Hacker News new | ask | show | jobs
by tenbino 2116 days ago
I would have thought giving a customer a faster than spec product isn’t a terrible sin.
12 comments

If I bought the drive for active work, yeah I'd probably be happy about the speed for the price (IF they don't artificially limit the speed!).

But if I buy these for NAS/backups, I want them to be quiet, cool and longer lasting - in my experience 5400 RPM drives do heat less and last longer than 7200 RPM.

Not sure if they still do, but WDs had this IntelliPark feature, which slammed the heads into parking a few times a minute (because otherwise your drive would last a decade, I guess), and their 7200rpm drives were clicking so loud I often thought "that's it, it failed". But no, just "normal" operation...

So I'd rather get the specs I'm paying for.

For the people who can't hear the vacuum cleaner noise from their PC because they use headphones yes, getting 7200 rpm at the price of 5400 might be nice.

For the people who want to run cool, quiet and less power hungry devices, they'd like to get exactly what they ordered. Nobody buys a 5400 rpm hard drive for speed.

Stick a 7200 rpm drive in a console and see how the cooling system copes, for example. Or a low power PC with a PicoPSU. Or a NAS.

Some portable hard drives can be powered off USB. If you get a 7200 advertised at 5400 drive, it won't run because it's not getting enough power.

The terrible sin was putting SMR (shingled magnetic recording) disks in drives sold as conventional magnetic recording. This may not be nearly at that level, but it's concerning that the habit of mis-labeling drives continues.

e.g. https://www.tomshardware.com/news/wd-lists-all-drives-slower...

In this case it is. High capacity 7200 rpm drives can be unbelievably noisy, to the level that they are not recommended for home theatre NAS as you can hear the seeking sound during a movie or from the next room at night.
Speaking of seeking, if the data was laid out sequentially on disk, and the computer was not accessing any other files on the disk in question (spinning disks for movies and other data, SSD or m.2 for OS itself), then it would be very silent. But my impression is that a lot of the time we end up with files fragmented more than what is desirable. That's another wish of mine, that I would have a system that would keep this in mind and which would be better tuned to my use. Maybe one day.
> if the data was laid out sequentially on disk, and the computer was not accessing any other files on the disk in question (spinning disks for movies and other data, SSD or m.2 for OS itself), then it would be very silent

That's not true. Even a spinning sound of a 5400 rpm disk makes quite loud.

Whether you can hear it due to casing or distance is another matter. Most of the time fan noise will make you ignore it. When fan is not spinning I can hear laptop drives clearly.

Windows had (has?) a defrag tool for this purpose. With SSDs, it's not an issue.
Windows 10 had a bug this year where it defragged the SSD (yes) on every boot instead of once a month.
Wow first I've heard of this, do you have a source? I'd like to read more about it.
7200rpm is a bit noisy compared to 5400rpm but if 7200rpm is unacceptable, it's better to use SSD or NAS rather than 5400rpm HDD.
The noise specifications for the drives are available from the WD prior to purchase.
It is very hard to trust hard drive manufacturers these days. I have a few of these drives mentioned here https://www.seagate.com/www-content/datasheets/pdfs/ironwolf... with the specified "typical" noise level of 25 dB for seek and it is definitely much much more than that.

Synology forums have several threads like this: https://community.synology.com/enu/forum/1/post/130843

And I understand that these are different companies, but why I even went for Seagate was because of the CMR issues.

Yeah but they'd probably lie about that too, based on their recent track record.
It is. 5400rpm drives have specific uses. It also hurts sales of competitors selling real 5400rpm drives. It may even hurt the whole 5400rpm drive market, if WD competitors can't match the price of the fake 5400rpm drives and decide to cease production.
I also heard that it might cause problems mixing 5400 rpm and 7200 rpm drives in a single NAS box due to vibration interference. Not sure if true. But the general trend that Western Digital seems to be OK with lying on specifications is concerning.
It is when it blows through your power budget and heating budget. Also, some array performance degraded significantly when using different speed drives, to much below the slowest drive (that’s not common anymore, but used to be an issue and might still be in some setups)
It matters because they don't give you what they said they would in the specs.
It's also becoming a habit for them to miss-label drives (see smr)
Because “faster” is not the only variable involved. If customers wanted 7200 they’d buy that, but the difference in power and heat between a 5400 and 7200 can reach 100% which is an issue when you have sized and enclosure for one type and get the other.

Noise is also an issue, if you were looking for an HTPC drive for instance you don’t need the increase in performances but the noise pollution is a problem.

WD provides a data sheet with the power and noise specs for their individual drives. If you are buying to use in a situation where these matter shouldn't you be consulting the published specs instead of just choosing 5400 vs 7200? After all these can be greatly variable even between drives of the same RPM.
I mean... doesn't that same sheet of published specs state that it's running at 5400 RPM?

If they lie about that on their spec sheet, how can you believe the power/noise levels written on there?

Also, I've always had a fair amount of scepticism for digital product spec sheets. I've been looking for a projector with low input lag for a while, and the measured value has constantly been way off the claimed value (> factor of 2). Admittedly, this seems to most often be the case with some of the cheaper chinese brands, but definitely not exclusively.

They don't actually claim it is running at 5400 RPM. They claim performance is in the 5400RPM performance class. In other words it will perform similar to a drive running at 5400RPM. Nowhere do they mention that it runs at 5400RPM.
I stand corrected, you're right. Ref from link

> Looking at the spec sheets provided by WD, it seems they don't really list the rotational speed

I still think it's somewhat disingenuous, but my comment above was wrong.

According to the linked post, "5400rpm class" was a lie too.
> giving a customer a faster than spec product

IIRC from older threads - seek times are faster but sequential speed is capped.

These aren't home drives, where you have ample cooling in a PC case and power.

These are for NAS's/Servers where power and heat is a controlling factor, if admins wanted/could they would buy the 7200rpm drives and there would be a market for it, but they chose the 5400rpm because of constraints.

This is WD being lazy/cheap. Instead of selling them as 7200rpm drives, they're shovelling them into the 5400rpm lines as a presumably cost cutting measure.

They did not learn their lesson with the SMR/CMR scandal. Label your drives clearly.

It matters because I tend to choose 5400rpm for bulk storage - for power and longevity reasons, mostly.
For a certain segment you're right, but its not that simple. Another segment would specifically want the lower RPM.