Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
stevewm
May 10, 2005
Wanted to add.. I have a Toshiba/OCZ RD400 NVMe drive...

Finally got a motherboard with PCiEx 3.0 so now the drive works at full speed. (2600MB/sec read, 1500MB/sec write). The RD400 is a 2280 sized m.2 drive that comes stock with a PCI Express adapter board. I recommend ALWAYS using it in the adapter board. The adapter board has a thermally conductive pad on it that contacts the back of the main controller IC. Without this thermal pad (as it would be in a motherboard m.2 slot) the operating temperature shoots up to 75°C and higher in seconds under load. On the adapter board with the thermal pad, it never exceeds 50 under the same conditions.

When mounted in my mobo's m.2 slot, OCZs own SSD monitoring utility would throw overheat alerts during disk benchmarks and any time it was under heavy load for more than 10 seconds or so.

Adbot
ADBOT LOVES YOU

stevewm
May 10, 2005

Potato Salad posted:

I think you're suddenly the most interesting person in the room :)

So, what's the boot time like on that thing? Got screenshots of those benchmarks you ran?

At work now... Later when I get home I'll get some screenshots together and post them. :D

stevewm
May 10, 2005
As promised...

From cold boot power button push to desktop is 18.7 seconds total. However, measured from the Windows loader (spinner appearing) to desktop is 3.7 seconds. So my motherboard takes 15 seconds to go through its initialization. On a reboot instead of cold shutdown, Windows is often already booted back to the desktop before my monitor wakes back up, so I cannot really time it..


Screenshots:

CrystalDiskMark


CrystalDiskInfo



Edit: Wanted to add... Using a Asrock Z170 Pro4 mobo, i7-6700. The SSD is mounted on the PCIEx adapter board in the second 16x slot. The drive was stuck in 1x mode until I updated the mobo BIOS, which fixed that problem.

stevewm fucked around with this message at 02:48 on Sep 2, 2016

stevewm
May 10, 2005
The older platforms are definitely a pain in the rear end when it comes to NVMe drives, or PCI Express devices in general.

My previous mobo was a z97. Due to the PCI Gen 2 limits with that chipset, my OCZ RD400 was limited to 1500MB/sec. And because of the limited PCIEx lanes available on that platform, installing the RD400 also disabled all the other PCIEx slots since there were not enough available lanes. I could only have my graphics card and the RD400 SSD.

My new mobo (Asrock z170 Pro4) didn't have these limitations, but needed a BIOS update to work properly. The drive was stuck at 1x until I updated the BIOS. Now I get around 2600MB/sec read and 1500MB/sec write. :circlefap:

stevewm
May 10, 2005

apropos man posted:

Is there a noticeable difference just opening a browser from startup with these nvme drives? Firefox with about 6 or 7 extensions can take a second to load on my laptop and desktop. Both have an 850 Evo in them.

Just trying to get an appreciation of what sort everyday performance to expect. I'll have a look at some YouTube reviews later.

Can't speak for Firefox. But I know there is zero wait for Chrome to load with my RD400 drive. On a fresh startup it is just as fast as opening a new window with the browser already running.

stevewm
May 10, 2005
When you add the RD-400 to the OP, I would definitely make a note that it runs REALLY hot if used in a M.2 slot.. I experienced this on 2 different mobos and the previous owner had the same issue. The thermal pad on the included PCI Ex. adapter is really needed IMHO. Any sustained activity of more than 10 seconds or so without the thermal pad setup on the adapter will see it hitting temps in excess of 75 C. With the pad/adapter it never exceeded 55 in my tests.

It may possibly be designed for that, but I'm not comfortable with it.

stevewm
May 10, 2005

Knifegrab posted:

OK so after doing some reading it would seem that I can boot from a 950 as well (as I would probably go for that over the 960), but apparently because my m.2 slot only has two lanes it will be a bit slower in terms of perforamcne. That's fine really, I am limited in chassis space currently.

You will also be limited by the m.2 slot being PCIex 2.0.. If it's only 2 lanes, that limits you to around 750-800MB/sec. Also as with other Z97 mobos, some of your PCIex 1x slots will be disabled if you use the m.2 slot due to the limited amount of pciex lanes the Z97 platform has.

stevewm
May 10, 2005

Hikaki posted:

I have a z97-a mobo which only supports pcie2x on the m2 slot. I need more space and I want to get an m2 ssd in particular because my sata slots are all full. Is the 600p a good idea in my case? Looks like it has poo poo numbers but my mobo won't be able to run something like a 960 evo/pro at full speed anyway. I could also just get a big sata ssd and just get rid of a smaller one but hearing about the better access times on nvme drives has me leaning towards the 600p.

Just be aware if you are using any of the PCIe 1x/4x slots on your mobo. The z97 platform is limited enough in PCIe lanes that on most using the m2 slot disables one or more of the 1x/4x slots.

I ran into this issue with my Gigabyte z97 mobo.

stevewm
May 10, 2005

Hikaki posted:



drat, I didn't know that. Mine has 2 pcie1x slots and I'm using one, and it looks like using the m2 slot disables both of them. There goes that idea!

Yeah, I found out the hard way.. Had a PCIE Wifi card in one slot, and thought I fried it because it stopped working after I had installed my SSD. The mobo was powering the card, but it was no longer being detected. I completely disassembled the entire machine and started putting parts back in one by one. Spent like 2 hours troubleshooting it. Never once did I consider that could even have been a issue. Finally it dawned on me when by chance I had removed the SSD yet again and realized the wifi card started working.

I used it as excuse to upgrade. Went to Microcenter and bought a shiny new z170 mobo and processor.

stevewm fucked around with this message at 15:06 on Nov 16, 2016

stevewm
May 10, 2005

Lum posted:

The keying is on the opposite side to the slot that has the SM951 in it, so I think it's only 2 lane. Here's a picture of the motherboard in question

That slot appears to be a "B" keyed slot. Which means it technically could have the following interfaces: PCIe ×2, SATA, USB 2.0 and 3.0, audio, UIM, HSIC, SSIC, I2C and SMBus. However it is not required that all these interfaces are present. Looking at the spec sheet for your laptop, it specifically says Slot 3 is for 3G/LTE cards with a USB interface. The system block diagram in the service manual confirms this. It shows that slot as being connected to "USB 3.0, Port 8".

stevewm
May 10, 2005

Harik posted:

Question about the samsung 850 - does it have a severe problem reading trimmed (zero) sectors?

I was trying to clone a 250gb to another disk, it was a nearly fresh install and I was putting it on a 500 instead of a 250. I thought (at the time I started) that a SSD copy would be faster than waiting around for w10 to get it's head out of it's rear end and do an agonizingly slow install. Instead it took north of 2 hours to clone.

It started at 250MB/sec but pretty quickly dropped down to 25-30MB/s when (I'd assume) it copied the 12+GB of windows 10 trash and started copying empty space.

It's either that or my 840 has sustained write problems, and that's a bad sign.

Neither have anything interesting in SMART.

Its your 850 a Pro or EVO? The 850 EVOs had a problem with earlier firmwares where older files would read slower. The 840s have the same problem to a lesser degree. This was fixed in newer firmware updates.

Also what tool are you using for cloning that would bother with empty space? Use something like Macrium Reflect... It only copies occupied sectors, making clones much faster.

stevewm
May 10, 2005

Harik posted:

dd.

I figured it would be fast enough from SSD to SSD that it wasn't worth downing the whole machine to run a standalone cloning program. I probably should have just to save the worthless writes, but oh well.


Macrium can clone a live system thanks to shadow copy support. Easy enough to use. Plug in new SSD, clone old SSD to new SSD, unplug old SSD and boot from new SSD.

stevewm
May 10, 2005

Volguus posted:

A lot of lovely software has "C:\Program Files" hardcoded and explode if such a folder does not exist.

The Windows XP installer had a nice little bug in that it would sometimes enumerate non-fixed storage devices first before getting to fixed disks. The result of which was that your main disk ended up being a letter other than C. This would happen even if the non-fixed devices where not readable by the installer. This commonly occurred with built in memory card readers.

The only way around this is to disable or unplug the additional storage devices until XP is installed.

stevewm
May 10, 2005
Intel SSDs and me have not been getting along recently...

At work I had 1 Intel 730 fail a few months ago; went completely undetectable. Then at home I had a Intel 530 fail; started showing disk errors in event viewer, read/writes would take forever to complete. And now just a few days ago the Intel 520 boot drive in my home server/3d printing/do everything machine started failing as well; the computer will hard lock and the SSD will be undetectable on a reset. Only way to get it working again is to remove power from it and wait a bit.

Out of all the SSDs I've used, my ONLY failures have been Intel SSDs. I've not had another brand SSD fail yet.

stevewm
May 10, 2005
I just find it funny that everyone used to say Intel SSDs where a good bet.. Yet that is the only SSD brand I've ever had fail.

While the fleet of cheap random SSDs from Kingston, ADATA, Crucial, OCZ, etc... have kept right on trucking for far longer than the Intel SSDs.

stevewm
May 10, 2005
For those of you with Intel 53x series drives. I ran into something interesting.

I was having issues with a Intel 535 in my home server, it kept disappearing randomly.

Turns out the 5xx series drives from Intel have 2 issues that can cause problems in some cases.

#1 - They don't handle "DevSleep" properly, which means they will sometimes go to sleep and never wake back up. This bug only appears on particular combinations of drivers and motherboards. Intel provides a tool to disable DevSleep support entirely, but it requires burning to a CD or USB and booting with it to make the change. (https://communities.intel.com/thread/106343)

#2 - They use a internal power saving mechanism that trades high write amplification for power savings. In some cases the amplification was extreme (30x or more!) and it is enabled by default. This is also why my 1.5 year old drive is already down to 40% health. They provide a special firmware that disables this. (https://downloadcenter.intel.com/download/26452/Write-Amp-Firmware-for-Intel-SSD-530-535-Series)


#2 is really lovely..... Some people where seeing a few TB of writes in just a couple weeks on brand new drives. It would also explain why the 530 drive I had in a older laptop died so quickly. It went read only in less than 2 years.

stevewm fucked around with this message at 18:54 on Mar 16, 2018

stevewm
May 10, 2005

JnnyThndrs posted:

I’m not going with NVMe until we get enough PCIe lanes on consumer chipsets so that running one doesn’t translate into “using a NVMe drive disables two PCIe 1x slots, two SATA ports, your USB 3.1 ports and a partridge in a pear tree’ written in 8 point type on the last page of your mobo manual.

Yeah, that was a fun one on my old Z97 mobo... Putting my RD400 m.2 into the 4x slot (with included adapter, the mobo itself didnt have a m.2 slot)), disabled ALL all the other PCI-E slots except for the 16x graphics slot. I wondered why my wireless card stopped working. I thought I had fried something. I didn't realize the Z97 platform had such limited PCI-E lanes.

stevewm
May 10, 2005

apropos man posted:

I've had one of the blue, OCZ Arc 100 drives since... since they were on the market.


Same here.... We have a small "fleet" of the 120GB variants at work. Maybe 15 or so. Never had a failure. I personally have a OCZ/Toshiba RD400 NVMe drive that is great too.

Meanwhile, nearly every Intel 5xx/7xx series drive at work has failed. Including the the 2 Intel 530's I had at home. In fact Intel is the only SSD brand I have ever had a failure with that I can remember at this point.

stevewm
May 10, 2005

Lambert posted:

I had an Intel 320 (80 GB) fail on me as well as two Corsair Force, the F80 and the F120. Since then, I haven't had any problems. But the early ones were quite unreliable.

Agreed on that point. All the new ones seem to be pretty reliable, even the weird off/no name brands that many OEMs are using these days.

stevewm
May 10, 2005

Phone posted:

I moved the Intel 535 between computers again and tried different power cables and SATA cables with no luck. I take it that I'm completely hosed on this and the drive is dead?

This is the standard way for that drive to fail. It just stops appearing. Those series of drives seem to have quite a high failure rate. Based on my own experience between the fleet of them we had at work, and some I had at home, they basically have a 100% failure rate after 2-3 years. Every single 53x drive we had at work died within 2 years. I got Intel to warranty swap some, but the swaps failed within months. Both 535s I had at home failed as well. I just gave up on them entirely.

There was also some pretty nasty firmware bugs.. One in particular could cause VERY high write amplification in some cases. (https://www.intel.com/content/www/us/en/support/articles/000027328/memory-and-storage/consumer-ssds.html) It was a bug in the drive's internal power management. I had several affected by this one. After a year of use they where at less than 50% health. Intel provides a tool to disable the feature that causes this, but you have to dig for it on their forums. Another bug causes the drive to disappear randomly when it is put to sleep by the host computer. Originally the fix for this was available only as a special firmware update given by Intel support. But it was rolled into the mainline firmware at some point I believe. (See: https://www.google.com/search?q=int...QIoBDAAegQIBRAL)

Anyways, dead drives can sometimes be revived, but its only temporary.

Leave the drive powered up for 20-30 minutes sitting in your BIOS. And then attempt restarting without powering down and see if it gets detected. With any luck the drive will show up. Once you get it working again, run the Intel SSD toolbox and let it update the firmware if there is an update available. But it WILL fail again, and next time it does, it will not come back.

I have revived several Intel 52x and 53x drives doing this. The drive may go for days or months, but don't rely on it. It should be replaced ASAP.

stevewm
May 10, 2005
Had my first non-Intel SSD failure at work today.

Sandisk Z400s... Was a OEM drive inside a Lenovo system.

Will get detected if you let it sit for awhile, but shows no file systems, nor can you write to it.

stevewm
May 10, 2005

Eikre posted:

My understanding is that M2 SATA is just another SATA form-factor. Is this also true of NVMe, with respects to PCIe? In that, for example, a converter card would be purely passive,

This is correct on both fronts. M.2 slots can have SATA, USB 3, and PCI Express connections in them. The slot will have keys that denote its configuration and prevent the wrong type of drive being plugged into a slot that doesn't support it. Its kinda complicated really: https://en.wikipedia.org/wiki/M.2

NVMe is a protocol that uses PCI Express as its physical layer. Thus it can work on any PCI Express interface port with the appropriate passive adapter.

stevewm
May 10, 2005
Thought my OCZ/Toshiba RD400 was dying.. Recently started having problems with it not being detected on boot, or causing the machine to fail to come out of sleep.

I had been using it on the included m.2 to PCIex adapter board since I got it. Well this board is not entirely 100% passive, it appears to have a small switching power regulator on it. I am guessing this might be failing because since putting the drive in my mobo's m.2 slot, my problems have all but disappeared.

So yay!

stevewm
May 10, 2005

BobHoward posted:

Also it may surprise you to know that in virtually all SSDs made the past several years, the on-device encryption is on all the time.

Its not just SSDs... Some external hardrive brands (WD in particular) use encryption on-board the USB-SATA adapter inside the drive casing, even if it was not specifically enabled. So if this adapter board dies, your data on the still working hardisk is now inaccessible.

stevewm
May 10, 2005
Just retired a laptop at work that had a Crucial M4 SSD in it. I think this is one of the first SSDs I ever bought at work...



4.7 hours of power on time, still 84% health.

I remember this one suffering the 5,128 hour bug.. One it hit that amount of power on time, it would fail to detect. You had to leave it powered for 20 minutes, and then warm reboot. It would be detected then and you could apply the firmware that fixed that issue.


Edit: I guess its not so healthy after all...

stevewm fucked around with this message at 20:58 on Aug 6, 2019

stevewm
May 10, 2005

ItBreathes posted:

NVMe is hot swappable?

U.2 NVMe is supposed to be.

stevewm
May 10, 2005
Well gently caress...

Was putting a new power supply in my computer and managed to mix up some EVGA and Corsair modular power cables. Resulting in me plugging a Corsair modular SATA cable into my EVGA power supply. Cable fit fine, but as I found out the pinout is different.

Fried all 3 SATA SSDs in my gaming machine. (480, 240, and a 120)

Strangely enough the one mechanical drive plugged into that same cable didn't seem to mind, it was working all the while I was smelling something burning.

Looking at the pinouts, the Corsair cable put 12v where 5v is supposed to go.

They where only holding game installs, but still quite an expensive mistake.

stevewm
May 10, 2005
While trying to figure out what happened, I went down a rabbit hole with modular power supply pinouts. 4 different manufacturers appear use the same 6-pin connector for their modular power supplies. However every one of them is a different pinout in some way.

Some manufactuers even have different pinouts across their own product lines. EVGA appears to have 3 different ones, and Corsair seems to have 4!


I took the dead drives apart hoping I could possibly repair them...Looking at the datasheets, they only use the 5v. 12v and 3.3v lines are not connected. I measured a dead short between 5v and ground on all 3 drives. On the 120GB I found several shorted filter caps and a blown diode. After I knocked off all the shorted caps and diode, the short to ground remains. So I gave up on them.

stevewm
May 10, 2005
My luck with SSDs has been terrible this week...

After the power supply debacle frying all of my SSDs in my gaming machine. The Intel SSD boot drive in my plex server/lab PC failed after a power outage Sunday night. I forgot I had one of those time bomb Intel 5xx series SSDs in that computer.

Anyways, the disk gets detected as "Sandforce 20002BEAB" and does not function. I went ahead and reinstalled windows on another SSD. After playing around with the dead drive a bit in a USB3 dock, I found I could get it to come back to life by turning the power on/off a few times in short succession. I wanted to copy my Plex data directory off it, as I really didn't want to setup Plex from scratch all over again. (rescanning my library and resharing with everyone would take days). Launched robocopy and let it go to work. The drive died a couple times during the copy but I could get it going again by just cycling the power to it, and robocopy just continued where it left off.

Managed to copy the 90GB worth of Plex data and a few other folders this way.

Adbot
ADBOT LOVES YOU

stevewm
May 10, 2005

Klyith posted:

(At least for Corsair .... who keep things consistent


Guess again... they don't. They have at least 3 different pinouts that I could find. One of the PSU cables I fried my SSDs with was of the "Type 4" variety.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply