MBS Productions - MBS Productions – Josh and Jason Diamond

So GLYPH contacted me recently to say my unit was able to be upgraded to some new features based on my and others feedback about the initial unit I blogged about HERE.
I sent it back and it returned with a nice new feature:

REMOUNT

That now means that when you turn off the unit it no longer destroys the temp drive set info prompting you to erase the drives you had been using in that set.

So to be clear, you can now power cycle the drives or the Triplicator unit and your drive set of 2 – 3 drives will remain intact data and all.
The only thing, which has zero impact on workflow, is that once the drive set is created, the drive that sits in the #1 slot must always be there since that’s the drive that hold the stripe info. Drives in the #2 and #3 slots can me switched to either port with no issues. I would still recommend keeping each drive in its initial slot after you create the set for organizational purposes.

The unit came back with a new faceplate:

and a new Backplate:

Overall this is a solid, robust unit def worth the price. Whether you use it on-set or back at the lab to clone drives for a client, it will pay for itself pretty fast.

I recently bought the Glyph Triplicator as a possible new tool for our digital media arsenal. We shoot a lot of digital media and mostly RED which means we generate a lot of data that needs to be copied, maintained and kept secure at all times and generally on set. Copying to multiple drives is a time consuming process that we’re always looking to accelerate but there are definite specifics we need. We need to verify that the data copied is bit for bit correct. The verification process is called a Checksum and it tells us that everything is or isn’t ok. We also need to be able to move the drive set(s) as the production day dictates so being mobile is a big factor. It also comes with a nice solid plastic Glyph protective/carrying case.

Glyph Triplicator

On to the Triplicator:
When you set it up its pretty self explanatory. There’s a power button on the front and one small LED on the bottom right.
On the back there’s 2 sections. the HOST section has 1 eSATA and 2 x FW800 to choose from to connect the Triplicator to your computer.
The other section simply has 3 identical eSata ports for the Drives in your set. So make sure your drives are eSATA before you leave for your gig or you’ll be making manual copies the whole day. You can use 1, 2, or 3 drives with the unit, it’s up to you and how many identical copies you want. It obviously wouldn’t make sense to use just 1 drive but hey that’s up to you.
There’s also a fun, big red button on the back but I’ll get to that in a second.

Glyph Triplicator back panel

Plug the drive(s) into the Triplicator and then the Triplicator into the computer and turn it on. The LED on the front will blink blue telling you it’s time to push the big red button on the back. When you do, the internal dual RAID controllers do their thing creating a RAID 1 drive set. When it’s done, the blue LED will flash red, blinking a number of times to let you know how many drives you have in the set.
This seems a bit obvious but if for some reason one of the drives stops working or is removed from the set, that light will actually update to show you the remaining drive number. A little subtle in the warning department but it’s status info nonetheless.

You will get a warning dialog on screen when it’s done that the volume cannot be mounted and the standard: Initialize, Eject, Ignore options. Just hit Initialize and it will take you to Disk Utility where you name and format the volume as you see fit. You can also use any program you normally use for this function. Once it’s done you now have a single mounted volume that is creating 3 discreet copies FROM THE ORIGINAL FILE when you put data on the drive, not copying from drive to drive as would happen if the drives were just looped through.

Transfer speeds are as you would expect from a single drive, bonus is you’re getting 3x the data. If you unplug the drives from the Triplicator and plug them into a computer one at a time you will find that each one bears the exact same name as you named the set and each logically has the exact same media.

If you haven’t turned off the Triplicator you can actually plug the drives back in and the set will remount, ready for action with data intact.
BUT
Here comes the pain.
If you turn off the Triplicator and back on, the drive set won’t be recognized as it’s a temporary RAID set.
If you push the red button again with those same drives, YOU WILL LOSE YOUR DATA REBUILDING A NEW SET.

That’s right, so if you were on set and turned it off you would need up to 3 more erasable drives to use, otherwise you would erase ALL your data from the ones you used for your initial set if you wanted to use the same drives again Also since it’s using RAID 1 if you tried to checksum the data using say R3D Data Manager you would only actually be reading from 1 of the drives not all 3 as that’s how RAID 1 achieves its performance so you could not verify a checksum against all 3 copies.

After all my testing and such I reached for the feedback phone number handily provided on the documentation. Apparently they’ve only made 100 of these units so far. I got mine at Tekserve in NYC and mine was only the second unit purchased of the 20 they had in stock.

I had a long and very informative conversation with the product manager at Glyph. They are well aware of the issue and he acknowledged the main feedback note they’ve gotten is to be able to create Drive sets that will maintain themselves regardless of the how many times you power cycle the Triplicator. He said he had seen it working in the lab so hopefully that feature is not far from being released through a firmware update that needs to be applied in their lab. I’ll have to send the unit back to upstate NY because they apparently have to update the actual firmware on the RAID controller cards.
We also had a long discussion about checksums and data verification and they’re looking into it and most likely will have to provide it internally in the unit based on the RAID 1 issue.

End result is this is a great first step and with these minor changes they already have in the pipeline this unit will be a great addition to any data centric workflow and I know it will save us quite a bit of time on set.