Why use DLP-Link in a LCD display? The reasoning is beyond me.PressBot wrote:Since this is both Nvidia and AMD compatible, our bet is on DLP-Link technology between the monitor and the glasses because it's something that does not require deep GPU synchronization or a proprietary USB connector.
It's a computer monitor, not a 3DTV, so it doesn't have to use consumer electronics protocols such as HDMI 3D.but HDMI 1.4A does not include LCD shutter glasses synchronization technology, and this is the only option we know of that is beyond the direct control of the graphics card manufacturers.
Still not supporting the HD66, why they list "Adds support for Optoma GT720 , Optoma HD67, and Optoma HD66 projectors
" is beyond me. There should be some legal disclosure about this,it's so frustrating. Even the guy on phone says that it supports new firmwares from 8/2010 for the HD66, which after sending mine off and just getting it back with latest firmware we know is not true.......
Oh well, luckly people on here are smarter then the "techs" at Nvidia and the Acer.Inf hack still works great even with new 1.34
Yes, this is another video transmission mode possible with Dual-link DVI, besides using DisplayPort 1.1a for stereo 3D.Neil wrote:If the GPU is passing on a left/right view to the display in a standard format (e.g. side by side), then all the synchronization could be between the display and the glasses which the software has no control over.
No, they haven't.I don't know why I thought that DLP-Link worked beyond DLP projectors and DLP checkerboard televisions. Has it ever been announced for anything beyond that?
No, these are certainly not Bit Cauldron glasses, most likely CMO as well.Dom wrote:Maybe they are using bit cauldron glasses and will have a usb wireless adapter to transmit to the glasses.
I'm sorry, I don't quite understand that part. This is a 120 Hz frame alternative monitor with active shutter glasses, not line-interleaved monitor with passive polarized glasses.it seems like izd3d will have to use interlaced on that monitor because currently they don't have core kernal usage of the nvidia and ati glasses wavelengh and such what you guys said.
) I say they might have to use interlaced mode because as of now their frame sequential mode don't work and loses sync because Nvidia and ATI would not give them their driver kernal to fix it. So going less than 60fps in game at 120hz lost the sync indefenitly.DmitryKo wrote:Yes, this is another video transmission mode possible with Dual-link DVI, besides using DisplayPort 1.1a for stereo 3D.Neil wrote:If the GPU is passing on a left/right view to the display in a standard format (e.g. side by side), then all the synchronization could be between the display and the glasses which the software has no control over.
No, they haven't.I don't know why I thought that DLP-Link worked beyond DLP projectors and DLP checkerboard televisions. Has it ever been announced for anything beyond that?
AFAIK no 120 Hz LCD panel is fast enough to insert a bright frame in between left-right frames as DLP-Link does, it will basically take a 240 Hz panel to do that. As of now, only DMD mirrors (and plasma pixels) are fast enough to do it - however it is impractical to implement for plasma panels since it will significanly rise power consumption.
No, these are certainly not Bit Cauldron glasses, most likely CMO as well.Dom wrote:Maybe they are using bit cauldron glasses and will have a usb wireless adapter to transmit to the glasses.
If you further read ATI interviews linked in my post above, they say that syncing directly with the display is the best option (which I tend to agree), so it's probably either built-in emitter or standard VESA Stereo port with a compatible emitter.
I'm sorry, I don't quite understand that part. This is a 120 Hz frame alternative monitor with active shutter glasses, not line-interleaved monitor with passive polarized glasses.it seems like izd3d will have to use interlaced on that monitor because currently they don't have core kernal usage of the nvidia and ati glasses wavelengh and such what you guys said.
BTW, what is "wavelength of glasses" and how is it connected with interlaced signal, exactly?
Dom, you can't "use interlaced mode" on a progressive frame-sequential display like this one, because, well, it just does not support interlaced video at all. BTW no modern LCD computer monitor ever does since at least 2000s, and certainly no stereoscopic computer monitor supports interlaced 1920x1080 @120 Hz (that is 120 fields or 60 interlaced frames per second).Dom wrote:I say they might have to use interlaced mode because as of now their frame sequential mode don't work
What you call "wavelength" is basically called "frequency"; sync pulses cannot have "wavelength" because they are not periodic sine waves, but rather square waves which are typically characterized by frequency and duty cycle, or pulse width (in split seconds).What I mean abot the wavelengh of the glasses is that the pulses lost the sequence from the iz3d drivers and nvidia
I don't think iZ3D can make eDimensional glasses work with DVI-D displays because eDimensional are using analog VGA dongle, and most video cards do not support 120 Hz on the analog VGA input.it sure will be nice if iz3d can make e-dims work on a 120hz panel with dual-dvi-d
Maybe not iZ3D, but other manufacturers should be able to sell small devices plugged in a USB or DVI port which could be connected to the eDimensional glasses. The team behind SoftGenLock did such an adapter to plug 3 pin DIN stereo glasses on the parallel port.DmitryKo wrote:I don't think iZ3D can make eDimensional glasses work with DVI-D displays because eDimensional are using analog VGA dongle, and most video cards do not support 120 Hz on the analog VGA input.
LCD's are nowhere near fast enough to switch the DLP link sync signal from what I've read of it (I don't know how fast it is, but I'd guess sveral kHz.) What is fast enough are the LED's used for the backlight. It would not be too difficult to strobe the backlight LED's (the driver circuitry would have to be overhauled) at several kHz to send a signal to DLP link glasses.DmitryKo wrote:
AFAIK no 120 Hz LCD panel is fast enough to insert a bright frame in between left-right frames as DLP-Link does, it will basically take a 240 Hz panel to do that. As of now, only DMD mirrors (and plasma pixels) are fast enough to do it - however it is impractical to implement for plasma panels since it will significantly rise power consumption.
DmitryKo wrote:Dom, you can't "use interlaced mode" on a progressive frame-sequential display like this one, because, well, it just does not support interlaced video at all. BTW no modern LCD computer monitor ever does since at least 2000s, and certainly no stereoscopic computer monitor supports interlaced 1920x1080 @120 Hz (that is 120 fields or 60 interlaced frames per second).Dom wrote:I say they might have to use interlaced mode because as of now their frame sequential mode don't work
So it's either 120 Hz frame sequential stereo mode, or no stereo at all (well, besides color anaglyph which should work on every monitor).
What you call "wavelength" is basically called "frequency"; sync pulses cannot have "wavelength" because they are not periodic sine waves, but rather square waves which are typically characterized by frequency and duty cycle, or pulse width (in split seconds).What I mean abot the wavelengh of the glasses is that the pulses lost the sequence from the iz3d drivers and nvidia
iZ3D sync issues are not really about high frequency, so lowering it to 60 Hz will not help.
I don't think iZ3D can make eDimensional glasses work with DVI-D displays because eDimensional are using analog VGA dongle, and most video cards do not support 120 Hz on the analog VGA input.it sure will be nice if iz3d can make e-dims work on a 120hz panel with dual-dvi-d
I don't understand what are you talking about. The video driver and the rendering API don't render "interlaced", they work with the concept of frame buffer(s) and "begin scene"/"end scene"; refresh rate and interlacing/progressive only matter after the scene has been completely rendered and frame buffer has been copied to the back buffer for display through RAMDAC or TDMS.Dom wrote: most of the stuff I am talking about does not work with current configurations, but looking at the system configuration I have now I assume that. E-dim software actually reproduces the interlaced and so does the iz3d driver.
Yes it does. If the monitor can not display interlaced signal, it just won't display it and will tell you "unsupported video mode" instead.Does it really matter if the display is meant for full 3d? If the driver can replicate the mode in an overlay I don't really see how the monitor will stop that.
Did I say that video cards don't support 120 Hz through analog VGA? I know I did.The easiest route to enable a vga dongle into a dvi-d connected monitor would be to use a "loopback driver" with a output connection dvi-port
If LCD pixels can't change fast enough, strobing the backlight will not help because the pixels have to be fully transparent to pass these bright white flashes.skrubol wrote: What is fast enough are the LED's used for the backlight. It would not be too difficult to strobe the backlight LED's (the driver circuitry would have to be overhauled) at several kHz to send a signal to DLP link glasses.
We've discussed it before; I'd think the easiest way would be to only mark the start frame somehow and then the projector will autosync forever. The technical documentation is not freely available though, so it's all only a guess.I've had a lot of trouble finding much info on DLP link. It may just mean that whatever signaling 120Hz DLP projectors receive on the input cable to tell it whether the frame is left or right for DLP link, is the same signaling this display uses.
I don't think so. The flashes might be brief, but they have to be bright enough to get a tight sync. The glasses probably do care about color and intensity, because they specifically say "white flashes" in the marketing materials, and 10% intensity is anything but "white", not to mention that currently displayed picture can be anything but white. With DMDs, producing intensive white flash in an instant is not a problem since all the mirrors can take the "on" position simultaneously.skrubol wrote:The glasses don't care too much whether the flashes are at 100% intensity or 10% intensity, and I doubt the color matters that much either.
Some signal degradation does not matter as long as you can receive vertical sync information, which are low-frequency square strobe pulses that can't be so easily corrupted, even when the horizontal sync is not tight enough to display the picture.It's too easy for noise or some sort of interruption to corrupt a few frames and make the projector lose sync for it to rely on an initial flag.
I don't quite understand why and how does it need to be this way, so I'll take your word for that. However what you describe is NOT an interlaced signal, which by definition contains two half-resolution subfields instead of a full frame, it's just standard progressive signal where each other line is black. Might call it "line skipping" if you wish.As far as interlaced goes, it's a very common 'workaround' mode when the 3d driver software doesn't have access to the underlying hardware API's.
They might support 120 Hz in low-res modes up to 1024x768, but AFAIK no video card supports 1920x1080 @120 Hz on analog VGA, which means you can't use the video card's clone mode to provide sync to the glasses. And 400 MHz RAMDACs were designed to handle 2048x1536 @85Hz, not 1920x1200 @120Hz. Anyway it's all only theory since I doubt iZ3D can be bothered enough. If the new displays will have built-in emitters or VESA Stereo ports, as ATI intends, there will be no need for tricks like that.skrubol wrote:every video card I know of supports 120Hz through the analog ports.
I think most cards have 350-400MHz RAMDAC's. That can push full HD at 120 Hz. That can push full HD at 120 Hz.
I'd rather trust that the ATI-compatible monitors will feature VESA Stereo port which you could use to plug the wireless transmiter and an USB port for power, that would eliminate most issues with imprecise sync and input latency in the monitor.Fredz wrote:but other manufacturers should be able to sell small devices plugged in a USB or DVI port which could be connected to the eDimensional glasses. The team behind SoftGenLock did such an adapter to plug 3 pin DIN stereo glasses on the parallel port.
Sure, that would be the ideal situation if manufacturers would integrate VESA Stereo ports on their displays, but history has shown that they've moved on and all go for proprietary solutions. Samsung had such a port on their old Plasma models (PN42A450P and PN50A450P) but that's not longer the case on their new LCD and Plasma sets. I don't see anyone integrate this port in the future, and ATI surely isn't relevant enough to change this trend.DmitryKo wrote:I'd rather trust that the ATI-compatible monitors will feature VESA Stereo port which you could use to plug the wireless transmiter and an USB port for power, that would eliminate most issues with imprecise sync and input latency in the monitor.
Something new is cooking in the iZ3D camp!
I'm afraid we know less than more, but the promotional page promises compatibility with both Nvidia and AMD graphics cards, and is based on LCD shutter glasses technology. Our understanding is that Chi-Mei (CMO) makes Viewsonic panels, which explains why the drivers are under the CMO name. This is also a potential indicator for how wide the iZ3D license agreements can potentially spread (i.e. who buys CMO panels?)!
