David Battistella
Forum Replies Created
-
walt,
you beat me by six seconds on that last post.
geez.
david
Peace and Love 🙂
-
Automatic duck.
It is expensive and worth it.
you can also try XML.
David
Peace and Love 🙂
-
Uli,
To really test this you should capture about a half hour to fourty minutes of uncompressed 10 bit hd clips and then report back the speed and whether you are getting dropped frames.
Many drive tests will report adequate speed, but you are perpetrating a bit of a myth by reporting raid 0 speeds on empty drives on an internal array.
I am not trying to be a hard ass or rain on anyone’s parade, but the only HD that raid is really good for is DVCPRO HD.
David
Peace and Love 🙂
-
I need you to post your system details, without this information, posting answers to these questions is futile.
David
Peace and Love 🙂
-
Much of this depends on your set up. You can work with 10 bit uncompressed SD material on a SATA raid, or you can try to push that much data on a G-RAID firewire 800 Drive, but if you are woking off firewire, I would ask for the material in DVCPRO50 format.
There are many variables depending on your configuration but it is definitely an external drive solution.
Remember the single largest factor for RT in FCP is drive speed. The faster the drives the more RT you will have. The processor, the ram, etc factor, but none as much as the speed of your hard drive system. You want a fast FCP system, by the fastest drives you can find.
David
Peace and Love 🙂
-
Yes,
In SD land we all tried to run sonnet cards and internal rive and such, but the rality with 10 bit hd is that requires drives that will sustain a data rate of about 165MB/sec. That is a full SIX TIMES faster than uncompressed SD at 27MB/sec. Drives are still running at 7200 or 10000 rpms. Even striped across many drives in a raid zero (which is just a stupid way to operate from a business standpoint) and even if you are aprroaching these speeds, you will still need headroom. Yes, serial ATA in the box is cheaper, we all want it to be a solution, but can you bank you busines on that? I won’t risk that anymore. No way!
There is the “well in theory” approach and the “what actually works” approach. If you want things to work and you want security for your data then I can’t say that this is a smart approach. It might “work”, “sort of”, but HD is much more taxing on the whole system.
David
Peace and Love 🙂
-
The HUGE 4210 is a TANK. It goes and goes. It has plenty of speed, even when close to capacity and paired with a Celerity card it pushes a lot of data.
Take a look at their new drive system as well.
David
Peace and Love 🙂
-
This is like cooking a steak with a candle. It might work, but how good is it going to be.
You want to play in HD land, you need to spend $.
my two cents
David
Peace and Love 🙂
-
On the second part of your post.
FCP requires that you edit in ONE codec only. Every other codec introduced will require rendering to the codec that the sequence is in.
The only time you could get away with multiple resolutions is if you were to use different levels of photo J-peg compression anywhere from 1000-to 10000K, similar to AVID’s 20:1 or 2:1 but youcan not put 2:1 together with 1:1
David
Peace and Love 🙂
-
Annimation and None are 4:4:4 codecs that will only play back with a very fast drive array.
If you want to edit in these formats you have to understand that they are not RT enabled (as the uncompressed codecs or DVCPRO codecs are).
So to answer the question in the most basic way, you are using codecs and a color system that is not optimized for use in FCP. You will always have to render a dissolve when editing in annimation or none. In fact, you would not even be able to edit in this color space in and AVID.
Neat, huh?
David
Peace and Love 🙂