhaps for a reason? As a general rule if one model of a CPU family is supported, they all are. Perhaps there's an exception going on with that model?

BTW, welcome to TPU. Hope we can help you find solution.

1680v2's are pulled from mac workstations, they were never sold retail so they aren't in the qualified list of processors, as long as they aren't faulty and you don't try to push the imc too hard (definitely don't try 1n CR) they work on the RIVE no problemo.

Update on my 1680v2/R4E: Tried every official BIOS from 4502 to 4901 and every one of them failed under the same workload of MSFS 2020 downloading and installing game files. Now some lasted longer than others, but the all eventually resulted in the same crash.

Current plan is to try a different seller for the 1680v2 and if it continues I will probably throw int the towel for at least a couple of months and just use my 3930K again.

Thanks guys for the help you provided.

this is indicative of IMC weakness, but can also happen when certain network booster tools are installed, Killer Network Manager Suite" or NAM, or similar.

https://www.reddit.com/r/intel/comments/mpzqgv
this motherboard using x79 or c602 chipset? suddenly it is not x79 but c602 maybe that's why it worked RDIMM we need to confirm this
but if it is x79 like in rampage then there is a chance it will work here too

Asrock X79 Extreme6 work RDIMM 8 sticks of 16gb each total 128gb​

chipset is irrelevant, ECC being available or not is a matter of the ecc pins being wired up on the mobo, as far as things go, they are on most of the high end X79 motherboards.

X79 and/or rampage IV OC'ing thread for those of us still left....

I have an Asus Rampage IV with i7-4960X processor. On the Ram slot's there are 4 piece of GSkill Trident 2666 Ram's. When I trigger, the mainboard automatically adjusts processor speed to 4000 MHz. Also increases BCLK to 125 MHz. 2666 / 2400 = 1. 1108333 if I multiply this result with original...

www.techpowerup.com

The RIVE and RIVG both support ECC memory,

 

[Edwired]

So far the evga gtx 1060 6gb sc seems to be holding up with two 2mm thermal pads one on the vrm and the other on top of the base plate connecting to the heatpipe but still need to order replacement pads for the memory chips thinking 1.5mm should cover that including another connecting to the heat pipe in the same layout as the memory so the card can run cooler for longer and will try to bench it to see if it solves the artifact issue. Pretty sure it a overheating issue on one of the memory chips,

Almost forgot to mention ordered a set replacement fans for the asus gtx 1050 ti oc expedition esport since the fin I repaired broke off and disappeared so I treat it for a set of these so it can match the fan shroud which is black and red

 

[Edwired]

Thinking the evga gtx 1060 6gb sc is on it last leg since it crashing repeatedly from a cold and hot boot same with unigine heaven. Strangely enough if I push up by the pcie power plug the artifacts changes colours or patterns during the crash same with touching the resistors/cap nearest to the pcie slot. But it will boot if I hold up the pcb. It pretty much have cracked solders joints. Might well try a hot air gun method as I have amtech flux as I have nothing to lose but before I do that I will try the digital multimeter (dmm) to see if anything shorting or have strange readings

 

[Edwired]

Found no short or strange readings on the evga gtx 1060 6gb sc so that ruled out. My guess was right the solder joints were broken between the gpu die to the six memory chips couldn't tell which one. So I reflowed all the chips with amtech flux and a heatgun. Started with the low heat to preheat the pcb front and back with the flux flowing under the chips for a few minutes in circular pattern then crank up the heatgun to high heat on per chip for 10 to 20 seconds repeat this for a few more minutes till the flux gets bubbles popping out once that was done you should see flux burning off from the chips. Let it cool down for a half an hour. Then clean off as much flux off the pcie gold contacts and the pcb and repaste the gpu and replaced the thermal pads.

Being testing it for an hour on unigine heaven no sign of artifacts appearing with the pre-set settings temperature on the gpu die peaked at 50c core clock flat lined at 1999.5mhz.

So it a possibility it fixed

 

[Edwired]

New fans install these runs a lot faster topped out about 3400rpm and stays there compare to the old one where it ramped up to 3100rpm then falls off to about 2800rpm. There's abit of a different shade of red but it kinda makes it pops on the shroud

 

[Ruru]

Paid the customs of the X5675 @MachineLearning sent me from USA, so I'll try it out when it finally arrives, should arrive next week

I was surprised what a boost from HD 7970 to R9 290X was, never had those at the same time.

 

[CheapMeat]

Downgraded my 2 x dual CPU storage boxes to 2 x E5-2637 v4's (each box 8C/16T 3.5GHz base); they have a combo of 2 x 10C/20T and 2 x 12C/24T. But sidegraded up my dual CPU compute box to 2 x E5-2699 v4 (so 44C/88T). The two storage boxes help "feed" the main compute box.

 

[Edwired]

So far the reflowed Evga Gtx 1060 6gb sc haven't crashed or artifacted yet so it looking good as it being tested in variation of games in high settings and pcsx2. Side note Asus p6x58d-e version 1.01 seems like a better board more stable compare to my old broken board but seems to run hotter on the northbridge even so most of the voltages are locked down to the first setting after auto. Wondering if the frequency for the northbridge is too high causing higher temperature as thermal paste is done on it unless the northbridge runs hot without overclocking

 

[Ruru]

So far the reflowed Evga Gtx 1060 6gb sc haven't crashed or artifacted yet so it looking good as it being tested in variation of games in high settings and pcsx2. Side note Asus p6x58d-e version 1.01 seems like a better board more stable compare to my old broken board but seems to run hotter on the northbridge even so most of the voltages are locked down to the first setting after auto. Wondering if the frequency for the northbridge is too high causing higher temperature as thermal paste is done on it unless the northbridge runs hot without overclocking

I have the same board on my HTPC, just one hella good X58 board.

I have the additional radial fan blowing to the NB/VRM cooling as my setup is watercooled so no much airflow for the heatsinks.

 

[Edwired]

Cool to know it just Noctua NH-C12P SE14 with 140mm fan should cool the northbridge unless the fan speed I set in fan control software is too low as not enough airflow is what I am thinking since I'm trying to make the pc abit quieter or come up with a few modifications for better cooing methods

 

[Ruru]

Cool to know it just Noctua NH-C12P SE14 with 140mm fan should cool the northbridge unless the fan speed I set in fan control software is too low as not enough airflow is what I am thinking since I'm trying to make the pc abit quieter or come up with a few modifications for better cooing methods

Change the thermal paste if it gets too hot

 

[Edwired]

Change the thermal paste if it gets too hot

Done the thermal paste about 3 weeks ago as it Arctic mx-4 I used. Suppose I look around for a small fan have to find a size that fits the heatsink without fouling the cpu heatsink as space is an issue. Or a second option to drop the gpu to the next pcie slot below the primary slot to give it more room for a bigger fan to cool the northbridge.

But if I was cleaver enough to modify the broken board heatsink file down the fins till it a flat surface and drill four holes and screw on a bigger heatsink to the existing base it would cool the northbridge even further

 

[Ruru]

Done the thermal paste about 3 weeks ago as it Arctic mx-4 I used. Suppose I look around for a small fan have to find a size that fits the heatsink without fouling the cpu heatsink as space is an issue. Or a second option to drop the gpu to the next pcie slot below the primary slot to give it more room for a bigger fan to cool the northbridge

Weird. Though X58 DOES get hot, so put some airflow there.

 

[Zyll Goliat]

So far the reflowed Evga Gtx 1060 6gb sc haven't crashed or artifacted yet so it looking good as it being tested in variation of games in high settings and pcsx2. Side note Asus p6x58d-e version 1.01 seems like a better board more stable compare to my old broken board but seems to run hotter on the northbridge even so most of the voltages are locked down to the first setting after auto. Wondering if the frequency for the northbridge is too high causing higher temperature as thermal paste is done on it unless the northbridge runs hot without overclocking

Nice.....at least it's working my GTX 1070 TI definitely died from static electricity....it was checked in service but no luck.....well it's not end of the world I guess I still have R9 Fury and that card still doing just fine in most of the games especially with that Nimez modded drivers........

 

[Edwired]

Nice.....at least it's working my GTX 1070 TI definitely died from static electricity....it was checked in service but no luck.....well it's not end of the world I guess I still have R9 Fury and that card still doing just fine in most of the games especially with that Nimez modded drivers........

Were you on the carpet when it died?

 

[Zyll Goliat]

Were you on the carpet when it died?

No...'tho i did wear a socks and no shoes.....well that's the first time that happened to me and I was always like nah static electricity even if it happens not going to kill any hardware that easily....well I guess I learn my lesson now.......

 

[Edwired]

No...'tho i did wear a socks and no shoes.....well that's the first time that happened to me and I was always like nah static electricity even if it happens not going to kill any hardware that easily....well I guess I learn my lesson now.......

Unless you were doing the Micheal Jackson's Moonwalk to build up static electricity and dapped the graphic card

 

[johnspack]

Waiting to bring this monster home from a buddies... but I just inherited an ibm system 3650 m3, with dual x5680 cpus, and 8x8gb ram. Has 4x savvio 15.3k sas drives of 146gbs each.
Also has a dedicated gpu added that has hdmi and dp outs. I have no idea what they must of used this thing for....

 

[Edwired]

So far evga gtx 1060 6gb sc holding up testing with pcsx2 1.7.0 nightly build having the resolution at 1440 (2k) made a few games I used to play look stunning on the Samsung curved monitor compare to the old tube tv back then. While my system is mildly over clocked with the cpu at 3.73ghz non turbo @ 1.208v llc enabled, ram 1417mhz @ 1.62v the dmm tells me the ram is close to 1.64v to 1.65v, the uncore at 3200mhz, qpi/dram voltage locked to 1.25v. Will have to adjust the frequency for dram to bring it back up to the rated speed which is 1600mhz the three rams rated for 1.65v and other three rams is rated 1.50v only difference between both set is the trfc timing the rest match up. As for a side note between both rams three have dual rank channel and the other three is single rank channel not too sure what sort of impact it have in performance due to mixed channels in fully populated state

 

[RadeonProVega]

Has anyone manage to use a GPU that requires a 8pin and 6pin for their workstation xeon desktops?
using a 5810 and planning to buy a 7810 and was wondering how do use a gpu that requires a 8pin and 6pin?
for my 5810 i have two 6pins.

 

[Edwired]

Might be tricky unless you changed the power supply to meet the requirment but I wouldn't trust the adaptors that covert 6 pin to 8 pin

 

[Sora]

the adapters are fine, the additional pins are ground.