Live data from Hacker News

Samsung's 970 EVO SSDs Reviewed – 500GB and 1TB

anandtech.com

21–30 of 68 posts

Re: Samsung's 970 EVO SSDs Reviewed – 500GB and 1TB

#22
post #15

Earlier quoted context omitted.

Volatile is ok, as long as it honors barrier commands.

I think the bigger issues are bandwidth and power consumption. If software is doing heavy IO, it doesn’t really want to have to contend with the HMB for PCIe and memory bandwidth. Also, a DRAM chip is a DRAM chip. I’d rather pay a couple dollars for a local DRAM chip in the NVMe device than a couple dollars for extra host RAM that will be permanently reserved for the HMB.

The heavy IO case is addressed by write-back characteristic of block devices. In the end, the cache is in host RAM regardless of the underlying device type.

Re: Samsung's 970 EVO SSDs Reviewed – 500GB and 1TB

#23

Do these drives feature power-loss capacitors? I'm still wondering why most manufacturers think it's okay to offer SSDs that lose data on power loss (!!)

I really wonder why a $10 capacitor add-on (a capacitor, a diode, two connectors) plugged in between main board power supply and a disk power receptacle is not a common item.

Or maybe it is, but I have no idea?

Re: Samsung's 970 EVO SSDs Reviewed – 500GB and 1TB

#24

Do these drives feature power-loss capacitors? I'm still wondering why most manufacturers think it's okay to offer SSDs that lose data on power loss (!!)

I wish reviews started taking reliability into account. The only report on power loss effects on actual hardware I could find is from 2014 [0].

http://lkcl.net/reports/ssd_analysis.html

Re: Samsung's 970 EVO SSDs Reviewed – 500GB and 1TB

#25
post #10

For those looking, here's the full anandtech review that also came out today. [0] @dang maybe change the link as the review is a superset of the announcement. [0] https://www.anandtech.com/show/12670/the-samsung-970-evo-ssd...

Sure, changed from https://www.anandtech.com/show/12674/samsung-announces-970-p.... Thanks!

(It's best to email hn@ycombinator.com to tell us things, though — I only saw this by accident.)

Re: Samsung's 970 EVO SSDs Reviewed – 500GB and 1TB

#26
post #23

Do these drives feature power-loss capacitors? I'm still wondering why most manufacturers think it's okay to offer SSDs that lose data on power loss (!!)

I really wonder why a $10 capacitor add-on (a capacitor, a diode, two connectors) plugged in between main board power supply and a disk power receptacle is not a common item. Or maybe it is, but I have no idea?

A common minimum voltage requirement for the nominal +5V supply is 4.5 V. Let's assume a voltage below that won't damage the drive, but it won't operate.

The diode is going to drop 0.6V, leaving you with only 4.4 V max for the drive. Then the capacitor is going to immediately start dropping in voltage as soon as you start pulling the stored energy out of it.

So, in practice, to do a quality job of this it will probably need a DC-DC converter. These aren't necessarily expensive, but definitely more complicated than the old diode trick.

Re: Samsung's 970 EVO SSDs Reviewed – 500GB and 1TB

#27
post #23

Do these drives feature power-loss capacitors? I'm still wondering why most manufacturers think it's okay to offer SSDs that lose data on power loss (!!)

I really wonder why a $10 capacitor add-on (a capacitor, a diode, two connectors) plugged in between main board power supply and a disk power receptacle is not a common item. Or maybe it is, but I have no idea?

I worked on SSD power loss implementation in the past. You also need a signal into the controller indicating power failed. At this point the controller will minimally stop all less critical tasks, prioritize the buffered user data to be written and make sure the FFS is in consistent state for recovery.

Re: Samsung's 970 EVO SSDs Reviewed – 500GB and 1TB

#28
post #2

I just changed my work video editing workstation order. I will have two 970 EVO NVMe for my OS and Project drives. There actually is a bottle next with raw 4K, 5K and 8K Video and playing it in an editor.

It doesn't seem like the 970 Evo is much faster than the 960 Evo. They're still faster than all of the competition (aside from super-expensive Intel Optane). So its not a big deal. But there are some reviews that show the 960 Evo winning some performance tests out there.

PCIe SSD cards can provide more bandwidth and IOPS, but for those prices you might as well RAID0 several M.2 sticks.

Re: Samsung's 970 EVO SSDs Reviewed – 500GB and 1TB

#29
post #23

Earlier quoted context omitted.

I really wonder why a $10 capacitor add-on (a capacitor, a diode, two connectors) plugged in between main board power supply and a disk power receptacle is not a common item. Or maybe it is, but I have no idea?

A common minimum voltage requirement for the nominal +5V supply is 4.5 V. Let's assume a voltage below that won't damage the drive, but it won't operate. The diode is going to drop 0.6V, leaving you with only 4.4 V max for the drive. Then the capacitor is going to immediately start dropping in voltage as soon as you start pulling the stored energy out of it. So, in practice, to do a quality job of this it will probab…

This makes sense! OTOH low-drop diodes do exist [1], and the drive likely needs just a (small) fraction of a second to flush the caches; it's SSD, not need to wait for the disk to rotate to the right position or move heads. I suppose the power loss signal from the controller still propagates, even if the computer's PSU plug is just yanked out of socket.

[1]: https://electronics.stackexchange.com/a/282471/9909

Re: Samsung's 970 EVO SSDs Reviewed – 500GB and 1TB

#30
post #23

Earlier quoted context omitted.

I really wonder why a $10 capacitor add-on (a capacitor, a diode, two connectors) plugged in between main board power supply and a disk power receptacle is not a common item. Or maybe it is, but I have no idea?

A common minimum voltage requirement for the nominal +5V supply is 4.5 V. Let's assume a voltage below that won't damage the drive, but it won't operate. The diode is going to drop 0.6V, leaving you with only 4.4 V max for the drive. Then the capacitor is going to immediately start dropping in voltage as soon as you start pulling the stored energy out of it. So, in practice, to do a quality job of this it will probab…

You also need to limit in-rush current without dropping that voltage and account for capacitor degradation over the life of the device.
Post reply on HN