Talk:RDX (disk)

RDX technology is _not_ similar to REV technology

This has recently turned out to be an important practical point for users of the Retrospect Windows backup software, which still allows definition of Backup Sets with a legacy Removable Disk device type that was intended for "super-floppy" devices such as REV. Until the last major release of that software, the text for the appropriate UI dialog said that RDX devices as well as REV devices should be defined as Removable Disk—which caused problems for users of RDX devices.DovidBenAvraham (talk) 00:20, 25 September 2016 (UTC)

Put back in an explanation—which is obvious from comparative statements in the two articles—of the key difference between RDX technology and Iomega REV technology. This shows that RDX technology should not be considered as a type of "super-floppy", which is the most important point for Retrospect users. I'll deal with the fact that RDX cartridges are no longer more shock-resistant than some modern 2.5-inch hard disk drives later; I'll have to use references to concrete-drop-test results for some hard disk drives, and that's not so important as the explanation I've already put it. DovidBenAvraham (talk) 04:36, 23 December 2016 (UTC)

REV is not a superfloppy, it's a removable hard disk. Superfloppies have a soft, 'floppy' medium and the heads touch the medium. --Zac67 (talk) 09:06, 23 December 2016 (UTC)

Now that I've put in links to this article and the REV article from Backup#Storage media, I'd like to point out that there really is a close conceptual relationship between Iomega's REV—as well as the Iomega Jaz preceding it—and the superfloppy. Both the REV and the Jaz continue the floppy/supperfloppy concept of physically sliding components into a media-containing rigid (since Sony's 3.5-inch floppy) cartridge in order to write and read the media. "Let's make the medium rigid, and slide in floating heads instead of a rolling head" gave Iomega the Jaz. "Let's leave the motor in the cartridge, so we at least don't have to slide in a spindle" then gave them the REV. "Bugger it, let's leave the heads in the cartridge so we can eliminate all physical holes" was the concept behind RDX technology, but it took ProStor Systems Incorporated—not Iomega—to come up with that true removable version of a modern hard disk. DovidBenAvraham (talk) 14:57, 15 April 2018 (UTC)

... and you can easily extend this chain of thought to toolless disk bays where you just slide in any generic HDD – I'm using a few of those. Then you've made the full circle from floppies to Winchester-type harddisks. It's completely all right to mention these technologies as being similar in handling from a user's POV', but technically, they are vastly different. There are clear distinctions between a) soft medium, touching heads, and b) rigid medium, floating heads, removable medium, and c) rigid medium, floating heads, hermetically sealed unit. It's not by chance that only the latter has survived by today. Yes, there were hybrid types like Bernoullis and Zips, but these only had comparatively short lives. --Zac67 (talk) 17:13, 15 April 2018 (UTC)

But we should take a user's POV into account in the article, because I hope it's going to be read by users needing to know what RDX is. Let me remind you what motivated me to expand the article's second paragraph in September 2016, and also to start this section of the Talk page. We had several administrator users of the Windows variant of the Retrospect backup application who ran into trouble because they defined their RDX drives as members of "removable disk" backup sets. One administrator user even referred to his Dell RD1000 cartridges as "tapes", and had great trouble correcting his terminology even after my friend pointed out that RD1000 is Dell's trade name for RDX. Thus I conclude that, for non-technical administrator users—especially ones with long memories, the pop-in/pop-out switchability of RDX cartridges makes them conceptually have a lot more in common with floppies/superfloppies/tapes than with Winchester hard disks. DovidBenAvraham (talk) 21:09, 15 April 2018 (UTC)

What happened is that, one day before the official major-version release date of Retrospect Windows 11.0 on 1 March 2016, the head of Retrospect Technical Support put up a YouTube video saying that RDX backup sets should be defined as type "disk" instead of "removable disk". That didn't stop administrator users from trouble-causing erroneous definition; therefore in September 2016 the Retrospect developers defaulted the point-release Retrospect Windows 11.5 version's "Can create Removable Backup Sets" option to Disabled, changed the text of the Create New Backup Set dialog to show RDX under the Disk option instead of the Removable Disk option, and wrote a wishy-washy Knowledge Base article that begins "Disk backup sets are a better choice for RDX drives than removable backup sets". The head of Retrospect Tech Support then posted "A large number of customers use Removable Disk Backup Sets and have existing removable disk backup sets", and he was writing 7 years after the last Iomega REV drive was manufactured. My overall point is that this is an example of applications developers needing to take account of sociological/legal factors as well as purely technical factors. If you look at my Backup#Enterprise client-server backup section, you'll see that the major reason for a majority of features described there is because backup administrators—as defined in the last sentence of that section's lead—need them rather than because the features are necessary for purely technical reasons. We editors of Wikipedia articles on computer facilities need to take into account the sociological/legal POV of users of these facilities, not just the technical POV of developers. DovidBenAvraham (talk) 03:12, 16 April 2018 (UTC)

Another way to appreciate the sociological/legal POV of users of these facilities is to consider the comparative prices administrator users are willing to pay. A quick Google search today shows me that, for 1TB native capacity (which ATM seems to be the smallest non-closeout size sold), you will spend more than twice as much for a new RDX HDD cartridge as you will for a portable USB3 HDD. Add to that the cost of a USB3 RDX docking station, which is an extra cost roughly that of one new 1TB RDX cartridge. So why are administrator users, or their bosses, willing to pay the price premium for RDX over ordinary portable USB3 HDD? It is unlikely to be for extra durability, since the better portable USB3 HDDs can now also survive a 3 foot drop onto a hard floor overlaying concrete and have a mechanism also proven to last 10 years. It appears instead to be for a sociological reason; administrator users without a technical background just feel more comfortable inserting and ejecting a cartridge from a docking station than they do inserting and removing a USB3 cable from a portable HDD. For this very reason, my friend was originally considering switching from DAT tape to RDX when he re-activated his use of Retrospect 3 years ago. Other posters on an Ars Technica forum convinced him that using USB3 portable HDDs would be just as easy as well as cheaper, and he can now swap the USB3 cable between two portable HDDs in less than a minute. It could also be for a legal reason; auditors (with whom my friend doesn't have to contend) may have been brainwashed years ago into thinking that RDX was more durable for archival purposes. DovidBenAvraham (talk) 02:48, 17 April 2018 (UTC)

Cartridge lock

The article claimed the drives have a custom firmware, citing the video as source. On closer inspection, the video never actually examines the drive in such detail - merely attempts to access it, with the speaker then speculating that the drive may have some custom firmware, configuration or formatting without any attempt to determine which of these possibilities is. I've disassembled one myself, and it appears to be a perfectly ordinary drive but subject to an ATA security lock, which is perfectly consistent to what the video observed. As this is original research though, I've not included my own observations in the article - merely removed the mistaken conclusion. 82.24.112.131 (talk) 08:07, 25 August 2018 (UTC)