| Author |
Message |
Tomwhite
Joined: 07 Jan 2017 Posts: 30 Location: Commack ny
TV/Projector: Barco 800 Sony 1270
|
| Posted: Sun Feb 27, 2022 4:56 am Post subject: 4k crt |
|
|
|
Hey every body in crt projector world, 4K is 2160 at 60hz. They didn’t double the refresh rate, it should be 120hz. Basically 4K is 4KI, and most decent crt projectors can handle that with no problem. The problem is converting 4k to rgbhv. Please moome or hdfury give us a way out. Give us a way to scale to 1440 or 1600 or some thing in the ball park and we’ll figure it out from there.
|
|
| Back to top |
|
 |
Dancrt
Joined: 16 Sep 2017 Posts: 88
|
| Posted: Sun Feb 27, 2022 5:22 am Post subject: |
|
|
I got all excited about the idea too
But came to the conclusion you’d never
Get the convergence well enough for it to
Really be a thing 1440p maybe
But at 1080p with moome card the
Input is so noise free and perfect
Useing any other dac would probably make it look
Worse ?
I use a vp50pro to add slight sharpening to 1080p
Again highest res the video proccessor supports
And now have bought a Panasonic ub-9000 bluray
Player and the result 4K hdr movies look fantastic
And the way the colour channel resolution works
They look sharper than standard bluray and Apple TV
With better colour than Apple TV which also looks great.
But the Panasonic ub9000 hdr to sdr conversation is epic
And more colour from my experience.
So I gave up on the idea but if you have success awesome
But after meticulously setting up 1080p
I can’t see a way of getting high quality signal or accurate enough
Convergence to be a any improvement.
My barco 909 has scan lines in the green and red tubes
So definitely could get higher resolution out of it
But think limiting factors is it’s all too hard ??
Have you ever done it ?
Anyone seen it done with success ?
At 3m wide screen the slight unsharp mask of the vp50pro
I no longer sit there thinking it could be sharper
I now sit there thinking this movie is epic
The sharpening added for me
What I was looking for by adding more resolution
For instance without the proccessor the (c) logo
On Marcel trailers was blurry with +15 fine details
Tge (c) logo is now so sharp you can see the pixels
And it’s clear
|
|
| Back to top |
|
 |
Tim in Phoenix
Joined: 21 Oct 2006 Posts: 4409 Location: Phoenix
|
| Posted: Wed Mar 09, 2022 3:29 pm Post subject: |
|
|
Guys
The DVX8022 needs to be tried across two nine inch machines at 4K but this begs the question: Is there an HDCP stripped or converter to RGB made anywhere? It may not output the full depth, but should impress. The Analog Nextage units likely would do better, but costly.
|
|
| Back to top |
|
 |
Dancrt
Joined: 16 Sep 2017 Posts: 88
|
| Posted: Thu Mar 10, 2022 1:28 am Post subject: |
|
|
Tim I think from what I’ve seen your the only one
Who would try this and I think it would be fantastic!!!
Think red fox is working in a solution for ultra hd hdcp
Would love to see results !!!
If I won tattslotto I’d love a 4K blend but maybe one day
Hmm for me jvc nx9 or a 4K crt think I’d go the crt !!!
That’s said if I could afford it I’d do both ahahha
And get a 5 m wide screen
Untill then 3m wide barco 909 will have to do
|
|
| Back to top |
|
 |
redfox001
Joined: 16 Mar 2009 Posts: 2257 Location: The Netherlands
|
| Posted: Sat Mar 19, 2022 3:16 pm Post subject: |
|
|
Because I think the whole 4k solution is simple a way to shift digital noise to twice the 1080p sampling frequency I think that oversampling like done in audio would result in a less disturbed video experience. In practice a sharp focused CRT does oversampling between lines and up sampling in a line. Analogue to the insertion of black frames in digital to improve the motion resolution. CRT has better motion resolution between the frames. That is because the dot has a short afterglow pulling it to zero between the frames. No need to go higher in resolution unless you have super eyes or are to close to a ultra big screen.
Improving the amplifier bandwidth on a CRT is a dual thing. It also makes the high frequency digital noise more visible. Might not be to good. The up sampling in a line becomes more like a sample and hold system. It becomes too perfect. If I could realize that bandwidth I would also invent a grid mute oscillator that mutes the tube between the pixels
I am experimenting with 150% zoom to 720p resolution material on a 1080p plasma. Plasma has nice 'oversampled' zeros between the dots in the plasma cells. I mean the dots are focus points in the cells that are dark on the edges. The zoom shifts the aliases to something like 1440p in virtual memory. What remains is that window like look on 1080p.
Love plasma and CRT
_________________ 701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
 |
Dancrt
Joined: 16 Sep 2017 Posts: 88
|
| Posted: Sat Mar 19, 2022 10:39 pm Post subject: |
|
|
Wow you really have loooked into that I have to say every new tv I’ve seen seems to have noisey
Shadows even with the smart processing turned off
So I get what your saying you lost me at “invent a grid mute oscillator“
But yeah motion resolution is bad on so much stuff I tested a test
Video on you tube and filmed it on my iPhone on slow motion
And my barco 909 did it perfectly and my jvc x55re everything from
A midtones and up was blurry !!!
|
|
| Back to top |
|
 |
redfox001
Joined: 16 Mar 2009 Posts: 2257 Location: The Netherlands
|
| Posted: Mon Mar 21, 2022 2:04 pm Post subject: |
|
|
To better understand digital noise, and how it disturbs the 'through a window like look' I will try to explain my own understanding of it.
It all starts with the sampling theorem from Shanon.
Shannon stated that it is possible to make a perfect reconstruction of a filtered input signal with just samples and a reconstruction filter.
Mathematically sample spikes produce a distortion that when seen in frequency space has a lot of very strong aliases, so it is not perfect.
When you multiply (reconstruct) the samples with a sinc function (sharp bell like shape) then all the aliases in frequency space are muted and you have a perfect reconstruction. That is the Shannon theoretical solution.
When you multiply with a rectangular function like digital does (a pixel) than all the aliases are weakened but also the high frequency signal is dimmed. So this is not prefect.
When you multiply with a gaussian clock shape distribution (like CRT/Plasma) you weaken the aliases but not so much the higher frequencies in the signal. So this gives a better resolution, closer to sinc.
So when you improve the bandwidth of a CRT you more and more multiply the sample spikes with a rectangular pulse and you loose high frequency information. When you could insert black dots (oversampling) this would improve the high frequency information that is why I mentioned the mute of the tube (not too serious)
But the beauty of a round Gaussian is also in two dimensions better because it improves also the diagonal resolution. So why did they spoil a perfect gaussian reconstruction with the digital rectangular. That is just stupid. And now they improve with subpixels and 4k. In the end you get something that looks like a Gaussian with 8k or more dimmed random pixels That can be done simpler with the Gaussian noise of a proper focused CRT beam.
----
To mathematical understand this all and remember it you have to know that a rectangular function becomes a sinc function in frequency space so it modulates the frequency response in a bell shape dimming higher frequencies. A sync function becomes a rectangular function in frequency space so it modulates the frequency response as a perfect brickwall filter.
A Gaussian looks something like the sinc and is also a Gaussian in frequency space. In fact the smaller the Gaussian peak focused in time the broader the Gaussian frequency response will be. You really need some HF filtering after this width the electronic bandwidth I think.
To improve on the high frequency response in a Barco 909 I would not improve the bandwidth because what you win in contrast you loose because of mathematics and you end up with digital harshness because those aliases look very bad So probable settle for something in the middle.
Anyway I saw this in the real world as I observed the image from my modified Marquee and compared to my theoretical inferior Barco. Not
_________________ 701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
Last edited by redfox001 on Wed Mar 23, 2022 1:16 pm; edited 2 times in total
|
|
| Back to top |
|
 |
Tim in Phoenix
Joined: 21 Oct 2006 Posts: 4409 Location: Phoenix
|
| Posted: Mon Mar 21, 2022 9:42 pm Post subject: |
|
|
Guys
I have sent an email to Moome requesting his comments here
Tim Martin, E-Tech Systems Scottsdale
480 847 9448 please text me first
or ehometech8022@yahoo.com\0
|
|
| Back to top |
|
 |
kal Forum Administrator
Joined: 06 Mar 2006 Posts: 18114 Location: Ottawa, Canada
TV/Projector: JVC DLA-NZ7
|
| Posted: Mon Mar 21, 2022 10:46 pm Post subject: |
|
|
Scalers already exist to take 4K to any resolution you like: http://www.curtpalme.com/Radiance.shtm
It does that plus a whole pile of other features.
The problem is that people probably want something cheaper, but you're not going to see anyone putting any effort into something that is strictly for CRT. That would (to be blunt) a completely stupid business move as any sales related to CRT are completely dead. That died many many years ago. To design something this advanced requires volume orders, something that would not have happened 10 years ago for CRT let alone today. If a specialized product is to be designed, it has to have a larger audience than CRT projector enthusiasts.
I'm not trying to poo-poo anything CRT related. (Remember that I was the one who put together many hundreds of pages of info for this website, much more than was required - I wasn't paid for it). I'm just trying to set some realistic expectations here.
Kal
_________________
Support our site by using our affiliate links. We thank you!
My basement/HT/bar/brewery build 2.0
|
|
| Back to top |
|
 |
redfox001
Joined: 16 Mar 2009 Posts: 2257 Location: The Netherlands
|
| Posted: Wed Mar 23, 2022 1:11 pm Post subject: |
|
|
Would perhaps be nice if Moome could use a higher frequency video dac that produces an upsampled 1080 signal (with just zero's), that would be great. The electronics of the CRT would be the low pass filter needed to get an upsampled 4k output The analog video signal would have the same energy and contrast as the original but far less pixels noise.
_________________ 701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
 |
Tim in Phoenix
Joined: 21 Oct 2006 Posts: 4409 Location: Phoenix
|
| Posted: Wed Mar 23, 2022 2:46 pm Post subject: |
|
|
The Radiance may work if you hang a 4K flat screen on the output; what I need is HDCP stripping.
DVX8022 specs "Line frequency up to 130khz." on the input. If 4k is twice 1080P/60 that would be 134khz. The Analog NeXtage units would handle it but possibly still need HDCP stripping.
One may also need to take a class how to set it up!
| Description: |
|
| Filesize: |
127.28 KB |
| Viewed: |
4062 Time(s) |

|
|
|
| Back to top |
|
 |
kal Forum Administrator
Joined: 06 Mar 2006 Posts: 18114 Location: Ottawa, Canada
TV/Projector: JVC DLA-NZ7
|
| Posted: Wed Mar 23, 2022 5:29 pm Post subject: |
|
|
| Tim in Phoenix wrote: | | The Radiance may work if you hang a 4K flat screen on the output |
No, it'll work just fine if you want to convert/downrez 4K to something lower for CRT use:
Source -> HDMI cable -> Any Radiance Pro -> HDMI cable -> CRT projector with a Moome HDMI input card
Only caveat is that the output resolution on the Radiance Pro has to be set to something the Moome input card will accept of course.
The Radiance Pro and Moome HDMI cards are available here at our discounted prices: http://www.curtpalme.com/Products.shtm
Kal
_________________
Support our site by using our affiliate links. We thank you!
My basement/HT/bar/brewery build 2.0
|
|
| Back to top |
|
 |
|
|