Panny 50"ers -- topping out the vertical refresh rate

StooMonster

Prominent Member
Joined
Nov 20, 2002
Messages
4,975
Reaction score
308
Points
991
Location
Kent
Just got the manual for Panny TH-50PHW5 out and noticed the PC signals section:
Horizontal scanning frequency 15.6-110kHz
Vertical scanning frequency 48-120Hz

Previously I've only read the PC resolution input table (that doesn't mention 1366x768) and saw 85Hz as highest listed; which is why it's the highest I've ever set the screen.

100Hz or 120Hz is very interesting, anyone tried to run 1366x768 at this kind of vertical refresh rate?

Would be great for gaming, and I'm thinking a 4:4 pulldown on PAL material at 100Hz would remove any trace of flicker (or 96Hz for NTSC at 4:4 pulldown).

StooMonster
 
I read something about this sort of thing a bit ago, can't remember what thread it was though. The problem that was raised was something about getting a scaler to correctly do 4:4:4 play back of PAL material, with even 3:3:3 being difficult to do correctly with out introducing problems, I figure this would be even worse with NTSC as the 2:3 pull down would have to be done first then doubled up to the 4:4 sequence....

I think it might have been a discussion on the HD-Leeza and CRT PJ's maybe....

This might all make no sense at all, if not just ignore it, its late and I'm tired :boring:, I'm sure someone a lot more knowledgable will no doubt be along soon (Gordon etc..).

Cheers

Ryan :smashin:
 
I'm going to experiment after work tonight, if I ever finish that is.

Some people claim plasmas resample all vertical refresh rates to 60Hz.

But when I display say 1366x768 @ 75Hz on this Panny5 the display readout says "H:60Hz V:75Hz"; and I don't seem to have any judder issues.

StooMonster
 
Everything is converted to 60hz except pal at 50hz (they can't even get the subfield processing to work well with pal correctly with banding and flicker). If they could except different refresh rates everyone would be trying to get 48hz or 75hz when watching DVDs.

Its the same with many DLP projectors.
 
Stoo - please let me know how you get on with this, I've just done a few tests of my own, described below. It'd be interesting to compare notes (I have a panny6). FWIW mine seems happy enough at 90Hz but I couldn't get it too stable when I pushed it up to 95 or 100 (picture ok, but with flickering ghost-image appearing). Didn't try that for long though... might be doable.

Timmy - do you have a definitive source for this? I'm not sure how or why this would happen, and it doesn't tally with what I'm seeing.

Anyway...

I've been doing my own (albeit rather brief) tests tonight, and my results seem to indicate that the plasma is matching input frequency and not doing anything too clever about resampling it.

What I did is to run frame-locked output on the PC and try different refresh rates to judge the difference visually. If the output was forced to 60 there ought to be visible problems (most likely stuttering).

What I saw seemed pretty smooth to me at all rates. I tried a bunch of different rates, and tried quite hard to find ones that would be pathologically bad if resampled to 60Hz.

Additionally, dropping much below 60 I started to pick up a flicker, and although the effects are very subtle I'm fairly sure it was getting worse towards 50.

All of this relies on trusting my eyes, but I've usually got a pretty good eye for this kind of thing. And my eyes are what I usually watch my panel with so they're the things that have to be satisfied!

However I should also point out that it's been a long hard day and I may just be hallucinating, combined with a panel doing more processing than I'm giving it credit for.

Also, the set was storing seperate settings for each scan-rate it detected, because it eventually ran out and now my previously perfectly locked resolutions are all over the place... so I'll now be setting those back to normal and hitting the sack...
 
Furthermore, isn't there a Service Menu setting in Panny5 to make them display PAL at 100Hz and NTSC at 120Hz -- like the Panny4s -- for flickerless operation, with the downside being the greyscale is dropped to 8-bit.

StooMonster
 
BertG and Gordon Frazer have both posted references to plasma panels converting inputs internally... Panasonic to 60Hz and Pioneer to 70ish..?
 
Bert's source seems to be a response from a panasonic techie over the phone - and I'm going to take that with a pinch of salt for now because experience tells me thats not the most reliable source of information. And also because it wasn't actually backed up with any kind of explanation, which I'd like to see really.

I honestly couldn't find a post from Gordon about this though. I did see a thread where it was *hinted* that this sort of thing happens, but I couldn't find the original source. He's obviously a clued up chap so I'd certainly like to find that. Any idea what to search for?
 
I questioned Gordon about stuttering pans after a demo at Event II... Search for 'Lumagen stutter on pans'... Don't think it answers your question though...
 
Oh dear, I'm quite concerned about this. I'm buying a 42HD Panny in a few weeks. I had visions of buying an iScan HD at a later date and perhaps feeding it 3:3 pulldown at 72hz NTSC and 75hz PAL etc. It now seems possibly any refresh rate beyond 60hz is pointless (or is it?). I've put this question to Gordon over on the Video Processing and Prog Scan section. I hope he'll enlighten us on this issue.

Joe, can your Panasonic contacts possibly give us an accurate answer on the internal refresh rate conversion issue?

Kind regards,

JP.
 
Originally posted by StooMonster
Interestingly, with the series 5 Panasonics it's possible to edit a setting in the service menu to automatically refresh PAL and NTSC at double rates of 100Hz and 120Hz, however, in this mode the greyscale depth drops to 8-bit.

Okay, read around a bit and did some research. Panny5 plasmas have 50Hz and 60Hz internal frame buffers and the above works; for the SD versions that is.

The HD versions of the Panny5 -- like my 50" model -- run a double sample rate frame buffer of 100Hz or 120Hz all the time, and keep the colour depth.

So what KDB4 says about frame buffers may have some element of truth, however, the numbers are apparently incorrect.

Can we test this?
An example of "tearing" is where one sees a picture broken up because a vertical refresh can not keep up with the physical output of a graphics card. One commonly sees this in computer games where a PC may be capable of outputting say 180 frames per second, but the vertical refresh is set to 60 Hz, without vertical refresh lock set when there is signficant movement the screen with tear into three sections 180/60 = 3. If the vertical refresh lock is on the picture will display 60 frames per second because it is locked to the output; and you waste all that lovely rendering power. Only lamers leave their games running at 60Hz.

So, how can we utilise this to test that screen is not locked to 60Hz? We run a game that my PC can generate say 100 frames per second, we set graphics cards vertical refresh rate to 85Hz and set the vertical refresh lock to on; thereby limiting and locking the output to 85 frames per second.

If the screen were locked to 60Hz internal frame buffer the effect of playing 85 frames a second with rapid movement would generate "tearing" in vertical lines. If there is no tearing the screen has either a higher refresh rate than 85Hz or is displaying 85Hz.

Perhaps one could argue that the frame buffer is sampling only a few of the entire frames presented to it, but this would result is stutter and judder.

Result of EA's "Battlefield 1942" running at 85Hz with vertical refresh lock selected, not one tear or stutter, silky smooth. Same with 75Hz, 72Hz; I'm going to see if I can run UT2003 at 100Hz later and see if I get any tearing then.

Therefore suggesting that my research at the beginning of this post was correct, and this screen does not appear to have an internal frame buffer of 60Hz more like one of 120Hz.

Hrm... we like science: theorise, experiment, conclude.

Back on topic...
Therefore it's going to be very interesting to see how iScan HD performs with 3:3 at 72Hz and 75Hz on this screen.

Stuart

PS: Don't know what the deal is with any other brand of plasma or generation of Panasonic, this is UK Panasonic 5 series.

PPS: still reading though service manual.

I get the same results as _Sin
 
That's interesting StooMonster... I'm still going to hold off my 50" purchase a little longer though... I'm also very interested to see how iScan HD performs with 3:3 at 75Hz on this screen...
 
I found that sending 60Hz SD/ED and HD resolutions to Pioneer433MXE resulted in an image which looked like it was probbaly refresh ing at 60Hz while sending PC resolutions at 60Hz ade it appear to do some frame rate correction stuff itself (probably up to 70.xxxHz). It is a strange phenomenon.

Gordon
 
Pioneer PDP 503 MXE UK brochure states "...has eliminated problems with flicker on displays, due to 100Hz progressive scanning. Flicker from PC signals will also disappear even in lower refresh rates."

This sounds very much like the internal frame buffer of some Pios is also not 60Hz.

StooMonster
 
I've been doing some more testing...

I wrote an application to display a vertical bar moving horizontally at exactly 1 pixel per frame. It also displays a fps timer so I can see what windows actually thinks it's displaying. It's a little rough and ready but it should let me test things out.

On a standard resolution, even one far in excess of what the display can do natively, I can change the refresh rate to whatever I like and my application is always smooth, and always displays (more or less) the fps to be the same as the refresh rate.

This tells me that the display is probably refreshing at the rate it's being told to, and not a hard-coded internal rate.

However, at the custom resolution I set up in powerstrip, I can only use rates of 60Hz or below. At rates above this, my application still reports back a rate around 60Hz and the movement starts to appear to stutter. The fact that windows itself is reporting the wrong update rate for the display tells me that windows doesn't actually think the display can be running any higher than 60 - I suspect something is wrong with the way I've set this up in powerstrip (or powerstrip itself hasn't done it right).

One caveat is that at certain "normal" refresh rates, my application is reporting a FPS a few Hz out and is actually stuttering. I think windows has got the timing wrong - but it only happens on a few combinations.

If the driver lies to windows about the refresh rate, or windows refuses to believe a correct one, that may explain why my DVD playback is not as smooth as I would expect either. The stuttering here is evidently happening on the PC and not at the panel.
 
Just a final word from me - I'm interested in a scaler myself at some point, but no amount of my playing (and playing is all I'm doing!) with the panel will convince me to buy one without seeing it working on my model of panel first...
 

The latest video from AVForums

TV Buying Guide - Which TV Is Best For You?
Subscribe to our YouTube channel
Back
Top Bottom