Lolight

Can you elaborate more on problems with shucked WDs? I've never shucked those drives before. All of my shucked ones are Seagates - of the NAS and Enterprise variety which are Ironwolf Pro and Exos. As far as I know, all external Seagates come in those flavors, as long as you get in them in high enough capacities starting with 12TB.

Congrats. You've finally brought this thread down to a point of name calling. Surely you got the strongest point all around.

According to who? Can you share studies? If you refer to Backblaze and their blog, then it's not a study - mostly speculation designed to promote their services. In other words - a hugely successful marketing campaign that doesn't cost them a penny.

I find BackBlaze's reports to be interesting, at most. Though not particular useful for anything. They provide absolutely no grounds for drawing any conclusions and should not influence your purchasing decisions. Some of their blog posts are absolutely ridiculous, like the one where they come to a conclusion that there's no correlation between HDD operating temperatures and failure rates. https://www.backblaze.com/blog/hard-drive-temperature-does-it-matter/ Dispelling Backblaze's HDD Reliability Myth - The Real Story Covered "We chronicle Backblaze's failed attempt to provide credible HDD reliability data. Read on to find out why you should pay no attention at all." https://www.tweaktown.com/articles/6028/dispelling-backblaze-s-hdd-reliability-myth-the-real-story-covered/index.html

Changes don't happen overnight. But they can happen over a year or two. Every acquisition is different in scope and can not be treated as a uniformed process. Processes can be partially kept so is the personnel. Or be completely disregarded for a new tech and people. Some companies are being acquired strictly for the value of brand recognition or patent portfolio or customer base. Everything else being secondary and expandable. Suppliers could be kept, partially changed, or switched to new ones. The workforce might get let go too, at least in part. There's no template whatsoever, especially in the current fully automated environment where even car factories can be fully (completely) retooled with brand new equipment in a year or so.

Current Newegg are no comparison to what they used to be in the past.

.-..-. .-.-.- .-.-.- .-.-.- --- -.-. -.-. .- ... .. --- -. .- .-.. / .-.. .. - - .-.. . / ..-. ..- -. / -.-. --- -.. . .-.-.- .-.-.- .-.-.- --. --- - / .-.. .. - - .-.. . / --- ..- - / --- ..-. / .... .- -. -.. ..--.. # -... ..- - / .. - .----. ... / ... - .. .-.. .-.. / ..-. ..- -. / ..- -. .-.. . ... ... / -.-- --- ..- .----. .-. . / .- / .... .- - . .-. .-.-.- .-..-. # .- -. -.-- / -- --- .-. . / .-.. .- -... . .-.. ... ..--..

Potentially problematic since you won't know their prior history - each drive or batch will differ from another in their prior history of handling, usage, storage conditions etc, etc. The savings might be worth the risks as long as you understand what you're dealing with.

.. ... -. .----. - / .. - / . ...- . -. / .-.. . ... ... / .--. --- .-.. .. - . / ..-. --- .-. / - .... . / - .... .-. . .- -.. .----. ... / -.-. .-. . .- - --- .-. / - --- / - . .-.. .-.. / . ...- . .-. -.-- --- -. . / - --- / -... .- ... .. -.-. .- .-.. .-.. -.-- / ..-. -....- --- ..-. ..-. / .-- .... . -. / - .... . -.-- / .- ... -.- / - --- / .-. . ..-. .-. .- .. -. / ..-. .-. --- -- / .- -. -.-- / -- --- .-. . / --- ..-. / -- --- .-. ... . / -.-. --- -.. . / ... .. .-.. .-.. .. -. . ... ... / --- -. / - .... .. ... / .--. .- .-. - .. -.-. ..- .-.. .- .-. / - .... .-. . .- -.. / -.. ..- . / - --- / .. - ... / . ...- --- .-.. ..- - .. --- -. / .. -. - --- / .- / -- --- .-. . / -... .-. --- .- -.. / -.. .. ... -.-. ..- ... ... .. --- -. --..-- / ..- -. .-.. .. -.- . / .- / ... .. -- .--. .-.. . / .- -. -. --- ..- -. -.-. . -- . -. - / ... . .-. ...- .. -.-. . / --- -. / - .... . / -... . - .- .----. ... / -. ..- -- -... . .-. ... / .--. .-. --- --. .-. . ... ... .. --- -. ..--.. # # .-- .... -.-- / -.. --- / .-- . / . ...- . -. / .... .- ...- . / - --- / ... - .. .-.. .-.. / -.- . . .--. / .- ... -.- .. -. --. / - --- / -.- . . .--. / - .... .. ... / - .... .-. . .- -.. / ..-. .-. --- -- / -.-. --- -. - . -. - / .--. --- .-.. .-.. ..- - .. --- -. ..--.. # # -. --- / --- -. . / .... .- ... / . ...- . .-. / . -..- .--. .-. . ... ... . -.. / .- -. -.-- / .--. .-. --- -... .-.. . -- / .-- .. - .... / .... .. ... / -.. . .-.. .. ...- . .-. -.-- / .-..-. ... - -.-- .-.. . .-..-. / --- -. / .... .. ... / .--. .-. .. --- .-. / - .... .-. . .- -.. ... / -.-. .- ..- ... . / - .... . -.-- / .... .- -.. / -. --- / .. -- .--. --- .-. - .- -. - / -.-. --- -. - . -. - / .- -. -.. / .-- . .-. . -. .----. - / --. . -. . .-. .- - .. -. --. / -- ..- -.-. .... / --- ..-. / - .-. .- ..-. ..-. .. -.-. / .- - / .- .-.. .-.. .-.-.- # # / # # .- ... / ..- -. ..-. --- .-. - ..- -. .- - . / .- ... / .. - / .. ... / ..-. --- .-. / -. --- .-- --..-- / - .... . / -- --- -.. . .-. .- - --- .-. ... / ... . . -- / - --- / -... . / - --- - .- .-.. .-.. -.-- / ..- -. .- -... .-.. . / - --- / .-. . -.-. --- --. -. .. --.. . / .- -. -.-- / .--. .-. --- -... .-.. . -- / -.. . ...- . .-.. --- .--. .. -. --. / .... . .-. . / .-- .... .- - ... --- . ...- . .-. .-.-.- # # .. -. / - .... .- - / -.-. .- ... . / - .... . / --- -. .-.. -.-- / .--. .-. .- -.-. - .. -.-. .- .-.. / ... --- .-.. ..- - .. --- -. / .-- --- ..- .-.. -.. / -... . / - --- / .. -. - . -. - .. --- -. .- .-.. .-.. -.-- / ..-. .-.. --- --- -.. / - .... .. ... / - .... .-. . .- -.. / .-- .. - .... / -- --- .-. ... . / .. -. / . ...- . .-. -.-- / ... .. -. --. .-.. . / .--. --- ... - .-.-.- # # .-- .... .- - .----. ... / --. --- --- -.. / ..-. --- .-. / - .... . / --. --- --- ... . --..-- / -- ..- ... - / -... . / --. --- --- -.. / ..-. --- .-. / - .... . / --. .- -. -.. . .-. .-.-.- .-.-.- .-.-.- .-.-.-

A 24 hours long test is not really required to isolate bad RAM. 10 passes without errors would be sufficient at first. Run each stick separately for 10 passes. If errors are encountered try a different slot. If no errors then run the sticks together for another 10 passes. If errors are encountered then the memory controller or memory slots could be at fault. In that case try switching RAM slots and do it again.

Translation: "Hello, I currently have a terramaster NAS with only 1GB of ram, but it is very limited in all its functionality. It is possible that with the following features the OS unraid works with: 1GB RAM OS: ARM v8-64 1.4 2 bays I will be attentive if they can affirm me, this in order to buy the license. Thank you." Welcome! If your unit has an ARM based processor in it then no - it can not be used with Unraid. Unraid can be only installed on the x86 (Intel) Terramaster units that also have an HDMI port.

Welcome! So how's your experience so far? What's your usage case in Unraid?

As @caplam has previously mentioned, 13 gen on 680 chipset. But if you really, really want it and are willing to spend... Because you're concerned about your data integrity. ...Just don't go that route before you got budgeted and set for proper backups of your critical data, correctly configured and segregated networking, quality UPS, high-tier, community tested PSU model and HDDs of the NAS/Enterprise varieties, purchased brand new (not used, refurbished, renewed etc unless you're willing to take chances but then you might take chances with everything else and save a real bundle). Shucked drives are fine when you know which external models contain the above-mentioned HDDs (no a difficult task to figure it out). Only after all that is done you should go ahead and spend on ECC if the desire is still there.

Wrong questions for a home user environment. ECC won't help you in any way against a user error (the biggest risk by far), followed by a general non-RAM related hardware failure, electrical spikes, lightning strikes, floods, fires, earthquakes, hacking etc, etc, etc. Most of which are more probable to happen, with far more devastating effects on your overall data integrity in comparison. Instead of spending on ECC, at first make sure to have proper backups of your critical data, then proper networking segregation plus enterprise quality HDDs. btw, I consider the above-posted clip to be misleading and exaggerating the actual danger, designed to over sensationalize the subject for the purpose of attracting the most possible number of views.

You might be facing a point where it's just not worth your time, electricity and effort to keep re-using certain old, obsolete but still functioning hardware. A fresh start with new, efficient and up to date components will make your project simpler, less problematic and cheaper in the long run.

by using similar to this 9-pin motherboard header to USB 2.0 adapter

Understood, yes, it's your thread, but it happened to evolve into a very informative discussion on possibly the most important at the moment development in Unraid's progression. I'd agree with others. The childish back and forth are greatly diluting the discussion and shouldn't be encouraged at this point.

If you plan on transcoded streaming then yes, Intel is the clear choice thanks to its Quick Sync tech. If strictly direct-played then it doesn't really matter, even though Intel is still considered to be a more stable and efficient (when mostly idled) platform overall.

Intel iGPUs are super-efficient hardware transcoding powerhouses thanks to intel's proprietary Quick Sync. Quick Sync is fully supported in Plex, Jellyfin and Emby. AMD graphic cores are not compatible with Quick Sync.

No problem, thank you for the clarification

There should be no difference between iGPUs of the same part number within the chip families. Stay away from those "gaming" motherboards with tons of MOSFETS and many power stages - a waste of money and electricity. https://pcpartpicker.com/list/ Get a more basic motherboard from your favorite brand with as many of SATA ports as you can find but not before checking its manual on the .M2 expansion slots configurations. Many MBs will disable one or several SATA ports when any of the .M2 slot(s) are in use. Consult the following link for a PSU. https://cultists.network/140/psu-tier-list/ Get a single rail, listed in the Tier A, High-End group. Pay attention to the model years (if applicable), some models are not recommended if out of the specified year range. Don't try to save on a PSU.

You haven't mentioned the size of HDDs. If they're old and small then you could easily improve your efficiency by replacing them with just a few of the current crop large HDDs (16-18TBs). Which will allow you to get rid of the HBA card - another electricity consuming device.

Hmm... Can you point me to a recent (or any) example of such recommendation?

Thanks for the link. The article has been posted a quite a while ago but it's still very relevant to the current situation. "...I sense that people not well-versed into Linux or FreeBSD may sometimes feel pressured into building a NAS that they can't handle when problems arise. In that case, opting for ZFS could cause more trouble than it would solve." So true!!! TrueNAS proponents keep trying to steer as many n00bs as they can towards TrueNAS Scale now by spreading FUD about n00b-freindly Unraid, basically setting them up for an unnecessary expense at first that often leads to a disappointment and then an eventual failure, especially for those who've never wandered off the Windows ecosystem and have no interest and/or time in learning anything Linux.