SD vs HD: Tables have turned
 

For those of us not willing to upgrade our equipment/software/workflow to HD there seems to be a new alternative. Red Giant (makers of Magic Bullet) have released a new product called InstantHD. This up-rezzes SD video to several flavors of HD. And from the examples it seems to do an amazing job. Here's a link for those interested: http://www.redgiantsoftware.com/instanthd.html .

If the results do prove favorable (and I'll be trying this out as soon as work is over) this could be an incredibly low cost alternative to upgrading to HDV. The plugin only costs $99 and will uprez SD all the way up to 1080p. One thing to keep in mind is that this only works with progressive footage. So you'll either have to deinterlace your footage before you uprez or shoot progressive. I already do shoot progressive with my DVX100A, so this kind of plugin could really change things up. If a delivery method for HD is available and becomes widespread soon, this may be a good alternative to getting all new equipment. I do think that HD is the future (not trying to fight that) but personally I want something beyond HDV before I commit to it.

I'd love to hear other comments on this, especially from those using HDV.

Curious,

Does it uprez EACH FRAME or does it do a compression with GOP's?

I too believe that HD is the ultimatte end product. But I'm not yet convinced that HDV isn't a transitional format. WHen all is said and done, I'll bet that HDV as a format won't have a lifespan as long as DV does/did. SO this might be the way to go as we watch and wait for the following developments.

!) TRUE HD uncompressed cheap storage solutions. (Watching the 3d media closely) The winner will probably be some form of tapeless storage.

2) RED Camera (Counting on Jim to shake things up completely) Variable framerates, 35mm sensor, choose your codec... yeah sounds great.

We know Red is coming soon, and I think the storage solution is less than two years away. SO if my gut instinct is right, and the R&D curve holds... I'm guessing everyone will be shooting/editing a form of HD uncompressed by 2008/9.

My xl2 meanwhile, is churning out great pix, and has paid for itself.

Still, If a BODY ONLY version of the H1 were available, I'd consider buying that now. I'd use my 16x manual lens, and the fu-1000 viewfinder on it.

Just my ruminations as I plan to head off on a LONG project and shoot. I'll be off board for the next two to three months.

C-ya.

(Bummer I see on the link that they don't support Avid. Too bad.)

I did a test with the FCP version, and it looked no different to uprez in FCP without the plugin. And the workflow was terrible.

On the Red Giant website there's no pictures showing how it's better than other methods - why?

Graeme

Graeme,

Do you have any comparison shots we could look at? I don't use FCP (PC User) but it would be interesting to see them side by side. From my understanding this is a relatively new product. Maybe they will have comparisons later? I know they have comparisons against other products on their Magic Bullet pages.

Also, if uprezzing has been around for a while does anyone use it? Obviously the resolution isn't as good as native HD. But what other downfalls are there to doing this?

No, I just used some progressive footage and it looked the same as FCP's scaling (best) in FCP5.

Graeme

If I was a gambling man I'd probably take you up on that bet. DV hasn't been in widespread use for ten years yet and is essentially obsolete as we head into the HD future, whereas HDV will continue to be useful until current standard HD formats are replaced by even higher resolutions. It took ~50 years for SD to be replaced by 2K HD, so if it takes half that time to move from 2K to 4K then HDV is good for the next 25 years. Keep in mind that MPEG2 at HDV data rates will be the primary broadcast and disc-based HD distribution format for at least the next few years, so HDV is a useful fit for that.

We could soon see a move toward MPEG4-based recording and distribution at bit rates somewhat lower than HDV, using the internet and flash-based memory for distribution. But HDV fits well into that because then it becomes the "high bandwidth" acquisition solution for low-bandwidth output, with higher bandwidth recording only needed in more demanding situations. If people like MPEG4 recording better than MPEG2 recording then maybe HDV becomes a transitional format, but it won't cease to be useful for quite some time after that. (DV may still be in use then too, but will be considered a nostalgic oddity.)

Fair enough, but all other HDV arguments aside, does anybody uprez their footage now? My "lens" on the situation is this. I live in an SD world. The local market doesn't demand HD yet. I work mainly in event videography. Weddings, family gatherings, community events. This Friday I am filming the flag ceremony at the beginning of a Grizzlies hockey game. Nobody asks for HD here. But as delivery options become more mainstream (I mean HD-DVD and BluRay) I think that someone, sometime may want it. As I don't want to upgrade all of my equipment, would it make sense to use a product like this to up-rez my footage to HD for those few times when it is needed? I'd love to hear a straight answer to this question, not another attempt to baptize me into the HDV ranks. No offense intended to HDV users. HDV is fine and I'm glad that it's out there. But I'm just trying to determine if this is a useful tool for rare occasions or not.

KEVIN WROTE - "DV hasn't been in widespread use for ten years yet and is essentially obsolete as we head into the HD future"

OBSOLETE is a tricky word. That's why I never use it.

Perhaps I should have said "Capture format" in terms of lifespan... In point of fact, most media already captured, that is of any value will continue to 'exist' and be relative for the forseeable future. Especially with technology making it possible to uprez and 'clean up' old footage.

While working on a documentary, the only footage I had of a particular piece was in VHS. Now then, VHS is not an 'obsolete' format... plenty of tapes left on shelves, plenty of cheap players still available... but as a capture format, especially in terms of cameras, one can effectively say it is 'obsolete'. (As I take YOUR meaning of the word Kevin.) Very very difficult to find a VHS camcorder being manufactured today. Hi-8 and Digital 8 are fast dissapearing to.

As to claiming DV is 'obsolete', I think that's premature. New DV cameras are still being manufactured and sold. For the sake of arguement, lets not hang the obsolete sign out untill the cameras and decks are no longer being made.

In that respect, I still maintain that the 'working' lifespan of HDV as a capture format will be shorter than that of DV. In fact, I don't think it will be a capture format in ten years... say 2016. Will you still be using tapes that were captured in HDV? Sure. I transposed tapes in VHS, Betamax U-Matic and even SUPER 8 footage as needed. Like I said, the final delivery will be some form of High Definition... but HDV as a capture format will be relatively short lived. I think the technological advances moving us towards HD are exponentially growing.

Just my opinion.

Mike: proselytization aside, I haven't seen any SD up-rezzing solution at any price which makes upsampled SD look like anything but upsampled SD. You can't create detail where none exists in the source image (contrary to what Hollywood would have us believe), and if this did work we wouldn't need HD cameras in the first place. Plus HDTVs are supposed to upsample SD footage reasonably well anyway, so there's little point in spending time doing it upfront.

One thing which might be useful is that SD content delivered at HD resolution using a bit rate higher than that of traditional DVDs should look a little better than those DVDs. Plus if you can start shooting in widescreen mode using an anamorphic lens or a true widescreen SD camera, that will give you a leg up for HD output versus shooting 4x3 SD. In fact the 4x3 versus widescreen issue may be as significant as the change in resolution, because the change in aspect ratio makes converting SD footage that much harder.

I suppose I see your point, but what I'm saying is that today's HDV cameras will still be useful production tools in ten years when HD acquisition has become a de facto requirement of professional videography. They may be relegated to "B roll" status or otherwise marginalized, but they'll still be usable until they cost too much to maintain and/or miniDV tapes are no longer widely available. Plus if blue-laser DVD players catch on and become as ubiquitous as red-laser players are today, then distribution of HD content in MPEG2 format will be a widespread option for at least a decade from now. If HDV does get sidelined, it won't be by higher-end solutions but by MPEG4-based video cameras, probably using even lower recording bit rates.

I'd almost bet good money, that 3-5 years from now, most professional wedding videos will be shot in HDV (more than 50%), and a significant number even delivered in HDV (essentially), on HD-DVDs or Blu-Ray disks. I'd be pretty surprised if my FX1 becomes obsolete before wearing out. I expect the FX1 and Z1 to become workhorses of the industry, much like VX2100s and PD170s are now.

or u could just use Vegas, throw down ur SD 4:3 footage on a HDV timeline, let it do all the scaling for you, all u need to do is reframe your shot if u want to... and then u can render faster than realtime to Cineform.. or even straight to M2t.. or H.264 or whatever..

I do this all the time, but i only scale up to 1280x720 from PAL 720x576 using DVX100a's as the source. Wrks a treat and not as "soft" as most people think.. if it is soft, just run a sharpening filter. The different aspect (from DV to square) will usually help clean this up anyway..

Ok, I'm a Vegas user and I think that's a fair idea. But how about the aliasing? Does Vegas do much for smoothing those harsh DV edges? I guess I should run some tests myself. I haven't tried that yet, but I suppose I shall. Thanks for the idea.

mike, the fact that he aspect ratio is ALSO interpolated cleans these jadggies quite nicely.. from DV rectangular, to the square pixel aspect of HD, the upscale is surprisingly clean... very clean in fact..

just throw some DV footage in a 720p HDV project.. youll see what im talking about.. jsut make sure youve got it set to interpolate fields, not blend.. also prefereably your seource material is native Progressive (higher res) but u can always convert interlaced footage to progressive, althoug it wil be ever so slightly softer... which could be a bonus actually...

there is so much mis-information in your posts that it's almost too much to address... but starting with the above, mpeg2 is already being dumped as a broadcast distribution format all over the world, because h.264 is so much better... take a look at any satellite network, especially in europe; nobody wants to use mpeg2, period.

claiming that mpeg2 is going to be the primary disc-based hd distribution format is absurd, because there aren't any hd dvd players in the hands of consumers yet... so everything is wide open at this point.

shooting hdv for weddings is a joke at this point in time, because the workflow is so crippled, and there is no delivery format for the resolution that you are shooting at... not to mention the marginal low-light performance you get with hdv, vs. dv.

mike, based on my years of codec rendering experience, i would say don't count on an up-res solution for critical work... just keep shooting dv, until there is an actual delivery format for hd product.

ATSC broadcast specifications are based on using MPEG-2 compression. A change from ATSC, for OTA broadcasts, in the US (or anywhere else for that matter) doesn't seem even remotely likely for years and years to come.

There are three compression formats written into the HD-DVD and Blu-Ray specs, and MPEG-2 is one of them. MPEG-2 may not prove to be the most widely used, especially as time goes on, but it would surprise me if MPEG-2 isn't used significantly for encoding the new disks, at least for a few years (any of the three compression formats is quite viable for full length, feature films). HD-DVD players will begin being shipped within weeks. Blu-Ray players aren't far behind.

Shooting HDV is pretty much the only viable method for acquiring HD footage within anything close to the kind of budget most weddings are professionally produced at (whether it's edited as MPEG-2, or using an intermediate codec like Cineform, or via a method such as using a product like Gearshift). Panasonic's P2 technology (to shoot DVCPro-HD) may become competitive, but it isn't really quite yet, and is not even a certainty to, at this point. Nothing else is even on the horizon.

HD content can be delivered currently on red laser disks (standard DVD disks). There are a couple of DVD players currently being sold in the US that can play HD content from red laser disks. DVHS tape is also viable for delivery of HD content (not particularly attractive, but it is viable).

Low light performance of the HDR-FX1 and HVR-Z1U is quite comparable to (or better than) that of many SD miniDV (and DVCAM) cameras currently in widespread usage for professional wedding videography (DVCs, DVXs, GLs and XLs - essentially, all but the PD and VX camera lines, also from Sony).

“As we look at transmission going to NAB, bandwidth management is becoming a huge part of our future,” says John Wallace, NBC Universal executive VP of television operations and production services. “We are interested in exploring MPEG-4 compression for more optimization in our bits for distribution, and that includes the NBC television network as well. A project for 2006-07 is to convert Skypath [NBC’s satellite distribution of programs to affiliates] to high-definition MPEG-4. Right now, it’s MPEG-2.”
http://broadcastingcable.com/article...Special+Report

"Sky's HD service will also utilise the advanced compression technology MPEG4. MPEG4 is a more efficient means of compressing the data in a TV picture."
http://www.digitalspy.co.uk/article/ds19617.html

"DISH Network Introduces Nation's Largest Package of HD Channels
ENGLEWOOD, Colo.--(BUSINESS WIRE)--Feb. 16, 2006--EchoStar and its DISH Network(TM) satellite TV service announced today it will launch local high definition TV channels via its satellite TV service to customers in Chicago... To make this possible, DISH Network will begin transmitting newly added HD channels in MPEG4, a signal compression standard developed by MPEG (Moving Picture Experts Group). MPEG4 allows DISH Network to maximize the bandwidth available on its satellites and offer the most robust lineup of HD channels in the nation."
http://biz.yahoo.com/bw/060216/20060216005172.html?.v=1

everything you shoot in hdv will have to be re-compressed for distribution, so there is no advantage to using mpeg2.

"The appeal of the FX1 is compelling – it’s high-def and it’s today. But that high-def comes at a steep price in other ways – poor audio connectivity/controls, narrow latitude, weak low-light response, and a video-only look, as well as significant loss of resolution on moving shots, and the dropout issue."
http://www.dvxuser.com/articles/shoot3/

that is not correct... the silicon has already been created for h.264 video acquisition by the ambrella corporation, and the first hi-def mpeg4 video camera is already on the market: http://www.dvinfo.net/conf/forumdisplay.php?f=115

that camera is but a toy, when compared to what will be here within the next couple of years:

"Ambarella Corp. has a working 17-mm2 packaged system-on-chip that handles H.264 encoding at data rates of 15 Mbits/second and up while consuming just 1 watt."
http://www.videsignline.com/showArti...leId=174919683

Don't hold your breath waiting for Congress to act again, and change the standards for OTA broadcasting. Cable and Satellite will eventually go to delivering content using MPEG-4 compression, but it won't happen overnight.

Currently, if you distribute on standard DVD, you will recompress, unless you shoot with a pretty odd camera (and you'll also go from 4:1:1 to 4:2:0 color sampling, if the source material is DV).

With HD-DVD and Blu-Ray, there actually is the potential to distribute content acquired with an HDV camera, without recompressing (you should be able to simply repackage the MPEG-2 video and audio streams, if desired, and you can stay at 4:2:0).

Frankly, I expect to encode HD-DVD and Blu-Ray disks using MPEG-2, for delivering wedding videos. There are some advantages to doing that, and no downside that I can see (25 or so gigs is plenty of room for very high quality HD MPEG-2 content - even a DL red laser DVD, at a little over 8 gigs, is workable with).

Reading from the same article you quoted, regarding low light performance of the three cameras compared, you might notice that they also state that the FX1 is clearly the more noise free of the cameras compared, when gain is used. They mention less detail in blacks with the FX1. Using black stretch on a Z1 should provide a considerable improvement there.

There are all kinds of things being researched and developed (more than you can shake a stick at - that's always been true, and always will be). Nothing to compete with HDV (and possibly Panasonic's HVX) is on the horizon. There are no inklings of plans for a major company (Sony, Canon, JVC, Panasonic) to release a high definition camera, using another compression codec (like any type of MPEG-4) or even using a different primary recording medium (like SD or CF cards), that would be economical for professional wedding videography, in anything remotely resembling the foreseeable future.

This has been a very interesting and well informed discussion. I am left with one impression:

Wait a bit, things are in flux.

Mike

waiting is indeed the smartest thing to do.

that is true, but i just proved to you that hd satellite in chicago is currently using mpeg4 to broadcast over the air, directly to consumers... i also proved that nbc television programming will be distributed first via mpeg4, then mpeg2 for the so-called "final mile".

i would suggest that you look into iptv, among other things... there is a community in texas that is currently receiving outstanding picture quality on their tv sets, via mpeg4 iptv... your fixation on fcc mpeg2/ota is but one part of the pie these days.

i did think about that, but i haven't bothered to look at any detailed hd dvd specs to see if it's even possible, because hd dvd is pure vaporware at this point.

how about if you and kevin tell us how many hd dvd's you guys have authored so far? hmm?

obviously, none of those companies have a history of talking about products in the engineering stages, because it would ruin the sales of current products... people would hold off buying current cameras, in favor of what's coming in a year or two.

and believe it or not, there is a definite connection between the current availability of new silicon, and how it will affect products being designed today... i didn't expect you to appreciate the impact of that ambarella h.264 chip.

as for your fx1, eventdv compared it to the xl2: "in low-light tests, the dv camcorder produced a brighter image than the fx1 in both hdv and dv modes, although it does look a bit faded."

as i understand it, the problem with hdv cameras with 1/3" chips is that they use small ccd's for a large frame size(aka higher resolution), so there is essentially less surface area on the ccd's to pick up the light... that's not a fault of the hdv mpeg2, but i think that it will be improved on as better chips become available.

Okay, distribution will be either MPEG2 or MPEG4...at data rates less than or equal to those used by HDV cameras. If anything, switching to MPEG4 distribution may extend the lifespan of HDV as a recording format, because then HDV will be higher bandwidth than typical delivery solutions and hence offer some overhead for production imperfections.

Sorry, make that "a" primary distribution format. Sony has already announced that they and their studio partners will use MPEG2 for the first round of HD movies, in part because that's easier to do effectively for now -- and coincidentally more of a nuisance for bootleggers to duplicate.

My workflow for HDV production is essentially identical to working with DV, except I have to use my best computer for full-resolution projects and I have to implement one workaround for making widescreen SD output from an HD timeline in Edius. HD delivery is available now for those customers willing to work with cutting-edge options, and will be mainstream within a few weeks when the Toshiba HD DVD players start shipping. Low-light performance is a bit of a challenge, but with 20-40 watts of on-camera lighting and a little post-production enhancement it's not that big a deal. I've actually reduced the wattage of my on-camera lights since I started shooting HDV.

On that much we can agree. I did some tests recently which indicated that true widescreen SD footage can be upsampled tolerably well to 720p resolution, but 4x3 SD converted to widescreen anything looks like crap. Any camera without a proper widescreen sensor or a good anamorphic lens is headed for obsolescence for professional videography purposes. That's the key point to take from this particular discussion: HD/widescreen will take over and 4x3 acquisition is doomed.

There will always be something better coming in the future, but HDV is here now and works admirably well for the price and considering its technical limitations. By the time something better is widely available and supported today's HDV cameras will long since be paid for, and will continue to serve as functional backup cameras for many years after that. Meanwhile, most footage being shot in SD is effectively outdated, and will be useless for your HD demo reels a couple of years from now. (Unless you have a really good widescreen SD camera.) Wake up and smell the coffee folks, HD is a reality today and will only become more so in the future. If you don't feel like investing in HD yet that's probably a prudent business decision, but you need to figure out when and how you're going to make that investment.

All this is true but one thing that keeps getting glossed over is the low light issue.

I have yet to shoot a wedding that did not require a good low light camera. I don't believe in shooting with lights at weddings. I don't want to dive into that discussion, it is just my preference and the preference of my clients. That said, HDV is effectively ruled out for everything except outdoor deals. For now.

I expect a HDV camera with low light capabilities similar to a PD-150 soon. Most likely sooner than any viable true HD solutions. It was put forth that the SD guys need to figure out when they are going to make the move and I suggest this is the point for me: When a low light HDV solution is available.

Everybody has different requirements and if you shoot with lots of lights then HDV may be the way to go right now.

By the way, did you see the press release about the SHD format? Super High Definition. They say it will be replacing HD soon. Any takers??

Mike

i have to agree to diasgree mike, some of your comments made me want to reply with a myriad of technical jargon but its almost 4am and i need sleep :)

one thing however is that until a format is STANDARDISED all were going to see is a mish mash of formats vying over each other. In the end it will cause more confusion for our clients and more work on our part to educate these said clients. Not to mention the fact that we still need to sell these formats to said client.

As for 4:3 acquisition, i tend to agree, mostly... but there are many solutions out there which handles the 4:3 to 16:9 issue so IMO its a non issue. I wouldnt go out spreading fear of the almighty widescreen, but then again, there are clients out there who have no idea.. as an example, a client wanted her wedding in widescreen, BUT when she viewed the widescreen footage she noticed the black bars on top and bottom (she was watching a 4;3 tv) You and I and most peopel out there know that this is normal. To her though, she thought it was a defect.. so 1hr later after going over the reasons why its the way it is, she wanted it 4:3
Each to their own, but in the end, its up to the client as to which way we as producers should be heading.

I honestly dont think Low Light performance will dictate the evolution across to HD.. far from it, even in DV standards most 1/3rd cameras find it hard to acquire GOOD footage in low lit environments. In this case you mentioned you dont like to use light, and if that works for you, great. I on the other hand, use indeirect lights and usually nothgn over 35-50w for dancfloor use. If im shooting with teh Z1, i change the globes to 50 and 100w. The difference a little light makes is very noticable to colour, gradation, and contrast (as wellas focus "appearing" sharper due to said colour reproductions) and like photographer, without light u have no picture.
Its a true fact that even in low light, what u might get without using a light, u might lose in detail.
IMO that detail is too priceless to sacrifice. But thats me and my opinions differ to most and i accept that. I also respect peoples choices about lighting as many times people say they dont want light, but to me, image quality is more important and i would recommend a light to any professional. This of course is if theyre using a 1/3rd CCD camera.
Larger CCD units can obviously get away with this

Either way, teh point im trying to make is that from where i sit as a distributor, low light performance IS NOT going to be the turning point of the industry. Low Light cameras MAY dictate the way HD acquisition evolves... but it wont be the turning point for SD shooters to jump onto the HD wagon..

Delivery formats, delivery media, NLE support, DVD Authoring support, cost of post production, and more importantly, education will dictate how this format is absorbed within the industry THEN absorbed to the general public

HDV works fine in reasonable indoor lighting if you know how to use it. The Sony HDV cameras are definitely not as sensitive as some of the best SD cameras and that can be a challenge for wedding work, but you could shoot HDV when the lights are on and then switch to your best SD camera when the lights are off. I suppose most of us shooting HDV at weddings are used to using a light at the reception anyway, so that's not a problem for us. I've tested the FX1 to be one stop less sensitive than my GL2 in low light but with cleaner footage, so I can make up some of the difference in post.

Yet again please note that HDV is as much "true HD" as most other HD recording solutions, and the Canon XLH1 offers a full uncompressed HD output for those who want that. The first camera to meet your low-light needs will probably be the Sony XDCAM HD, with a retail price around $17K plus lens when it's released sometime in the next few weeks. For those of us who don't have $20K to spare for a wedding camera, see note above about options for low-light HD recording.

Not gonna happen for event videography for at least the next few decades. We're just now getting to the point where we can shoot and deliver 720p or 1080i resolution to customers with (mostly) 720p HDTVs, and it will be a long time before either we or customers have maxed out the potential of 1080p. Besides, 720p may turn out to be "about right" for wedding work because it's enough more detail than SD to be visually satisfying, but doesn't reveal every last blemish as much as SHD would do. Even at 720p we've got customers worrying about how they'll look on video, which is absurd because that's a small fraction of the resolution that their photographers are using. But that's the way people think, so even 1080p is likely to be a tough sell unless future generations want more realism.

It's pretty difficult for me to imagine that televisions with greater than 1080p resolution are going to become even akin to commonplace, in living rooms across the country (or elsewhere in the world for that matter, aside perhaps possibly in Japan), for many years to come (think decades). Greater than 1080p video image acquisition is unlikely to become economical for all but ultrahigh end professional wedding video acquisition for quite awhile either. We'll probably see actual resolution of HDV cameras gradually shift closer to (and exceed a little) that of the Canon H1, but the actual realized resolution of cameras, such as the A1, FX1, Z1, HD100U and HVX200, are certainly reasonable (noticeably sharper, to anyone who isn't all but legally blind, than the sharpest SD), into and at least somewhat beyond the foreseeable future.

Jeez you guys, SHD was a joke. You guys need a hobby.....

The sad thing is that it's not a joke, and some of us would probably love to have an "8K" video camera:

http://www.studiodaily.com/filmandvi...ssue/6159.html

lol... as mpeg4 continues to take over distribution, the popularity of the format will solidify it's use as an acquistion format, shortening whatever minimal lifespan hdv gets.

that is complete rubbish... mpeg2 dvd's have been copied(aka "bootlegged") since dvd's first came out, it's the most widely copied format in the world... there are hundreds of software programs out there right now that can copy mpeg2.

as for distribution, sony owns a bunch of mpeg patents, which is why they backed a lossy, inefficient mpeg2 format like hdv in the first place.

whether sony uses mpeg2 for it's movie distribution is completely irrelevant to this wedding forum anyway... if anything, people will not use it, because they can get a lot more footage onto a dvd with superior codecs like vc-1 and h.264.

that's not relevant to this discussion... we are debating the use of hdv cameras here and now, not years down the road.

you don't have the slightest idea what cameras are in the engineering pipeline, so spare us the ranting.

i have noticed that you are attempting to discount the importance of low-light performance in shooting weddings... i hope that any video newbies reading this thread will know better.

many wedding ceremonies, in particular, take place in a sanctuary that does NOT allow additional lighting... so the low-light performance of the camera is mission-critical in getting a decent picture.

Actually, the discussion is whether or not you can upsample SD footage to HD footage with current software offerings and create decent footage. I started this thread, and while it has matured to a rather detailed argument of HD/codec specs, it has gone far afoul of the original thread.

I have learned several things from this thread. One, upsampling footage to HD is ok at best. Two, never start a thread with the term HD in it unless you want a debate about HDV. And last, but not least, I have reaffirmed my decision not to switch to HDV. These are all my personal opinions and should not be taken as facts or as further fuel for the fire. I do appreciate the passion that both sides have taken in presenting their arguments, and if you wish to continue the debate please do so. I just thought I'd step in and remind everyone of what the original post was about.

"Actually, the discussion is whether or not you can upsample SD footage to HD footage with current software offerings and create decent footage. "

well if thats the case here, then the simple answer is yes.
Its very possible, its very plausible and its very real.

as for codecs, well one must remember that when disucssing this topic, delivery of said formats WILL dictate the way WE work as producers

Its all relative

Mike Oveson - Did you get a chance to try Red Giant's InstandHD? What did you think of it? Do they indicate what type of technology they are using for resizing (spline, fractals, etc.)?

Robert,

I did try out the demo and the results were rather favorable. Rendering speed is quick. I have seen some other apps that use spline methods to scale up the image. In comparison this had better results than those. Someone has mentioned that Final Cut Pro does better than InstantHD. I don't have FCP (Vegas user here) so I can't say on that. I did try upscaling some SD footage in Vegas. InstantHD did better, for sure. The demo puts a funky red/green dot watermark over the image but I'd be happy to post comparisons if anyone wants to see them.

I don't know what scaling Vegas uses, but I'm pretty sure FCP is just basic bicubic, and it looked no different to what InstantHD is doing. Indeed, now that they have examples on their site, they're deceptive as they're comparing their scaling to clearly deficient "nearest neighbour" scaling, which is so poor, nobody uses it any more.

Graeme

I only see these examples (http://www.redgiantsoftware.com/instanthdexamples.html) on their site and they don't really seem to be comparing different methods of upsampling. Maybe I am missing something. I can see what you mean about a company choosing an inferior method to compare to in order to boost the superiority of its own product. The company I work for (and many others) use a similar tactic. They give you the MSRP price and then tell you their price to show you JUST how much you are saving. The MSRP price is usually inflated to begin with. Anyway, I just wondered if I was overlooking something on their site. Thanks Graeme.

Dan: you're missing most of my points in your rush to bash HDV, but that's okay. The fact is that HDV has already become the most widely used HD acquisition format ever, and is being used by many wedding and event videographers today to attract premium clients at a premium price. The HDV format is currently the most cost-effective way to produce decent HD footage and makes SD upsampling look like a poor choice by comparison, since in the long run it doesn't cost much more to shoot HDV than DV.

As far as why MPEG2 makes HD DVDs harder to bootleg, that's precisely because it doesn't fit as much on a disc. What the bootleggers will probably do with the MPEG2 HD DVDs is crunch them down to MPEG4 and cram them onto red-laser discs, but that will cost them time and money and allow the legitimate distributors to claim they offer "full quality HD."

I agree that event videographers may opt not to use MPEG2 for distribution of HD content for practical reasons, and I suppose in the long run that means MPEG4 cameras, editing and output could take over. But this works out better for HDV than if distribution moved in the higher bandwidth direction, hence my statement that MPEG4 distribution extends the life of HDV cameras. If customers started expecting high bandwidth HD output, then HDV cameras would definitely become endangered. Not likely given current trends.

As far as the "engineering pipeline" is concerned, I welcome any insight on the prospects for an HD camera which is higher quality, easier to edit from and more cost-effective than HDV. So far I don't see clear signs of such a camera materializing soon, although it is interesting to see what's happening with MPEG4 technology options. If we get a big surprise sometime soon that's okay by me.

Agreed that low-light concerns are an issue for current HDV cameras for event work, but depending on your needs that's not necessarily as bad as some people make it out to be. I did some tests recently and was able to make processed low-light footage from a Sony FX1 look about as bright as unprocessed footage from a DVX100, so with a little care it can be used in dim settings. I definitely look forward though to better low-light HD cameras in the future, and will consider buying one someday when they're available at a reasonable price. Again, I don't see clear signs of such a camera materializing soon, but would welcome an unexpected surprise.