Sliders DVD Releases (Universal, Mill Creek, SD blu-ray, Restoration) (Page 13 of 15)

Seems to be fine, although I'm not familiar with GaussResize.  The wiki says it has an adjustable sharpness parameter p so if you wanted to control final sharpness you could do eg.

nnedi3_rpow2(rfactor=4, cshift="GaussResize")
GaussResize(1440, 1080, p=30.0)    #1.0=blurry, 100.0=very sahrp

I'm just using the standard LanczosResize() with default settings which I believe is 3 taps. 

I don't normally do the upscale in Avisynth as it uses CPU so I prefer to let the media player do it on the GPU.  There's nothing particularly special about Lanczos, I just like it cause it's inexpensive and doesn't look objectionable to me.  Some people complain it's too ringy (overshoot artefacts) but I don't seem to notice it.

Depends on how good your TV is at upscaling 1080p to 4k - I'd expect it should be perfectly fine & I'd be surprised if it was bad.   Although current model LG OLEDs have surprisingly low sharpness at max sharpness setting.

@ireactions your artistic choices have always been the right ones.

All that being said, as a general note to both our artistes.   I may not have the familiarity or possibly the aptitude to match what you are doing.  But I do have the similar nit pickyness.   On my bucket list is editing an ultimate version of the best of whatever we can pull out of these files with the magic of technology.  That means splicing close up shots / medium shots / wide shots on a timeline (which I know I can do, as I've done that stuff before).   And possibly adjusting color or even sharpness to individual scenes.   That will be a project I hope to do one day, when I have the leisure time [not something my life affords at the moment].  It's something I would quite enjoy.

Interesting thoughts @ireactions.

One thought.  I did find an algorithm made in the academic world to add grain to an image.  It's open source and I posted the code here and corresponded via email with the creator.

I'm gonna see at some point if I can get it running on a server.   It would be interesting to try a two minute clip with it processing to add grain and see if topaz has more luck with it.

The algorithm is supposed to really not just layer a flat dots all over the image.  It's really supposed to be better but testing is seeing.  I'm not optimistic though.

One small step for sliders, one giant step for sliderskind

So checking ireactions samples so far....

I agree, this, while not an order of magnitude of a "level-upping" is definitely a strong and impactful incremental improvement.  It's the general fuzzyness, wobblyness, distortion that is gone.

I'd say that one of the big tests, Prince of Wailes, is definitely definitely better.  And  I see it with King is Back.  Even Fever is definitely better though  I have to say I was disappointed that the celebration ("you're rich!") scene at the beginning isn't as clear/sharp as I've always hoped for.  That scene has always bothered me.

One note regarding the asthetic choices:  so for me close up, I find the grain a bit distracting as it's a little bit too much added noise.   On a laptop, when you are watching close up, I guess you could say it's a bit of a negative for me.

However,  I think at a distance, cast to a tv, if would show its value.  Living room distance.  So I can see arguments in both directions.

Overall, I'd say it's quite impressive what pneumatic and ireactions have managed to do.  Basically take away some of the harshest flaws of the S1 episodes, while leaving room in the future for new technologies to come in and try to extract more out of the frame, and perhaps build a more realistic proxy to how the film footage may have looked.   For now, our mad scientists have definitely extracted the best out of these babies as possible, and Universal should be commissioning them (or at least Peacock) to do a substitute for what a restoration would be.

I watched "Summer of Love" and "Prince of Wails" today via the blu-ray Turbine disc on my blu-ray player and TV and... it looks nice. I said that the blu-ray was very good DVD quality, much better than the Universal DVDs. And it certainly looked that way today. Except that reviewing the blu-ray files reveals them to be blurry and desaturated and low contrast.

It looks like the strong presentation in my living room is due to a variety of factors. First: the Season 1 Turbine image may be blurry, but it doesn't have all the field issues on the Universal discs that pneumatic has devoted his life to sorting out. It doesn't have flickering and jagged edges on every straight line; it doesn't have horizontal lines flashing across the screen; it doesn't have all the aliasing on the Universal S1 files that makes the image look pixelated and overcompressed.

The image, while blurrier than the Universal discs, has a very clean look without all the obnoxious distortions. And the clean Turbine video image without the seeming pixelation of Universal is of course going to look better.

Another factor seems to be the scaling from my blu-ray player. The player scaling up the Universal discs to a 55 inch TV magnified all the lines and jagged edges and made it look worse. The player scaling up the Turbine discs is scaling a very clean image and adding a bit of pixel contrast to create the illusion of sharpness. The cleanness of the image is actually blurriness, but the slight blur across all shots whether wides, mediums or closeups actually adds a certain sheen. Blurry wide shots don't seem glaringly poor when everything has a little blur. The player could offset that blur and the image wasn't riddled with distortions, so it looked sharper than Universal.

Another feature of the player and the TV: my blu-ray player seems to amp up the saturation levels for standard definition content and my TV's energy saver settings (which I use to watch SD content on low backlight) also have increased saturation. This filled in the colour that seems to be missing from the blu-ray files, so the Turbine video image seems reasonably colourful when the files themselves look really washed out.

RussianCabbie wasn't able to play the blu-ray properly on his hardware. It played at the wrong framerate because the blu-ray uses 25 FPS files whereas NTSC uses 29.97 FPS. My blu-ray player seems to be able to handle PAL video correctly.

All-in-all, it looks like the Turbine blu-ray's S1 files, while poor, are poor in a way that my home theatre situation can remedy and present effectively. The Universal S1 files, however, are poor due to the field misalignment issues which my hardware can't address and can only make worse.

It'd be interesting to try playing a 540p version of The Pneumatic Edition on my hardware (with no grain added) and see if The Pneumatic Edition looks better than the Turbine files on this hardware since pneumatic has cleared away the field issues.

I agree. The Pneumatic Edition is best.

The great results I'm getting from the Turbine blu-ray are due to a peculiar set of circumstances in my living room. I have no idea how it would turn out for someone else. When I have some time this coming weekend, I'm going to try playing the Turbine blu-ray on my TV using an old PS3 as the player and then a Windows laptop with an external blu-ray drive. That will tell me if the very impressive upscaling is due to the TV or the blu-ray player. I also recently asked RussianCabbie to pass on the Turbine blu-ray files to pneumatic in the hopes that pneumatic might review them and figure out what's going on with these mercurial files.

The one thing that really throws me off about The Pneumatic Edition: whenever we go to a wide shot, the image becomes so ill-defined and blurry in its downscaled analog videotape approximation of a film shot. It's really jarring and it completely knocks me out of the story because it's so mismatched to the otherwise adequate video quality; the video files can't seem to maintain consistent video quality.

In "Summer of Love", the wide shots of the sliders triggering the vortex are so hazy compared to the rest of the scene. The shots of Rembrandt and Wade standing before the hippies is like a watercolour sketch of the actual shot. In "Prince of Wails", the shots of the sliders walking into the Royal Suite look good, but when we go to a wide of the sliders looking around, the shot looks like it's VHS rather than DVD.

The problem is less distracting in the Turbine version because all the shots are a little blurry to begin with, so when we go to a blurry wide shot, it doesn't look out of place since the shot before and after it will be slightly oversmoothed too.

The Turbine version is poorer than The Pneumatic Edition, but it's consistent; you don't go from decent mediums and closeups to awful wide shots. Turbine is consistently very, very, very average (at least at living room distance).

But I think I might be making the problem worse in my files of The Pneumatic Edition: the added film grain texture that I put on the image may be making those wide shots look even more aged even as the grain adds some texture to closeups and mediums.

However, I also have to wonder if this is another issue where we're using Lanczos or nnedi3 or whatever algorithmic pixie dust of the week we're trying to bring these old videotape files to 1080p... and these files are not even really 480p. The pneumatic Avisynth+ scripts are definitely eliminating a lot of the interlacing issues of viewing a CRT-intended product on a modern LED, but what was underneath the interlacing issues offers very little for scaling up.

I'm currently attempting a nnedi3 upscale on the Turbine files just to see if I can replicate how it looks on my TV in the file itself. After that, I'm going to run The Pneumatic Edition again but with no added grain, and I'll output to 540p  and 1080p. It'll be interesting to know if one is more consistent than the other.

The Pneumatic Edition is definitely the best of the lot, especially in avoiding the obnoxious flicker. Even the Turbine version of those episodes have a small amount in some of the worst scenes on Universal.

Why do the Turbine blu-ray discs of Season 1 look so good on my budget blu-ray player (bought for $35 USD refurbished)?

My current theory: the player must be using a very simple scaler. There's no way this $35 player has any sort of neural net or AI scaling or so much as a noise filter. It's likely bilinear scaling, a low-power method of analyzing the surrounding four pixels of each pixel to add more for upscaling.

Bilinear is out of vogue these days since bicubic creates smoother gradients by analyzing the surrounding 16 pixels while Lanczos and nnedi3 have even more intricate methods to smooth-sharpen video. But I'm starting to think Lanczos, nnedi3 and Topaz AI may be completely overshooting on SLIDERS 1.02 - 1.09.

Those scalers are all about rebuilding detail and maintaining smoothness at higher resolutions to offset upscaling degradation. 1.03 - 1.09 don't have much detail to rebuild, so what we keep getting is an oversmoothed blur on video that was already lacking detail.

Bilinear is probably adding very simple pixel contrast that is well-suited to the fuzzy Turbine files. That and the HDMI Deep Color setting are probably making the files look like a good (enough) DVD.

I'll run a few bilinear scaling jobs and report back later.

As for the Universal files: pneumatic's QTGMC deinterlacing and detelecining have created a fairly clean set of files. I think my grain additions were a mistake.

I suspect that pursuing the illusion of sharpness with grain and increasingly complex scaling algorithms is trying to build on a foundation of sharpness that isn't there.

It may be better to just use pneumatic's QTGMC script but edit it to use bicubic scaling to prioritize smoothness over sharpness in these low-grain, low-detail videotape files. But I'm going to start with the Turbine files first since I've already seen how that can turn out pretty well from my $35 blu-ray player.

I agree, the grain was a bad move. This is the first time in recorded history that I have made a mistake. Now I know how everyone else must feel.

I'm currently going back to The Pneumatic Edition and running another encode on 1.02 - 1.09. I've elected to use the bilinear scaler because it worked so well on those smooth Turbine files. And I find that pneumatic-style QTGMC detelecined files have a certain smoothness that's actually very reminiscent of the Turbine files and might benefit from that simple pixel contrast increase. I also won't be adding grain to these files.

I was working with the Turbine files in Hybrid and the bilinear scaling was turning out some good results, but Hybrid seems like a poorly coded program that couldn't produce the colour settings that it was promising in its previews. I realized I'd need to use Avisynth+ to do it properly... and I felt I should use pneumatic's script on the Universal files first.

After I finish The Pneumatic Bilinear Edition, I'll use Avisynth+ to use bilinear scaling on the Turbine files. It'll make for a neat comparison.

Roku in the u.s. and Peacock (subscripition) has Seaquest, which was produced by Amblin/Spielberg, and got the blu-ray treatment as a result.  It's all in HD, with a true re-scan.  I'm watching off streaming but I am quite surprised how favorably the pilot of Sliders with topaz compares to it.

11% seems substantial and who knows - that may come through in the final results.  The currrent topaz pilot looks incredible to me but i'd say up close, right in front of the tv, it's clearer it's not HD. I wonder if you'd get more out of the lines of resolution now.

Yep that's right, except it's using BWDIF for deinterlacing and then applying QTGMC "repair mode" to all frames. 

If we use QTGMC for deinterlacing it looks bad cause QTGMC deint frames are so visually different in terms of the shapes of object outlines that it's noticeable when a QTGMC deint frame is occasionally substituted in.  Kind of looks like object outlines are popping slightly larger.     

I've been busy working on a script with presets to choose things like whether you want to output 24p/60p, whether to apply QTGMC repairing to deint frames or all frames (in case you don't want it suppressing film grain all the time), specifying how much of the source contains video camera content like the Eggheads episode so it uses lower combing detection thresholds for those scenes, and some other things.

@ireactions -- that's interesting that bilinear may be better for some vs. bicubic for others.  Obviously, you are still evaluating but as we've talked about in this thread before, it has always felt like there's some lack of consistency between episodes -- and it was never clear if it had to do with how they were  shot and just the characteristics of the filmmaking, or some post production process.  I mean, some episodes just seemed really awful in the old VHS looking style but I never understood if that was just because of lighting and cinematography, or more at play.

The fact that you have had such good results wiht the bilinear does feel like a breakthrough.  Quite exciting actually.

If you are using a video card for processing now, I would think about doing a sample of gaia vs. artemis for anything with topaz.  I don't know if the processing time will still be multiple times what artemsis would be (and if so, forget it), but in general, at  least with the pilot, and perhaps something else, there may be enough "good" source data in there for gaia to take advantage.  It seems to be designed for content that does have enough foundation of that data.  A limited 30 second  clip may tell you.

I have some questions for pneumatic because I still don't fully understand his learnings and some of the underlying technologies.

What exactly is the difference between deinterlacing interpolation and deinterlacing weaving?
You've specified that QTGMC doesn't weave but instead interpolates.

My (shaky) understanding is that interpolation is when the deinterlacer takes an even or odd field that's only 50 percent of the image (stretched to fill 100 percent of the frame) -- and takes a guess at what's in the missing 50 percent of the of the frame based on analyzing previous and subsequent frames as well as the existing information in the current frame.

My (shaky) understanding is that weaving is when the deinterlacer combines the even and odd fields and then doubles that frame in order to meet the video's existing framerate (or the framerate would otherwise be cut in half).

My (shaky) understanding is that interpolation can create image degradation if the missing data is filled in with bad estimations, and bad estimations happen when the current and surrounding frames don't have helpful information for the algorithm.

My (shaky) understanding is that weaving creates motion artifacts when the weaved frames of two fields don't line up precisely due to movement in the scene.

My (shaky) understanding is that most modern blu-ray players will use progressive scan on a DVD, which combines two fields into one frame and then duplicates that frame by three or two for an alternating 3-2 pattern that reduces the motion artifacts and provides a full image without the guesswork of interpolation.

My (shaky) understanding is that when dealing with video like SLIDERS 1.02 - 1.09 where the fields were scaled separately and misaligned, progressive scan produces all ugly visual anomalies on the Universal discs (as does weaving). Meanwhile, interpolation to remove those artifacts depends on analyzing current and surrounding frame-fields to fill in the gaps; since the fields are misaligned, that analysis yields poor data and image degradation.

My (shaky) understanding is that QTGMC's motion analysis yields better data to fill in those gaps which is why the pneumatic scripts have yielded better results than handbrake or progressive scan players.

It's a shaky understanding that I'm hoping pneumatic can correct when time permits.

Interpolation is really just a synonym for upscaling - taking some samples like pixels and making a guess at what the pixels in between them might be based on the surrounding pixels.

The simplest form of deinterlacing is Bob deinterlacing which just takes each 720x240 field (60 of these per second in a 30fps interlaced file) and upscales them to 720x480, typically using Bicubic or Bilinear resizing in the vertical direction.  If you want to see what it looks like type Bob() in Avisynth.  The result is soft and flickery, but frame rate is perfect and combing artefacts are impossible.

QTGMC deinterlacing is like Bob deinterlacing but NNEDI3 upscaling each field from 720x240 to 720x480, followed by some advanced techniques to smooth and antialias the image.  I believe it looks at the previous/next field as part of this process.  But it never weaves fields together - a test pattern of alternating 1px white/black horizontal lines always becomes solid white/black in each output frame, just like with Bob.  I would characterise QTGMC deinterlacing as the best possible version of Bob deinterlacing.  It tremendously repairs the field misalignment issue in the Mill Creek release.

QTGMC "repair" mode doesn't deinterlace at all, it just looks at progressive frames and applies its antialising magic to it.  Film grain is slightly suppressed but the 1px black/white lines are fully resolved.  Field misalignment issue is not repaired. 

It depends what the cadence is.  1:1 cadence is where every field is a temporally unique image captured from a new moment in time, like the Mindgame scenes in Eggheads.  Sports, news, talk shows and soap operas tend to use it.  If we weave every pair of 2 fields together on a 1:1 sequence the resulting image is combed:

top field        A-C-E-G-I-K-M-O-Q-S
bottom field     B-D-F-H-J-L-N-P-R-T
combed?          Y Y Y Y Y Y Y Y Y Y

(letters indicate temporally unique image)

To avoid combing on 1:1 we can use Bob, but a better solution is BWDIF (bob-weave deinterlace filter) which uses Bob deinterlacing for moving parts of the image and weaves any static parts.  Result is 480p60 same as Bob.  From memory this method of deinterlacing became mainstream around the late 2000's and was touted as "per pixel deinterlacing".

If the cadence is 2:2 (every 2 fields = a temporally unique image) like the spinning earths in the intro, weaving them results in 480p30:

cadence          2 2 2 2 2 2 2 2 2 2
top field        A-B-C-D-E-F-G-H-I-J
bottom field     A-B-C-D-E-F-G-H-I-J
combed?          N N N N N N N N N N

If the cadence is 3:2 (99% of the show) we get:

cadence            3 2 3 2   3 2 3 2
top field        A-A-B-C-D-E-E-F-G-H
bottom field     A-B-C-C-D-E-F-G-G-H
combed?          N Y Y N N N Y Y N N

TFM field matching applied to the above results in:

cadence            4 2 2 2   4 2 2 2
top field        A-A-B-C-D-E-E-F-G-H
bottom field     A-A-B-C-D-E-E-F-G-H
combed?          N N N N N N N N N N
duplicate?         Y         Y 

TDecimate can then remove the 1 in 5 duplicates to produce 480p24.

So we have a mix of 480p24, 480p30 and 480p60 coming out of the original 480i30 container.  If your goal is to preserve these original rates, they would fit best inside a 480p60 container, which is what I'm currently working on in Avisynth.  99% of the show is 3:2 though, so most people would probably be fine with putting it all inside a 480p24 container and tolerating the odd bit of stutter/combing on 2:2 and 1:1 sequences.

Yes, although I think the field misalignment issue is not too severe on the Universal NTSC discs.

Mill Creek.

Which script was it - this one?  Because that one doesn't repair the field alignment - only QTGMC in deinterlacing mode does that (InputType=1 is repair mode, InputType=0 is deinterlacing mode).

Maybe the improvement you're seeing is the antialiasing provided by QTGMC repair mode?  It does work wonders at times.

Very interesting reactions.  The previous topaz pilot as you know I think is stellar but it definitely had smoother textures so it's interesting to hear you say you are getting more detail.  Although maybe that's partly the extra resolution rather than upscale method. 11% is a lot.

I think you tried double upscaling (lanzos/topaz?) in the past and it was a dud.  I wonder if there is any possibility of gains here since you mentioned more detail in the image (presumably for topaz to work off of).  But maybe it won't be true grain info to work off of. Maybe a straight topaz would be better than nnedi3 to topaz (though possibly not better than nnedi3 itself).

Thanks to a forum member for lending me their copy of Turbine BD, here are some comparisons of NTSC DVD vs PAL DVD vs Turbine BD.

All are remuxes from the original discs, all were inverse telecined with TFM to 576p then Bicubic upscaled to 1080p in Photoshop (otherwise the screenshots are tiny).

101 [NTSC DVD] [PAL DVD] [Turbine BD]
104 [NTSC DVD] [PAL DVD] [Turbine BD]
206 [NTSC DVD] [PAL DVD] [Turbine BD]
305 [NTSC DVD] [PAL DVD] [Turbine BD]
405 [NTSC DVD] [PAL DVD] [Turbine BD]

According to MediaInfo the Turbine BD is MPEG2 576i25 @ ~5mpbs, the same ancient technology as the PAL DVD (MPEG2 released in 1996).

imo the NTSC DVD is still the best and the PAL DVD is perhaps slightly better than the Turbine BD.

This lensdump site doesn't work for me. I've tried three different browsers and the images don't load.

Turbine blu-ray for S2 to S5 episodes look significantly better in my living room than the NTSC DVDs, but if pneumatic says Turbine is actually poorer, then I wonder if my living room TV/bluray scaler is somehow working well with the blu-ray but not working on my DVDs. I also have to wonder if playing the raw MKV files on my Android TV without any inverse telecine is having some effect on our differing results.

But thinking about it, I now suspect my comparison was flawed because I compared the raw MKV blu-ray files to Handbrake processed DVD files (which has a haze-inducing detelecine process and reduces the image resolution from 540 to 480). There's also perceptual bias on my part due to my frustration with Universal.

Why don't the Turbine files have all the jagged edges and flicker of the NTSC version?

In other news, I've looked over the bicubic and bilinear versions of The Pneumatic Edition and the bilinear version has a little more edge-pixel contrast that I prefer for these fuzzy videotape files. But I have to be honest, with pneumatic informing us that the Turbine blu-ray is actually poorer than the DVDs, I'm not really filled with confidence in my assessments anymore. I'm going to take a GPU-powered nnedi3 run at these files one last time.

Well I'm a bit confused too, because I seem to recall evaluating the blu-ray and thinking later production episodes were better. I was using a junky blu-ray player too.  I have to look back in this thread to find the comments, just to make sure I was comparing to dvd files.

Another relevant comment, perhaps somewhat related

I have no doubt that pneumatic is correct to determine that Turbine uses poorer files than the Universal set. Why have RussianCabbie and I gotten better results from Turbine in many cases?

I have wondered about bias, but given that RussianCabbie too got better results from Turbine disc playback, I wonder if the blu-ray players may be the factor.

RussianCabbie and I watching the Turbine discs and Universal discs on our blu-ray players would have been seeing the files played with progressive scan to convert 25fps (Turbine) or 29.97fps (Universal) to 24fps TV display. Progressive scan combines separate fields to form full progressive frames and duplicates or drops frames as needed for playback.

Is it possible that the progressive scan combining separate field-frames will create some image degradation from minor field misalignments? Is it possible that a 25fps file, with fewer frames for progressive scan combination, with fewer misalignments, then has less degradation? Could that be why RussianCabbie found Turbine's "Rules of the Game" clearer than Universal's? (I don't know.)

pneumatic used TFM inverse telecine to progress the files for screenshots. As I understand it, TFM engages in field matching to recover the original progressive frames, in contrast to progressive scan simply combining two fields into one frame. Field matching frame recovery shows that the Universal files have more image data than the Turbine files.

So what if the lack of field matching frame recovery on our old blu-ray players is what tripped up Universal even as Turbine looked good?

Is it possible that in terms of blu-ray player playback, the Turbine files are better suited to progressive scan and upconverting than the Universal files?

There's no doubt that my previous process of converting the DVD files -- MakeMKV followed by Handbrake detelecining -- was producing fuzzier files at 640x480 that were inferior to pneumatic using MakeMKV and inverse telecine and outputting files at 720x540. And that certainly skewed my comparison.

But RussianCabbie also found the Turbine discs had a better image than Universal under his playback conditions. I have to wonder if PAL 25fps is just easier for our blu-ray players than NTSC 29.97fps.

However, my understanding of the difference between progressive scan and inverse telecine is very uncertain; I think it's that progressive scan combine field-frames while inverse telecine drops unnecessary fields and rebuilds progressive frames. That could be wrong.

Yes!   My Sony DVD player for instance, when playing a PAL 2:2 cadence DVD (pretty much all PAL DVDs) it inverse telecines (weaves) to 576p.   But with NTSC 3:2 cadence DVDs like Sliders it just deinterlaces using BWDIF style deinterlacing (weaving only static elements, but anything that moves drops to 240p).  So I think that could explain the difference you were seeing.

Also for NTSC DVD's there are ones that are authored as "soft telecine" where the video is stored as 24p on the disc and there are "repeat field" flags in the video stream telling the DVD player when to add fields to make a 3:2 pattern for 480i output.   On these discs my Sony DVD player appears to ignore the repeat flags and just outputs 24p.  But Sliders DVD uses "hard telecine" where it's stored as 30i on the disc with the 3:2 pattern baked in, and the only way to recover 24p is to use inverse telecine (eg. TFM) which my Sony DVD player doesn't appear to support, so it just deinterlaces every frame BWDIF style.

I'm having the Lensdump issue as well - seems to be a problem with their site.  Disabling my adblocker seemed to fix it but now it's playing up again.  The reason I use Lensdump is cause they don't compress the images.   I'd imagine they are aware of the issue and will fix it soon.

@pneumatic -- right click the broken image and then click "Load image" on the browser menu.

Looking at those shots, the best two shots

pneumatic TFM + TDecimate + QTGMC output to 1080p via nnedi3

pneumatic TFM + TDecimate + QTGMC (540p) with Topaz AI Artemis upscaling to 1080p.

This one is somewhat interesting, but I don't have a definitive conclusion:
pneumatic TFM + TDecimate + QTGMC + nnedi3 (540p) with Topaz AI Gaia upscaling (1080p).

It gives it an interesting, almost intriguing texture but I'm not sure if it'd come across as additive or distracting in video form?  I am kinda surprised you don't maybe enjoy it more because it does feel like there's a grain / exposure to it.

The sharpened one is bad compared to the others.

What's interesting is how much each version "colors" the content in terms of I feel like it would have a real impact on the viewing experience and the vibe.   This is probably why hollywood has people who pay so much attention to a film's "finish" in the post production process.  It almost frames the story, or serves as a reference to the viewer on the voice of the storyteller.

I guess I have to run the full Topaz Artemis job now for the eventual fan convention where I will supply the digital files for the screening but pneumatic will introduce it.

I like grain if it's part of the texture of the film image or if it's filling in for detail in a video that's lacking in definition. But I think grain can easily become a needless distraction. The TV show CHUCK is horrifically grainy to the point where it looks like static. The 4K77 fan reconstruction of a STAR WARS theatrical print is so obnoxiously grainy that I had to run it through Topaz AI to tone it down to enjoy it. And the grain in the Gaia upscales is just ugly and oversharpened.

**

I have run pneumatic's script on 1.02 - 1.09 to output 1080p files with bilinear, with nnedi3. After painstaking analysis, careful review, a 12 point system of screenshot comparison and a live viewing on four different TVs, I have come to the conclusion that I have no idea what difference choosing either scaler would actually make.

These are blurry videotape files. There comes a point when we might just conclude that no scaler is going to turn these analog recordings into the crisp renderings we want. We have to settle for below average files that at least have been shorn of those jagged edges and flashing horizontal lines.

Until pneumatic comes up with a new script or until we reshoot all the episodes with stand-ins and deepfake the actors' faces onto our photo doubles, we've probably come as far as we can. nnedi3 was the most powerful scaler pneumatic encouraged me to have my gaming laptop use, so I'll accept that this is what we've got. The Pneumatic NNEDI3 Edition.

There's certainly a cooler, bluer look to the LOIS AND CLARK HD rescan with a higher brightness level. And all the effects are just the videotape sequences that have been stretched, so any time there's any kind of video effect, it goes from HD to terrible looking SD. I only saw the first season, but it looks like they didn't even try to upscale the SD footage.

**

I thought the pneumatic Avisynth+ script had dealt with the jagged edges in this Pilot shot when reviewing a 540p version, but looking at it in 1080p output, the jagged edges are still there:

How would pneumatic address it?

Ah, thanks. I'm now letting RussianCabbie make remote use of my gaming laptop to upscale his pneumatic inverse telecined 540p Pilot to 1080p via Topaz Artemis. I myself will stick to the pneumatic nnedi3 edition.

**

The thing is: I remember at least Season 1 of LOIS AND CLARK on DVD perpetually having a mildly fuzzier video image whenever special effects were present. If Clark were using superspeed, firing heat vision or flying, the image would get a little softer. It's like the effects in Season 1 were being done on a different video editing suite from non-effects sequences. The drop in video quality for effects shots is in the original SD DVD release, and in the HD release, the small gap between effects and non-effects shots has become a massive gulf.

The S1 effects shots were actually so fuzzy that I'm not sure they can be upscaled fully. And I too would choose SD DVDs over the HD release for this show. Yes, SD isn't as sharp, but at least it's consistent instead of shifting from HD to SD every time Superman uses his superpowers.

I think the unwillingness to redo the effects shots is because the STAR TREK: THE NEXT GENERATION HD blu-ray was a sales failure. It was a high budget remastering with rebuilt effects that came out just in time for streaming to wipe out the home video market. Paramount and CBS lost a ton of money. Since then, HD releases of SD shows have sought to cut costs by simply rescanning the original film but using the master tapes for any effects footage rather than recreate those effects.

BABYLON 5 had the same treatment as LOIS AND CLARK, but BABYLON 5's effects always looked like Playstation 2 graphics on original broadcast, and those untextured computer models actually upscale to 1080p pretty well.

STARGATE SG-1's non-digitally shot seasons were upscaled from SD to HD for HD home video release and... while it actually looks okay, it has an overly noise-reduced image that borders on blurry. The distributor called that a feature of their 'Super Clean Image' feature. And at least its effects and non-effects scenes have the same video quality throughout.

With the L&C Season 1 effects shots, I'm not sure any upscale was even attempted on those fuzzy videotape sequences. They may have been too poor to upscale.

I had a quick look just now at the HD version of Season 2 of LOIS AND CLARK and the Superman sequences look a bit better than in Season 1, still not full HD, but upscaled to HD approximations. It looks like the master tapes from Season 2 onward had the effects shots at a better SD video quality that could be raised to a better HD approximation. It matches closely enough for living room viewing, although it's certainly less-than-ideal.

Personally, I think it's best to be consistent and it's better to be SD throughout than an awkward blend of SD and HD. It's very jarring when HD footage gives way to SD effects. For an HD Season 1 of LOIS AND CLARK, I'd probably advise going the STARGATE SG-1 route of an SD upscale... albeit with nnedi3 rather than whatever they were using.