Serious Magic Announces the DVCProHD Decoder for Sony® Vegas® and Adobe® Premiere®

That's no different than a 6-pin to 4-pin cable -- if the cable (or the adapter) are defective and let current get onto the data pins, presto, you've got a blown DV port.

If the computer only has a 4-pin Firewire port and has no means of sending current out, then the problem is eliminated.
 
I read your post too quickly and missed the mention of 6-pin to 4-pin cables. What I'll do is use an adapter but test it first by buying a 4-pin firewire cable, cutting it in half, stripping the wires, and using a voltmeter to make sure there's no power-level voltage across any of the wire pairs. The power voltage should be easy to differentiate from the signal voltage. Since it's a rigid adapter, I'll trust that if it works at first, then it'll keep working. (I'm making the assumption that the flexing wire is part of the failure mode of the 6-pin to 4-pin cable problems.)
 
I tried the 3 day trial on a Core 2 Duo e6400@2.13ghz, 1GB RAM. The clip was 720p, 24fps. I would have tried a 1080 clip, but I didn't have one. Rates are with no effects applied.

Vegas 5d:
best auto (640x360x32): 20fps
best full (1280x720x32): 5fps
good and below: 23.976

Vegas 6d
best auto (640x360x32): 20fps
best full: 5fps
good and below: 23.976fps

Vegas 7 Trial:
best quarter:23.976fps
best half: 20fps
best full: 5fps
good and below: 23.976fps

With no effects in 'Good', framerate drops ~8-10 frames during dissolve transistions. With a short stack of glow, color curve, color correction FX, they drop about half, but workable in draft mode. Overall, it's still very usable, and this is on a lower-end Core2... imagine the virtual quad-core Core2 Extreme Edition.
 
qazwsx said:
imagine the virtual quad-core Core2 Extreme Edition.
That's assuming that your software can take advantage of them. I have a P4D 840EE, which has two HT-enabled cores (four logical processors total), and most software can't take advantage of it.
 
gazwsx, what were your project settings? Did you match the project to the clips, or just use a default project setting?
 
THoff, true, but as far as I know, Vegas 6 and up have dual core and HT support and that's the software I want to use, so it's all gravy - well for me at least.

Karl, I'm not sure if I did this correctly as I haven't done any HD editing in Vegas, but I used a 1280x720/23.976/Progressive/1.0PAR project - I had to change the PAR of the clip to 1.3333 though. If this is somewhat or totally incorrect, please let me know - these settings seemed correct based on the performance I was getting.
 
Barry, on Serious Magic's web site they have a graphic showing that it makes QT files for FCP, which lead me to believe there was a Mac version of this converter.

Now I go back and re-read, and it's somewhat misleading. It's not, as I read more carfully, realted to FCP, except that you can copy the QT files back to a Mac. So, why even mention FCP if it's not a Mac product? Sorry for the confusion, but since they keep mentioning FCP on the site so many times, I thought there was some benifit for FCP users, MOST of whom, in my experience, do not use Windows ever.

BTW, I don't run BootCamp nor Parrallels, I've never had a need to.

And I've never experienced the "glitch" you keep refering to. The FCP/HVX users I work with personally have not had the "glitch" you keep refering to. As far as I can tell, it's a small minority of users, and not a universal problem. I've only read about it here, never actually seen it happen. I know from my previous career that every computer system ever made has a "gotcha" somewhere. In the word of computers, there's no such thing as 100%.
 
gazwsx, my point is that most of the software doesn't scale well. Most of the video tools today will to varying degrees take advantage of a second processor, second core, or a logical HT processor. Beyond two processors, all bets are off, and that is the situation I am encountering with my P4D 840EE. Even if you are used to seeing 100% utilization with Vegas on your current system, you may still see only 50% utilization when you get a Kentsfield processor.

Any software that's more than two years old and was designed for HT-enabled processors is in fact very unlikely to be able to take advantage of additional processors beyond the first two.
 
Sorry for the confusion, but since they keep mentioning FCP on the site so many times, I thought there was some benifit for FCP users,
Well, first of all, it allows users to use a PC laptop but save files that are Mac compatible. For a lot of us that is a benefit. Serious Magic has never made a piece of Mac software, as far as I know, and they're not claiming it to be so. There's nothing "misleading" about it, there was just a misconception on your part.

Second, regarding the glitch -- there are reports on the web, there are screen captures, and nobody knows what's causing it. We only know that it's related to importing cards on a mac. Turning a blind eye to it is, I believe, a big disservice. I'd rather have people actively trying to figure out what it is that causes it so we can get it FIXED. Pretending it's not there, or just saying "well, doesn't happen to ME" doesn't advance the cause of getting it fixed.

I have had occasion to work on several mac-based large-scale P2 projects recently, and the attitude boggles my mind.

Them: "I'll just import the footage straight off the card."
Me: "yeah, but won't that cause the glitch?"
Them: "Yeah. I'll just cut around it."
Me: WHAT?

I don't get it. Why don't people take this more seriously, and FIX IT? Ben, what is your workflow? You say you haven't had it -- are you importing direct from the card into FCP? Are you offloading to an external hard disk using P2 Genie? What's the combination? What WORKS reliably, and what causes the glitch?

That's the part I'm interested in. There is no doubt that there's something going on out there, there are too many reports of it (just like there are reports of G4 PowerBooks getting slower and slower and then locking up when using cards directly). This isn't a "pc is superior" thing, this is a "there's something wrong in the mac side of things, let's figure out what it is and fix it" type of thing.
 
Regarding distribution resolution:

Barry_Green said:
No one? What about everyone with a DVCPRO-HD tape deck? Lots of people distribute to broadcasters and post houses at 960x720. And I'd rather stay in the native codec throughout, than transcode & scale upon ingest, and transcode and scale upon export. There are benefits both ways. You've made your point, and for many it will be valid, but it's certainly not a universal one-and-only way to go; for someone mastering out to DVCPRO-HD (whether on tape or on the cards) your method would involve a minimum of two scale and transcode/recompress stages. That's not ideal for all workflows.

I still stand by "no one" because broadcasters, HD-DVD formats, don't distribute content at 960x720 to the viewer. If you are to producing a 720p product, 1280x720 is your target resolution. If I use 1280x720 elements (titles, graphics, other format media) native means a 33% loss in resolution (which means quality loss), no intermediate loses even a small fraction that amount. Using the DVCPRO-HD deck or P2 media as a distribution format will lose 33% of you target resolution. Probably not reason to throw you DVCPRO deck out the window, but it kills any perceived advantage native editing has. I know we don't see eye to eye on this, but I fully understand native workflow, but your concern over transcoding losses is somewhat out of touch we what are achieving. In the end, post production is all about enhancing image data though color, timing and graphics, we help and do that well. If you keep to mainly cuts and minimal post elements native is fine, just not for everybody.

The flipping thing is nice for those who use the adapters, and where you do it is probably the ideal stage in the production chain to do it. Everyone else does the 24pN conversion and removes pulldown from 24pA, so that's nothing new. (But does yours remove pulldown from 3:2 24P also? If so that's unique.)

Yes it does.

Interesting -- I was wondering how you got around the lack-of-codec issue, and that's rather brilliant -- just tie into the existing freely-distributed P2 Viewer one. On other boards someone keeps complaining about the licensing fees for DVCPRO-HD, yet I haven't found one developer who's paid a penny to license it... clever going there.

Clearly I think it was a mistake for Panasonic to charge developers (a huge fee) for a codec when they gave the decoder for free to all there customers. It is why PC products where delayed on the PC. Particularly native solutions were hurt by this.

I commend you for coming up with the product, and for some users I think it definitely fills a valid need. Especially for those who continually render out their images for transport between programs; in cases like that the 10-bit wavelet with no scaling is probably of benefit.

"Probably" -- Why be cagey? It is a benefit, no harm in openly acknowledge a clear advantage. I will acknowledge that a simple production mastered to DVCPRO-HD native works best. We just believe (agree or not) that this type of production is in the minority.

I hadn't looked at your website recently; frankly you know that I was rather turned off by the way you had marketed your product before; we had a bit of a to-do about it as I felt it was borderline dishonest (you know that to which I refer, and I believe you disagree so be it.) But I wish you would just go for more open disclosure instead of portraying things in ways that just aren't true. For example, there are some factually incorrect things on your website, specifically when you say "do you need 24p workflow for your HD100 or XLH1" and then answer that it's ONLY available in cineform; that's not true, both Vegas and EDIUS support 24p from both of them right now, and FCP will have it in the next 20 days. You also claim that HVX support in Premiere Pro is only available with CineForm, that also is not true, Raylight and Serious Magic and Matrox Axio all offer shipping solutions today that work with Premiere Pro. I think if you'd go in and remove such statements it'd go a long way towards me seeing you & your company in a better light.

Done! That one sentence, on one page, was absolute true at the time we posted it. But thank you for pointing it out, it has been updated. We are a very open a honest company, the fact that both the CTO (me) and CEO spend time in forums talking with customers helps prove that (there are plenty of opportunities to call us out.) We don't have a dedicated web guy (CEO does that when he has time) so it can get stale in areas. But I still stand behind the CineForm vs DVCPRO-HD pages that ruffled your/Panasonics feathers so long ago, we clearly state our workflow, even point out the rescaling that occurs (in the deck, not by us, this is an HDSDI limitation -- if HDSDI had a 960x720 or 1280x1080 mode we would have used it.) Plus we did buy the deck, so we aren't totally adverse to give Panasonic money, and it works great for testing. But then you should know in you heart, if we where to post a 960x720 "native resolution" multiple generation test, CineForm would still be the clear winner (I have done this test.) It would go like this -- generation '0' DVCPRO-HD has the mathematic (not visible) edge, generation '1' CineForm takes a subtle lead, generation 2+ CineForm lead increasing. Not a dig at DVCPRO-HD which a good acquisition codec, it just wasn't designed for post.
 
Barry_Green said:
Well, first of all, it allows users to use a PC laptop but save files that are Mac compatible. For a lot of us that is a benefit. Serious Magic has never made a piece of Mac software, as far as I know, and they're not claiming it to be so. There's nothing "misleading" about it, there was just a misconception on your part.

Second, regarding the glitch

I never said, nor suggested anyone turn a blind eye to the "glitch". Don't put words in my mouth I never said. I only said it's not universal, which is a true statement.

As you know, I work almost exclusively with the FS-100. I rarely use actual P2 cards. And when I have, I've never seen the glitch. That does not mean "turn a blind eye", that means it's not universal.

And as for Serious Magic, they are stating there is an advantage to FCP users. I still don't know clearly what that advantage is, aside from some folks working around the "glitch" that we hope the new 5.1.2 upgrade will fix. But going to a PC, then back to Mac, that defeats the purpose of tapeless workflow. That means you're spending lots more than real time to capture you footage.

So, and please correct me if I'm wrong, but the one and only advantage to FCP users with this Serious Magic product is to work around the "glitch" some users are finding.

And please do not insist I am suggesting in any way that anyone turn a blind eye to the FCP "glitch". I just read your posts as insiting it is a universal problem, which it is not. It still needs to be fixed, but there are those of us who are NOT having that glitch. That in no way suggests anyone turn a blind eye to it. Don't put words in my mouth I never said, thank you.
 
Well, thanks for coming back and filling in more, that's all good.

As for "native editing", all I can say is that I think your approach is out of touch with the approach that something on the order of 90% of DVCPRO-HD editing is done as. FCP is a 960x720 (or 1280x1080) editor, and I'd say the overwhelmingly vast majority of Varicam and HVX footage is being cut on FCP systems. And of the small percentage that isn't, it's being cut on primarily Avid systems (which is also 960x720 or 1280x1080) or a small but growing percentage is on EDIUS (which is also 960x720 or 1280x1080). And for delivery to a final rendered master, if one is using DVCPRO-HD tape, it's going to 960x720 or 1280x1080.

As far as the codec comparison, again, different workflows for different folks. I like to get the image right in-camera, and I don't do a lot of grading and stuff. When I finish an edit it's in 100% original quality. Saying "DVCPRO-HD isn't designed for editing" is an extension of saying "DV isn't designed for editing", and I think you would find many disagreeing with you on that statement! DV ushered in the desktop video revolution (yes we had MJPG cards before DV, but ... seriously...) DV is an ultimately editable format (barring the difficulties of anamorphic pixels, of course.) DVCPRO-HD is exactly the same thing but at HD resolution.

I think your product served an incredible purpose during the early days of HDV -- HDV is a format that's not in any way designed for post, and basically everyone admits that. You turned HDV into something that COULD be edited, and that's great. I just don't see it the same way at all for DV100, I think DV100 works extremely well in post and has for the last couple of years since FCP 4.5 came out.

But again, different workflows for different folks.

Regarding the codec -- I certainly don't think it was in the HVX's best interest to set up a licensing fee type of system, that sure didn't help the NLE development side of things. But it does tickle my rebellious programmer's heart to see that in the end it didn't really make a lick of difference; Marcus and Serious Magic and Canopus and Apple and Avid and MainConcept and Imagine Software you all released products in extremely short order; heck, Apple and Canopus had functional solutions before the camera even shipped. So in the end the "licensing" issue turned out to be a bunch of bluster over something that didn't turn out to be an issue after all.

Anyway, thanks for the discourse, and I look forward to a better relationship in the future!
 
BenB said:
I never said, nor suggested anyone turn a blind eye to the "glitch". Don't put words in my mouth I never said.
Not trying to. I'm referencing a large cross-section of the Apple-using public who seem to not notice or care, and that's where the problem is -- people AT LARGE should be demanding a fix.

I only said it's not universal, which is a true statement.
Well, see, I think it is. There's something that isn't working, and everyone with the same circumstances is going to run into the same issue. I don't know what that set of circumstances is though; whether it's importing onto a G4 through the slot, or through the firewire port. But it's there, it's happened to people on different forums at different times. That means there's a snake in somebody's boot somewhere, and I want people to start emptying out their shoes so we can find that snake before it bites somebody else. The problem is real. It exists, it's (to some degree) repeatable, now it needs to be clearly identified and exterminated.

As you know, I work almost exclusively with the FS-100. I rarely use actual P2 cards. And when I have, I've never seen the glitch.
Well, this is just another piece of the puzzle. If you've never had the issue with the FS-100, that goes to back up my assertion that FCP is not at fault. Now, IINM the FS-100 is FAT32 as well, which implies that it's not Apple's implementation of FAT32 that's necessarily at fault. And we know it's not the cards themselves since the footage is always good on the cards, and it imports flawlessly into PC systems through either the card slot or the camera as card reader, so it's neither of those. It points the finger at either the PCMCIA slot itself, or the driver for the P2 devices. What I'd like to know is: does it ever happen when importing footage directly off the card when using the camera as a card reader?

And as for Serious Magic, they are stating there is an advantage to FCP users. I still don't know clearly what that advantage is, aside from some folks working around the "glitch" that we hope the new 5.1.2 upgrade will fix. But going to a PC, then back to Mac, that defeats the purpose of tapeless workflow. That means you're spending lots more than real time to capture you footage.
I do not follow you at all on that one, unless you're only referencing capturing to a laptop's internal drive. If you're using an external drive to capture to, there's no difference between importing on the PC or importing on the Mac, you'll end up with MXF files on an external drive either way. Serious Magic's product lets you use a PC laptop, avoid the potential glitch, and convert to QT files too. Maybe not a huge benefit, but of benefit to those who want the absolute surest measure of footage integrity. It's just as tapeless as the capture-on-a-G4 workflow (unless your capture-on-a-G4 workflow has been to discard the MXF files and just import directly off the card straight into your capture scratch disk using FCP; in that case there's a benefit to doing it directly on the Mac but at the peril of perhaps getting some glitched frames.)

The main thing that SM's product offers for FCP users is not the decoder. It's DV Rack 2.0 itself. That's the thing; live monitoring at full screen, comprehensive DDR, etc, and the ability to save files as direct QTs means you don't have to do any conversions anywhere along the way. Run DV Rack in BootCamp for live capture, swap over to OSX for instant editing.

So, and please correct me if I'm wrong, but the one and only advantage to FCP users with this Serious Magic product is to work around the "glitch" some users are finding.
Well, with the decoder that would be about the only thing, not even necessary though because you're supposed to be making disk images and you can always use FCP to do the conversion. The advantage to FCP users would be (as explained above) that DV Rack can be used as your live Video Village (under BootCamp) and the files it creates are directly importable to FCP without having to do any sort of file conversion.

And please do not insist I am suggesting in any way that anyone turn a blind eye to the FCP "glitch".
Ben, I apologize, I never intended to single you outnor did I, I have no disagreement with you, I think you are the bomb and I very much appreciate all you bring to DVXUser and its members. This is a frustration I have had with many FCP users who just do not seem to acknowledge or care. And they seem content to put up with lengthy workarounds, when it doesn't have to be that way. I don't get it. Why not just demand it be done "right"? Why not DEMAND MXF support on the timeline in FCP? Why not DEMAND the glitch get fixed? Why are people content to live with things the way they are, when they could be so much better? I just don't get it. As long as people don't clamor for a fix, no fix will happen. And sooner or later it's going to bite more people, and they will be unhappy when it does. I would rather see Panasonic and/or Apple just FIX IT NOW than wait for someone else to get "bit".

Maybe it's just a "pet peeve" and I should drop it. But it's soured me enough on the FCP workflow that I've reconsidered whether to bother getting a Mac again or not.
 
Hi Guys,

We never meant for the marketing message of the Decoder to be misleading or confusing. Serious Magic has always made PC-only products. With DV Rack, we do see a significant portion of Mac editors using DV Rack to shoot with. The original plan with the standalone Decoder was to just market it as a PC-only tool, showing the benefits for Vegas and Premiere Pro users. However, a lot of our Macbook beta testers told us that the conversion to QuickTime that we are doing was better or more reliable that what's currently in Final Cut. So, that's what you currently see on the web site.

We just launched both products 6 days ago, so there was bound to be some web site tweaking required. Please understand that Serious Magic is listening. I understand BenB's point of view, and based on his feedback, I am working on 'tweaking' the Decoder web site a bit to make it more clear to FCP users. There's also a DVCProHD Decoder FAQ in the works that will also make things clearer.
 
Barry, I do very much agree that FCP users need to give Apple more feedback. I know for a fact that Apple does listen, and acts quickly, to end user feedback. The problem is, most FCP users don't give it. In FCP, under the FCP menu, there's an option there that clearly states, "Provide Feedback". It amazes me how many FCP users in my classes and consulting have no clue it's there. I do push all FCP users to take as much advantage of that feature as possible!

Is it FCP, is it QT? I'd love to know. It would be great to find someone who has "The Glitch" regularly, and have them ingest the same footage via card slot on a MacBook, and via FW. Is it something that happens on towers, or only on laptops? It's all something that needs to be addressed. Again, Apple won't address issues that end users don't tell them (en masse) about.

So all you FCP users, send Apple your feedback on the DVCPRO-HD workflow issues, for sure! If you're FCP machine is not on the Intenet, there's a feedback page on the web site.

http://www.apple.com/feedback/finalcutpro.html

Write them, let them know that the ingest is too slow, there's a glitch somewhere, give them your full hardware set up, software versions, etc. The more they hear from us, the more they'll do. And I konw for a fact that Apple does listen and work very hard to meet end user requests. They just have to be ligitemate requests, and they have to get enough of them to justify the work.

FCP has not always been the Top Dog NLE. It's had periods of being it, periods of not being it. But then so did Liquid, Avid, etc. Every NLE will have it's moments to shine, and it's moments to lay low and re-tool. No one NLE has been the Top Dog forever, none ever will. Especially with HD being so "standardized" now, yeah, right, and the IEEE is keeping everyone on the same page, yeah, right...

Wasn't XML and MXF and all that supposed to standardize things and make NLE's and such more universal? But look at the mess it actually made. Oh well, that's the industry. FCP did DVCPRO before anyone else, had a period where it was the only game in town, shortly. Avid was king of the hill at one time, now I just read an article where big Hollywood studios are throwing out the outdated large systems (we won't name names here) to replace them with larger numbers of smaller desktop systems that together do much more much faster.

My best guess is, in ten more iyears, desktops will be on the iway out as we'll all be iediting in the iairport waiting on a delayed ifilght on our i27th generation iPodPro-MegaHD with a new version of iFCP running on it. And no one will use the same codec for anything. The big money makers will then be file format translation software... oh wait, it already is...

So, FCP users need to fill out the feedback forms for Apple and get on it!

Barry, you can still buy an Intel Mac, as it runs all your Windows stuff at least as well as a Windows made PC. And you can keep up with the Apple side of things at the same time! Two machines for the price of one! So, not using FCP is not a reson to not buy a Mac, or something like that.

Mac, Windows, UNIX, MXF, XML, MPEG, my head spins from all of the standardization that's taking place! Why is the IEEE still in business, anyway?

As for Serious Magic's web site, I may have mis-read it at first. But when I was in collage (educational psychology) taking a "linguistics and psychotherapy" class, the professor gave us a really neat quote from someone. It should be the motto of all film makers/story tellers, and advertising designers, "The meaning of your communication is the response it elicits, it has nothing to do with your intentions." Glad to see SM (is that "SM", or "S&M", or, oh, nevermind...) is rewording the web site. Cause on first glance, it does look like something that interfaces with FCP, which it isn't exactly. Thanks for helping me understand what this translator was all about.

So, back to the main focus; tapless workflow is so we don't have to wait an hour for an hours worth of footage to get into our NLE's belly. We should all push our NLE makers to do just that. When it takes an hour to ingest 5 minutes of footage, what's the point of tapeless?

Again, FCP users, let Apple know what you want from them!
http://www.apple.com/feedback/finalcutpro.html
 
Last edited:
Barry_Green said:
As for "native editing", all I can say is that I think your approach is out of touch with the approach that something on the order of 90% of DVCPRO-HD editing is done as.

That like saying Windows is the best operating system just because 95% of the world uses it. My argument is that for anyone wishing to use color correction (not everyone gets it right in camera) or other post elements, that native is limiting their output quality. I'm not talking native vs CineForm, I'm talking DVCPRO-HD native vs everything else (uncompressed, Sheer, M-JPEG, Avid DNxHD, Canopus HQ and CineForm.)

FCP is a 960x720 (or 1280x1080) editor, and I'd say the overwhelmingly vast majority of Varicam and HVX footage is being cut on FCP systems. And of the small percentage that isn't, it's being cut on primarily Avid systems (which is also 960x720 or 1280x1080) or a small but growing percentage is on EDIUS (which is also 960x720 or 1280x1080). And for delivery to a final rendered master, if one is using DVCPRO-HD tape, it's going to 960x720 or 1280x1080.

Which is 25-33% lower resolution than you potential output res, if you step beyond this limitation of "native", you can have a higher quality master, which is the point that Jim made. If 960 out of 1280 or 1280 out of 1920 (or out of 1440) is acceptable for everyone, then cool. However, it is a compromise than can be avoided for those want to. We just offer that choice for a higher quality post for many workflows.

As far as the codec comparison, again, different workflows for different folks. I like to get the image right in-camera, and I don't do a lot of grading and stuff. When I finish an edit it's in 100% original quality. Saying "DVCPRO-HD isn't designed for editing" is an extension of saying "DV isn't designed for editing", and I think you would find many disagreeing with you on that statement!

Maybe those how don't know what "designed" actually means. :)

DV ushered in the desktop video revolution (yes we had MJPG cards before DV, but ... seriously...) DV is an ultimately editable format (barring the difficulties of anamorphic pixels, of course.) DVCPRO-HD is exactly the same thing but at HD resolution.

DV has the same issues, although it was less compressed, so a little less loss, DV is in-fact better than DVCPRO-HD (do the math -- 5:1 is less compressed than 7:1.) But DV success wasn't because its ability to hold up in post -- there are plenty of uncompressed SD systems out their because of DV's poor render quality (and 4:1:1 / 4:2:0 issues.) DV was very difficult to be used safely as a post format, there are websites dedicated to showing DV artifacts by each vendors implementation (faults like green shift, blocking, mosquito noise & chroma smear.) It was far short of an ideal post format, convenient yes, but not ideal. DVCPRO-HD is less convenient than DV, as it is not as dominant (only one camera vendor) and don't have a resolution that matches "final" distribution (DV did get that part correct.)

I do take the position that the source bits are not sacred, if you can make a better output, who cares if your source was DVCPRO-HD, AVCHD, M2T, DV, or even our own CineForm RAW in-camera compression. That is the approach we are taking. "Native" is good marketing spin, that sometimes has its advantages, yet if it bandwidth/storage wasn't an issue, everyone would abandon compression completely. CineForm is just a step on the way to uncompressed, less compression, higher quality and more flexibility.
 
The early reports are that the new firmware for the MacPro fixes the SATA issue. The RAM limitation of 1 Gb still stands for the app, however.
 
Back
Top