VideoHelp Forum
+ Reply to Thread
Page 3 of 3
FirstFirst 1 2 3
Results 61 to 78 of 78
Thread
  1. Member Cornucopia's Avatar
    Join Date
    Oct 2001
    Location
    Deep in the Heart of Texas
    Search PM
    What you show in that clip appear to be standard WSS, in the normal vbi portion, on ~line22. Not in the actual "visible image area" For well designed devices, this would still be recognized, even if it didn't pass it along to be stored in the capture (e.g. in the case of 480 line units).

    I'm not sure that your representation of the switch between WS and 4:3 is correct however, as the frame never changes...and it should. I am thinking that the capture device doesn't understand the WSS and you had to manually set the aspect? And normal anamorphic video allows you to see more of the sides, but this doesn't appear to, though it is a bit difficult to follow and know for sure. If you are using this for historical curiosity purposes, that device is probably ok, but if you are trying to get good captures for preservation/archival, etc, I would not recommend this route. If this avenue of investigation is that much importance to you, it is worth it to get a desktop and a PCI card like those Ospreys.


    Scott
    Quote Quote  
  2. Originally Posted by dellsam34 View Post
    Originally Posted by Videogamer555 View Post
    The Black Magic Design Intensity Shuttle for USB-3 is the only (at this time) product I'm aware of that could capture all 486 active image lines of an NTSC video signal. Do you know of any alternatives that also capture 486 lines and will run from USB-3 (doesn't require a desktop PC with expansion card)?
    I'm not sure if the BM IS can capture 486 lines, And I'm not aware of any good capture device that uses USB 3.x let alone captures 486, You will need a pro capture device, it comes in two flavors, PCI or SDI (then SDI2USB3). and no I don't know where you can get one of those besides just searching online.
    The BM-IS absolutely can capture 486 lines, and yes it uses USB-3.x (there was also another version, which was for Mac instead of PC, and it used a Tunderbolt port instead of USB-3, though I don't have this Mac version, just the PC version). Problem is it is no longer being made, so I can't recommend it to someone as a product they should use if they are in the same situation as I was before I bought it. That's why I hope to find an alternative.

    You said PCI or SDI. That doesn't make sense. SDI isn't a motherboard, while PCI is. SDI is a digital video interface, much like HDMI or DVI. And SDI doesn't at all have anything to do with the BM-IS which is an analog video capture device.
    Quote Quote  
  3. Originally Posted by Cornucopia View Post
    What you show in that clip appear to be standard WSS, in the normal vbi portion, on ~line22. Not in the actual "visible image area" For well designed devices, this would still be recognized, even if it didn't pass it along to be stored in the capture (e.g. in the case of 480 line units).

    I'm not sure that your representation of the switch between WS and 4:3 is correct however, as the frame never changes...and it should. I am thinking that the capture device doesn't understand the WSS and you had to manually set the aspect? And normal anamorphic video allows you to see more of the sides, but this doesn't appear to, though it is a bit difficult to follow and know for sure. If you are using this for historical curiosity purposes, that device is probably ok, but if you are trying to get good captures for preservation/archival, etc, I would not recommend this route. If this avenue of investigation is that much importance to you, it is worth it to get a desktop and a PCI card like those Ospreys.


    Scott
    If you read the description on the thread I linked to, you would see how I was using a computer-based oscilloscope (the Picoscope 2204A) to get a RAW capture of the analog signal, and then converting it into a stream of fields through multiple complex steps, including using audio processing software to restore the correct DC level, and writing my own software to sync with the sync pulses and demodulate the color signal from the chroma carrier. I have a very in-depth technical description in that thread, and you really should read it, to understand why the video looks like it does.
    Quote Quote  
  4. Member Cornucopia's Avatar
    Join Date
    Oct 2001
    Location
    Deep in the Heart of Texas
    Search PM
    I read and understood that. But your mention above alludes to it as if it is operating like a standard capture card, and I think you made assumptions based on that which don't apply with normal cards.

    Scott
    Quote Quote  
  5. Originally Posted by Videogamer555 View Post
    By card do you mean like a PCI-e card? I don't have a desktop, only a laptop, so it would have to be a USB-3 based device to work on my laptop.
    Thunderbolt PCIe enclosures are a thing. I generally recommend Sonnet for Thunderbolt expansion chassis, but OWC and Startech have some OK ones as well. The Highpoint RocketStor 6661A has issues.
    Last edited by energizerfellow; 30th May 2022 at 17:24.
    Quote Quote  
  6. Capturing Memories dellsam34's Avatar
    Join Date
    Jan 2016
    Location
    Member Since 2005, Re-joined in 2016
    Search PM
    Originally Posted by Videogamer555 View Post
    The BM-IS absolutely can capture 486 lines, and yes it uses USB-3.x (there was also another version, which was for Mac instead of PC, and it used a Tunderbolt port instead of USB-3, though I don't have this Mac version, just the PC version). Problem is it is no longer being made, so I can't recommend it to someone as a product they should use if they are in the same situation as I was before I bought it. That's why I hope to find an alternative.

    You said PCI or SDI. That doesn't make sense. SDI isn't a motherboard, while PCI is. SDI is a digital video interface, much like HDMI or DVI. And SDI doesn't at all have anything to do with the BM-IS which is an analog video capture device.
    I'm aware BM IS is a USB 3 device, I just didn't think it could capture 486, You will have to see the native driver for it, MediaExpress will probably captures 486 regardless what capture card in action, it just blanks the extra 6 lines.

    -PCI Card is a capture card that has analog inputs and fits into a PCIe slot for desktops. Some have combined analog, SDI and HDMI inputs.
    -SDI device is an external capture device that has analog inputs and only SDI output, to connect it to the laptop you would need a SDI2USB3 adapter, It can also be used with a desktop using a PCI-SDI card.

    Capturing is a legacy task, there is no good quality new capture cards being made other than the chinese crap, and certainly no USB3 devices, Anything good is 2000's.
    Quote Quote  
  7. Originally Posted by dellsam34 View Post
    Originally Posted by Videogamer555 View Post
    The BM-IS absolutely can capture 486 lines, and yes it uses USB-3.x (there was also another version, which was for Mac instead of PC, and it used a Tunderbolt port instead of USB-3, though I don't have this Mac version, just the PC version). Problem is it is no longer being made, so I can't recommend it to someone as a product they should use if they are in the same situation as I was before I bought it. That's why I hope to find an alternative.

    You said PCI or SDI. That doesn't make sense. SDI isn't a motherboard, while PCI is. SDI is a digital video interface, much like HDMI or DVI. And SDI doesn't at all have anything to do with the BM-IS which is an analog video capture device.
    I'm aware BM IS is a USB 3 device, I just didn't think it could capture 486, You will have to see the native driver for it, MediaExpress will probably captures 486 regardless what capture card in action, it just blanks the extra 6 lines.

    -PCI Card is a capture card that has analog inputs and fits into a PCIe slot for desktops. Some have combined analog, SDI and HDMI inputs.
    -SDI device is an external capture device that has analog inputs and only SDI output, to connect it to the laptop you would need a SDI2USB3 adapter, It can also be used with a desktop using a PCI-SDI card.

    Capturing is a legacy task, there is no good quality new capture cards being made other than the chinese crap, and certainly no USB3 devices, Anything good is 2000's.

    Thanks for tte info. I would argue though that capture isn't legacy. It us specifically ANALOG video capture that is legacy. There's plenty of HDMI digital video capture cards around.
    Also there's at least one analog capture device still being manufacturered. I've heard people recommend it as good quality too, even though it does seem to be from a Chinese company. It's the Hauppauge analog video capture card. It's consumer level, not professional, so it only does 480 high video, not 486 high. But it is still being made and sold by the original company. I don't need to go to ebay to look for it.
    Quote Quote  
  8. Capturing Memories dellsam34's Avatar
    Join Date
    Jan 2016
    Location
    Member Since 2005, Re-joined in 2016
    Search PM
    Yes, I meant capturing analog tape based formats. Yes there are few exceptions but really depends on the OS. Windows 7 has the widest support for consumer capture cards. But with an SDI interface in place whether PCIe, USB3 or Thunderbolt the OS is taken out of the question, SDI is a professional port and has a wide support from all major OS's on the market, no need to fiddle with drivers like you do with consumer gear and their USB 2.0 drivers that are diminishing by the day.
    Last edited by dellsam34; 31st May 2022 at 03:26.
    Quote Quote  
  9. Originally Posted by dellsam34 View Post
    Yes, I meant capturing analog tape based formats. Yes there are few exceptions but really depends on the OS. Windows 7 has the widest support for consumer capture cards. But with an SDI interface in place whether PCIe, USB3 or Thunderbolt the OS is taken out of the question, SDI is a professional port and has a wide support from all major OS's on the market, no need to fiddle with drivers like you do with consumer gear and their USB 2.0 drivers that are diminishing by the day.
    The Hauppauge analog video capture device at https://hauppauge.com/pages/webstore2/webstore_usblive2.html is still a thing. It's still being sold and still being manufactured. And no mention on their website that it's getting close to being discontinued. I've heard people recommend this as one of the best consumer-level analog video capture USB dongles. I don't think it does 486 line capture, just 480 line capture, but it proves that analog video capture isn't completely dead yet. The website on this page, https://hauppauge.com/pages/products/data_usblive2.html , even advertise it as good for digitizing your old tape collections by saying "USB-Live2 is the easiest way to record your old VHS video tapes". Apparently there's enough people with old VHS tapes laying around, that selling a device like this is good business for the company.
    Quote Quote  
  10. Member
    Join Date
    May 2005
    Location
    Australia-PAL Land
    Search Comp PM
    Originally Posted by Videogamer555
    The Hauppauge analog video capture device at https://hauppauge.com/pages/webstore2/webstore_usblive2.html is still a thing. It's still being sold and still being manufactured.
    Do you have one of these and does it work, actually, on Windows 10 (I see Hauppauge says it does but...). And what capture software do you use?
    Quote Quote  
  11. Originally Posted by Alwyn View Post
    Originally Posted by Videogamer555
    The Hauppauge analog video capture device at https://hauppauge.com/pages/webstore2/webstore_usblive2.html is still a thing. It's still being sold and still being manufactured.
    Do you have one of these and does it work, actually, on Windows 10 (I see Hauppauge says it does but...). And what capture software do you use?
    I have been using it with Windows 10 all the time, works fine here.
    Capture to lossless 720x480 (or 720x576 for PAL) with a codec like Huffyuv, Lagarith, UTVideo, MagicYUV. Connect it via a suitable DVD recorder in passthrough to stabilize ("TBC") the picture for eliminating the flagging/line wiggle. Set the video levels right. There are plenty of post about this. Do the rest (filtering, compression...) in post processing.
    Forget about the 486 lines, stay with the standard 480 for NTSC.
    Quote Quote  
  12. Member
    Join Date
    May 2005
    Location
    Australia-PAL Land
    Search Comp PM
    Thanks Sharc. I was "asking for a friend".
    Quote Quote  
  13. The crux of this thread deals with the leap from analog to digital.

    The analog NTSC subcarrier frequency is 5*63/88 or 3.579545455 MHz. This is ironclad.

    The period of one horizontal scan line (1H) is 63.55555556 μS. This is also ironclad. (227.5 subcarrier cycles per H) 1H is the period from the leading edge of sync of one scan line to the leading edge of sync of the next scan line.

    The nominal horizontal blanking period is 10.7 μS. This leaves us with a visible scan line of 52.86 μS

    If the pixel clock is derived from the subcarrier frequency, this gives us a pixel period of 69.84127 nS

    This gives us 756.8 pixels for the visible portion of each scan line. 756.8 is inconvenient as it is not evenly divisible by 16.

    In analog video, the H blanking period can be fudged a little bit but the H frequency remains ironclad. This means there is a fudge factor in the duration of the visible part of a scan line.

    If you want a pixel count which is evenly divisible by 16 for each scan line, you have:

    768 pixels = 53.64 μS per visible line = 9.92 μS H blanking

    752 pixels = 52.52 μS per visible line = 11.03 μS H blanking

    720 pixels =50.29 μS per visible line = 13.27 μS H blanking

    Note that this does not give you a square pixel.
    Quote Quote  
  14. In analog broadcast NTSC there are 483 active scan lines:

    525 lines per frame

    525 - 483 = 42 lines

    42 / 2 = 21 lines per field

    Line 19: GCR (ghost cancelling reference)

    Line 21: captioning data
    Quote Quote  
  15. Originally Posted by chris319 View Post
    The crux of this thread deals with the leap from analog to digital.

    The analog NTSC subcarrier frequency is 5*63/88 or 3.579545455 MHz. This is ironclad.

    The period of one horizontal scan line (1H) is 63.55555556 μS. This is also ironclad. (227.5 subcarrier cycles per H) 1H is the period from the leading edge of sync of one scan line to the leading edge of sync of the next scan line.

    The nominal horizontal blanking period is 10.7 μS. This leaves us with a visible scan line of 52.86 μS

    If the pixel clock is derived from the subcarrier frequency, this gives us a pixel period of 69.84127 nS

    This gives us 756.8 pixels for the visible portion of each scan line. 756.8 is inconvenient as it is not evenly divisible by 16.

    In analog video, the H blanking period can be fudged a little bit but the H frequency remains ironclad. This means there is a fudge factor in the duration of the visible part of a scan line.

    If you want a pixel count which is evenly divisible by 16 for each scan line, you have:

    768 pixels = 53.64 μS per visible line = 9.92 μS H blanking

    752 pixels = 52.52 μS per visible line = 11.03 μS H blanking

    720 pixels =50.29 μS per visible line = 13.27 μS H blanking

    Note that this does not give you a square pixel.
    Just a note: I think you refer here to the Composite 4fsc sampling principle, which gives the 14 7/22 MHz = 14.3181818... MHz sampling rate for NTSC and 17.734475 MHz for PAL, leading to "unusual" Pixel Aspect Ratios, like 6/7 for NTSC for example.
    It's quite different from the Rec.601 standard which specifies a harmonized luma sampling rate of 13.5 MHz for both NTSC and PAL.
    Quote Quote  
  16. I think you refer here to the Composite 4fsc sampling principle, which gives the 14 7/22 MHz = 14.3181818... MHz sampling rate for NTSC and 17.734475 MHz for PAL, leading to "unusual" Pixel Aspect Ratios, like 6/7 for NTSC for example.
    It's quite different from the Rec.601 standard which specifies a harmonized luma sampling rate of 13.5 MHz for both NTSC and PAL.
    Yes, I had to work out those calculations in the 1980's when I worked with the Amiga computer and other graphics devices including the Apple ][, VIC-20 and C64, which were based on U.S. analog video standards so the user could see the output on his home TV set. You could get a little RF modulator to put the output of your device on TV channel 3 or 4. I know of a business in San Diego that was being run on a C64 as recently as 2013, so I guess you could say it was "Y2K compatible".

    I actually called the U.S. FCC to ask about the blanking widths I was contemplating and was told by a nice fellow named Gordon Godfrey that they likely wouldn't cite a broadcaster for broadcasting those H blanking widths. On the VIC-20 there was a bit you had to set to get interlaced video.

    As I said, everything at the time was derived from the 3.58 MHz fsc.
    Quote Quote  
  17. Originally Posted by chris319 View Post
    I think you refer here to the Composite 4fsc sampling principle, which gives the 14 7/22 MHz = 14.3181818... MHz sampling rate for NTSC and 17.734475 MHz for PAL, leading to "unusual" Pixel Aspect Ratios, like 6/7 for NTSC for example.
    It's quite different from the Rec.601 standard which specifies a harmonized luma sampling rate of 13.5 MHz for both NTSC and PAL.
    Yes, I had to work out those calculations in the 1980's when I worked with the Amiga computer and other graphics devices including the Apple ][, VIC-20 and C64, which were based on U.S. analog video standards so the user could see the output on his home TV set. You could get a little RF modulator to put the output of your device on TV channel 3 or 4. I know of a business in San Diego that was being run on a C64 as recently as 2013, so I guess you could say it was "Y2K compatible".

    I actually called the U.S. FCC to ask about the blanking widths I was contemplating and was told by a nice fellow named Gordon Godfrey that they likely wouldn't cite a broadcaster for broadcasting those H blanking widths. On the VIC-20 there was a bit you had to set to get interlaced video.

    As I said, everything at the time was derived from the 3.58 MHz fsc.
    why would you be broadcasting out-of-spec signals anyway? Isn't that just for the purpose of making it easier to digitize, presumably to output it to a remote TV via composite video cable?
    Quote Quote  
  18. Originally Posted by Videogamer555 View Post
    why would you be broadcasting out-of-spec signals anyway? Isn't that just for the purpose of making it easier to digitize, presumably to output it to a remote TV via composite video cable?
    If you are broadcaster and you have license for broadcasting then you need to follow set of rules and standards. If you are produce video signal transmitted to your TV by cable then you are not broadcasting. Difference from standard can be explained in many ways - for example circuitry simplification.
    Quote Quote  



Similar Threads

Visit our sponsor! Try DVDFab and backup Blu-rays!