Has anyone seen the Lumagen Vision or Vision Pro? - Page 7 - AVS Forum
Forum Jump: 
Closed Thread
 
Thread Tools
post #181 of 230 Old 04-01-2003, 01:25 PM
jrp
AVS Special Member
 
jrp's Avatar
 
Join Date: Jan 2003
Location: Portland, Oregon, USA
Posts: 1,049
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 39
Quote:
Originally posted by VideoGrabber
Jim,
thanks for the clarification. Was I correct that 720p at 72 Hz would be ~1500?

- Tim
At 720p at 71.93 Hertz (72/1.001) we oversample to 1553 pixels.

Jim Peterson
Lumagen, Inc.

Jim Peterson
Lumagen
jrp is offline  
Sponsored Links
Advertisement
 
post #182 of 230 Old 04-02-2003, 05:48 PM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
frawau offered:
> Any particular movie/scene you'd like to see? <<br />
Yes. I'd like to see this frame from TFE, since I have a PC upscaled VMR9 1440x960 version to compare it to. Thanks, François!

- Tim
LL

- Tim
VideoGrabber is offline  
post #183 of 230 Old 04-04-2003, 06:56 PM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
P.S. It's not necessary to take a shot of the whole screen, or the entire monitor surface. Just a closeup of one segment should be revealing (say in the area of the Federal Building, or the Nucleo Lab. There's a lot of detail there... in the windows, and the lights along the balcony on the building, etc. VMR9 does a really clean job of scaling this, partially because both axes are exact multiples (x2) of the source.

If you run your monitor at 960p (@60 Hz), that would put the Lumagen into a mode with a similar 1440 horizontal resolution. If you take a segment closeup, I'll provide the matching fragment for comparison.

- Tim

- Tim
VideoGrabber is offline  
post #184 of 230 Old 04-06-2003, 10:25 AM
AVS Special Member
 
Allan Jayne's Avatar
 
Join Date: Aug 1999
Location: Nashua, NH USA
Posts: 1,721
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Just re-ran the AVIA 200 TVL test comparing the Lumagen to PC monitor with a direct connect to an interlaced monochrome NTSC monitor using the same DVD player and composite output. (the best equipment available now) The latter monitor does display over 540 TVL and does reproduce the 6.75 MHz circle quite well.

The half starburst, or fishtailing always? present at the narrow end of the upright horizontal resolution wedges is much less prominent in the Lumagen output. This should be the benefit provided by oversampling.

Remember when lines and logos burned the TV screen? I was at a concert where a musical selection made extremely heavy use of about four of the keys of the piano.
Allan Jayne is offline  
post #185 of 230 Old 04-07-2003, 07:21 AM
Member
 
gerni g.'s Avatar
 
Join Date: Mar 2002
Location: Germany
Posts: 94
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
hi,

also have questions about this scaler.
how does it perform in video mode?
is faroudja´s dcdi almost the best or does the lumagen come close?
i´m asking this cause my main source is watching tv and that every day.
i´m thinking bout upgrading from my extron lancia linedoubler.
the faroudja nrs seems the best choice but the drawback is the fixed resolution. the bigger faroudjas are not suited for my wallet.
also a htpc doesn´t fit in our livingroom cause of the low waf.
all i want to know is the videoquality and if filmode is autodetect or on/off switchable.

thanx,

gerni

-crt or die!_
gerni g. is offline  
post #186 of 230 Old 04-07-2003, 08:21 AM
AVS Special Member
 
Allan Jayne's Avatar
 
Join Date: Aug 1999
Location: Nashua, NH USA
Posts: 1,721
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
The video/film mode is fundamentally autodetect, which IMHO all the best de-interlacers are. The Lumagen does have a two position menu selection for this which alters the autodetect threshold, not force film or video mode. A review (Stacy Spears et al) of another de-interlacer using the same chip (Sil004) said that it is better to leave it in video mode. Silicon IMage de-interlacer units (iScan's) do not offer the user this selection!

Although Faroudja chips with DCDI (which I have not auditioned) may have better diagonal processing, I have not seen any problems in sportscasts or other live video with the Lumagen nor with the iScan v1 I had before. Then, my TV is not the best so nothing may have been proved.

Remember when lines and logos burned the TV screen? I was at a concert where a musical selection made extremely heavy use of about four of the keys of the piano.
Allan Jayne is offline  
post #187 of 230 Old 04-07-2003, 08:34 AM
AVS Special Member
 
Dale Adams's Avatar
 
Join Date: Nov 2002
Location: Anchor Bay, CA
Posts: 1,886
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Allan,

The reason you won't find a film/video mode switch in an iScan is because you're virtually always better off leaving it in video mode. This is exactly what Stacey (et al) pointed out.

'Film' mode was introduced at the request of DVD player manufacturers using the Silicon Image chips. They wanted a way to bias the source type detection toward film for DVD applications, since the majority of DVDs are film-based. This mode essentially desensitizes the logic which detects source problems and film sequence breaks, which makes the deinterlacer more likely to stay in 3:2 PD lock. Unfortunately, many (most?) DVDs also have mastering problems such as bad edits which the film mode is less adept at detecting. Because of this, it's better to just always use 'video' mode (a bit of a misnomer) since it is much more reliable with problem sources and works well with quality sources as well.

For an application such as a video processor which can receive as input a very wide variety of sources, it's almost always best to just use 'video' mode, so that's what the iScan products do. There may be a few sources out there that would behave a bit better with 'film' mode on, but they're very few and far between.

- Dale Adams
Dale Adams is offline  
post #188 of 230 Old 04-07-2003, 09:53 AM
AVS Special Member
 
yubyub's Avatar
 
Join Date: Dec 2001
Location: NY
Posts: 1,515
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
gerni g.,

This is all my opinion, so do take it with a grain of salt:

I can comment on the performance of video mode v.s. DCDi, but only in comparison to the Sil503 chip (although IIRC the 504 is similar, but cheaper and a bit more integrated - I don't remember there being any difference in the algorithms used between the two chips).

I can compare my IScan Pro, Faroudja NRS, and a friend's progressive DVD player with a Faroudja chip inside (not sure which kind of chip). My NRS is scaling to 1024x768, while my IScan and my friend's player are simply deinterlacing to 480p.

The IScan combs a bit on video, which is to be expected. It is not bad, and is definitely quite an improvement on straight interlaced video. The NRS and the Faroudja equipped player faired a bit better on certain video sourced DVDs - certainly less combing. The NRS did well on sports from my cable box (Time Warner digital cable), while the IScan did an okay job.

All three still had problems with video based sources, but the Faroudja appeared to be a bit better. By no means were video sources substantially cleaned up, but there was a marginal improvement with the Faroudja.

If you spend most of your time watching video based sources, then choose carefully here. If you spend most of your time watching film based sources, I personally believe the Sil504 chip has a _slight_ edge here.

-Jon

-Y
"First things first, but not necessarily in that order." - Dr. Who
yubyub is offline  
post #189 of 230 Old 04-07-2003, 11:04 AM
Member
 
gerni g.'s Avatar
 
Join Date: Mar 2002
Location: Germany
Posts: 94
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
hi yubyub,

thanx for your reply.
it´s a little bit complicated. a friend came by with his htpc, holo3d and mp1.
dcdi is very visible when it comes to formula one or concert videos.
that was the first time guitar strings were straight and glasses were not flirring. the only thing i noticed was, if there are just 2 horizontal lines you could see some movement.
on the other hand, pal looks better in 864p and ntsc in 720p.
so the nrs has a fixed scan rate but dcdi and we´re watching almost every evening tv.
so now, dcdi with 720p or no dcdi and 864p.
here in germany it´s a little complicated to get a nrs and a lumagen just for testing.

bye,

gerni

-crt or die!_
gerni g. is offline  
post #190 of 230 Old 04-12-2003, 09:23 PM
Advanced Member
 
DerekFSU's Avatar
 
Join Date: May 2002
Location: Tallahassee, FL
Posts: 724
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I have one question: I can't stand the resolution from Directv on my Mitsu 73909 RPTV. DVD's look good via a progressive scan megachanger by Kenwood and HD looks great through my Hughes receiver. Anyone use the lumagen to address a satellite standard definition picture? If so, how did it do?

Thanks,
Derek

Go Noles
DerekFSU is offline  
post #191 of 230 Old 04-13-2003, 10:00 AM
Member
 
philnort's Avatar
 
Join Date: Mar 2001
Location: Tujunga, CA
Posts: 120
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Anyone use the lumagen to address a satellite standard definition picture? If so, how did it do?
My results are excellent. Having been an owner of an ISCAN Ultra previously, I anticipated getting great deinterlacing from the Vision on SD DirecTV signals and that's what I got. Smooth, clean and clear, with the usual caviat of lesser quality on those channels which are notoriously "soft". However, the majority of Showtime,HBO, news and PPV channels were clean.

The icing is being able to scale them up to 1080i. Some of the movie channels yielded a smoothness, dimensionality and color palette which previously was the province of DVDs. (I've also had my Philips 9352 grayscale calibrated...so that certainly doesn't hurt)
philnort is offline  
post #192 of 230 Old 04-13-2003, 05:47 PM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Time for some more technical discussions... :)

Back in February, Patrick wrote [my emphasis added]:
> The video is deinterlaced (using a silicon image chip), then processed and bilinearly scaled (using a Lumagen proprietary scaler) to create anything from 480p to 1080p (or 1080i) in scanline increments. <

My understanding is that bilinear scaling is toward the lower end of the scaling spectrum (other algorithms being bicubic, spline, and sinc), with the least computational overhead. Could you comment on the tradeoffs (pros and cons) that result from this selection?

The Radeon video boards use a bicubic scaling algorithm in their GPU chips that yields good results, but has known "sweet spots". I.e., horizontal upscaling to 1080, 1440, and 1920 are artifact-free; while many other resolutions result in banding anomalies. This is a result of the (limited) internal bit-precision maintained by the scaler, resulting in rounding error and quantization noise.

Given that the Vision provides an almost unlimited number of horizontal resolutions, all automatically selected based on the 108 MHz sampling output, does the Lumagen scaler avoid "sweet spots" altogether? Or are there some combinations that provide better/worse results? Thanks!

- Tim

- Tim
VideoGrabber is offline  
post #193 of 230 Old 04-14-2003, 11:20 AM
Senior Member
 
patrick harkin's Avatar
 
Join Date: Sep 2002
Location: Portland, Oregon
Posts: 409
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Tim, the Vision & VisionPro are using a propietary scaling algorithm developed specifically for video. I was oversimplifying things when I mentioned the scaling was bilinear, its 2 dimensional like bilinear but its significantly different. Graphics chips typically use the scaling abilities that they have built in for 3D texture applications. So I'm not surprised that they tend to have sweet spots as it wasn't really designed just for video.

The optimal resolution for a LCD or DLP is the native vertical resolution .

The optimal setting for a CRT projector would balance the vertical & horizontal resolution. That is, as our output vertical resolution is increased, the horizontal oversampling is reduced (since we output at a fixed 108 MHz pixel rate). We have found that the range from 720p to about 840p on most 16:9 projectors provides such a balance.

-Patrick

Patrick Harkin, Lumagen Inc. engineer
patrick harkin is offline  
post #194 of 230 Old 04-15-2003, 06:47 PM - Thread Starter
Member
 
papaek's Avatar
 
Join Date: Aug 2001
Location: ny
Posts: 123
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Patrick, Jim,

Congrats on the WSR review!

Ken
papaek is offline  
post #195 of 230 Old 04-16-2003, 01:02 AM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Patrick,
thanks for the technical followup.

Ken,
I'm glad to see a review finally surface! Thanks for alerting us.

But the "sneak peek" on the WSR web site really doesn't say all that much, and I'm wondering if I go out and track down a copy of the mag is there a lot more meat to the actual review? Thanks.

- Tim

P.S. Duh... I guess if I bothered to check out the front page for VidProc's, I would have seen Patrick's posting. With pointers to all the detailed information, no less. :eek: Thanks for waking me up anyway, Ken. :o I'm off to read now. (I found yesterday pretty taxing.)

- Tim
VideoGrabber is offline  
post #196 of 230 Old 04-16-2003, 01:11 PM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
An overall excellent review, with several goodies I wasn't even aware of. I was impressed to hear about the tailoring of the gamma curves (tweakable R/G/B gains at half a dozen IRE levels) in the service mode menus. Very cool. And Greg did a nice job.

In the interest of keeping Vision discussions focused here, and giving Jim and Pat an opportunity to comment, I'd like to echo some material here. Over in another thread, I was extolling the virtues of the Lumagens, and commented...

Quote:
I haven't seen the Vision either, in person, but it's strong point appears to be almost unlimited flexibility. I.e., the VLS-2000 does strictly 60 Hz, while the Lumagens can do anything from 48-75 Hz in 0.01 Hz (or less) increments. So 48 Hz might be useful in PJ environments not prone to flicker issues. Or 72 Hz might provide smoother, more judder-free motion imaging.
Dale Adams agreed with my comments about its flexibility, but warned...

Quote:
You do need to be careful, however, in assuming that just because a scaler will produce a 48 Hz or 72 Hz output rate that you're actually going to get smoother motion. This only occurs when the frame rate conversion is done with respect to the original 24 Hz film source and not the 60 Hz 3:2 PD video signal.

The SiI504 deinterlacer chip used in the Vision does[sic] produce any indication of what the 3:2 cadence actually is, and therefore with 72 Hz output you're as likely (and probably more likely) to get a 2:4:2:4 output cadence (which will be worse than the 60 Hz 3:2 cadence) than the 3:3:3:3 cadence you expect. This is probably true of the Focus Enhancements CS-1/2 products as well, unless they're doing something special to recognize the 3:2 cadence of the SiI504's output signal and modify their frame rate conversion accordingly.
My response was... 'Rats. This seems to have been confirmed by Greg Rogers in his WSR review. "...the output will slowly shift between a 3-3 and 4-2 pattern, but 60% of the time it will produce a 4-2 pattern, which makes judder appear worse." '

I already knew that the output of the unit wasn't phase-locked to the input, but previous comments here led me to believe that the drift was slow enough that only 1 frame would be affected over a rather long interval. I was hoping (and still am) that Greg's comments were in reference to the strict 72 Hz frame rate, and that when corrected to SMPTE timings of 71.928 Hz, the Visions would provide mostly (99.99%) clean 3-3 tracking.

From Dale's comments and Greg's review, this may not be true? In which case, having anything other than the 59.94 Hz vertical rate may be of very limited value. Comments and clarifications from Lumagen would be most welcome.

- Tim

- Tim
VideoGrabber is offline  
post #197 of 230 Old 04-16-2003, 01:32 PM
AVS Special Member
 
Dale Adams's Avatar
 
Join Date: Nov 2002
Location: Anchor Bay, CA
Posts: 1,886
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Tim,

You wrote:

Quote:
I already knew that the output of the unit wasn't phase-locked to the input, but previous comments here led me to believe that the drift was slow enough that only 1 frame would be affected over a rather long interval. I was hoping (and still am) that Greg's comments were in reference to the strict 72 Hz frame rate, and that when corrected to SMPTE timings of 71.928 Hz, the Visions would provide mostly (99.99%) clean 3-3 tracking.

From Dale's comments and Greg's review, this may not be true?
We probably need to hear from one of the Lumagen folks to be absolutely sure, but from what I know and my understanding of how this all works, I'd have to say that you can't count on a clean 3:3 output sequence.

Using a 71.928 Hz output frame rate should keep you in sync with the source for the most part. The problem is you don't know where each film frame starts in the deinterlacer's output, and therefore you don't know which frames need to be repeated how many times. Using a conventional frame rate conversion mechanism, you'll be wrong a lot of the time. As long as you start out in the right phase relationship to the source, the output should be in 3:3 sequence. If you're then producing output frames at exactly 6/5 times the input rate, you'll then stay locked and keep putting out the correct 3:3 sequence. However, if you start off at the wrong phase, or if you're not producing exactly the right output frame rate, then you're guaranteed to be off sequence at least part of the time. You may actually drift back and forth between 3:3 and 4:2 sequencing, which might be pretty weird.

- Dale Adams
Dale Adams is offline  
post #198 of 230 Old 04-16-2003, 04:30 PM
Senior Member
 
patrick harkin's Avatar
 
Join Date: Sep 2002
Location: Portland, Oregon
Posts: 409
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Hi, with regard to 3:3 output it just can't be done perfectly using the Sii504 chipset and does move from 3:3 to 4:2. So you can try it out but it is probably best to leave it at 59.94 Hz output.

-Patrick

Patrick Harkin, Lumagen Inc. engineer
patrick harkin is offline  
post #199 of 230 Old 04-17-2003, 08:24 AM
Member
 
ipandmusic's Avatar
 
Join Date: Sep 2002
Location: Nothern Virgina area
Posts: 72
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
To us just getting started with scalers what would drifting between 3:3 and 4:2 look like on screen? Blur? Freeze Frame?
ipandmusic is offline  
post #200 of 230 Old 04-17-2003, 10:17 AM
Senior Member
 
patrick harkin's Avatar
 
Join Date: Sep 2002
Location: Portland, Oregon
Posts: 409
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
I've rarely been able to see it myself at 72 hz. Smooth, slow panning in movies is where you'd be most likely to see it. It wouldn't be blurry or a freeze frame. If you noticed it, the pan would not be as smooth in motion--somewhat jerky. If you just leave the refresh at the default 59.94 Hz you don't have to worry about it of course.

Patrick

Patrick Harkin, Lumagen Inc. engineer
patrick harkin is offline  
post #201 of 230 Old 05-13-2003, 04:33 PM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Patrick wrote:
> I've rarely been able to see it myself at 72 hz. <

In Greg Rogers' review in WSR, he mentioned that the default setting was initially 60 Hz (later updated to 59.94 Hz), and that at that rate, an additional frame was generated every ~17 seconds. "The repeated frame created a very noticable break in the motion as the camera panned..."

When you were testing at 72 Hz, Patrick, were you using 72 or 71.93 Hz? And would implementing the enhancement you suggested you would to increase the timing resolution to 0.001 Hz (to enable exactly 71.928 Hz) minimize the frequency of the anomalies?

When Greg tested 71.93 Hz, he reported "...60-percent of the time it will produce a 4-2 pattern, which makes judder appear worse." His conclusion was that "I can't envision a reason to use any setting other than 59.94 FPS." This was particularly surprising to me, since the Vision User Manual explicitly recommends, "For film-based material in these formats (NTSC), 72-Hertz is preferred to reduce motion-judder."

Does the same limitation also apply to PAL, where "75-Hertz is preferred for video and film sources" should now be revised to indicate that 50 Hz is recommended?

- Tim

- Tim
VideoGrabber is offline  
post #202 of 230 Old 05-13-2003, 04:51 PM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
On a completely different topic, I've been wondering about the IR remote-control capabilities of the Vision, for automated macro-based operations.

1) can IR commands be fed to the Vision at high-speed, or do they need to be slowed down to a "user pressing buttons" rate?

2) when macro-command streams are being received, will they always be indicated on-screen? Or can turning the OSD off first conceal the process?

3) for commands in the menu hierarchy where you reach an adjustment that has only relative up/down control, is there a corresponding MENU 0,x,y,z access point that allows for absolute values to be set?

Thanks!

- Tim

P.S. 4) the Pro is way out of my price range, but (just for completeness) do commands received via the serial port on it cause the unit to respond on-screen any differently than IR commands do? Does the 9600 baud data stream accept commands any faster than IR does?

- Tim
VideoGrabber is offline  
post #203 of 230 Old 05-13-2003, 07:05 PM
Senior Member
 
patrick harkin's Avatar
 
Join Date: Sep 2002
Location: Portland, Oregon
Posts: 409
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Hi Tim,
IR commands can be fed to the Vision faster than you can press buttons but its not high-speed. It can go around 10-15 keypresses per second. Yes, you would see onscreen activity for some commands but others simply execute with no display (ie. setting 720p or 16:9 input aspect is a single keypress and no onscreen display occurs). The OSD disables showing the input when selecting inputs only.
We did have menu shortcuts at one time during development where you could enter numbers rather than arrows to get to an item but they are not in our current codebase. Some menu items do have built in shortcuts like output aspect or horizontal rate for example.
On the Pro the commands recieved via the RS232 do not act any differently than the IR commands but that wouldn't be hard to modify and is a good idea. At 9600 baud the serial port does take commands in quite a bit faster than the IR port.

-Patrick

Patrick Harkin, Lumagen Inc. engineer
patrick harkin is offline  
post #204 of 230 Old 05-14-2003, 02:19 AM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thanks, Patrick.

> you would see onscreen activity for some commands <

I suppose it would be possible to mute the display temporarily, while these commands were processed.

> We did have menu shortcuts at one time during development where you could enter numbers rather than arrows to get to an item but they are not in our current codebase. <

That's not the most critical part, though it would certainly be handy having direct access, and not having to navigate the menu tree to select a parameter. The key point is that once you've reached some of the settings, all the manual documents is use of the up/down arrows to relatively adjust the values. Since in general an automation system reconfiguring the Vision wouldn't know the current state, some absolute mechanism would be needed. (Unless of course if where the manual indicates only up/down key usage, numeric selection is also available... then it's a non-issue.)

The only way I can see around this (other than a Lumagen enhancement) would be to force a Factory reset each time to establish a known base, then send command macros for all the deltas to the desired states.

> At 9600 baud the serial port does take commands in quite a bit faster than the IR port. <

Since there's no handshaking, that makes it ~960/second, which IS a bit faster than 10-15. ;)

- Tim

P.S. Has Lumagen ever put together a chart of the full menu command tree, as an overview document? This could be pretty helpful, especially to new owners getting familiarized with the unit. Plus provide a bird's eye view of its extensive capabilities for those considering a purchase.

- Tim
VideoGrabber is offline  
post #205 of 230 Old 05-14-2003, 03:34 PM
Senior Member
 
patrick harkin's Avatar
 
Join Date: Sep 2002
Location: Portland, Oregon
Posts: 409
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
> The only way I can see around this (other than a Lumagen enhancement) would be to force a Factory reset each time to establish a known base, then send command macros for all the deltas to the desired states.

The menu key always restarts at the root of the menu no matter if you're already in the menu somewhere or not so you wouldn't have to go so far as to do a factory reset to get macros working. Also we do have a couple of downloadable Pronto files on our website to get you started if you've got a Pronto remote.

>P.S. Has Lumagen ever put together a chart of the full menu command tree, as an overview document? This could be pretty helpful, especially to new owners getting familiarized with the unit. Plus provide a bird's eye view of its extensive capabilities for those considering a purchase.

Sorry, its a big tree and the most thorough documentation is the manual which you can download on our site. It covers everything but the service menu items which we give to anyone that inquires.

-Patrick

Patrick Harkin, Lumagen Inc. engineer
patrick harkin is offline  
post #206 of 230 Old 05-14-2003, 10:40 PM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Patrick,

> The menu key always restarts at the root of the menu no matter if you're already in the menu somewhere or not so you wouldn't have to go so far as to do a factory reset to get macros working. <

Well, that's twice now that I've failed to get my question across. :) Let me take another shot at it.

For easy (direct access stuff) like Vertical RESolution, we can just do: VRES, value, ENTER. But in the more general sense, we do:
  • MENU --> OUT --> RES --> VRES --> value, ENTER
But what if I want to set the Input Sizing?
  • MENU --> IN --> ADJ --> SIZE --> TOPL --> ????, ENTER
My understanding (perhaps incorrect) is that you can't just enter a numeric value at the ???? point (you'd actually need 2 values here), but are limited to using the ^,v,<,> arrow keys to make relative adjustments to the scanline and pixel values? Key point... it's all relative. If this is true, how can you set it to a known desired destination setting if you don't know what the value you're starting from is? That's why I've kept mentioning "absolute vs. relative values". And the need for a hack like Factory Reset to establish a known baseline.

For anybody needing more than the two configurations provided (MEMA and MEMB) per input, loading all the necessary registers with the desired values looks like a difficult proposition.

> Sorry, its a big tree and the most thorough documentation is the manual which you can download on our site. <

Yes, it is. ;) If it was a small tree, there'd be no need for an overview (big picture). I just spent half an hour drawing it all out on paper. Greg Rogers' review provides a pretty nice view of the trunk of the tree, and your Owner's Manual (which I downloaded many moons ago) provided all the detailed branch information.

> It covers everything but the service menu items which we give to anyone that inquires. <

I'll PM you for that info. Thanks. I assume the service menu is the only way to access the gray-scale tracking settings, and stuff like the Freeze function, which is direct-access but has no key on the remote?

- Tim

P.S. Does the Vision accept IR commands when it's in STDBY mode? (other than ON) I'm pretty sure it doesn't, but thought I'd ask.

- Tim
VideoGrabber is offline  
post #207 of 230 Old 05-15-2003, 10:06 AM
Senior Member
 
patrick harkin's Avatar
 
Join Date: Sep 2002
Location: Portland, Oregon
Posts: 409
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Tim, sorry I didn't catch what you were driving at before. Yes, if your trying to set up pseudo-memories for inputs via a sophisticated remote to get more than the 2 per input that are built-in there would be some adjustments that are relative and problematic. A way to do it would be to let one of the built-in memories be used as a baseline so if you switched to that input setting you could make the relative adjustments from that point. The only caveat, of course, is that the baseline should not be saved after being altered. I'll check into adding a more direct method to do this and also hide the display as you suggested.

In standby (aka off) the Vision only accepts the power-on command.

-Patrick

Patrick Harkin, Lumagen Inc. engineer
patrick harkin is offline  
post #208 of 230 Old 05-15-2003, 03:04 PM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Patrick wrote:
> let one of the built-in memories be used as a baseline <

That's a good suggestion.

> I'll check into adding a more direct method to do this and also hide the display as you suggested. <

That would be wonderful. Thanks a lot.

- Tim

- Tim
VideoGrabber is offline  
post #209 of 230 Old 05-19-2003, 09:52 AM
AVS Special Member
 
VideoGrabber's Avatar
 
Join Date: Oct 1999
Location: SW Michigan
Posts: 2,799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Patrick, a while back (March) you commented...
> Macrovision is removed from the video signal and it does not effect the video quality. <

I see in the Service mode menus that there are adjustments for MV black offset and contrast offset. Are these to compensate for the level shift caused by the presence of MV, which on other units can result in loss of brightness or saturation?

Do these have to be calibrated by the user/installer? Or are the factory settings normally all that is necessary (and users don't need to be messing with them)? I assume that once set, MV will be detected, and the proper levels used automatically?

- Tim

- Tim
VideoGrabber is offline  
post #210 of 230 Old 05-19-2003, 10:50 AM
Senior Member
 
patrick harkin's Avatar
 
Join Date: Sep 2002
Location: Portland, Oregon
Posts: 409
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Tim, yes these are factory set and automatically selected. While the Vision does remove macrovision from the signal its presence does effect the DVD players output levels and these settings compensate for that. They almost never need to be changed from the factory setting but there was one case where an adjustment was needed so a backdoor is available if someone has a problem.

-Patrick

Patrick Harkin, Lumagen Inc. engineer
patrick harkin is offline  
Closed Thread Video Processors

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