Jump to content

Last Wednesday I was thinking that I may need to add extra storage to my Dell Precision T1650 workstation (Don't ask me why it's such an old rig, I found it near a litter bin in a Best Buy last year🤣). The motherboard only has 4 SATA ports. The manual and specifications are here: https://downloads.dell.com/manuals/all-products/esuprt_desktop/esuprt_dell_precision_workstation/precision-t1650_owner%27s%20manual_en-us.pdf

However, there's a PCI-E x16 (wired as x4) (should be 2.0) channel left, so I bought a PCI-E x1 to SATA 3.0x4 converter to connect more disks.

 

To begin with, I would introduce my current setup so that you can get some clues:

Motherboard chipset: Intel C216

Processor: Intel Core i7-3770 (with no K)

Memory: 4x DDR3 PC3-12800E 8GB Pure ECC memory (the label says it's PC3-12800U, but I found it actually ECC supported in memtest86+...)

(an extra word, according to the manual, the chipset only supports 32GB with ECC, and 16GB Non-ECC. so these sticks have to be ECC supported to actually run on the board...)

GPU: Zotac Nvidia 1060 6GB

Storage: before installing the converter, 1x 2TB + 1x 3TB HDDs, 1x 256GB SSD (with both Windows 10 and Ubuntu 19.10 installed). Fourth SATA is for the CD/DVD drive.

Power: Seasonic S12-III 500W Bronze

 

The adapter is this. https://www.amazon.com/gp/product/B07SZDK6CZ/

 

It was Wednesday night and I tried to make things work. And here's what happened in a strange manner:

0. I installed the converter and connected a disk to it. I firstly booted to Ubuntu and found no problems in the first 3 minutes.

1. Then I booted to Windows to see if there was any problems and went into a BSOD after running into the desktop.

2. I looked up the website and found that the converter was in ACHI mode but the booting disks were mistakenly set in RAID mode. Changed it in the BIOS.

3. Restarted to Windows and went to Safe mode (to make the disk loading modes reset to ACHI).

4. Restarted to Windows normally. The converter and the disk is shown properly in the BIOS and the system. 3 minutes after booting Windows went into a BSOD, this time with another error code.

5. Restarted to Windows Safe mode and checked that the disks were already loaded in ACHI mode. 

6. Restarted back to Windows normally, successfully visited the disk appended to the adapter. 3 minutes after booting the BSOD came again, now yet another error code. 

7. This made me think that maybe some mishandling of mine has made some memory sticks to crash. I booted to Ubuntu, tried pressurizing the momory, and it hung after 7 minutes.

8. I think there were definitely problems in the memory. I ran the memtest86+ in the computer and found errors in 6400~MB and 23000~MB. 

9. I removed the first and third memory sticks (thinking they might be broken) and booted back into both systems. They worked well and posted no problems.

 

Today I decided to do some extra experiments to see whether it is the memory issue or not. And something even more strange happened...

0. The converter and two 8GB memory sticks worked well for two consecutive days. I closed the computer and added the sticks back.

1. I booted the system up to the memtest86+ and found the memory errors to be persistent at the same aforementioned address. (Notice that the memory stick in Slot 1 should be the one that I marked "working" in the previous test).

2. I removed all memory sticks and moved the two "broken" memory sticks to Slots 1 and 2. Booted up, memtest86+ showed no errors. (So the memory sticks are actually working. But I am going to change them anyway, I have bought a new set of four 8GB RAMs.) Then I started the computer to Windows and did a filecopy through the adapter. Things are working as usual. 

3. I added a "working" memory disk to Slot 3. Booted up, memtest86+ showed errors at the exact 6400~MB address.

4. I began to think if the motherboard has caught problems. I called a friend and talked about the issue. In the meantime I removed the three sticks and changed with the computer's previous 4x4GB DDR3 PC3-12800U sticks. Booted it up, it passed tests of the diagnostics on board.

5. The friend searched through technical sheets and thought there should be no interference between memory slots and PCI-E ones, but advised me to remove that adapter anyway. I removed the adapter and installed all four 8GB memory sticks. 

6. I conducted a memtest86+ test, only to find the errors recurring at the same address. I thought it was doomed.

7. I bet with the friend that the machine would definitely fall to a BSOD if we boot into Windows. I did that, opened a game that occupied 4GB, and successfully exited it to the desktop, no problems. 

8. Then I booted to Ubuntu with the 32GB memory and all things resumed to normal as if nothing happened to the machine.

 

A conclusion to the bizzare adventure 

16GB RAM + adapter = works; 32GB RAM + no adapter = works; 32GB RAM + adapter = BSOD/hang. Any clues on this?

 

 

P.S. I looked through the Intel website, and it said i7-3770 doesn't accept ECC memory. So how come I ran this little workstation for nearly two months in a row without any problems?

Link to comment
https://linustechtips.com/topic/1179817-im-having-a-peculiar-question/
Share on other sites

Link to post
Share on other sites

Maybe 32GB is too stressful, try a lower clock speed.

 

27 minutes ago, ToniChan said:

P.S. I looked through the Intel website, and it said i7-3770 doesn't accept ECC memory. So how come I ran this little workstation for nearly two months in a row without any problems?

doesn't matter, ECC is just a feature that could be disabled if anything doesnt support it and it will run like normal memory. What cant be wrong is unregistered and registered memory, motherboards can only take 1 type at a time if the CPU supports both (in this case, Core series only supports unregistered DIMMs)

CPU: i7-2600K 4751MHz 1.44V (software) --> 1.47V at the back of the socket Motherboard: Asrock Z77 Extreme4 (BCLK: 103.3MHz) CPU Cooler: Noctua NH-D15 RAM: Adata XPG 2x8GB DDR3 (XMP: 2133MHz 10-11-11-30 CR2, custom: 2203MHz 10-11-10-26 CR1 tRFC:230 tREFI:14000) GPU: Asus GTX 1070 Dual (Super Jetstream vbios, +70(2025-2088MHz)/+400(8.8Gbps)) SSD: Samsung 840 Pro 256GB (main boot drive), Transcend SSD370 128GB PSU: Seasonic X-660 80+ Gold Case: Antec P110 Silent, 5 intakes 1 exhaust Monitor: AOC G2460PF 1080p 144Hz (150Hz max w/ DP, 121Hz max w/ HDMI) TN panel Keyboard: Logitech G610 Orion (Cherry MX Blue) with SteelSeries Apex M260 keycaps Mouse: BenQ Zowie FK1

 

Model: HP Omen 17 17-an110ca CPU: i7-8750H (0.125V core & cache, 50mV SA undervolt) GPU: GTX 1060 6GB Mobile (+80/+450, 1650MHz~1750MHz 0.78V~0.85V) RAM: 8+8GB DDR4-2400 18-17-17-39 2T Storage: HP EX920 1TB PCIe x4 M.2 SSD + Crucial MX500 1TB 2.5" SATA SSD, 128GB Toshiba PCIe x2 M.2 SSD (KBG30ZMV128G) gone cooking externally, 1TB Seagate 7200RPM 2.5" HDD (ST1000LM049-2GH172) left outside Monitor: 1080p 126Hz IPS G-sync

 

Desktop benching:

Cinebench R15 Single thread:168 Multi-thread: 833 

SuperPi (v1.5 from Techpowerup, PI value output) 16K: 0.100s 1M: 8.255s 32M: 7m 45.93s

Link to post
Share on other sites

1 hour ago, Jurrunio said:

Maybe 32GB is too stressful, try a lower clock speed.

 

doesn't matter, ECC is just a feature that could be disabled if anything doesnt support it and it will run like normal memory. What cant be wrong is unregistered and registered memory, motherboards can only take 1 type at a time if the CPU supports both (in this case, Core series only supports unregistered DIMMs)

Well, you know it's a branded workstation... I am just wondering the actual cause of the conflict

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×