Not to mention being completely at the mercy of forced updates which can ruin your favorite game at any moment. People bitch about having to install from disks and patch manually, but I value that freedom and control.
For example, for me, Company of Heroes peaked at 1.71. This is fine because I simply install it from disk, apply the two patches required to get it to 1.71, and play the game exactly how I like it forever. I wish I could do the same thing with some modern games, which have been ruined by developers who don't know when to stop tinkering with their work.
Only GOG seems to be allowing for the old way of installation for new triple-A releases. I'll never be forced to install any particular version of The Witcher 3 or Cyberpunk 2077 because I have the offline installers and all the incremental patches saved.
All my disks are imaged and the images are all I ever interact with. I only keep the physical disks as extra backups now. Gave up on the idea of "collecting" after facing the reality that packaging starts to disintegrate after a couple of decades, not to mention disk rot.
What do you mean by "stops the clicking noise"? IME having head parking enabled increases clicking noises because there is a loud click every time the heads park/unpark (applies to Seagate and WD).
I don't have Exos, but IronWolf Pro. I disabled head parking out of habit. I've been doing it for a decade at this point and never regretted it. I'm pretty much on the fence though.
The weird thing is that even after disabling Idle B, the heads still sometimes park. I see the load cycle count occasionally go up and I'm pretty sure I've heard it. It happens much less often than before disabling Idle B, but it's still happening, nonetheless. Obviously, I'm not counting the load cycles happening during restarts. I'd like to know why.
QNAP only lists one TR-002 on their website and I can't find any evidence of more than one major version of the product. I'd guess that the -A-US refers to the power cable it's bundled with.
There are plenty of terrible products with a lot of good reviews on Amazon. The TR-002 reviews are pretty inconsistent and contradict each other on basic facts (e.g. some people claiming the TR-002 doesn't pass SMART data and others claiming it does - it does, BTW) so I wouldn't put too much stock in those. It's worth noting that this device can be configured and used in several ways, so it might be the case that, for example, it works great in hardware RAID0 mode but sucks as a basic port multiplier (as per my experience). I haven't tested the RAID modes personally so I can't comment on those.
All I can say is that if you have space for the drives in your case, have the spare SATA ports on your mobo, and don't have any particular use-case to necessitate an external enclosure, then you'd be mad to not install then directly in your PC. You'd only be adding cost and complexity for reduced performance and lower reliability. You may have noticed that a common refrain on subs like this is to avoid connecting storage via USB whenever possible. There's a good reason for that! You can easily end up with an unstable connection and random disconnects if the USB controllers happen to not like each other. It's always a gamble. SATA is far less temperamental.
If you’re talking about the TR-002, the only time it spins drives down is when you disconnect/eject it or turn off the PC it’s connected to.
However, I do not recommend the TR-002. It seems to have major performance issues when accessing two drives at once (at least in Individual mode). If you can install the drives directly into the PC that will be a vastly superior option.
Gigabyte was putting U.2 ports on some consumer motherboards or a hot second. My z170 designare has one. Never used it though.
I put the drives back into a PC to triple-check the EPC settings with showEPCSettings, and they definitely have not reset:
Name Current Timer Default Timer Saved Timer Recovery Time C S
Idle A 0 *1 *1 1 Y Y Idle B 0 *1200 *1200 4 Y Y Idle C 0 6000 6000 60 Y Y Standby Z 0 9000 9000 150 Y Y
I also ran them in the PC for 24 hours and the load cycles went up another two points for both drives, so I guess that rules out the DAS. Again, the PC had HDD spindown disabled and did not restart during that time.
As for the wisdom of enabling/disabling head parking: You make a compelling argument there. The same thoughts have occurred to me and I'm pretty much on the fence about it. Maybe I'll reenable parking and see how rapidly it climbs.
I would still love to know what exactly why it's increasing when it's supposed to be disabled, just to satisfy my curiosity.
I have two IronWolf Pro (ST10000NT001) drives in a QNAP DAS. Before putting them in the DAS, I used SeaChest to disable EPCfeature and powerBalanceFeature, and verified that the settings had stuck. However, I've noticed that the load cycle count will still occasionally increase even if the system hasn't been restarted. For example, I might see the load cycle on both drives increase by +1 in 24 \~hours.
I'm NOT worried about the increasing load cycle count in and of itself. It's not like it's happening at a rapid rate. I'm just concerned that it might be the symptom of some other issue with the DAS, such as random disconnections (though I can't see any evidence of this in Event Viewer).
So, my question is: Is there any reason for occasional head parking even when head parking has been disabled? For example, maybe it's part of some housekeeping routine of the drive?
Other notes:
- I've observed this behavior with the DAS on two different PCs.
- The DAS does not appear to have any kind of power saving/standby feature that spins down idle drives. I've only ever seen it spin down drives when it's disconnected, ejected, or when the host system shuts down, and the manual states as much.
- The Windows "Turn off drives after..." setting is set to Never/0.
- I've disabled all the relevant USB power saving features in Windows.
- This is not a case of Windows Update restarting the PC. I know this for sure because a restart causes two load cycles due to the disks being spun up twice during POST, and I'm only seeing the drives increase by one load cycle count at a time.
- I don't think it's a problem with the drives themselves since it's happening on both.