View previous topic :: View next topic |
Author |
Message |
ElTopo
Joined: 07 Nov 2006 Posts: 1608
|
Link Posted: Sun Feb 21, 2016 2:54 pm Post subject: Moome card with HDMI 2.0 and HDR possible ? |
|
|
@Moome: May this be possible ?
cheers
ElTopo
_________________ Barco Cine 9 the one and only
|
|
Back to top |
|
|
cmjohnson
Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s
|
Link Posted: Sun Feb 21, 2016 5:34 pm Post subject: |
|
|
Tagged. Yes, I'm really interested in this.
I'm on a quest to push a Marquee into 4K capable territory. Specifics are outlined in the CRT forum on more than one topic.
In short, yes, 4K capable tubes exist. LUGs may even be able to do it. (Reference the specs for the Barco Cine 9.) Bandwidth capacity needed is achievable. But I need to be able to get UHD resolutions out of the Moome HDMI card or it's all for naught.
Pleasepleaseplease make a Marquee HDMI 2.0 card that will output UHD resolutions up to 4K. I'll make a deposit on one NOW.
|
|
Back to top |
|
|
ElTopo
Joined: 07 Nov 2006 Posts: 1608
|
Link Posted: Sun Feb 21, 2016 9:43 pm Post subject: |
|
|
Not only for Marquees there are also Barco users out there
_________________ Barco Cine 9 the one and only
|
|
Back to top |
|
|
cmjohnson
Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s
|
Link Posted: Sun Feb 21, 2016 10:19 pm Post subject: |
|
|
Yes. All the "top gun" machines are capable of better than 1080p. They should all be able to benefit from an input card that allows the possibility of higher than 1080p resolution.
A Cine 9 uses the same tubes as a LUG-equipped Marquee, and doesn't the G90 also use LUGs? We know the Cine 9's rated resolution approaches 4K requirements. So it stands to reason that any PJ using LUGs should be able to match that, although some modifications may be required to get there. (Run Barco yokes in the Marquee, for one.)
My suggestion would be for Moome to develop a 4K core input board that is built as a daughter board on projector-specific adapters that would have a low cost of manufacture. You buy the core board and the adapter board you need for your specific model of projector. This should be able to reduce the total cost of the boards. Also it makes it possible to adapt the board to other projectors at greatly reduced cost as you'd only have to make the adapter. The core decoder board would always be the same,
and built with a port to allow easy firmware upgrades. (I'd suggest via USB, just attach a USB with new firmware to it, power it on, and let it automatically load the update.)
|
|
Back to top |
|
|
moome
Joined: 01 Jun 2007 Posts: 497
|
Link Posted: Tue Feb 23, 2016 4:09 pm Post subject: |
|
|
yes, I plan make HDMI2.0 card, but it is early stage now~~
will update if any progress~~
|
|
Back to top |
|
|
cmjohnson
Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s
|
Link Posted: Sun Feb 28, 2016 1:51 am Post subject: |
|
|
Awesome! Looking forward to it!
Feature request: Remote-selectable output resolutions, so that if 4K is too much, but the machine will do better than 2K, then the user will be able to pick resolutions between 2 and 4K.
|
|
Back to top |
|
|
greg9518lc
Joined: 19 Apr 2016 Posts: 360
|
Link Posted: Tue Aug 02, 2016 8:12 pm Post subject: |
|
|
moome wrote: | yes, I plan make HDMI2.0 card, but it is early stage now~~
will update if any progress~~ |
How is the progress on this I am in for one of these if there is a preorder
_________________ VDC 9518LC modded: I do not sell or promote mods only interested in the best PQ possible......
|
|
Back to top |
|
|
kal Forum Administrator
Joined: 06 Mar 2006 Posts: 17860 Location: Ottawa, Canada
TV/Projector: JVC DLA-NZ7
|
Link Posted: Tue Aug 02, 2016 8:22 pm Post subject: |
|
|
cmjohnson wrote: | Feature request: Remote-selectable output resolutions, so that if 4K is too much, but the machine will do better than 2K, then the user will be able to pick resolutions between 2 and 4K. |
Why?
I'd recommend not doing this at all and keep it simple, just like Moome's HDMI cards today where you choose the resolution at the source. You'd be able to choose 1080p or 4K (if available).
No different than how Moome's cards work today.
You're asking for a scaler to be included which greatly increases cost and complexity. I don't see the benefit.
If you want to go nuts and upscale 1080p to 4K, I'd recommend a 4K Radiance: www.curtpalme.com/Radiance.shtm
It comes with a ton of other things that will be useful too such as multi-point greyscale and gamma adjustment, a full CMS, etc, and you take it with you if ever you change your projector. Building all that into a card that is meant for a single brand/type of projector (IMHO) doesn't make sense.
My 2 cents.
Kal
_________________
Support our site by using our affiliate links. We thank you!
My basement/HT/bar/brewery build 2.0
Last edited by kal on Tue Dec 06, 2016 2:30 am; edited 1 time in total
|
|
Back to top |
|
|
cmjohnson
Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s
|
Link Posted: Tue Dec 06, 2016 12:49 am Post subject: |
|
|
Any updates? It'd be nice to know if this product will actually be made.
I'm saving up money for it in any event. No reason NOT to save money anyway, right?
|
|
Back to top |
|
|
mp20748
Joined: 12 Sep 2006 Posts: 5681 Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
Link Posted: Fri Dec 09, 2016 11:07 am Post subject: |
|
|
It wouldn't make sense to do this, when CRT technology would never be able to handle 4K in either scan or bandwidth requirements.
CRT would better benefit when 4K is done right in scaling down the resolution to 1080P
|
|
Back to top |
|
|
CIR Engineering
Joined: 25 Aug 2008 Posts: 4264 Location: Chicago USA & Berlin Germany
|
Link Posted: Fri Dec 09, 2016 7:13 pm Post subject: |
|
|
You need to have an HDMI 2.0 card for the HDCP protection if you want to run any 4k or UHD sources. So no matter what resolution you want to run from a 4k source, you still must have the HDMI 2.0 chip.
The other thing is that Lumagen is considering providing a 4k interlaced output. This run at 48Hz would not really push bandwidth too far.
craigr
_________________ *NEW JETI 1501-HiRes 2nm Spectroradiometer
JETI 1211 Spectroradiometer
Photo Research PR-650 Spectroradiometer
Klein K10-A Colorimeter
X-Rite i1Pro2 Spectroradiometer & Spyder Colorimeters *For JVC auto-calibration when Klein & Jeti are not applicable
Murideo Fresco SIX-G HDMI 2.x Multimedia Generator
Murideo Fresco SIX-A HDMI 2.x Analyzer
*NEW Light Illusion ColourSpace XPT Version β Color Calibration Software
Light Illusion LightSpace XPT Pro Version 10.x Color Calibration Software
*NEW OMARDRIS JVC Software Patch to use K10-A and Jeti with JVC OEM AutoCal Software!
Sencore CR7000 CRT Tube Analyzer / Rejuvenater
Authorized Dealer for Lumagen & just about everything worth buying
www.CIR-Engineering.com - craigr@cir-engineering.com
Phone: 865-405-6892
|
|
Back to top |
|
|
mp20748
Joined: 12 Sep 2006 Posts: 5681 Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
Link Posted: Sat Dec 10, 2016 10:59 am Post subject: |
|
|
CIR Engineering wrote: |
The other thing is that Lumagen is considering providing a 4k interlaced output. This run at 48Hz would not really push bandwidth too far.
craigr |
That indeed would be interesting and could be the exception to CRT not being able to do 4K.
Make it happen Jim!
|
|
Back to top |
|
|
CIR Engineering
Joined: 25 Aug 2008 Posts: 4264 Location: Chicago USA & Berlin Germany
|
Link Posted: Sat Dec 10, 2016 5:03 pm Post subject: |
|
|
mp20748 wrote: | That indeed would be interesting and could be the exception to CRT not being able to do 4K.
Make it happen Jim! |
But even if we don't get that, we still need the HDMI 2.0 card if we want to use UHD sources for the HDCP decryption. UHD down sampled to 1080p would still provide benefit because of the 10-bit color on the UHD BD. So while resolution would not be increased, color fidelity would still benefit. And scaling artifacts should not be any (much) issue because UHD is exactly 4x the resolution of 1080p. So instead of using old school line doublers, now we may need to start using quad reducers.
craigr
_________________ *NEW JETI 1501-HiRes 2nm Spectroradiometer
JETI 1211 Spectroradiometer
Photo Research PR-650 Spectroradiometer
Klein K10-A Colorimeter
X-Rite i1Pro2 Spectroradiometer & Spyder Colorimeters *For JVC auto-calibration when Klein & Jeti are not applicable
Murideo Fresco SIX-G HDMI 2.x Multimedia Generator
Murideo Fresco SIX-A HDMI 2.x Analyzer
*NEW Light Illusion ColourSpace XPT Version β Color Calibration Software
Light Illusion LightSpace XPT Pro Version 10.x Color Calibration Software
*NEW OMARDRIS JVC Software Patch to use K10-A and Jeti with JVC OEM AutoCal Software!
Sencore CR7000 CRT Tube Analyzer / Rejuvenater
Authorized Dealer for Lumagen & just about everything worth buying
www.CIR-Engineering.com - craigr@cir-engineering.com
Phone: 865-405-6892
|
|
Back to top |
|
|
cmjohnson
Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s
|
Link Posted: Sun Dec 11, 2016 11:01 pm Post subject: |
|
|
Correct, we get better 10 bit color fidelity and have HDCP 2.2 compliance so if we get nothing else, at least our 4K blu-ray players will be able to deliver 1080p better than we've seen it before. Plus I do think there's potential in the "4K interlaced" kind of idea.
I don't have the programming skills, and I may NEVER have them, but with something that strips out HDCP protection and some suitably fast DACs and amplifiers, some bright boy could probably create a 4K to analog hardware interface that'd run.
They say that the ultimate bandwidth of 4K is 18 GB per second. But do the math.
Consumer 4K is 3840 by 2160 pixel resolution or 497,664,000 pixels per second at 60 Hz refresh. Add 20 percent for overhead and you have a need for 600 MHz of analog system bandwidth.
But take it down to half that, 30 FPS or even 24 FPS and use pulldown. Save more bandwidth by using 4:2:2 rather than 4:4:4 chroma subsampling and save more bandwidth. I think that at 4K you'd never really see any faults with using 4:2:2 anyway.
We can get this under 300 MHz bandwidth with a few tradeoffs!
Even then, in a scaler, who's to say that you can't run an intermediate resolution? If 4K can't be done, how about 3K? Or 2.5K?
2.5K would be....2880 by 1620. That'd require 268 MHz total video bandwidth at a 48 Hz refresh rate and it's certainly within the realm of possibility for any good 9" machine.
What I would really like to see is for Lumagen to include an HDMI loopthrough feature so that if you were to have one of those little gadgets that "accidentally" strips HDCP out of the HDMI signal, then ALL HDMI signals could be routed through it before any further processing is done.
|
|
Back to top |
|
|
CIR Engineering
Joined: 25 Aug 2008 Posts: 4264 Location: Chicago USA & Berlin Germany
|
Link Posted: Sun Dec 11, 2016 11:49 pm Post subject: |
|
|
cmjohnson wrote: | Correct, we get better 10 bit color fidelity and have HDCP 2.2 compliance so if we get nothing else, at least our 4K blu-ray players will be able to deliver 1080p better than we've seen it before. Plus I do think there's potential in the "4K interlaced" kind of idea...
They say that the ultimate bandwidth of 4K is 18 GB per second...
But take it down to half that, 30 FPS or even 24 FPS and use pulldown. Save more bandwidth by using 4:2:2 rather than 4:4:4 chroma subsampling and save more bandwidth. I think that at 4K you'd never really see any faults with using 4:2:2 anyway.
We can get this under 300 MHz bandwidth with a few tradeoffs!...
Even then, in a scaler, who's to say that you can't run an intermediate resolution? If 4K can't be done, how about 3K? Or 2.5K?
2.5K would be....2880 by 1620. That'd require 268 MHz total video bandwidth at a 48 Hz refresh rate and it's certainly within the realm of possibility for any good 9" machine.
What I would really like to see is for Lumagen to include an HDMI loopthrough feature so that if you were to have one of those little gadgets that "accidentally" strips HDCP out of the HDMI signal, then ALL HDMI signals could be routed through it before any further processing is done. |
The UHD BluRay discs are mastered at 4:2:0 10-bit and the Lumagen can accept 4:2:0 24Hz 10-bit, and then output whatever we want at 4:2:0. This will save more digital bandwidth too. As long as our HDMI 2.2 card can use 4:2:0 (which it should) than we will be good there.
That said, it will cave to be decoded to RGB analog so the analog bandwidth will be the same as any other digital encoding.
18 GHz is the maximum spec for HDMI 2.2. That is only required for 60Hz sources with color above 4:2:0, or if a source is doing something stupid like an UHD BD player using 4:4:4 (cough Samsung). UHD BD only requires just under 9 GHz for full resolution and color, or 1:1 disc reproduction.
The issue with Lumagen and interlaced is that they have only EVER been able to do 1080i and not other resoultion interlaced. If Lumagen makes a 4Ki output mode that will almost certainly be adjustable to almost any resolution you want to interlace or "re-interlace."
Regarding HDCP strippers, all you need to do is put one at the output of the Lumagen and all sources will work. The way the strippers work is to report up the chain that, "I am a display device," therby tricking anything upstream that HDCP 2.2 is being followed. So no, we don't really need an HDMI 2.2 card for HDCP resons, but I would rather not run a stripper as this can always add complications. Also, we really do need HDMI 2.2 if we want any hope of running above about 195 MHz which is where Moome caps out even after raising the voltage to his HDMI chip by 10% above specification (this is what I do for all my CRT 3D clients).
craigr
_________________ *NEW JETI 1501-HiRes 2nm Spectroradiometer
JETI 1211 Spectroradiometer
Photo Research PR-650 Spectroradiometer
Klein K10-A Colorimeter
X-Rite i1Pro2 Spectroradiometer & Spyder Colorimeters *For JVC auto-calibration when Klein & Jeti are not applicable
Murideo Fresco SIX-G HDMI 2.x Multimedia Generator
Murideo Fresco SIX-A HDMI 2.x Analyzer
*NEW Light Illusion ColourSpace XPT Version β Color Calibration Software
Light Illusion LightSpace XPT Pro Version 10.x Color Calibration Software
*NEW OMARDRIS JVC Software Patch to use K10-A and Jeti with JVC OEM AutoCal Software!
Sencore CR7000 CRT Tube Analyzer / Rejuvenater
Authorized Dealer for Lumagen & just about everything worth buying
www.CIR-Engineering.com - craigr@cir-engineering.com
Phone: 865-405-6892
|
|
Back to top |
|
|
jbmeyer13
Joined: 03 Dec 2010 Posts: 1135
|
Link Posted: Mon Dec 12, 2016 4:57 pm Post subject: |
|
|
cmjohnson wrote: |
They say that the ultimate bandwidth of 4K is 18 GB per second. But do the math.
Consumer 4K is 3840 by 2160 pixel resolution or 497,664,000 pixels per second at 60 Hz refresh. Add 20 percent for overhead and you have a need for 600 MHz of analog system bandwidth.
But take it down to half that, 30 FPS or even 24 FPS and use pulldown. Save more bandwidth by using 4:2:2 rather than 4:4:4 chroma subsampling and save more bandwidth. I think that at 4K you'd never really see any faults with using 4:2:2 anyway.
We can get this under 300 MHz bandwidth with a few tradeoffs! |
After Craig suggested I meet up with Mike we've spent the last few years working with Kurt to get the Marquee to the point of being able to run 200mhz 1080p/72 cleanly. It was A LOT of work to push it to from 170mhz to 200mhz and wind up with the desired outcome. It's hard to see 300mhz (even at 48hz) as an acceptable solution. Flicker city to say the least..
Quote: |
Even then, in a scaler, who's to say that you can't run an intermediate resolution? If 4K can't be done, how about 3K? Or 2.5K?
2.5K would be....2880 by 1620. That'd require 268 MHz total video bandwidth at a 48 Hz refresh rate and it's certainly within the realm of possibility for any good 9" machine.
|
Hard enough to resolve 1080p and even if Lumagen offers a 4Ki output the processor is $5K. Hard to see many CRT folks shelling out that kind of dough. Just like the blend function that Nash has been pestering Jim about for years; not enough demand means it likely won't see the light of day.
_________________ Projector: Modded 9501LC ULtra- MP VIM, Vold VNB, ETECH LVPS, Silver VIM Cables, HD10F's & a V1 case!
|
|
Back to top |
|
|
cmjohnson
Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s
|
Link Posted: Mon Dec 12, 2016 9:46 pm Post subject: |
|
|
I am FAR from being a software guy so I can't say anything about the complexity of any of these ideas, but to me there is logical sense in a scaler that can also divide the image between multiple projectors and also blend them. I would consider the holy grail of scaler processing to be one that can take a 4K input signal and use it to drive multiple projectors at their best quality resolutions and blend them at the same time.
It does seem to me, however, that once you have a scaler that can downres an incoming hig res image, adding more downrezzed modes to it should not be horribly difficult for the software developer.
Of course there are video bandwidth limitations associated with CRT projectors. But I'd like the freedom to be able to push my system to its limits, or even beyond, if I want to. Don't limit me to 1080p because these machines CAN do more than that.
Incidentally, the reason why the video bandwidth required for 4K is so high is that in the digital domain the color depth information has to be transmitted using the actual color values in digital format. So you have H resolution x V resolution x frame rate x color depth bits. And add overhead data. But once converted to analog, the instantaneous brightness level per RGB channel is a single voltage point, and not 10 (or 12) bits of data representing the same thing. So in this case analog video will actually end up
using less bandwidth at its full resolution capacity than the parent digital signal.
|
|
Back to top |
|
|
jbmeyer13
Joined: 03 Dec 2010 Posts: 1135
|
Link Posted: Mon Dec 12, 2016 10:36 pm Post subject: |
|
|
Quote: |
Of course there are video bandwidth limitations associated with CRT projectors. But I'd like the freedom to be able to push my system to its limits, or even beyond, if I want to. Don't limit me to 1080p because these machines CAN do more than that. |
4K interlaced with raster ringing three quarters of the way across the screen and flicker really isn't "doing more" than 1080p. Doing something and doing it well are two different things as the tradeoffs quickly crush performance.
Quote: | Incidentally, the reason why the video bandwidth required for 4K is so high is that in the digital domain the color depth information has to be transmitted using the actual color values in digital format. So you have H resolution x V resolution x frame rate x color depth bits. And add overhead data. But once converted to analog, the instantaneous brightness level per RGB channel is a single voltage point, and not 10 (or 12) bits of data representing the same thing. So in this case analog video will actually end up
using less bandwidth at its full resolution capacity than the parent digital signal. |
Even so it's still way beyond the BW capabilities of the most modified Marquees in exsistence. You'll need a $5K video processor just to downscale the 4K signal AND then require custom porch timings to tame the raster ringing which (even at 48hz) will be god awful. Then of course you have to get around the SMPTE 2082 gamma.
A moome card that could safely and consistently go to 200mhz would be a giant leap forward and is what's needed to complement the best modified Marquee video chains. Hey I hope I'm wrong and all of this can be a reality but it seems like "pie in the sky" to me.
_________________ Projector: Modded 9501LC ULtra- MP VIM, Vold VNB, ETECH LVPS, Silver VIM Cables, HD10F's & a V1 case!
|
|
Back to top |
|
|
cmjohnson
Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s
|
Link Posted: Mon Dec 12, 2016 10:52 pm Post subject: |
|
|
My experiments have shown that raster ringing is caused in large part by having an imbalance between the two coils that together comprise the horizontal deflection yoke coils. When carefully matched ringing can be reduced and even eliminated. But there are other causes of raster ringing other than coil mismatch.
There are some (older) analog video cards that can be used to push well beyond 1080p already. There's one guy here in these forums who has posted photos of him running 4K on a Marquee using one of those cards. Of course it's not a phenomenally sharp image as he's run out of real video bandwidth, but all that does is make me want it even more.
I know there's a limit. I just want to GET THERE. And not stop short of it.
|
|
Back to top |
|
|
CIR Engineering
Joined: 25 Aug 2008 Posts: 4264 Location: Chicago USA & Berlin Germany
|
Link Posted: Tue Dec 13, 2016 12:48 am Post subject: |
|
|
jbmeyer13 wrote: | ...Just like the blend function that Nash has been pestering Jim about for years; not enough demand means it likely won't see the light of day. |
Just have to say that making a blending processor is a dog gone difficult undertaking and requires immense work. Adding interlacing is not that difficult to do comparatively speaking. Jim already told me he's willing to ask Patrick to try and program it... just not right now because the HDR shaping LUT and other features have much higher priority (and rightfully so).
craigr
_________________ *NEW JETI 1501-HiRes 2nm Spectroradiometer
JETI 1211 Spectroradiometer
Photo Research PR-650 Spectroradiometer
Klein K10-A Colorimeter
X-Rite i1Pro2 Spectroradiometer & Spyder Colorimeters *For JVC auto-calibration when Klein & Jeti are not applicable
Murideo Fresco SIX-G HDMI 2.x Multimedia Generator
Murideo Fresco SIX-A HDMI 2.x Analyzer
*NEW Light Illusion ColourSpace XPT Version β Color Calibration Software
Light Illusion LightSpace XPT Pro Version 10.x Color Calibration Software
*NEW OMARDRIS JVC Software Patch to use K10-A and Jeti with JVC OEM AutoCal Software!
Sencore CR7000 CRT Tube Analyzer / Rejuvenater
Authorized Dealer for Lumagen & just about everything worth buying
www.CIR-Engineering.com - craigr@cir-engineering.com
Phone: 865-405-6892
|
|
Back to top |
|
|
|
|