XGA DLP BANDING MYSTERY SOLVED!!! - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 23 Old 04-22-2001, 06:15 PM - Thread Starter
jxo
Advanced Member
 
jxo's Avatar
 
Join Date: Jul 1999
Location: New York, NY
Posts: 999
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Sounds like a supermarket tabloid headline... Many of you may recall a raging debate on this forum about a year ago regarding xga dlp machines and their suitability for HT use. A few folks got their hands on an NEC LT 100 and noticed some nasty banding/posterization artifacts. These artifacts manifest as purplish blobs that appeared in high contrast scenes, eg dark scenes with lit faces; the posterization would appear, for example, in shadows on the face. Instead of a smooth transition from light to dark, the area would take on a cartoonish quality of blotched/blocked purple.

I have been working with Thumper to modify my Plus UP 1100. After a round of mods, I noticed these posterization artifacts appearing in scenes where they never appeared before (I use an htpc/ Radeon LE with WinDVD 2.6.4). They also appeared on HDTV signals. As the "Masters" got late in the day, the posterization began appearing in the grass in the shadows. This was really distracting and I understand why folks condemned these units.

Thumper didn't panic and quickly concluded that the artifacts appear if the color wheel is not properly alligned. He intentionally misalligned color wheels on some units he had around and was able to replicate the artifacts, which he dubbed the Barney-effect (damn purple @#$%). I crossed my fingers and sent the unit back to Thumper to have the color wheel alligned. The 1100 came back on Thursday and I have been through a dozen titles.... not a SINGLE instance cam now be detected. It may be psychological, but other image qualities also seemed to improve.

Bottom line: banding/posterization should not scare you away from xga dlp; if you got, it can be cured. Coupled with Thumper's other mods, you may have the best single-chip dlp image available. A Plus 1100+Thumper's mods+htpc=one of the best values in dlp image you can get.

------------------
Regards,
Jim

Regards,
Jim
jxo is offline  
Sponsored Links
Advertisement
 
post #2 of 23 Old 04-22-2001, 06:37 PM
AVS Special Member
 
kelliot's Avatar
 
Join Date: May 2000
Location: Thousand Oaks, CA
Posts: 3,945
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Nice post.

------------------
Ken Elliott

Ken Elliott
kelliot is offline  
post #3 of 23 Old 04-23-2001, 05:38 AM
Advanced Member
 
Joe Linn's Avatar
 
Join Date: Oct 1999
Location: Roseville, MN. USA
Posts: 838
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
If what you were seeing was related to a misalligned color wheel, that would certainly explain why some people had the problem and other's didn't.

Joe
Joe Linn is offline  
post #4 of 23 Old 04-23-2001, 07:17 AM - Thread Starter
jxo
Advanced Member
 
jxo's Avatar
 
Join Date: Jul 1999
Location: New York, NY
Posts: 999
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Joe: exactly... apparently it doesn't take much misallignment to cause the problems. Thumper believes that the misallignment would not be caught by any QC protocols for these units-- the artifacts just don't seem to come up other than on video sources. I am anxious to get Seleco and Dwin units in the setup to AB against the "Thumperized" 1100 and the Sony G90. Having spent a few hours with those other dlp units already, if I had to bet, I would put my money on Thumper. My sense is that the Seleco and Dwin units could have slightly more accurate colors (ie/read, slightly better reds) but Thumper's unit will prevail with regard to black level performance. I can't say enough how much better low level detail is. Dark scenes have lost most of the washed out look and the halo is practically gone-- further improving perceived black levels. I don't think lcd will perform like this (even the new Sanyos)... anyone considering lcd should wait for Thumper's impressions of the Sanyo units.

Prospective buyers should not be swayed by inflated stats. Lumens and resolution are important but at the end of the day (or, in my case, at the end of a year), much less objective and measurable characteristics are the most important. Color depth and accuracy and, perhaps, most importantly, low level performance are the areas where the most improvement is needed and where Thumper came through. Any 2000L lcd will great with TS2 but be sure to put it through a black level torture test to see if you can live with it in the long run.

------------------
Regards,
Jim

Regards,
Jim
jxo is offline  
post #5 of 23 Old 04-23-2001, 08:29 AM
AVS Special Member
 
MrWigggles's Avatar
 
Join Date: Aug 2000
Location: Houston, TX, USA
Posts: 6,240
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
JXO,

To be a little more exact, the color wheel sensor needed to be realligned not the colorwheel itself. Also note that the colorwheel sensor needs to be alligned precisely as you stated. However, there is a huge range for adjustment. It is a little like using a sun dial as a stop watch. It is hard to be accurate when the precision of the instrument or adjustment is so wide.

I am glad to hear of your success with the unit.

-Mr. Wigggles

------------------
The Mothership is now boarding.

[This message has been edited by MrWigggles (edited 04-23-2001).]

The Mothership is now boarding.
MrWigggles is offline  
post #6 of 23 Old 04-23-2001, 08:37 AM
Advanced Member
 
Jeff Lemke's Avatar
 
Join Date: Jul 2000
Location: Grand Blanc, MI USA
Posts: 695
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I just sent my Up-100 out to have it "thumperized". I hope it is a significant improvement.

Glad the banding fix was spotted, lucky timing on my part.

Jeff
Jeff Lemke is offline  
post #7 of 23 Old 04-23-2001, 08:48 AM
Member
 
ChrisO's Avatar
 
Join Date: Jan 2000
Location: McKinney,Tx,USA
Posts: 66
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I have a Sharp Notevision 7, but I assume the alignment problem/fix applies to it as well. So what is involved in aligning the sensor? Is it a do it yourself kind of thing? I already need to open the unit up to clean out some dust spots, might as well align the sensor as well.

Chris
ChrisO is offline  
post #8 of 23 Old 04-23-2001, 10:30 AM
Advanced Member
 
dlptheater's Avatar
 
Join Date: May 2000
Location: Calgary, Alberta, Canada
Posts: 576
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Wow this is great news. I'm sure in the next few month LT100's/1100's will start dropping in price even more. One of these with Thumpers mods would be a great entry into the XGA world http://www.avsforum.com/ubb/smile.gif

------------------
lance
I'm looking for a job!
Resume online at http://resumes.dice.com/lkstitch
dlptheater is offline  
post #9 of 23 Old 04-23-2001, 11:19 AM
Member
 
ChrisO's Avatar
 
Join Date: Jan 2000
Location: McKinney,Tx,USA
Posts: 66
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Peter,

I agree that the banding problems on the NV7 can be horrible. We almost didn't buy one until my wife and I discovered that the banding problems are greatly reduced if you run s-video into a DVDO Iscan into the projector via component.

We tried several options and I would rate them as:
s-video direct to projector - banding
s-video to Iscan to projector as RGB - worst banding, worse that s-video direct
s-video to Iscan to projector as component - very mild banding, still see some on occasion but rather rare.

I haven't tried the projector with a HTPC or thru the DVI interface.

Hope that helps.

Chris
ChrisO is offline  
post #10 of 23 Old 04-23-2001, 03:02 PM
AVS Special Member
 
indygreg's Avatar
 
Join Date: Feb 2001
Location: indianapolis
Posts: 1,109
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
i am thinking the same thing as dave was. could this be the root of the rainbow effect? i have a compaq 2800 and for the life of me cannot see it. if i move my head in front of an led clock i can see all kinds of funkiness but i have never seen this rainbow effect with my projector nor has anyone in my family. maybe it is genetic or maybe the compaq just does something right but i cannot see it no matter what i do.

greg

------------------
-------------------------
Why does he shake hands with that guy right after he sexed the alligator

-------------------------
Big Red Cinema Build - In Progress
indygreg is offline  
post #11 of 23 Old 04-23-2001, 06:25 PM
Member
 
Robert Payne's Avatar
 
Join Date: Apr 2000
Location: Kingsland, GA
Posts: 62
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thumper has stated in the past that he does not currently, nor does he have any plans to, modify the Proxima DX3. That is unfortunate for me since I happen to own a DX3.

Is there any way that I could perform this alignment myself? I happen to notice the banding problem quite evidently during "The Rock" and "Armageddon". Specificly, what would I need to do?

Thanks.

-robert

------------------
"Of Course I'm at home. I'm un-cool."

"Of Course I'm at home. I'm un-cool."
Robert Payne is offline  
post #12 of 23 Old 04-23-2001, 06:30 PM
Senior Member
 
Thumper's Avatar
 
Join Date: Jul 2000
Location: Maryland, USA
Posts: 397
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Mr. Wigggles is correct, the proper name is the color wheel sync (timing) using the "sensor arm" position adjustment (It's my fault, I don't try to get to technical with these things in casual side correspondence). There is a WIDE adjustment range possible, but only a very small proportion for the proper adjustment.

As to the rainbow effect, I am not as sensitive as some folks are but I can see it on certain DLPs. I believe there are at least five major components which come into play with the rainbows which may explain some of the randomness:

1. An individuals sensitivity
2. Implementation of a clear section
3. Speed of the wheel (i.e. HTPC sourcing)and/or number of segments
4. Gain of the screen
5. Strobing of lamp thru segments (particularly clear) and/or mirror strobing as a sensitizing element

I think #5 has not been explored enough and also deserves some attention. Greater combinations of the above would seem to increase the odds of one having a rainbow problem.

Thumper

Thumper is offline  
post #13 of 23 Old 04-23-2001, 06:44 PM
Senior Member
 
Thumper's Avatar
 
Join Date: Jul 2000
Location: Maryland, USA
Posts: 397
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
As a followup to the previous post:

It is possible that the wheel sync timing adjustment may play into the rainbow mix as described above. This aspect will be studied more to see if there is any relationship or not.

BTW, I have aquired a U2-1130 and will be examining it over the next three-four weeks for any areas of improvement.

Thumper (again)
Thumper is offline  
post #14 of 23 Old 04-23-2001, 06:52 PM
AVS Addicted Member
 
RobertWood's Avatar
 
Join Date: Oct 1999
Posts: 17,185
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Did this have any effect on crawlies?

Bob Wood

------------------
~~The Sultan of Cheap~~
RobertWood is offline  
post #15 of 23 Old 04-23-2001, 07:06 PM
AVS Special Member
 
MrWigggles's Avatar
 
Join Date: Aug 2000
Location: Houston, TX, USA
Posts: 6,240
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Robert,

The crawlies that are generated by film grain are not exaggerated like they are when a projector has a clear section blazing away. Thumper's mod does a good job in getting rid of them.

Now, the crawlies that are present when you get a couple feet away from the screen are still there.

-Mr. Wigggles

------------------
The Mothership is now boarding.

The Mothership is now boarding.
MrWigggles is offline  
post #16 of 23 Old 04-23-2001, 09:14 PM
Senior Member
 
Dave McKean's Avatar
 
Join Date: Apr 1999
Location: San Diego, CA
Posts: 493
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I was just thinking, I could also see how a misalligned sensor could also cause or worsen the "rainbow" artifacts too. Did this get improved with reallignment?

It has to be something similar causing the rainbow artifacts. I've seen alot of DLPs and some have lots of rainbow artifacts and some have none.

[This message has been edited by Dave McKean (edited 04-23-2001).]
Dave McKean is offline  
post #17 of 23 Old 04-23-2001, 09:26 PM
Newbie
 
Peter from Germany's Avatar
 
Join Date: Apr 2001
Location: Hannover, Germany
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hi!

i have a shrp nv7 too, and the bending-artifacts are horrible! the other picture-elements, sharpness, color, contrast are the best picture i have never seen before.

however, this is a very bi problem and i don`t know how to solve it?!

can my dealer (sharp, gemany) adjust the color wheel or the sensor? an for what have i to look for during this "operation"?

sorry, for my bad english...
...i hope i`ll be better next time!

thanxx!


------------------
Peter (Hannover, Germany)

Peter (Hannover, Germany)
Peter from Germany is offline  
post #18 of 23 Old 04-24-2001, 08:25 AM
Newbie
 
Peter from Germany's Avatar
 
Join Date: Apr 2001
Location: Hannover, Germany
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
@ChrisO,

thanks.
On wich settings you run your NV7 on the Component Settings via I-Scan?

On Friday i went to Sharp Germany, i wrote them abaout the banding-problem. they acted very professional until this moment, an further i hope they can solve the problem by adjusting the color whell an the sensor!

i will post it next week!

Greetings,

------------------
Peter (Hannover, Germany)

Peter (Hannover, Germany)
Peter from Germany is offline  
post #19 of 23 Old 04-24-2001, 08:57 AM
Member
 
ChrisO's Avatar
 
Join Date: Jan 2000
Location: McKinney,Tx,USA
Posts: 66
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
The IScan can output the signal to the NV7 in 3 different ways:
1. RGB
2. component
3. component with sync on green

Only 2 of these settings will work with the NV7. If the NV7 is set to RGB then set the IScan to RGB. If the NV7 is set to component then set the IScan to component with sync on green. There is a switch on the IScan to switch from RGB to component. To select sync on green there is a jumper inside the IScan that you need to change. This jumper will not affect the IScan's RGB output so once you set the jumper you can leave it.

It won't hurt the NV7 if the settings are mismatched and it will be very obvious that something is wrong. If the IScan is set to component and the NV7 is set the RGB, everything will be very very red. If the IScan is set to RGB and the NV7 is set to component, everything will be very very green.

I set the NV7 to component and use the IScan to output the image from the DVD player. For HDTV I bypass the IScan, set the projector to RGB and go directly from my RCA DTC100 to the NV7.

Hope this helps.

By the way, the NV7 has been discontinued by Sharp, don't know how much effort they will put into fixing a dead product. Good luck.

Chris
ChrisO is offline  
post #20 of 23 Old 04-24-2001, 02:21 PM
AVS Addicted Member
 
noah katz's Avatar
 
Join Date: Apr 1999
Location: Mountain View, CA USA
Posts: 20,621
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 294 Post(s)
Liked: 167
Can anyone think of a reason why the SVGA machines wouldn't have the same banding problem?

Thanks

------------------
Noah

Noah
noah katz is online now  
post #21 of 23 Old 04-24-2001, 04:21 PM
AVS Addicted Member
 
krasmuzik's Avatar
 
Join Date: Oct 1999
Location: NewPort, VA
Posts: 11,270
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Can you throw up an Avia horizontal gray scale ramp and tell me what you see on this supposedly band-free box? I guess too late for a before/after shot? I have always seen bad banding in the middle. Cannot say I have seen purple blotches on faces though.
krasmuzik is offline  
post #22 of 23 Old 04-26-2001, 06:02 AM
Newbie
 
Peter from Germany's Avatar
 
Join Date: Apr 2001
Location: Hannover, Germany
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
@ChrisO & other I-Scan Users,

the resolution of a XGA-Panel is not the resolution of the signal which comes out of the I-Scan.

The I-Scan produces only (in my case 720/576 PAL), so the XGA-Projector had to scale down on 800/600. Is this right?

Why is there no Up-Scaling to Panel in the Projector! The Resolution from a Panel upper SVGA is lost!

I had an an Sony 7700 with Progressive Scan in PAL. The Board is configured to XGA, the Picture is very, very, sharp and there are much more details as the I-Scan shows, but the Banding Problem ist still existing...



------------------
Peter (Hannover, Germany)

Peter (Hannover, Germany)
Peter from Germany is offline  
post #23 of 23 Old 05-01-2001, 10:31 AM
Member
 
ChrisO's Avatar
 
Join Date: Jan 2000
Location: McKinney,Tx,USA
Posts: 66
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Peter, in my case the Iscan outputs a 480 progressive image that the projector (NV7) scales up to the full XGA panel resolution. The Iscan does seem to soften the image some.

Chris
ChrisO is offline  
Closed Thread Digital Hi-End Projectors - $3,000+ USD MSRP

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off