Anything else you noticed on this firmware that's a disadvantage over the older firmware? I know about the pci-e being downgraded to 2.0 already.
It's strange that sleep works fine in that bios but is then broken again with the latest 2016 bios ;(
Intel s2600cp Dual Xeon 2670 3D / VFX Rendering Node
- Thread starter jdartnet
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
It's strange that sleep works fine in that bios but is then broken again with the latest 2016 bios ;(
Yes, I noticed that as well!
In my case, a speedy pci-e was not a priority and it was key I had working sleep. I hope it isn't a case where we'll have to make a choice of one over the other
Also, out of curiosity, have you benchmarked the pci-e to see if it's downgraded? Forgive me for asking, but I wonder if it's been proven.
In my case, a speedy pci-e was not a priority and it was key I had working sleep. I hope it isn't a case where we'll have to make a choice of one over the other
Also, out of curiosity, have you benchmarked the pci-e to see if it's downgraded? Forgive me for asking, but I wonder if it's been proven.
you can run gpu-z and see what it says. i am going to run gpu-z before i upgrade the firmware later tonite. So other then pci-e downgrade, you didn't notice anything else that you don't like compared to the older firmware?
the GT 640 card is a pci-e 3.0 card so that confirmed the downgrade to pci-e 2.0.
I currently have a quadro fx 3800 in there right now, which is a pci-e 2.0 card. so I will have to remember to put in my gtx 950 in to test this out. even the quadro 6000 is itself a pci-e 2.0 card.
Here's the good news: the scaling effect between 2.0 and 3.0 and x16 and x8 is neglible:
AMD Radeon Fury X PCI-Express Scaling
I currently have a quadro fx 3800 in there right now, which is a pci-e 2.0 card. so I will have to remember to put in my gtx 950 in to test this out. even the quadro 6000 is itself a pci-e 2.0 card.
Here's the good news: the scaling effect between 2.0 and 3.0 and x16 and x8 is neglible:
AMD Radeon Fury X PCI-Express Scaling
sleep didn't work properly at all ;(... it would try to enter it and then almost right away try to wake up by itself but failing to do it gracefully and the whole thing just started beeping. I upgraded to the latest firmware and still same problem. going to have to do a bios recovery to downgrade firmware. I got several earlier ones that I will try out first from what I saw in the readme file.....
Btw, I am running windows 10. don't know if the sleep in windows 10 any different from win7. shouldn't be...
before the update, powercfg -a did not show s3, but after the updates it does show s3.
also before the firmware update, my gtx 950 show pcie2.0 and same after update, so I guess it doesn't make any difference in that either..
Btw, I am running windows 10. don't know if the sleep in windows 10 any different from win7. shouldn't be...
before the update, powercfg -a did not show s3, but after the updates it does show s3.
also before the firmware update, my gtx 950 show pcie2.0 and same after update, so I guess it doesn't make any difference in that either..
Seems to me that the issue on your end may be on the Windows side. If S3 is showing up, that's a sign that it's enabled and working from the bios. In my case, prior to updating and disabling the onboard video, my S3 wouldn't even show up.
Windows 10 has too many unstable variables currently going on, and I'd suggest either testing with an older version of Windows (if available), or a good old linux distribution live usb. If it sleeps well in linux, at least you know your bios is setup properly, and then move on to troubleshooting Windows 10.
Windows 10 has too many unstable variables currently going on, and I'd suggest either testing with an older version of Windows (if available), or a good old linux distribution live usb. If it sleeps well in linux, at least you know your bios is setup properly, and then move on to troubleshooting Windows 10.
I guess I could test it out with win 7. the dell t5500 and hp z800 worked fine with win10.
there's at least 1 bios that reportedly say something about fixing s3 resume problem in the readme. so I want to try that one too. my usb 3.0 card is here today. Bought a sound card for it last nite. So hardware wise, everything almost here.
there's at least 1 bios that reportedly say something about fixing s3 resume problem in the readme. so I want to try that one too. my usb 3.0 card is here today. Bought a sound card for it last nite. So hardware wise, everything almost here.
looks like there's additional issues as well . I plugged in the new usb3.0 card and windows 10 is not loading it at all. the driver is not loaded correctly. I tested on a dell inspiron 780 and the card works fine in windows 10 on there....
installed windows 7. I installed the chipset driver and got no exclamation mark in device manager with all drivers installed usb 3.0 card works also. but then ssame problem with sleep. it would attempt to go into sleep and then it would immediately by itself try to come out of the sleep only to crash and gives a bsod.
At this point I am open to just about any suggestion. don't know if it's anything inside the bios setting?
At this point I am open to just about any suggestion. don't know if it's anything inside the bios setting?
Usb 3 doesn't work. even though driver installed and looks good in device, it won't detect any usb drive connected to the one on the card.
Interestingly, an adaptec pci-e sas controller card in that same slot would work showing hard drive and copying files from the hard drive. Wonder what's up with the MB hating the usb card....
I am almost to the point now where I want to just rma everything back. I thought this was a good experience but this is a big waste of time. I almost would have rather spend $150 more to get the Lenovo d30 and just add cpu and rams and go.
There's an amber system status light that keeps blinking. is yrs doing the same? I tried to do the frusdr update but after putting in all the info, at the end, I got unknown error.
Interestingly, an adaptec pci-e sas controller card in that same slot would work showing hard drive and copying files from the hard drive. Wonder what's up with the MB hating the usb card....
I am almost to the point now where I want to just rma everything back. I thought this was a good experience but this is a big waste of time. I almost would have rather spend $150 more to get the Lenovo d30 and just add cpu and rams and go.
There's an amber system status light that keeps blinking. is yrs doing the same? I tried to do the frusdr update but after putting in all the info, at the end, I got unknown error.
Last edited:
Still can't get sleep to work. it would comes out of sleep by itself almost immediately and hard crash.
I have downgraded bios all the way down back to 01.02.0003. i even want to go back to 01.01 but couldn't find it. with the 01.02, no S3 sleep state was avail in powercfg. i then upgraded incrementally to the next bios, then next, then next. at 01.08.0003 now. do have S3 in powercfg now but still no proper sleep without crash. at least the blinking amber light in the back is now gone.
Also got a working usb 3.0 card finally. Seems like the ones with the renesas chipset works.
Got asus xonar dgx sound card which works great.
I know an adaptec sas card works.
I have downgraded bios all the way down back to 01.02.0003. i even want to go back to 01.01 but couldn't find it. with the 01.02, no S3 sleep state was avail in powercfg. i then upgraded incrementally to the next bios, then next, then next. at 01.08.0003 now. do have S3 in powercfg now but still no proper sleep without crash. at least the blinking amber light in the back is now gone.
Also got a working usb 3.0 card finally. Seems like the ones with the renesas chipset works.
Got asus xonar dgx sound card which works great.
I know an adaptec sas card works.
So it looks like sleep mode worked all along and i didn't know it. It takes mine about 1 min to go into sleep mode. It initially spin down for just a sec then immediately spin up again like it's come out of sleep by itself. Then only after about 1 min does everything gets spin down. is this how yours entered into sleep?
On all my other computers, when sleep is activated the computer enters in almost immediately without all this initial spin down and up and then finally down again. This is so weird. I don't have the hibernation enabled so i don't see why take so long.
On all my other computers, when sleep is activated the computer enters in almost immediately without all this initial spin down and up and then finally down again. This is so weird. I don't have the hibernation enabled so i don't see why take so long.
just want to check back with you regarding the sleep thing. How long does it take for yours to go into sleep? does it immediately power down everything or takes a while to enter it like mine? to wake it up, do you just need to press a key or mouse button or do you have to press the power button like mine? thanks.
I've now been playing around with this on a new s2600cp2j build this past week (a game box using a P4000 chassis, single e5-2670,32GB RAM and GTX 1080 with Win10pro ). It consistently takes ~4 seconds to power down to sleep, and ~40 seconds to 2 minutes to awaken.
only 4 sec to power down to sleep? that's normal. so i don't know why it takes like over a min for me.
But why does it take so long for yours to wake? are you able to wake it with a keypress or mouse press? what does it do initially when you try to wake it? what bios ver are you running?
...I"m using the currently available BIOS (2.06.0002). I boot from 1 SSD (Win10 pro), and then I have 12 2.5" SAS drives (connected to a 24 port SAS expander, whcih daisy chains from an 8 port SAS RAID card). So on a clean boot, it takes ~40 seconds just to spin up the drives and init the controller before the POST completes.
I have to awaken via the chassis power button; it does not wake upon USB keyboard interaction.
In sleep my system is currently pulling 8 watts:
I have to awaken via the chassis power button; it does not wake upon USB keyboard interaction.
In sleep my system is currently pulling 8 watts:
Code:
george@ubuntuvm1:~$ ipmiutil sensor -i 001e -N nas4-ipmi -U root -P xxxxxxxx
ipmiutil ver 2.97
isensor: version 2.97
idx = 0x1e
-- BMC version 1.25, IPMI version 2.0
_ID_ SDR_Type_xx ET Own Typ S_Num Sens_Description Hex & Interp Reading
001e SDR Full 01 01 20 a 0b snum 54 PS1 Power In = 02 OK 8.00 Watts
ipmiutil sensor, completed successfullyNice. 8W is better than mine. I have 1 SSD connected and that's it.
S3 on my lenovo d30 uses 3w. i am wondering if there's still something in S2600 that causes it to use more power.
So when you select sleep, it almost immediately goes into sleep in 4 sec? What does it do that take so long to wake up from sleep?
Yeah, so weird that both our systems can't be wake with keyboard press even though this is enabled for wake up.
Also, i noticed that under the usb section in the bios for mine it says, 2 keyboard, 2 mice detected. So weird bc I only have 1 mouse and keyboard connected. so where/what is this 2nd phantom set? Does yours say the same?
Hello. I want to buy this kit (deal from Natex: dual 2670 - 128GB) to build a rendering workstation, but rendering in GPU mode (Octane, basically). The xeons will help my work with 3D (viewport and simulations + after effects) and I like the 6 (5+1) PCIe slots.
I already have a gtx 1080 + 1070 + 980ti and probably will upgrade all to 1080 as add a 4th one.
You guys seem to know how this kit works, so do you think it would be a good idea to buy to GPU rendering?
Cheers!
I already have a gtx 1080 + 1070 + 980ti and probably will upgrade all to 1080 as add a 4th one.
You guys seem to know how this kit works, so do you think it would be a good idea to buy to GPU rendering?
Cheers!