[Cory Caplan]

3 entirely different decks made several years apart all exhibit the same behavior, specific to deck control being activated after the deck is turned on “late”

There’s always the possibility that this tech wasn’t able to find the problem either, but I’d think that these decks would exhibit additional issues, like random shutoffs or whatever, but like I said, this is the only “problem” I have with the deck, and I’ve seen it on 3 different models, working off both Avid PC and FCP systems..

It doesn’t happen every time, btw, just when the right set of circumstances arises… I’ve noticed it’s much more likely if I try to turn it on WHILE the computer is trying to access the deck (edit to tape is coming up) that’s what led to the original hypothesis, and for the past 3 years it’s stood.. what can I say?

Cory

[Cory Caplan]

yeah, dollars to donuts that’s it.

I actually exit out of FCP entirely now, drop AJA control panel out, then come back in to avoid this issue… As I said, I think it has to do with 422 protection. Like FCP just assumes its on, and sends it a control, but the deck was waiting for some sort of initialization protocol or something, so it shuts down to protect itself.

I’ll be darned if I can figure out how it “knows” to turn off and lock down for a period of time, even if unplugged. Some capacitor thing, I assume.

BTW, my colleague who had this issue 1st took his deck in 2x and it got a clean bill of health.. I wouldn’t sweat it.

Occasionally, I do forget though, and I actually had a client show up for dubs on the spot, and in my rush, I forgot, and this happened… Then I had to sit there and wait for the deck to “timeout” before I could restart it.. Awkward.

[Cory Caplan]

Yeah, I thought that was implicit in my answer, but I guess I shoulda said that more clearly…

[Cory Caplan]

I’ve had this problem with 3 different UVW-1800’s and it always manifests in the same way.. My question to you is, did you turn the deck on after FCP had initialized? I noticed a while back that these UVW-1800s all turned off, then could not be turned on for a few minutes ONLY when deck control was activated after a late turn-on.

I had this problem with AVID and FCP on different systems.. I assumed it was some sort of overload protection for the serial port.. My current deck is an 1800 and has done this since I’ve had it.. Last 1800 and a colleague’s reliably do the same thing.. (when turned on after “capture tool” is open in Avid, or after FCP/Aja initializes)

So the million dollar question is– when did you turn your deck on?

Cory

[Cory Caplan]

Hey man, can you email me the Aja Deck polling script fix? Having some “deck trouble” issues this morning and the search function sent me your way.

oops… email me at cory at spacecadet.com

[Cory Caplan]

Avid DV50 is a 4:2:2 codec, just like ProRes, unless I’ve been misinformed.. DV25 is chroma sampled 4:1:1, but DV50 codecs should be 4:2:2.

[Cory Caplan]

Good to know about DVCPRO HD v ProRes. Thanks for sharing the results. I suspected they would not be comparable, but it exceeded the scope, as I have no ability to to test that with uncompressed HD. Considering the bitrate for DVCPRO HD is about 1/2 that of ProRes HQ, and the resolution is 1.5 anamporphic, I’m not surpised there are big differences.

In my practical experience,most of us are still going to be dealing with primarily SD and pre-compressed 8-bit HD acquisition for a few years, as not many productions are conducive to lugging around a Mac & an IOHD.. 🙂

My point to all this was that I did the comparison specifically in SD, took extra pains to make sure I was testing accurately, and shared my results if anybody else was interested. I tried not to make assumptions or “interpret” the results when it comes to HD.

“Also, if using ProRes an acquisition codec as gets done with dv50 (ie recording straight out of the camera), there’s no doubt that ProRes wins in SD or HD.”

Keep in mind that DVCPRO 50 and DVCPROHD are very different animals… Their bitrates are almost identical (, but DVCPRO HD is crunching exactly 2x as much raw information into that space. (720×480)x(2) = (960×720) = 7MB/sec.

My test wasn’t perfect, but I think that it indicates there’s not much difference between ProRes & DVCPRO50 in the SD performance bitrate-for bitrate. I mean, if you have a modified 4:4:4 HVX200, go for it, but as stock HVX’s shoot 4:2:2 8-bit anyway, why go through the extra steps?? Or is there an SD test somewhere I haven’t seen?

And I defintely think ProRes is a great finishing codec for DVCPRO50 sourced stuff.. If you need your final SD output to be 10 bit.. AFAIK, color is 10+ bit internal, so it would actually work in 10 bit space, and even if you saved it ‘same as source’ you probably wouldn’t be able to detect any difference in the 8-bit world (DVD, qt file, Beta Sp, etc..) right?

My obviously opinionated stuff is obviously..well, obviously opinionated. 🙂

PS. Sorry about the rambles. After all of the testing and typing, I had no energy to go back and edit, and it clearly shows.

[Cory Caplan]

This is wrong. FCP is coded horrifically for rendering. On my machine, if Activity monitor says 35%, it’s 35% of ONE cpu. If it’s using all 4 to capacity it says 400%. (Quad 3.0)

If you think about it for a second, you’ll realize there’s no way drives are the bottleneck, unless you’re compositing many layers of uncompressed data.

DVCPRO HD, for example clocks in at 15-18 MB/sec– modern single drives have sustained throughput of 3-4 times this– yes it gets bigger once uncompressed, but this happens in memory & cpu– the drive is not the bottleneck here.

I’ve done the math many times on my timelines re: datarate of source material, and there’s no way drives are the bottleneck.

FCP is inefficiently written, renderwise. Once you fall out of the RT capabilities, they haven’t bothered to optimize it for multithreading, which is pretty inexcusable, considering how many generations of CPUs are multithreaded.

They didn’t bother before with G5s because of memory bandwith– the actual pipe going from bus to the memory to CPU was slow and tight– That’s why Avid Adrenaline on PCs kicked G5s to the curb– sure, the G5 could run a mean specint, but if you can’t get the data to the CPU, what’s the point– and if the developers were running up against a bottleneck, why bother making your cpu code efficient..

And now we’re paying the price. For the first time in a long time, the biggest limiting factor is not the hardware, but the bad code.

And they’ve taken their bad practices to Motion as well– Just throw the stuff to the GPU and let it handle it, then come up with some lackluster code for when you exceed the GPU capabilities. Aftereffects absolutely destroys motion on basic, basic things like moving text. How can you screw that up? Plus, layer a few effects or put a dozen instances of a (seemingly simple) motion template on the timeline, and the system slows to a deathcrawl… I hope Final Cut studio 3 is just final cut studio 2 that actually works well. (That includes a sane media manager)

[Cory Caplan]

I finally put two and two together. I just assumed that prores was another Apple reality distortion, but this makes sense… people love it for keying, sure– because it smooths out and erases detail.

I actually prefer DV50 vs prores SD– I ran lots of factory auto footage through both, and believe me, the dv50 has more detail– at a smaller footprint. Much crisper looking– and much more like the original tape. Of course dvcpro50 is only 8 bit not 10 bit, so that is something to consider, but I think considering.

1) DNXHD came out a year and a half before hand.
2) Prores visibily alters the grain in the footage (Who here is actually shooting on a “lossless” system anyway?)
3) Apple is stupidly protective of the codec, — I can’t send my PC Avids dnxhd footage, nor can you even send macs that don’t have FCS a QT with prores (after effects artists, etc..) This is just childish protectionist bs.

I can, however, generate a DNXHD quicktime with FCS but no Avid and send it to another mac with no avid, and they can use it in aftereffects etc..

DNXHD, with very similiar bitrates, hides data in grain (It seems to me anyway) I find this much more aesthetically pleasing when working with grainy sources (just about any modern HD video camera or film transfers)

Most of my HD stuff is DVCPRO HD these days– either varicam or p2, and I notice smoothing taking it through color, and then through prores (which is now the only compression option– they disabled ‘same as source’– uncool guys.)
However, because Red camera uses a wavelet based raw data compression technology, this might actually be the best application for a codec that doesn’t like “noise.”

Keep in mind, though you may think that this “noise reduction” is a feature rather than a detractor– this means that edges aren’t as crisp as the original footage either…

Cory

[Cory Caplan]

Actually, the irony about your implication about “quality product” is that the x1900XT does have significant issues, a la the ‘red ring of death’ on the xboxes previously mentioned.

You may recall a few rants from me earlier last year about instability, lockups, bugs etc. I recently ‘upgraded’ to leopard, figuring that with the most recent round of updates, the FCP team is going to be focusing way more heavily on 10.5 than 10.4, and I’d take my chances over here on leopard.

After installing Leopard, I experienced the ‘Screen Corruption’ issues people discussed specifically with the x1900xt. Furthermore, after a lot of digging, I’m hoping that this issue is key to my hard freezes that I was having on Tiger.

Apparently, gamers playing WoW on macs ran into this first, as it pushed the video hardware the hardest, and apparrently leopard pushes the x1900 harder than Tiger did (I suspect they increased the clock in the driver, to make leopard/Quartz extreme seem “snappier”) Which of course, revealed these defects in the cards to many.

After digging and digging, I found an internal apple reference number, that once mentioned, was like saying “open sesame.” They actually agreed to swap out my x1900 xt out of warranty, without my exploding beforehand.. (Which is good, because If I would have had to pay for a manufacturing defected 2+ year old video card at “full mac retail” I probably would have firebombed cupertino.)

So, anyway, I’ve had it in a few days now, and no more artifacts. I also turned off all sleep and power management, which I think has major issues on both Tiger & Leopard with FCP, especially if you’re running a wireless mouse. (infinite beachball, UI hangs, etc.

Unfortunately, many bugs and oddness remain in FCP, but I guess that is going to be the standard for “professional software” for the forseeable future.. (I just cancelled my 3D Studio contract, because I was tired of an annual release schedule that NEVER fixed the bugs…)

So, there is quite a decent amount of truth in the original posting. I have actually moved some of my workflow back to AE & PC/AVID because of the uber-crappy speed and lack of stability in motion (fcp integration, esp.), and the general bad behavior & lack of communication on Apple’s part with the Pro apps & never-fixed bugs… (QT 7.4, “recommended updates” that you can’t uninstall, but have no list of bug fixes, etc.) The point is, they don’t care.. they want to sell software to the iPhone crowd (wait, does that include me?) not truly “pro apps” to professional users, or else they’d be at NAB, ESPECIALLY in the absense of Avid.

I incorrectly thought having a single provider of (insanely limited spec) Hardware, Software, O/S would result in a more stable platform, but I was wrong– it just means that it’s too much under one house to keep track of– Or a wash, in other words.