Hackintosh for Resolve

[Nikolai Vavilov]

I think your supplier sucks.:emote_hippie:
I have a question for all of you: If raid 0 is so unreliable, what happens if Red drive fail after a day of shooting exclusive documentary material? Or you say, hold your thought Mr.President, we have to change the drive in order to make backup, or maybe you say to the capricorn to repeat that beautiful jump.

I'm also have some experience with big external 24/48 disk RAID systems and also can say - drive fails happens pretty often, doesnt matter which brand or model they are.

Red drives also looks pretty risky for any serious work - we used at least 2 RedDrives per camera and we try to change them back and forth and make backups as often as possible.

 

[Uli Plank]

Same here, the RED drives are risky. I think there's a reason why RED is getting away from them.

 

[Aleksandar Colancevski]

Hey guys, I don't want to go too technical, but if you use it on moto "don't be gentle it's a rental" then failure can be expected. Maintenence is the name of the game and nobody cares except the guy who gives the money.
For me it works until now.
Also I believe in Reds testing methods for realibility and never regret it.

 

[Uli Plank]

Well, the drives are not made by RED, and all drives fail sooner or later. A RAID-0 quadruples the risk of loss for the same amount of data, since one drive will kill all content, and the RAID doubles capacity.
If you have ever shot in a chopper or a fast car with stiff suspension, you will have experienced dropped frames with the RED RAID. It's just advised to backup as often as possible. I'm very grateful that SSDs will be soon here to get rid of that technology.

 

[Jan Prochazka]

I think it depends on workflow. While your everyday "two capucino" backup is fine for you, I can´t imagine it here.
Notice: you can hardly notice performance dropdown during on-the-fly rebuild of raid 6.

 

[Thor Melsted]

Maintenence is the name of the game and nobody cares except the guy who gives the money.

No, protection is the name of the game. While important, maintenance is not as big a factor, it just makes drives last a bit closer to their estimated life cycle. Hard drives are mechanical with parts that move very fast, and they will all fail eventually.

For me it works until now.

:banghead:
With all due respect, I find your attitude towards data redundancy pretty careless, and something one can only expect from someone who's never lost data. Just because you haven't lost a drive, doesn't mean you never will. Relying on luck and knock-on-wood is bound to come back to haunt you at some point, and most likely when you can least afford it.

Just my opinion...

 

[Aleksandar Colancevski]

I am not saying that using Raid 6 or X is stupid, I would use it too.
Just saying that for single CC- editing suite is not nessesery.
I went on training configuring XSAN so I really understand what you are saying.
But for Red R3D and ProRes codec, a single disk setup is a must.
I am using 2 different e-sata external drive backup for recorded R3Ds.
For complex multistation, tens of Terabyte of data serious disk array is a must.
But thinking of building yourself serious raid is not very serious.(in case you are not experienced engineer)
My mathematics has proved that 100 Capucino is much cheaper than good quality raid solution.(if have that luck to have over 50 failed drives)
I have already backed up my source, DaVinci is a realtime, backing up FCP project, edl or DaVinci project is a one second job, so I don't get you what is there to be afraid if drive fails.
And Yes Uli, I have over 50 commercilals and 2 feature movies recorded with my Red.
I am well aware of the Red drive sensitivity. But then again if you know how to balance on the string, you will not use CF cards so often. Ofcourse we are all happy that Red put SSD in the RedOne too.

 

[Aleksandar Colancevski]

Hint: I allmost forgot to tell you my recent event. A frend of mine had failure of the power supply into one of his computers. It resulted with fried electronics on all the drives in it. He had luck that one of the same drives series sits on the shelve so we changed only the electronic plate on the failed drives and managed to take out all the data from them.

 

[jake blackstone]

I can see, that this discussion veered off the main subject a while ago, so I'm going to add to this goal a bit more.
In many locations, Eastern Europe included, low quality of power distribution network leaves said equipment at the mercy of quality UPS. Power can and does go out at most unexpected times in duration frequently exceeding UPS power reservoir. Personally, I can't imagine using RAID 0 in such locations without major loss of data at some point...

 

[David Kellermann]

I can see, that this discussion veered off the main subject a while ago, so I'm going to add to this goal a bit more.
In many locations, Eastern Europe included, low quality of power distribution network leaves said equipment at the mercy of quality UPS. Power can and does go out at most unexpected times in duration frequently exceeding UPS power reservoir. Personally, I can't imagine using RAID 0 in such locations without major loss of data at some point...

You don't even have to go to Eastern Europe to have problems with the electric network. I'm living in Germany, close to the Swiss and French border... last week, out of nowhere, the power supply struggled for 30 seconds, with all the lights in the room flickering (like in a bad horror movie). I'm sure glad that all our electronical equipment is behind UPS. Especially since they don't cost a fortune, if you don't want to run without power for more than a couple of minutes. The problem is returning since then and the electrician is clueless where the problem lies...

 

[Aleksandar Colancevski]

I can see, that this discussion veered off the main subject a while ago, so I'm going to add to this goal a bit more.
In many locations, Eastern Europe included, low quality of power distribution network leaves said equipment at the mercy of quality UPS. Power can and does go out at most unexpected times in duration frequently exceeding UPS power reservoir. Personally, I can't imagine using RAID 0 in such locations without major loss of data at some point...

Right, we are way off. Because we are disscusing Hacks I think exchange experience in using certain parts is needed. Regarding your comment about the electricity here you are right. We tend to use quality UPSs. In the case I described it was the falty Asus video card that draw a lot of current from Chieftec power supply which inner protection failed to react.

 

[David Kellermann]

My Hackintosh finally works (see thread http://reduser.net/forum/showthread.php?t=54430). The Areca 1880i RAID0 with 8x 1TB Spinpoint F3 is flying... Blackmagic speed test gives out strange results, since I have no idea how to disable the OSX system cache. On the Aja benchmark, I get 3500MB/s read / 2000MB/s write speeds (the write speed is going down to 860MB/s which seems more realistic)

Right now, I'm securing all data on a daily basis on our server, running Windows Server 2008 R2, on a LSI 9260-8i (with BBU) RAID5 with 6x 2TB Samsung F4.

The transfer speeds are ok (Gigabit ethernet), but on the long run I have to find a better solution for backing up these huge data amounts.