Minimum specs for external drive for FCPX

Hey guys,

I',m a big user of Glyph drives and now want to pickup a seagate 2tb drive for just my Youtube vids I edit in FCPX. I'm not sure what specs you all recommend for basic editing on a one camera shoot. Please excuse my ignorance, but I have really learned a ton here. I've got my setup pretty much dialed in, but I want to get everything else in place, so that I don't have to think about it.

If you are curious, I was looking at the Seagate slim plus bus powered drive. 2tb, USB 3.0.

I think it runs about 120mb/sec, which probably wouldn't cut it with higher res editing, but for what I am using it for I think it should work. Please give me your thoughts. My Glyph drive is mainly for all of my audio sessions in Pro Tools and Logic. I'm really digging using FCPX though. It's great.

Thanks peeps

Well I don't know if bus-powered vs externally powered would make a difference on USB 3. Perhaps someone else can answer that.

But USB 3 or better yet a Thunderbolt 2 drive would be fine. I do not recommend Seagate though. Seagates love dying more than anything else. I have a bunch of WD external drives but I don't use them for editing, only storage and backups. I also have a SATA caddy where I can mount internal drives via a USB 3 connection. Putting a good quality WD drive on there will be plenty fine. For my main editing drives I have G-RAID Thunderbolt 2 drives. I believe they use Samsung hard drives internally. No issues with them yet.

I read sometime ago an article about disk speed with fcpx (maybe in this site, I don't remember). The author wrote that when you choose an hard disk you must pay attention not only to the "classic" reading/writing speed but also to the writing speed when accessing multiple part of the disk at the same time. This is because fcpx does a lot of operations in background, more than any other NLE (auto backup without interruptions, auto save, the auto loading of inspector without double clicking the clips, live update of thumbnails and waveforms...). On top of that I must admit that having just the "better performance/better quality" options is sometimes a limitations. When using Premiere I can choose 1/8 quality and playback fine Red raw 6K using a simple usb3 external disk while FCPX in better performance give me a better quality in preview but can't playback smoothly the same file.

My opinion is I cut HD on external usb3 drives not my choice but if I have a heavy edit I also was using a 2 drive raid 230/mbs but that is not really fast enough for filters and playback.

I recently picked up a usb3 ssd sandisk @600.00$ and its running 400-450mbs

The big issue you will see down the road is running two usb or TB externals you'll find random eject which from what I've read is based on voltage issues on osx computers. Back up a lot in case of failure.... I also use an external powered hub/akito? TB extender forgot the name it runs on TB gives you 2 usb3 and TB port via 12volt.

I've never had a random eject on my Late 2013 Mac Pro. Maybe it is only an iMac or laptop issue?

Gilbert13 wrote: ...want to pickup a seagate 2tb drive for just my Youtube vids I edit in FCPX. I'm not sure what specs you all recommend for basic editing on a one camera shoot...I was looking at the Seagate slim plus bus powered drive. 2tb, USB 3.0.

I think it runs about 120mb/sec, which probably wouldn't cut it with higher res editing, but for what I am using it for I think it should work...

That is an OK drive as little bus-powered drives go. You could probably edit a single stream of H264 1080p on it. I have a few of them, along with many others: photos.smugmug.com/photos/i-nKGfmzf/0/X3/i-nKGfmzf-X3.jpg

OTOH if you ever must create optimized media, this increases the I/O load by about 5x to 8x, so you can quickly encounter an I/O bottleneck. Likewise multicam editing requires more I/O.

There was a specific problem with a single model of 3TB 3.5" Seagate drives some time ago, but in general I have not found Seagate drives less reliable than others. However I have had periodic failures in most drive brands over the years, so everything must be backed up.

The 4TB Seagate Backup Plus Fast is significantly faster than the Backup Plus Slim, and is really a better editing drive: amzn.com/B00HXAV0X6 However it is internally RAID-0 so that theoretically could reduce reliability but it really depends on the demonstrated MTBF of the individual drives, so you never really know.

If 1TB is enough, the HGST Touro S is a little faster than the 2TB Seagate Slim drive. I have about 15 of these (not all pictured above). It is one of the few 7200 rpm bus-powered USB drives: amzn.com/B00IVFDQFM

RAID 0 does not "reduce reliability".

It just does not allow for drive failure like RAID 5 or other similar format does.

RAID 0 should just be looked at as a single drive and would be as reliable as any of the drives that make up the RAID array. It's really no different than editing with a single external drive except for the benefit of increased performance.

As always - backup what you can't afford to lose. Have at least one clone of your primary editing storage device and preferably two.

I use an external 250gb and a 500gb LaCie Thunderbolt SSD for FCPX project. Bunch of LaCie HD's (Raid) for backups (Thunderbolt / FireWire 800). Works fine for me.

anickt wrote: RAID 0 does not "reduce reliability"...It's really no different than editing with a single external drive except for the benefit of increased performance...

RAID-0 definitely reduces reliability with all other factors held constant. IOW a 4TB logical drive composed of two physical drives in RAID-0 will be less reliable than a 4TB logical drive composed of a single physical drive of the same characteristics.

The reason I said "theoretically could reduce reliability" is because:

(a) The real-world reliabilities of the underlying physical drives is generally not known. Manufacturer MTBF and "bit error rate" numbers are not predictive, but more like a worst-case guarantee. IOW it is easily possible a single physical drive could have worse real-world reliability than a two-drive RAID-0 array of the same size, if the underlying drives have different characteristics (which is often the case).

The difference between manufacturer bit error rate and real-world error rate was examined by database pioneer Dr. Jim Gray: "Empirical Measurements of Disk Failure Rates and Error Rates"
research.microsoft.com/apps/pubs/default.aspx?id=64599

(b) Even IF a two-drive RAID-0 array has lower reliability than a single drive of the same size (which is impossible to determine ahead of time), that "lower" reliability could nonetheless be so good that it has no practical impact.

(c) Offsetting the theoretical lower reliability of the two-drive RAID-0 array is the higher performance and lower operational time relative to a single drive of the same total size. IOW if the MTBF on the single drive was 2x the RAID-0 but it had 1/2 the performance which necessitate 2x the operational time to achieve the same work, there might be be little difference in overall reliability.

(d) The theoretical diminished reliability of RAID-0 vs a single drive of the same size is often used to say RAID-0 should never be used or restricted to narrow conditions. That is misleading, especially when using integrated two-drive RAID-0 units sealed within the same housing.

I personally would not use a four-drive RAID-0 for critical editing but I have no problem using two-drive RAID-0 for backups or other utility operations. With the increased data load of 4k content, selectively using RAID-0 often means the difference between backing up content vs not backing it up since single-drive systems can be too slow.

I edit off of a T'bolt RAID 5 (4 bay) and it is great with 4K.
Also a USB 3.0 bus powered, 7200rpm OWC portable drive, works fantastic.
And an old Firewire 800 bus powered, 7200rpm OWC portable drive, and it holds up great with HD and UHD footage.
So, that's about all I know about.