Sean Oneil
Forum Replies Created
-
What do you guys mean by “it’s not true shared storage?” Do you just mean that it’s one of those systems that only allows write-access to one person at a time, but everyone has read access?
Sean
-
[Peter Wiggins] “Take a look at this before going Terrablock
https://www.lairdtelemedia.com/ “
Just curious, why was everyone talking about that thing? 12TB ethernet sharing for $20k. It looks like a total rip-off to me. Am I missing something? The website doesn’t have shred of detail. Doesn’t say if it’s a NAS or a SAN. Doesn’t say anything.
Sean
-
Trying to do this with uncompressed is pretty hardcore. Here are some tips:
– Make sure Jumbo Frames is on. It has to be enabled on both ends, as well as the switch.
– Make sure hardware Flow Control is on. Again, both ends as well as the switch. Flow Control will actually slow your throughput, but throughput is more consistent and stable (meaning less dropped frames).
– Don’t even use a switch if you don’t have to.
– Run Leopard, not Tiger. The TCP stack is newer, better, and faster. And it uses a newer version of AFP.
Those things will improve your status quo, but honestly, AFP is probably not going to work well no matter what you do. Uncompressed is too demanding. Final Cut runs a ridiculous amount of background services during Log & Capture that makes it choke at the slightest hiccup. And it’s even worse for Edit To Tape.
You could try the Blackmagic Capture Utility. That might be better.
Sean
-
SD can can be 30fps or 60 half-frames a second – depending on how it was shot and with what camera.
If you shoot at 60p, you can still render it as 30p later on if you want, and it won’t look any different than if you had shot it at 30p.
Sean
-
Did you check the help guide in Toast 9? I’m sure that explains it. I don’t know anyone who has Toast 9 yet (did it just come out today?).
If it’s anything like creating DVDs on Toast 8, then it will encode it for you. You just feed it any Quicktime file.
Sean
-
This is a common problem with many SATA RAID controllers. Certain makes and models of disks will mysteriously lose sync and disappear.
It’s Sonnet’s responsibility, but if you’re desperate and impatient, it may be worth calling Seagate and seeing if there is a firmware update for the ST3750640NS model. And if there is, let Sonnet know about it and ask them how to proceed.
Sean
-
[gary adcock] “My tests at the time showed an average error of only a .5% ( 1/2 of 1 percent) difference that could be compensated for if very careful. When the original camera version captured via FW was overlaid with the “baseband converted copy” in “difference mode” The output of the combined file was then viewed on a Leader 5700 scope to highlight the issues. “
You’ve completely missed my point. I’m not arguing the validity of the workflow. I’m not saying it creates any noticeable loss.
My sole point was that it is NOT native. That’s all. You’re own test, showing a .5% difference proves it. Do I think .5% is a big deal? No.
[gary adcock] “That being said, I am “now” only a fan of camera codecs ( DVCPROHD, SxS, RED), for capture, at which they excel at. However after working in ProRes I will never go back to editing camera native content again. “
ProRes is the only codec I’ve ever used since FCS2 came out.
Sean
-
[Jeremy Garchow] “No. Never have. But I ask again, what does native get you? “
It saves a generation. Instead of decoding it, re-encoding it, and then decoding it again, with native capture you are just decoding it once. There is no re-encoding.
[Jeremy Garchow] “OK. So you trust a Quicktime to decode an already lossy format, and not a hardware Kona card to do it in an uncompressed space? “
You’re taking old knowledge of the Apple software DV decoder and applying it to the Panasonic DVCProHD Quicktime component. They are not the same. DVCProHD is 4:2:2. Decoding it to 4:2:2 U/C is a trivial task for the decoder. It’s not the same as when software DV (4:1:1) decoding cause poor chroma up-sampling.
And I’m pretty sure that the Quicktime component (made by Panasonic, not Apple, I believe) uses the same algorithms as the SDI option board on Panasonic deck. I’m not even sure a Kona has it’s own decoder. AFAIK, it still uses the Quicktime component. It just provides hardware assistance to reduce the CPU load. In fact I’m quite sure that’s the case. Their sales language says “hardware accelerated” for DVCPro HD. If it had it’s own hardware codec, it would just say “hardware” without the “accelerated” part.
[Jeremy Garchow] “Don’t get me wrong, DVCPro HD is an awesome in camera acquisition codec. It just doesn’t hold up to much editing/generation loss. “
Well I agree with that, but I haven’t brought it up because it’s not the point I was making. Walter has time and again made a fervent argument that the generation loss is trivial. I had no desire to engage in that discussion. Saying it’s just as good as native is one thing. Saying it IS native is another. That’s what I had a problem with.
Sean
-
[Jeremy Garchow] “Oh yeah? Why not? When talking about video, what does native get you? Why does editing in lossy DVCPro HD gain you anything at all? The point is not to lose anything further. “
If that’s what the point is, then you would want native capturing. But I fail to see your point. Do you consider “HD-SDI->DVCPro Re-encoding” a native capture?
[Jeremy Garchow] “So when Quicktime decodes it, it’s an exact duplicate?”
No, it’s not. But when you output via Firewire back to DVCProHD tape, it is.
Sean
-
[walter biscardi] “You can call it what you want”
That’s your problem. You’re calling it what you want, not what it is. Firewire capture is native. SDI capture is not. You’re free to say a BMW is as good or better than a Mercedes. But you don’t call a 525i a Mercedes, because it isn’t. It’s a BMW. You’re confusing semantics with actual defined terminology.
[walter biscardi] “I don’t consider it false information if we’ve used the workflow on about 100 HD Broadcast Masters to date “
What does that matter? I created hundreds of broadcast masters in the 90’s using a Media 100 and a BetaSP machine. The client always had the BetaSP masters dubbed to D2 at a local facility prior to creating hundreds of dubs that were sent all over the world. That didn’t give me license to say “My BetaSP masters are digital.”
[walter biscardi] “Comments like that about AJA annoy the crap out of me too. AJA is the first to tell you when you should and should not use their gear.”
In my experience they’re a great company. I was basing my comments on what they allegedly told you. I didn’t want this to get ugly. So let’s drop it. I’ve already said there’s nothing wrong with your workflow. My only problem is the misuse of terminology.
And if you want to know why I’m so militant about this, one time I rented out my AJA-1200A to someone for a week. It had the FW option but not the SDI. He brought it back the next day because his editor told him using Firewire is inferior because it re-compresses the video, and that you need a Kona to maintain the full quality. I would not be shocked if he was one of your readers.
Sean