Yes, most compressed video uses interframe compression that makes it tough to edit... DV uses Intraframe, so it takes less processing time. But Apple has pulled of software miracles before, just look at the result that they get using altivec and DVD encoding. Not to mention FCP's realtime rendering. I don't expect to see a camera like this tomorrow, but it will happen eventualy. You'd be a fool to disagree. Maybe it won't use a megnetic HD, but, someday, within the next few years, somone will make a removable-medialess camera.
why mess with something thats already decent enough?
Because that's what Apple does. That's what makes them so great.
<strong>I don't expect to see a camera like this tomorrow, but it will happen eventualy. You'd be a fool to disagree. Maybe it won't use a megnetic HD, but, someday, within the next few years, somone will make a removable-medialess camera.</strong><hr></blockquote>
The whole point of this thread was based around an idea that was not way in the future. To say that one would be a fool to disagree that "it will happen eventually" is a foolish statement. Once you abstract away the magnetic HD from the discussion, we are no longer having the same discussion. The main problems with the idea revolve solely around the proposed use of a HD. In three years, when the iPod sized HDs are at about 100 GB, MiniDV will no doubt have been replaced with a much higher-capacity fomat (think mini version of Super DLT - now we're talking capacity!!!).
While I wholeheartedly agree that MiniDV has some issues, and that something will come along to replace it, the proposed solution just would have more downsides than advantages. I'd love to have a non-linear system, but downloading all the footage to your computer and then editing out the stuff you don't want/need has a lot of advantages too. I have actually never experienced the dropped frames problem you mention once tapes have been written to a few times.
And as for MPEG 4, it's great, but for web video primarily. I personally would much rather have the original footage at a very high bitrate, and encode it later for web braodcasting.
Look at the iPod! It is a better way of doing the same thing, while retaining compatibility. This whole suggestion would involve completely breaking the mould, and compatibility with other products. This would be equivalent to the iPod only playing some proprietary Apple audio format, that only Apple software could encode or decode. Apple has finally smartened up and moved away from doing this exact thing. Instead, they are proposing new open standards and getting the competition on board, or acepting other standards. If they try to do this, it will only succeed if they get the other major hardware and software vendors on board. You're talking about a revolution in the way it's done, and it could only happen with a lot of support behind it.
I agree that an iPod-style HD would be great in a still camera with QT movie mode. As it is, my Olympus digital camera can do this, but is limited to the capacity of the SmartMedia card.
P.S. And Gizwald, why did you have to end your previously intelligent post with some junior-high insult involving nationality and intelligence? It only served to devalue your other comments.
[quote]a 20 GB HDD will only fit about 90 minutes of video. A MiniDV tap fits 60 minutes of video, and costs a few dollars. A 20 GB HDD the size of the iPod's will cost you a few hundred dollars.[/QB]<hr></blockquote>
<strong>The whole point of this thread was based around an idea that was not way in the future. To say that one would be a fool to disagree that "it will happen eventually" is a foolish statement. Once you abstract away the magnetic HD from the discussion, we are no longer having the same discussion.
</strong>
That's true. But you guys were shooting down the idea as if it was totally stupid and that it could never happen. I get the impression, from reading your post, that you think that tapes will never ever be replaced by some sort of fixed media. I disagree. It will happen. (in the comsumer market, at least) But I do think that some better technology needs to be developed first... Higher density storage, and maybe a better standard video codec.
<strong>
In three years, when the iPod sized HDs are at about 100 GB, MiniDV will no doubt have been replaced with a much higher-capacity fomat.
</strong>
I think it will take longer than 3 years for Mini DV to be replaced by something better. Standard consumer technology moves on a much slower cycle than that. DVD technology has been around for about 10 years, if I remember correctly. And it's just begun to catch on in the last few years.
<strong>
downloading all the footage to your computer and then editing out the stuff you don't want/need has a lot of advantages too.
</strong>
Such as....?
<strong>This would be equivalent to the iPod only playing some proprietary Apple audio format, that only Apple software could encode or decode.</strong>
What would be wrong with that? I would love it if Apple decided to use something more efficient than MP3 for the iPod and iTunes. As long as this was an option, not a requirement. If it meant that the tunes you ripped couldn't be distributed as easily, this makes the music industry a bit happier. Apple is firm on its "don't steal music" policy. M$ has its own proprietary format for music, that has better sound quality thatn MP3, and the only reason that people dislike it is because Media Player now doesn't provide full support MP3. If Apple used a new format for people that wanted to use it, yet still reatined full support for MP3, I don't see a problem. Same goes for Video. If Apple were to provide an option of a better solution, and still support the common standard, I don't see a problem.
<strong>
P.S. And Gizwald, why did you have to end your previously intelligent post with some junior-high insult involving nationality and intelligence? It only served to devalue your other comments.
</strong>
Sorry, I was drinking. I expected more from a Canadian.
<strong>Another problem I have with the 20gig iVideoCamera is that where the heck are you going to store this 20gig when you want the camera drive free again? certainly not on my 20gig iBook!!!!!</strong><hr></blockquote>
How much space will it take after you edit all the crappy footage out? Besides I thought that was what the Superdrive was for....
I still think the reality point now is a still camera w/ a hard drive (though being able to capture QT video would be a nice bonus). I have been using digital cameras exclusively for the last two years for our family photos and the big downers are: Storage space, battery life, and having to tote a battery charger rather than being able to charge in camera (AC brick vs. AC brick and charger). let me charge and transfer by the firewire port and I'm in love.
While I agree that there are some advantages with a HD for storage capacity in a dv cam, there is a MAJOR disadvantage that no one has mentioned. VIBRATION and CENTRIFUGAL FORCE.
To my limited knowledge ( ), not even IBM's Microdrive technology can spin and the speed needed to capture video. Heck, even people using 80GB Maxtor Firewire external HDs as video scratch disks have problems with dropped frames and such... the only thing that will help this situation are advances in nanotechnology in my opinion. Even if the industry did come up with something small enough and fast enough for capture video, is would be releasing some serious centrifugal force and would cause small amounts of vibration. Think of how slow a tape records...it's almost static in comparison and vibration is nearly untraceable.
A micro spinning HD does not seem to be a viable solution. I've got an iPod, and I wouldn't want its quiet wine next to my face if I were recording dv with a similar HD.
It would make way more sense to develop an nanotechnological advance in memory chips, such as the teracrystals (storage bonded onto the surface of a diamond via a pyradine compound) the U.S. government is using for satellite weather pattern recognition (among other things).
Besides memory chips are MUCH MUCH faster than Hard Disks.
A secondary video function on a premium quality digital still camera could probably make good use of a 5-20Gb HDD right now. A 3.6MB/s data rate for MiniDV could actually run on a drive spinning a lot slower than normal (with an adequate buffer to protect against dropped frames.) MPEG2-4, or QT would cut that bit rate down considerably and allow 2 hours or so even on a 5GB disc. That's good as a nice bonus, for web ready shots, e-mail attachments, even a home video DVD. This is the one-for-all solution for a tourist. Meant primarily to be a still camera, video tags along because [a]it can and [b] doing so reduces the need for you to bring a second device. 20GB and an MPEG recording scheme would give 6-8 hours of standard def resolution, more if you don't need 720x480. An expensive gadget for sure, but not without appeal.
I don't know what memory you're referring too, but all the tests I've seen show that even drives as small as the IBM microdrive are faster than typical flash memory. Unless cameras were to use some kind of big GB+ RAM disk while offloading to harddrives every so many seconds? I'm not sure where the video problems lie. Modern drives, even 5400 have throughput that is much greater than the data-rate of MiniDV. I think the problems come when you try to do real time editing/encoding/NLE. Simply recording the stream shouldn't tax a drive at all.
Pro cameras have the option to remove drive noise completely -- they can connect drives via firewire and have the camera automatically feed the data to the drive. There's no need to have ANY kind of physical storage on the drive itself. Perfectly quiet. Good for a studio, but possibly a headache in the field. Then again, a Native HD DV stream would require a whopping 150MBps (1920x1080@24p). You need an array to keep up with that, I wonder what George Lucas uses? I remember him saying that some things were captured directly to hard drives? The measly data rate of MiniDV is barely one 50th that of Native HD, we could certainly get by with hard-drives.
But tape has advantages too. MiniDV is lossless, and you have an instant (relatively stable) archival stock -- no need for down-loading and burning, the originals are there if you need to go back and re-edit something. The format is lossless (something you can't really do right now on small hard-drives unless you plan on running out of room fast.
This gives me an intriquing idea for a digital still/video convergence that combines the advantages of a small internal hard-drive with those of tape. You make a still camera with a very strong battery, good (fast) optical zoom lens, good controls, and an internal 1.8" (5-20GB depending on market prices) and a firewire port. It works as a still camera, or as a video camera, but obviously it's MiniDV format storage is limited.
Now look at a MiniDV tape. Pretty small, huh? Take that tape and wrap a slim walkman-type enclosure around it, add it's own li-poly battery. Via firewire, this little MiniDV deck snaps onto the bottom of your still/QT camera and turns it into a MiniDV camera. Still expensive, but a legit 2-in-1 device. When you want to dump vdieo to your computer for editing you don't even need to use the whole camera, just the deck. While someone edits, someone else can shoot stills. When you travel you decide how much digital camera you really need to take with you. The still camera being about the size of your average consumer digital camera, and the whole widget together being about twice that size (or still about as big as the smaller MiniDV cameras.)
Enough ranting, insomnia finally fading, me sleep now. bye
The drive comes off and can be plugged directly into an Avid, no digitizing needed. So a hard disk camcorder can be done, but do consumers really need it. Would they buy it?
Bah! Accidently deleted, now I have to retype! Anyone else think that IE should have an "undo" feature?
VIBRATION and CENTRIFUGAL FORCE.
No problems, the way I see it. As DaeargiMan points out, there already is a consumer camera that uses high-speed spinning disc technology. From my experience, optical drives are much louder than HDs. I don't own an iPod, but I doubt it's much louder than the motor noise in my mini-DV Cam.
It would make way more sense to develop...
Yeah, but since that technology is not really available, "we are no longer having the same discussion". Foresight is not allowed in this here discussion. Please refrain from using your imagination.
MiniDV is lossless
Ok... Uh... No. It uses intraframe compression. Good point about the instant backup, tho. And I like the idea of the add-on tape drive. Not very economical, but a cool idea nonetheless.
Kind of makes the idea of a consumer HD Camcorder silly doesn't it?
Not really... I like the idea of a camera with a huge internal capacty and no removable media. Those disks are less than 5GB. That's not very much video. I bet within a year of today, iPod drives will be at at least 30 GB. 3 Years... Who knows?
I agree with Matsu that the limitation of sequential access of a tape drives me nuts. I'd be willing to put up with somewhat lower capacity and shorter batterly life to be able to pull stuff of a disk in random sequence. The current miniHD of the iPod doesn't seem like a winner, though.
What about this possibility:
Rather than the miniHD of the iPod, what about a mini DVRW drive? Nintendo is using disks that are 80 mm (a tad over 3") across for the GameCube, which would be small enough for a hand-held camera. This would give you a replaceable/reuseable medium just as convenient as tapes, along with the random access that Matsu cites as a big advantage.
I'm no expert in the field, so maybe DVD writers can't write in real time? I don't know. CD-RW is up to 24x or 32x or higher on writing now, so maybe DVD-RW can keep up. With the new double-layer DVDs, the storage capacity should be pretty good. I suspect battery life would be a problem though - burning a DVD has got to consume significantly more power than writing to a HD. The rest of the camera consumes a lot of power too, though, so maybe the percentage increase would be relatively small.
Is this feasible? It at least seems in the realm of the possible using current technology. I'm not sure of its real practicality, though... <img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" />
A hard disk based camera is inevitable - that is my opinion after 13 years of working with (and trying to avoid) linear tape based cameras and recorders.
Since 1995 there has been the <a href="http://www.tvcameramen.com/equipment/equipment15.htm"; target="_blank">Ikegami Edit Camera</a> which allows such feats as looped recording - simply point the camera at the subject, and when the action has finished press ok - the HD stores the last few minutes of footage. This method captures things that would either waste hours of linear tape, or would simply be missed because of delayed reactions pressing the record button.
Linearity is one of those things that you have to get past - I remember my first experience with a Mini Disc Walkman - it was a little strange to be able to record anywhere on the disk, and to rearrange tracks according to taste - my mind still had metaphors from my record decks, and from my tape Walkman. The same will be true of the HD camera - we will wonder how we coped with linear tape, rewinding, recording 2 seconds over the end of the original footage to maintain irrellevant timecode, capture logs, etc. How much easier will a date stamped clip file list be!?
Foveon recently announced its <a href="http://www.foveon.com/X3_vps.html"; target="_blank">X3 technology</a> which allows for variable pixel size recording, with none of the post-processing problems of standard CCD based cameras. This potentially allows for clean recording of perfect digital video signals, and I would suggest that a HD is the best method for capturing this material. Someone will do it, and very soon.
Anyhow, thats my first-post rant, you can see my motion graphics work at <a href="http://elasticspace.com"; target="_blank">elastic space</a>.
It's good to see that I'm not the only one who thinks a HD based camera is not too far off. It's things like this that let me convince myself that I'm not crazy.
Yes consumers would buy it - anything that makes their lives easier. How many consumers get around to cataloguing or even labelling their tapes? How many remember where the long forgotten 'kids in the car, summer 98' scene was? This is easily solved by iMovie, iPhoto or iView Media Pro.
Sorry for repeating you DaeargiMan, your links to Ikegami and Hitachi are super. I am not sure that DVD-RAM solves everything though - it seems a very large physical format for a camcorder - but I would be happy to have all my rushes as DVD files rather than DV tapes.
Insomnia is back, 'cause I woke up at 5:30 this morning. WTF? 4-5 hours sleep... errr...
Anyway,
Sorry about the lossless thing. Ooops, my bad. MiniDV and Mpeg are both 'lossy' but DV streams have a generally higher bit rate. I think only QT is lossless in that you can play around with it without degrading it. It's just that with the MiniDV tape you always have the original tape stock to go back to if you screw something up on your hard-drive. A tapeless (HD based system) might lose that advantage.
What format do most people work in to avoid excessive processing degradation?
I really don't get why a hard-drive has to spin faster for MiniDV capture? Every commercially available HDD has a higher throughput than neccessary for MiniDV bit-rates (3.6MB/s). I don't get it?
In some sense working with DV is 'lossless' in that the original compression applied by the camera when recording the video signal to tape is not lost when editing. Due to the intraframe compression system taking the video through the editing stage in FCP or Premiere means no loss in quality, provided you have the right settings, right back onto tape.
The only time that DV loses its quality is when effects or transitions are applied, and the material gets recompressed.
Its a fairly nice codec, certainly better quality than the Hitachi DVD camera, or Mpeg2/4.
As for advantages of tape over HD (or linear over non-linear), I should rephrase that a bit, in that I simply don't see as many advantages to non-linear as others seem to. I really like having all my footage in front of me in clips, ready to rearrange, edit, etc. I would do exactly the same thing with a HD-based system; I'd download all the footage to my computer's HD and do all the editing there. The main benefit that I see from a HD-based system is being able to transfer it all to my computer faster. Consumers seem to love to edit their footage into shorter, more entertaining pieces, add music, titles, etc. Therefore, I'm not sure there's much of a market for a camera that would write straight to DVD or a HD in MPEG, ready for "broadcast." Not many people have a lot of different kinds of footage (for different projects) on the same MiniDV tape, so being able to jump to the appropriate section quickly and grab a specific clip I think has limited functionality. Now, granted, once capacity gets to the point of holding 5 hours of footage, then linearity will become more of a problem.
Archivally, you are looking at more time to transfer from your HD-based camera to PC, then burn onto DVD or CD-RW, than to just slap a label on your Mini-DV tape and throw it in the drawer (not to mention tape being a better anchival medium than discs). Now, for non-consumer products, I agree that there is a benefit to straight-to-HD, in that they will also have an automated DLT or other backup system for their footage onto good archival media. Also, the end footage often gets tranfered to film anyway (Apple's profile of Soderburg's upcoming is a good example).
As for why HDs can have trouble keeping up with DV streams, I believe it is because while the drive supports a transfer speed adequate for DV, that speed only applies to a linear, continuous segment, and once the drive head has to move, it needs to do that fast enough before the drive buffer overflows. Even on a freshly defragged HD, I believe that the drive has to update it's catalogue/index (can't remember the technical terms) every time it writes to a new segment, so the write head still needs to jump around. This is my guess based on my limited knowledge of how HDs work. Anyone with more knowledge that can shed more light on this?
Gizwald, get a hold of yourself (and reality). No one is shooting down hypothesis or imagination. It was just pointing out major flaws in the idea (from the original post) which was conveyed as a product that was being sold to Samsumg, based on current, specific technology. I have no doubt that MiniDV will be replaced, and in fact don't doubt that current hard drive technology will be replaced in the not-so-distant future, with something more akin to RAM than HDs. You make good points and I appreciate the discussion, but don't take it personally when someone disagrees. Just try to convince them that you are right! :-)
FYI: CCIR 601 video, which is what digital video is recording as, is a little over 20 MB/sec. The 3.6 MB/sec is a compression put on it during capture (the process of getting it off of the tape and putting it on your drive) in order to edit w/out dropping frames.
<strong>It was just pointing out major flaws in the idea (from the original post) which was conveyed as a product that was being sold to Samsumg, based on current, specific technology.</strong><hr></blockquote>
Oh, so that's what you were getting at. The guy who wrote the first post is BSing us. If that is what you were saying, I agree. Seems pretty unlikely, also, that the guys as Samsung never heard of the iPod.
Archivally, you are looking at more time to transfer from your HD-based camera to PC, then burn onto DVD or CD-RW, than to just slap a label on your Mini-DV tape and throw it in the drawer (not to mention tape being a better anchival medium than discs). Now, for non-consumer products, I agree that there is a benefit to straight-to-HD, in that they will also have an automated DLT or other backup system for their footage onto good archival media. Also, the end footage often gets tranfered to film anyway (Apple's profile of Soderburg's upcoming is a good example).
</strong><hr></blockquote>
I agree that tapes are the lesser of two evils, I am willing to work with capture lists and edls for the moment. But I always end up storing clips and snippets for re-use on the HD, and returning to re-create an old project is always a slow, non-frame-accurate nightmare. Having an old project backed up on DVD or DLT in its entirity at the end of the project, without the unused scenes, is always an easier prospect.
Actually I think a HD camera would cause a revolution in the motion graphics industry, being able to go and capture short snippets of moving textures and graphical elements, without transfer times, would be a dream.
Given an Xserve RAID, with an abundance of 120GB IDE disks, hopefully I will never have to back up a project again.
Comments
why mess with something thats already decent enough?
Because that's what Apple does. That's what makes them so great.
<strong>I don't expect to see a camera like this tomorrow, but it will happen eventualy. You'd be a fool to disagree. Maybe it won't use a megnetic HD, but, someday, within the next few years, somone will make a removable-medialess camera.</strong><hr></blockquote>
The whole point of this thread was based around an idea that was not way in the future. To say that one would be a fool to disagree that "it will happen eventually" is a foolish statement. Once you abstract away the magnetic HD from the discussion, we are no longer having the same discussion. The main problems with the idea revolve solely around the proposed use of a HD. In three years, when the iPod sized HDs are at about 100 GB, MiniDV will no doubt have been replaced with a much higher-capacity fomat (think mini version of Super DLT - now we're talking capacity!!!).
While I wholeheartedly agree that MiniDV has some issues, and that something will come along to replace it, the proposed solution just would have more downsides than advantages. I'd love to have a non-linear system, but downloading all the footage to your computer and then editing out the stuff you don't want/need has a lot of advantages too. I have actually never experienced the dropped frames problem you mention once tapes have been written to a few times.
And as for MPEG 4, it's great, but for web video primarily. I personally would much rather have the original footage at a very high bitrate, and encode it later for web braodcasting.
Look at the iPod! It is a better way of doing the same thing, while retaining compatibility. This whole suggestion would involve completely breaking the mould, and compatibility with other products. This would be equivalent to the iPod only playing some proprietary Apple audio format, that only Apple software could encode or decode. Apple has finally smartened up and moved away from doing this exact thing. Instead, they are proposing new open standards and getting the competition on board, or acepting other standards. If they try to do this, it will only succeed if they get the other major hardware and software vendors on board. You're talking about a revolution in the way it's done, and it could only happen with a lot of support behind it.
I agree that an iPod-style HD would be great in a still camera with QT movie mode. As it is, my Olympus digital camera can do this, but is limited to the capacity of the SmartMedia card.
P.S. And Gizwald, why did you have to end your previously intelligent post with some junior-high insult involving nationality and intelligence? It only served to devalue your other comments.
Yup and a blank CD is $0.39
I've still got an iPod.
Go figure.
<strong>The whole point of this thread was based around an idea that was not way in the future. To say that one would be a fool to disagree that "it will happen eventually" is a foolish statement. Once you abstract away the magnetic HD from the discussion, we are no longer having the same discussion.
</strong>
That's true. But you guys were shooting down the idea as if it was totally stupid and that it could never happen. I get the impression, from reading your post, that you think that tapes will never ever be replaced by some sort of fixed media. I disagree. It will happen. (in the comsumer market, at least) But I do think that some better technology needs to be developed first... Higher density storage, and maybe a better standard video codec.
<strong>
In three years, when the iPod sized HDs are at about 100 GB, MiniDV will no doubt have been replaced with a much higher-capacity fomat.
</strong>
I think it will take longer than 3 years for Mini DV to be replaced by something better. Standard consumer technology moves on a much slower cycle than that. DVD technology has been around for about 10 years, if I remember correctly. And it's just begun to catch on in the last few years.
<strong>
downloading all the footage to your computer and then editing out the stuff you don't want/need has a lot of advantages too.
</strong>
Such as....?
<strong>This would be equivalent to the iPod only playing some proprietary Apple audio format, that only Apple software could encode or decode.</strong>
What would be wrong with that? I would love it if Apple decided to use something more efficient than MP3 for the iPod and iTunes. As long as this was an option, not a requirement. If it meant that the tunes you ripped couldn't be distributed as easily, this makes the music industry a bit happier. Apple is firm on its "don't steal music" policy. M$ has its own proprietary format for music, that has better sound quality thatn MP3, and the only reason that people dislike it is because Media Player now doesn't provide full support MP3. If Apple used a new format for people that wanted to use it, yet still reatined full support for MP3, I don't see a problem. Same goes for Video. If Apple were to provide an option of a better solution, and still support the common standard, I don't see a problem.
<strong>
P.S. And Gizwald, why did you have to end your previously intelligent post with some junior-high insult involving nationality and intelligence? It only served to devalue your other comments.
</strong>
Sorry, I was drinking. I expected more from a Canadian.
<hr></blockquote>
[ 05-14-2002: Message edited by: Gizwald ]
[ 05-14-2002: Message edited by: Gizwald ]
[ 05-14-2002: Message edited by: Gizwald ]</p>
<strong>Another problem I have with the 20gig iVideoCamera is that where the heck are you going to store this 20gig when you want the camera drive free again? certainly not on my 20gig iBook!!!!!</strong><hr></blockquote>
How much space will it take after you edit all the crappy footage out? Besides I thought that was what the Superdrive was for....
I still think the reality point now is a still camera w/ a hard drive (though being able to capture QT video would be a nice bonus). I have been using digital cameras exclusively for the last two years for our family photos and the big downers are: Storage space, battery life, and having to tote a battery charger rather than being able to charge in camera (AC brick vs. AC brick and charger). let me charge and transfer by the firewire port and I'm in love.
To my limited knowledge (
A micro spinning HD does not seem to be a viable solution. I've got an iPod, and I wouldn't want its quiet wine next to my face if I were recording dv with a similar HD.
It would make way more sense to develop an nanotechnological advance in memory chips, such as the teracrystals (storage bonded onto the surface of a diamond via a pyradine compound) the U.S. government is using for satellite weather pattern recognition (among other things).
Besides memory chips are MUCH MUCH faster than Hard Disks.
Any comments?
[ 05-14-2002: Message edited by: xServer ]</p>
A secondary video function on a premium quality digital still camera could probably make good use of a 5-20Gb HDD right now. A 3.6MB/s data rate for MiniDV could actually run on a drive spinning a lot slower than normal (with an adequate buffer to protect against dropped frames.) MPEG2-4, or QT would cut that bit rate down considerably and allow 2 hours or so even on a 5GB disc. That's good as a nice bonus, for web ready shots, e-mail attachments, even a home video DVD. This is the one-for-all solution for a tourist. Meant primarily to be a still camera, video tags along because [a]it can and [b] doing so reduces the need for you to bring a second device. 20GB and an MPEG recording scheme would give 6-8 hours of standard def resolution, more if you don't need 720x480. An expensive gadget for sure, but not without appeal.
I don't know what memory you're referring too, but all the tests I've seen show that even drives as small as the IBM microdrive are faster than typical flash memory. Unless cameras were to use some kind of big GB+ RAM disk while offloading to harddrives every so many seconds? I'm not sure where the video problems lie. Modern drives, even 5400 have throughput that is much greater than the data-rate of MiniDV. I think the problems come when you try to do real time editing/encoding/NLE. Simply recording the stream shouldn't tax a drive at all.
Pro cameras have the option to remove drive noise completely -- they can connect drives via firewire and have the camera automatically feed the data to the drive. There's no need to have ANY kind of physical storage on the drive itself. Perfectly quiet. Good for a studio, but possibly a headache in the field. Then again, a Native HD DV stream would require a whopping 150MBps (1920x1080@24p). You need an array to keep up with that, I wonder what George Lucas uses? I remember him saying that some things were captured directly to hard drives? The measly data rate of MiniDV is barely one 50th that of Native HD, we could certainly get by with hard-drives.
But tape has advantages too. MiniDV is lossless, and you have an instant (relatively stable) archival stock -- no need for down-loading and burning, the originals are there if you need to go back and re-edit something. The format is lossless (something you can't really do right now on small hard-drives unless you plan on running out of room fast.
This gives me an intriquing idea for a digital still/video convergence that combines the advantages of a small internal hard-drive with those of tape. You make a still camera with a very strong battery, good (fast) optical zoom lens, good controls, and an internal 1.8" (5-20GB depending on market prices) and a firewire port. It works as a still camera, or as a video camera, but obviously it's MiniDV format storage is limited.
Now look at a MiniDV tape. Pretty small, huh? Take that tape and wrap a slim walkman-type enclosure around it, add it's own li-poly battery. Via firewire, this little MiniDV deck snaps onto the bottom of your still/QT camera and turns it into a MiniDV camera. Still expensive, but a legit 2-in-1 device. When you want to dump vdieo to your computer for editing you don't even need to use the whole camera, just the deck. While someone edits, someone else can shoot stills. When you travel you decide how much digital camera you really need to take with you. The still camera being about the size of your average consumer digital camera, and the whole widget together being about twice that size (or still about as big as the smaller MiniDV cameras.)
Enough ranting, insomnia finally fading, me sleep now. bye
<a href="http://www.hitachi.com.au/publishfiles/product_3_23_98.htm"; target="_blank">http://www.hitachi.com.au/publishfiles/product_3_23_98.htm</a>;
Kindof makes the idea of a consumer HD Camcorder silly doesn't it?
Ikegami makes a broadcast grade Hard Disk camcorder (for news gathering and the like).
<a href="http://www.ikegami.com/201w1.html"; target="_blank">http://www.ikegami.com/201w1.html</a>;
The drive comes off and can be plugged directly into an Avid, no digitizing needed. So a hard disk camcorder can be done, but do consumers really need it. Would they buy it?
VIBRATION and CENTRIFUGAL FORCE.
No problems, the way I see it. As DaeargiMan points out, there already is a consumer camera that uses high-speed spinning disc technology. From my experience, optical drives are much louder than HDs. I don't own an iPod, but I doubt it's much louder than the motor noise in my mini-DV Cam.
It would make way more sense to develop...
Yeah, but since that technology is not really available, "we are no longer having the same discussion". Foresight is not allowed in this here discussion. Please refrain from using your imagination.
MiniDV is lossless
Ok... Uh... No. It uses intraframe compression. Good point about the instant backup, tho. And I like the idea of the add-on tape drive. Not very economical, but a cool idea nonetheless.
Kind of makes the idea of a consumer HD Camcorder silly doesn't it?
Not really... I like the idea of a camera with a huge internal capacty and no removable media. Those disks are less than 5GB. That's not very much video. I bet within a year of today, iPod drives will be at at least 30 GB. 3 Years... Who knows?
What about this possibility:
Rather than the miniHD of the iPod, what about a mini DVRW drive? Nintendo is using disks that are 80 mm (a tad over 3") across for the GameCube, which would be small enough for a hand-held camera. This would give you a replaceable/reuseable medium just as convenient as tapes, along with the random access that Matsu cites as a big advantage.
I'm no expert in the field, so maybe DVD writers can't write in real time? I don't know. CD-RW is up to 24x or 32x or higher on writing now, so maybe DVD-RW can keep up. With the new double-layer DVDs, the storage capacity should be pretty good. I suspect battery life would be a problem though - burning a DVD has got to consume significantly more power than writing to a HD. The rest of the camera consumes a lot of power too, though, so maybe the percentage increase would be relatively small.
Is this feasible? It at least seems in the realm of the possible using current technology. I'm not sure of its real practicality, though... <img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" />
As a full-time motion graphic designer and filmmaker I have been using HD capture in all my projects, first of all using an <a href="http://www.dps.com/custserv/products.nsf/418b34bb465fcab38525651c0080fa0f/a93c35827d193eb6852569430079900b?OpenDocument"; target="_blank">Amiga PAR</a> in 1993, then on to SGI realtime capture cards, and eventually G4 Ti lugged around after a Canon 3 Chip CCD, and using Sony digital stills cameras for stop-frame animation.
Since 1995 there has been the <a href="http://www.tvcameramen.com/equipment/equipment15.htm"; target="_blank">Ikegami Edit Camera</a> which allows such feats as looped recording - simply point the camera at the subject, and when the action has finished press ok - the HD stores the last few minutes of footage. This method captures things that would either waste hours of linear tape, or would simply be missed because of delayed reactions pressing the record button.
Linearity is one of those things that you have to get past - I remember my first experience with a Mini Disc Walkman - it was a little strange to be able to record anywhere on the disk, and to rearrange tracks according to taste - my mind still had metaphors from my record decks, and from my tape Walkman. The same will be true of the HD camera - we will wonder how we coped with linear tape, rewinding, recording 2 seconds over the end of the original footage to maintain irrellevant timecode, capture logs, etc. How much easier will a date stamped clip file list be!?
Foveon recently announced its <a href="http://www.foveon.com/X3_vps.html"; target="_blank">X3 technology</a> which allows for variable pixel size recording, with none of the post-processing problems of standard CCD based cameras. This potentially allows for clean recording of perfect digital video signals, and I would suggest that a HD is the best method for capturing this material. Someone will do it, and very soon.
Anyhow, thats my first-post rant, you can see my motion graphics work at <a href="http://elasticspace.com"; target="_blank">elastic space</a>.
//m
Sorry for repeating you DaeargiMan, your links to Ikegami and Hitachi are super. I am not sure that DVD-RAM solves everything though - it seems a very large physical format for a camcorder - but I would be happy to have all my rushes as DVD files rather than DV tapes.
Anyway,
Sorry about the lossless thing. Ooops, my bad. MiniDV and Mpeg are both 'lossy' but DV streams have a generally higher bit rate. I think only QT is lossless in that you can play around with it without degrading it. It's just that with the MiniDV tape you always have the original tape stock to go back to if you screw something up on your hard-drive. A tapeless (HD based system) might lose that advantage.
What format do most people work in to avoid excessive processing degradation?
I really don't get why a hard-drive has to spin faster for MiniDV capture? Every commercially available HDD has a higher throughput than neccessary for MiniDV bit-rates (3.6MB/s). I don't get it?
The only time that DV loses its quality is when effects or transitions are applied, and the material gets recompressed.
Its a fairly nice codec, certainly better quality than the Hitachi DVD camera, or Mpeg2/4.
Archivally, you are looking at more time to transfer from your HD-based camera to PC, then burn onto DVD or CD-RW, than to just slap a label on your Mini-DV tape and throw it in the drawer (not to mention tape being a better anchival medium than discs). Now, for non-consumer products, I agree that there is a benefit to straight-to-HD, in that they will also have an automated DLT or other backup system for their footage onto good archival media. Also, the end footage often gets tranfered to film anyway (Apple's profile of Soderburg's upcoming is a good example).
As for why HDs can have trouble keeping up with DV streams, I believe it is because while the drive supports a transfer speed adequate for DV, that speed only applies to a linear, continuous segment, and once the drive head has to move, it needs to do that fast enough before the drive buffer overflows. Even on a freshly defragged HD, I believe that the drive has to update it's catalogue/index (can't remember the technical terms) every time it writes to a new segment, so the write head still needs to jump around. This is my guess based on my limited knowledge of how HDs work. Anyone with more knowledge that can shed more light on this?
Gizwald, get a hold of yourself (and reality). No one is shooting down hypothesis or imagination. It was just pointing out major flaws in the idea (from the original post) which was conveyed as a product that was being sold to Samsumg, based on current, specific technology. I have no doubt that MiniDV will be replaced, and in fact don't doubt that current hard drive technology will be replaced in the not-so-distant future, with something more akin to RAM than HDs. You make good points and I appreciate the discussion, but don't take it personally when someone disagrees. Just try to convince them that you are right! :-)
<strong>It was just pointing out major flaws in the idea (from the original post) which was conveyed as a product that was being sold to Samsumg, based on current, specific technology.</strong><hr></blockquote>
Oh, so that's what you were getting at. The guy who wrote the first post is BSing us. If that is what you were saying, I agree. Seems pretty unlikely, also, that the guys as Samsung never heard of the iPod.
<strong>
Archivally, you are looking at more time to transfer from your HD-based camera to PC, then burn onto DVD or CD-RW, than to just slap a label on your Mini-DV tape and throw it in the drawer (not to mention tape being a better anchival medium than discs). Now, for non-consumer products, I agree that there is a benefit to straight-to-HD, in that they will also have an automated DLT or other backup system for their footage onto good archival media. Also, the end footage often gets tranfered to film anyway (Apple's profile of Soderburg's upcoming is a good example).
</strong><hr></blockquote>
I agree that tapes are the lesser of two evils, I am willing to work with capture lists and edls for the moment. But I always end up storing clips and snippets for re-use on the HD, and returning to re-create an old project is always a slow, non-frame-accurate nightmare. Having an old project backed up on DVD or DLT in its entirity at the end of the project, without the unused scenes, is always an easier prospect.
Actually I think a HD camera would cause a revolution in the motion graphics industry, being able to go and capture short snippets of moving textures and graphical elements, without transfer times, would be a dream.
Given an Xserve RAID, with an abundance of 120GB IDE disks, hopefully I will never have to back up a project again.