Please explain video bandwidth to me!?

Well, the simple answer is that they're NOT good for 1080p, but it doesn't necessarily make much difference, except for test patterns!

Bandwidth usually refers to the upper roll-off frequency where the signal power has fallen by 3 dB, which is a half. That means that even at 110 or 130 MHz, the amplitude of the signal is significantly down on lower frequencies, and gets progressively lower at higher frequencies still. There's no sudden brick wall effect; there is still response beyond the bandwidth, but the finest detail in the picture will be reduced in contrast. This will give a softer picture than where the BW is sufficient. In theory and in practice, you can continue to push video to even higher frequencies, but you progressively lose detail.

I don't get 186MHz, though. 1080p is usually 2200 x 1125 x 60 = 148.5 MHz.

Nick

One Bad boy Barco 1209s was still running at over 400MHZ bandwidth

But your right the picture was a little soft

Total lines x total pixels x refresh rate.

Total includes non-displayed (blanking) lines.
Blanking lines can vary with format and refresh rate.
For example, with 1080p60:

Visible resolution: 1920 x 1080
Total resolution: 2200 x 1125
Refresh rate : 59.94 Hz

Nick

Go to the archives and look up CWM9's post about bandwidth. It was about a year and a half ago. Kal may have the info over on Curt's site.

Had I been the first poster after the OP, I was going to write something up about picture width, widescreen vs 4:3. But you ratbastards ruined that now, didn't you..

Sparkle inquired:
> http://www.myhometheater.homestead....calculator.html <<br />
Nick's comments are correct, but the big problem is that the calculator at that site is bogus and misleading... and has confused many CRT owners.

I've pointed this out to Chris to no avail. That calculator comes from Extron sources, and is intended for components that have to process a signal and pass it on to other stages, while doing minimal damage (losses) in the process. It's not designed for a device at the end of the chain (your PJ), or for dealing with real-world film or video signals (i.e., not sharp-edged computer text transients).

If you follow up on Eric's suggestion, CWM9's posts should provide all the info you seek.

- Tim

P.S. PJ, nicely done.

Am I missing something?
I thought RGB Bandwidth was calculated with the formula BW = [(TP x Vt) /2] x 3

Which for 1920x1080p@60Hz would be...

1920 x 1080 = 2073600

2073600 x 60 = 124416000

124416000 / 2 = 62208000

62208000 x 3 = 186624000

~ 186.7MHz

That's just visible pixel clock pls 50%, which is probably just a rule of thumb.

Or else it is meant to use total resolution instead of displayed, then add a bit extra to account for the -3dB at the upper roll-off.

Nick

Well, there is rarely anything that I can add to Tim's posts, so let me take this opportunity.

Those Extron calculation are quoted here all the time and despite many people including Chiem trying to explain it, it persists.

Those calculation are based upon the follow assumption: A component that has to process a signal and pass it on to other stages requires TRIPLE the bandwidth of the signal it is processing to avoid loss. As Time said, you PJ is not one of those devices. Chiem's posts that Eric suggested are pretty good.

Dave

Unfortunately, I think it receives a lot of hits. This is not unlike Projector Central misinforming people dozens at a time. If Chris is the owner of the site and refuses to change it or clarify it, then shame on him.

Here is Chiems thread:
http://archive2.avsforum.com/avs-vb/...ight=bandwidth

Thanks for finding that.

Hey, since you are "credits boy" you going to give me credit for spurring the creation of that thread with my "Why is there no visible difference between a Barco 808 with 75 MHz bandwidth and a 1208/2 with 120 MHz bandwidth" thread that he commented on then wrote that post.

Dave

Also Remember some tubes get very soft at 1080p too much for the spot size.

Where the 180DVB tubes in the 808 and 1208/2 would not really do 1080p regardless of the Bandwidth.

IF you use P16 tubes in a 1208 vs P16 in a 808 then you will see a difference of the BW. Without the upgraded tubes the differences will not be as good.

The aperture of the gun makes all the difference when running 1080p because you have a smaller spot size. So the projector can do its work.

The real improvement of most of the S models was the high rez guns P19 and P16 tubes. If you put a high rez gun in a 1209/2 and a 1208/2 you will see high rez pictures that are better than the S models because of the neck cards. The major draw back of the S models was the neck cards.

The only problem is the back end amps on the 1208 1209/2 are a little soft and they can be upgraded.

However the software on the S models was better. I have seen a few rare models of early S that used the old neck cards. These are the best projectors to use.


--I have not seem too many S model projectors pull off a perfect 1080p image. The only two units I have seen it on are the 1208 1209(Non S) with high rez tubes and upgraded back end amps, and the 909 or cine 9.


--Remember if using P19 tubes in your 1209 to link the Grids tougher or your picture will be soft.


Greg

Only the 1208s/2 uses P16 tubes. The 1208s and 808s do not.

Greg,
would you/ can you put the older 1209/2 neck cards in a 1209s then?

No you cant that is the major problem. Some have tried and you get an IC2 error

The back ends of the S models are much better and the tubes are better on the S models but the neck cards are not. The S uses a 5 GHz input stage with 800 MHZ op-amps.


That all being said. The did make a few S models with the Sony Tubes that used the old generation neck cards True VPA-13 chips and old back end. The advantage of this model is it had the S software to work with the old Neck Cards.

I have this software but you cannot just plug it into the older units because they moved to a Motorola processor computer that uses different instruction set. So basically you have to replace the entire computer and put it into an old 1208 or 1209.

If you upgrade the computer in an older 1208 and upgrade the back end. Its very easy to upgrade the back or input stage on the older 1208 1209 because it has discrete components and you can move up the Semiconductors to a larger BW. Then the final picture is quite good.

If you want to do this email me and I can try to help you.

Greg

Hi Greg.
thank you for the information.
I find your posts very helpful and informative and I'm sure others do to.

Question:

Is there a way to test bandwidth of CRT projector and get graphical readout on some piece of equipment?
Reason why I ask is I remember a magazine from 80-early90's that injected a high bandwidth known test video signal at input stage of monitors/projectors and then used a scope at cathode (I think?) to display output right before tube. I remember that there was a huge difference between equipment tested and they actually could prove actual bandwidth of a device(+ distortions or other issues) Is this possible or do I not remember correctly.

thanks again,
Nate

No problem, as long as I get to trademark the name Kimcoder!

Nick,

You indicate that total lines and total pixels should be used for calculation of bandwidth. I found this website, which has a table containing the total lines and total pixels for different resolutions. Are the values in that table correct? I can figure out which values are appropriate for 1080p, 720p, and 480p. But I'm not clear on which values are for 1080i. What total lines and total pixels should be used for 1080i; 2200 x 562.5?