Sliders DVD Releases (Universal, Mill Creek, SD blu-ray, Restoration) (Page 7 of 16)

So does that mean that THE DEAD ZONE content for S1-3 looks better after topaz processing than S4-6?

I assume on the dvds itself, the 1-3 don't look as good at 4-6?  Or maybe that's a draw.

With THE DEAD ZONE: Seasons 1 - 3 after Topaz upscaling look absolutely perfect. 10 out of 10. They look like a digital film scan.
It's because Seasons 1 - 3 are grain based images from 16mm film, edited on high definition digital videotape and downscaled for broadcast and DVD and with all the film grain still present and available for an upscale.

However, THE DEAD ZONE's Seasons 4 - 6 after an AI upscale look like an upscaled DVD. It is an amazing upscale and no blu-ray player could do this sort of upscale during playback because all the compression artifacts are gone and the resolution has been increased through smooth pixel blending specific to the image textures to ensure no new artifacts or imperfections. But fine details like pores and grass have a slight blur to them. It's because these seasons are pixel based images from digital cameras, edited digitally, then downscaled for broadcast and DVD.

The DVDs themselves look pretty much the same across all six seasons of this show in terms of video quality and characteristics (grain, sharpness, clarity, detail). There's no weird discrepancy like Season 1 of SLIDERS being denoised and blurry and Season 2 being sharp and grainy. THE DEAD ZONE made some effort to retain its visual identity even though its cameras and filming location changed.

However, THE DEAD ZONE's look certainly shifted with digital cameras, but where SLIDERS benefitted greatly from the digital videotape format, THE DEAD ZONE suffered a little for the transition into a fully digital format. The first three seasons of DEAD ZONE benefitted from film because for film, productions light everything with thought as to how that light will show up through the film chemical process.

In contrast, digital videography captures video 'as-is,' and lighting and colour are often adjusted afterwards. There is less thought put into lighting a scene as it's adjustable in post, there are also artistic limitations. Seasons 1 - 3 were very artful and deliberate in lighting. Johnny Smith's house in THE DEAD ZONE had, in Seasons 1 - 3, a cozy warmth with a sense of amber comfort where every room seemed to be lit like it had a fireplace even if it didn't. From Seasons 4 - 6, Johnny's house just becomes 'dark.' It could be fairly said that THE DEAD ZONE's later seasons' use of digital cameras just make it look cheaper and shabbier much like SLIDERS in Seasons 4 - 5 not really bothering to use colour or its absence for storytelling and just looking beige.

Nothing to do with video quality itself, of course.

Bah! Medium quality has been a disaster on every episode aside from "Fever," it all looks like watercolour animation. I give up! Life moves on! I want 1080p versions of all the masterpiece episodes of THE DEAD ZONE!

I've been rewriting the first post of this thread, updating it with a summary of the video quality of each DVD set.

I've come to realize something: the Universal DVDs are not bad. I've said repeatedly that they are, but they aren't. Yes, DVD compression and interlacing can diminish video quality, but upon further review and comparison, the Universal SLIDERS DVDs don't have particularly small file sizes (1.6 - 1.8 GB per episode). The non-Pilot Season 1 episodes look terrible because the video masters are blurry, because they were film transferred to analog low resolution videotape. The DVD compression and interlacing don't do it any favours, but lifting off the compression and decombing have only further revealed the poor quality of the image underneath.

Seasons 2 - 5 on the Universal DVD sets look good. DVD compression and interlacing have marred them, but the underlying video image is sharp, detailed and defined; the film to digital videotape transfer has maintained a scaled down version of the film's crisp detail. An upscaling blu-ray player and/or HDTV upscaling can make them presentable; AI upscaling can make them look like impressive HD approximations.

Mill Creek, however, is bad; Season 1 looks terrible, blurry and artifacted from overcompression, Seasons 2 - 3 are hovering around poor to okay, looking fuzzier than the Universal versions and covered in blockiness and noise. Seasons 4 - 5 look average; the 16mm film image is grainy to begin with and a decent blu-ray player and/or HDTV upscaler can smooth out Mill Creek's damage while AI upscaling can remove it completely.

I think Season 1 looking so bad on Universal just made me feel biased against the rest of the Universal discs, but I've come to realize that SLIDERS' video quality on the Universal DVDs is as good as it can be within the limitations of DVD and the masters.

Looking at the discs through MakeMKV.

Universal: 1.6GB - 1.8GB per episode

Mill Creek: Ranging from 1GB - 1.2GB per episode

Turbine: 1.7GB - 1.8GB per episode

Why do Turbine's SD files look clearer than SD DVD (if you can access them directly off the disc)? I think it's likely that the DVD authoring techniques of 2004 and the DVD format itself created the image degradation.

Turbine likely used more modern-to-2015 video codecs to create the blu-ray files from the PAL masters.

I was trying to watch the ArtemisHQ upscaled version of "Prince of Wails" on my TV (Hisense 55H7608) via my AndroidTV box earlier today. I ended up turning it off and deleting all of my upscaled Season 1 files.

The issue I have with it: the closeups look really good, but all medium and wide shots have a distractingly inconsistent effect where the AI is trying to sharpen the outlines of all people and objects and creating jagged edges that flicker into blurriness and back into jagged edges. It's because the algorithm is seizing on what little grain there is to upscale and it doesn't have enough to work with.

Out of interest: I tried watching the DVDs and blu-ray of this episode with an upscaling blu-ray player. I started with the Mill Creek disc of "Prince of Wails" and watched Prince Harold and Wade's scenes. This is the first time I've watched SLIDERS on an HDTV with an actual blu-ray disc player (a Samsung BD-J4500R). Before, I had just a PlayStation 3. While the PlayStation 3 was DVD and blu-ray capable, its DVD 'upscaling' was little more than deinterlacing and scaling the image (like any SD DVD player) and applying various noise filters that blurred the image. And my TV is just a budget 55 inch with zero upscaling features.

The Samsung player is a bargain basement budget model, but it still extends some effort to upscale DVDs to 1080p during playback. Comparing the output and what I've seen on the raw disc files, the Samsung player employs progressive scan on DVD (an interlaced format) to blend the odd and even fields more smoothly than standard deinterlacing. The player also adds a moderate amount of pixel contrast to the edges of images to sharpen up the image for 1080p. It deblocks the video to reduce the DVD compression. It mildly tones down jagged edges that result from the scaling. It leaves film grain intact.

On the Mill Creek disc, all this is useless. The picture is blurry with no detail to sharpen. The colour is shockingly washed out due to Mill Creek compressing the file too much; the vivid blue of the boiler room where Prince Harold is locked up is instead a dull grey. Wade's purple sweater with its texture of squares looks smoothed out in the medium shots.

I switched to the Universal DVD of "Prince of Wails" and found the colours really strong: the blue of the boiler room and Quinn's sweater, Wade's purple top -- it all came through along with the texture of the clothes. My blu-ray player did a good job of scaling the image to look presentable on HDTV, sharpening edge details but leaving surfaces within any outlines alone. There remained a bit of haziness to the image, but unlike my AI upscale of "Prince of Wails," it was consistent throughout and therefore not distracting. Despite the deblocking, there was still quite a bit of noise from the DVD compression, but the blu-ray player toned it down a touch in the upscale. My blu-ray player upscaled the Universal DVD to look presentable on an HDTV.

Out of curiosity, I put on the SD Turbine blu-ray of "Prince of Wails" and found the video quality to look pretty much the same as the Universal DVD version except there were none of the DVD compression artifacts. This is very odd. I know from playing the raw file on the SD blu-ray on my comptuer: Season 1 episodes are blurrier than the Universal DVD versions and have washed out colours. But that blur and desaturation isn't showing on my HDTV.

My blu-ray player's upscaler has an automatic setting for "Deep Colour - HDMI" that may be amending that flaw by increasing saturation for blu-ray disc playback. And my player is likely blurring the Universal DVD a little in lifting off some (but not all) of the DVD compression. The SD blu-ray file has a blurriness that is native to the file. End result: my blu-ray player's upscaling improves both the Universal DVD and the Turbine blu-ray for Season 1 episodes to the point where both versions look presentable on an HDTV and both look about the same.

It's interesting. I said awhile that SLIDERS DVDs were no longer something I wanted to watch on my new 55 inch screen TV, but my blu-ray player seems to be able to improve the Universal DVDs and the Turbine blu-ray's issues. I guess my recommendation is to get a blu-ray player with a descent upscaler.

Another peculiar (and strictly anecdotal) situation: the Turbine blu-ray files offer superior results on my Android TV. I don't know why this is.

I tried playing the raw Universal DVD files and the raw Turbine blu-ray files of "Luck of the Draw" on my Android TV box (hooked up to my HDTV) with the MX Player app.

My Android TV box did something strange: it oversaturated all the colours, making Quinn and Arturo look weirdly orange. Everything was too rich. There seemed to be some oversharpening as well: the raw DVD file looked blocky, jagged and the amped up colour accentuated the DVD compression and artifacts. It looked hideous.

I tried playing the file in VLC and got the same results. It must be something within the MiBox 3's SoC and decoding settings that's adding increased saturation. I've never noticed oversaturation with my other video files on my MiBox 3. Maybe they benefit from it and Season 1 of SLIDERS doesn't.

I have no idea why it did this; my HDTV has the same colour profile for the blu-ray player. I played the raw file on my blu-ray player and my blu-ray player didn't increase the colour or sharpen the video so unflatteringly.

I played the Turbine version of "Luck of the Draw" on my MiBox 3 -- the blu-ray file as MKV. The colour saturation was increased here as well, but because the raw file was discoloured, the increase made it look normal. In addition, there seemed to be some mild sharpening and the raw blu-ray file looked okay -- certainly not HD or even at the level of the post-Season 1 episodes, but adequate and acceptable as an upscaled image on an HDTV. It looked as good as the upscaled Universal DVD on my blu-ray player.

It's very strange. Speaking only for the equipment I have here: for Android TV playback, the Turbine versions turn out best. For upscaled DVD playback, the Universal versions are best.

One thing to also remember is how different Goodfellas looked for you vs. me.  I had what seemed to be a lot of pixelation (compression artififacts?).  I know it's S2, and you're talking about S1 above but the image looked terrible.  That said, I mostly found S2 episodes better on turbine than universal.

I'll re-try the S1 on an xbox blu-ray player which has a lot of upscaling techto take a second look, but iirc it will still look pretty bad.  The turbine release looked pretty bad (maybe  the PAL messed with the upscaling tech).

FWIW, I found out the MillCreek Release Made about 200k domestically, in revenue (not talking profit here).  I don't now, maybe we can convince Universal to spend 100-150k upscaling S1 if they felt it could make the money back in a blu-ray release and enhance peacock's offering.  But they may expect Peacock to pony up some cash for that and I doubt they would (maybe 20k at most).

It's a tricky thing because everybody's equipment is different, plus the post-processing in real-time isn't totally a reflection of the disk.  Although all blu-ray players and tvs to a certain degree do upscaling, as you said.  So I guess it's OK to factor that in.  But then in your case you said S1 looked better because the upscaling didn't over do it.

The Lanczos algorithm turned out pretty well for taking the Topaz Artemis HQ deblocked-denoised 480p file of "Summer of Love" and increasing the resolution to 1080p (although it took a 1.8GB file and made it 5.5GB). The Lanczos 1080p video file looks a little better than my blu-ray player or Android TV bicubically scaling the 480p file to 1080p; the added pixels are smoothly integrated into the image to avoid jagged edges.

My upscaling blu-ray player, when playing the disc, toned down the DVD compression artifacts, and bicubically stretched it to 1080p, creating some pixelation at the edges of any objects or people. Topaz AI has been able to remove remove that compression from the DVD file completely. Lanczos has taken that denoised, deblocked file and smoothly stretched it to 1080p without creating any pixelation.

The cleaned-up/Lanczos stretched version doesn't look HD, but the low quality SD flaws of compression and pixelation have been removed. The AI issues of jagged, flickering edges and watercolour distortion are avoided. Admittedly, the underlying image remains a 250 line analog videotape instead of a 540 line digital videotape, but the artifacts and pixelation issues were a massive distraction from the story. Now they're gone.

I did have to turn off all Lanczos sharpening functions.

Lanczos on my old home theatre PC was fine for upscaling early 2000s standard definition TV because all those shows were shot on film or digital HD and simply downscaled to SD, so Lanczos was merely restoring the sharpness inherent to the file. SLIDERS in Season 1, unfortunately, is a pretty fuzzy image devoid of grain detail, so there isn't much for a sharpening filter to build upon.

Most blu-ray players and TVs will scale an SD video to HD with a bicubic algorithm that increases pixel contrast at any edges within the image in adding more pixels. Lanczos blends the added pixels into the image more smoothly, and is a more forgiving scaler, but because it's CPU-demanding, it's more than what a blu-ray player or Android TV box may be able to offer.

My old HTPC used an i5 processor and a dedicated GPU to make it happen. Sadly, I don't have it any more after a power surge. In most use cases, Lanczos is best applied when encoding a video so that the gains are contained within the file itself.

The AI deblock-denoise takes about seven hours per episode and Avidemux using Lanczos in stretching the video to 1080p takes about 90 minutes per episode, at least on my hardware. I wonder how much that would factor into a future upscaling effort from a home video distributor or streaming service.

Hmm. Lanczos proved effective for closeup and medium shots, bringing them from 480p to 1080p without degradation (and the videotape image is already pretty degraded). However, for wide shots where the actors are standing at a distance and there was no real detail in the standard definition image -- well, Lanczos can't seem to figure out what colour pixels to add and what blending and blurring is needed to make it smooth. As a result, those shots have jagged edges that are really jarring when all the closeup and medium shots look okay.

The 480 version, when played at 1080p, had a certain layer of haziness from the bicubic scaling that obscured the problems with the wide shots. Lanczos, however, makes those problems really apparent, so I think the best thing is to either let the HDTV scale the image or create the 1080p file bicubically after all.

But not everything needs to be a 1080p file, especially when the file itself is not particularly improved by being encoded at 1080p.

I might just leave the Universal files as Topaz-cleaned 480p SD files. I mean, for Seasons 2 - 5, my best versions are just the plain Turbine 576p SD files that scale to an HDTV just fine and I won't be boosting those any further.

Topaz Artemis HQ for removing DVD compression (noise, artifacts), but leaving it at a 480P file.

Close-up and medium shots from this denoised, deartifacted version are effectively upscaled to 1080p via Lanczos.

Wide shots are best upscaled bicubically.

I don't see why you couldn't have a Lanczos-created 1080p file and a bicubically stretched 1080p file, made from a Topaz-cleaned version. And then assemble the best versions of each shot from both.

I just... don't have time for that. I need to go teach my mother how to use a Google Nest and an Instant Pot. And I need caulk for my windows.

It's interesting: I was reading that the blu-ray and HD streaming re-release of THE X-FILES (Seasons 1 - 9), despite being composed of rescanned film, has many shots and entire scenes that are unfortunately fuzzy SD footage stretched to HD resolution for at least Seasons 1 - 2. Some are establishing shots, some are special effects -- but some are entire scenes that production would have shot on 35mm film. Why are these in SD videotape for an HD release? Why weren't they at least upscaled?

I imagine FOX must have lost multiple reels of film and could not fully reassemble every episode from film. Also, the effects and any stock footage are upscaled SD videotape. THE X-FILES' first two seasons were 1994 and 1995, so both would have, like SLIDERS, been film transferred to analog videotape for editing and effects. For the missing sequences, the HD versions of THE X-FILES would use SD footage that, like SLIDERS in Season 1, lacks the lines of resolution and/or film grain needed for an effective upscale.

Are there missing film sequences for Seasons 3 - 9? Possibly. But I imagine that by Season 3, THE X-FILES would have switched from analog to digital videotape and any missing sequences (as well as effects and stock footage) could be upscaled effectively, if not via AI, then through Lanczos and spline algorithms.

**

Back to SLIDERS: Just to see, I encoded the Universal-disc copied, Topaz deblocked / denoised "Prince of Wails"  from 480p to 1080p with bicubic scaling. Then I played the 1080p file on my Android TV at 55 inches and watched the first 10 minutes. Then I played the 480p file on my Android TV and watched the first 10 minutes. I couldn't tell the difference. The cleaned up DVD file at 480p looks a bit hazy, but without the compression artifacts getting in the way.

The Lanczos version looked even clearer for close-up and medium shots, but wide shots looked jagged and ugly because the SD image had no detail for effective pixel blending. The bicubic version hides these issues with a pixelated fuzziness, but that fuzziness is present on all the shots, so if you get used to it after a few minutes, it isn't jarring.

I think I'll just leave the files at 480p; there's no point inflating the file sizes from 1.5GB or so to 3 - 5GB if the video output results are the same.

In terms of process: the only reason to run Topaz AI over these fuzzy Season 1 episodes is to remove DVD compression. However, I don't think the original NTSC master tapes would suffer from DVD compression; the blurry quality is due to 250 line analog videotape and the PAL masters are blurrier due to being copied via an analog process and stretched from 480 pixels high to 576.

I don't believe there's any gain to running AI processes over non-DVD compressed masters; they aren't effective on a 250 line videotape image. If they have to be put into a 1080p video file, they could be scaled bicubically. The Lanczos version is a little nicer than the bicubic version for medium and close shots -- but really not by much.

The pilot and Seasons 2 - 5 can be dragged and dropped into Topaz and upscaled via Artemis HQ to 1080p. They have sufficient film grain in the digital videotape for the AI to render the grain detail into pixel detail at a higher resolution.

Now this is interesting. "Prince of Wails," scaled to 1080p via Lanczos, had really messed up wide shots which looked like broken stained glass due to the algorithm being unable to blend in the new pixels effectively. But "Summer of Love," upscaled to 1080p via Lanczos, doesn't have the same problem; the wide shots look about the same as they would through the bicubic algorithm while the medium and close up shots look okay / forgivingly scaled from 480 to 1080.

This speaks to what RussianCabbie pointed out: the Season 1 episodes are in poor video quality, but even within being poor, there are degrees of visual failure with some episodes being worse than others. "Summer of Love"'s wide shots weren't as bad as "Prince of Wails" for pixel blending.

Avidemux only takes about an hour to encode each episode to 1080p bicubically or via Lanczos, so I'm going to set a queue to encode each of the low quality eight Season 1 episodes to 1080p twice, once bicubically, once in Lanczos. And then I can compare and see which one yields better results and whichever ones look good in Lanczos can be retained.

Just one other piece here.  I think it's really strange that SLIDERS, even on Mill Creek discs, looks not half bad at an old (small) tube tv from 12-15 feet away.  I continue to think the HD era (along with larger tvs) really exposed the flaws of the original image, which was fine for the general specs and usage at the time. 

It's very weird looking at SLIDERS at an ancient tv and thinking it looks better than on a modern one (although my tube one is only 12 inches or so, which plays a role here).  It's also helpful older tvs are natively 4:3.  I have noticed if you watch SLIDERS on a 16:9 in a dark room, the cut off sides aren't as annoying, because you really are only seeing the image area, and nothing else.

Loading the raw Universal DVD files into AVIDemux now to try upscaling them to Lanczos at which point I'll try running them through Topaz Artemis HQ to see if that creates any improvements. My soul can't rest without knowing.

interesting idea.  another variable might be that such an ordering of things might behave differently on different episodes (I know, headache).  I wonder if it would be more beneficial for ones that worked well with lanczos or more beneficial for ones that didn't (although in the latter case the end result may still not be good enough to justify using lanczos).

ha!  it's a rabbit hole.

I wonder what the characteristics exactly are that would cause some to benefit vs. some to not.  Maybe it's not even a issue of something in production or post production but rather the general nature of the content and the composition of the shots.

I watched a bit of "Prince of Wails" (Universal DVD, deinterlaced in Handbrake, upscaled to 1080p via Lanczos). The result looks good at 1080p; Lanczos deartifacted the image a little but left a layer of noise that looks like film grain. The grainy quality actually helps: the video image under all that grain is quite fuzzy and bare. I think I actually prefer this to all previous upscales of these episodes because it's the closest the low-res videotape image has come to looking like the original film. The blurry haziness is offset by the noise.

I did run a few shots of "Prince of Wails" through Topaz and -- I don't know if it's feasible because the AI needs 20 - 22 hours to denoise and deblock each 1080p scaled episode. Topaz did lift off the grain, but removing it just shows how dull and undetailed the image is without the grain.

The reduced compression texture might be best left there for these analog videotape Season 1 episodes. What's under the compression isn't very good anyway and would take almost a day per episode to unearth, at least on my hardware.

I've mostly been using Topaz Artemis HQ which is mostly about rendering film grain into pixel detail and removing compression artifacts. The medium and low quality Artemis settings are more aggressive with noise and compression artifact removal, but they create an overly airbrushed look.

The noise in the Lanczos upscaled files is compression noise, not image forming grain, so Topaz just removes it. It's not really worthwhile especially when Topaz needs 20 - 22 hours per episode. Usually, it only takes 7 - 9 hours to upscale a 480p image to 1080p in Topaz.

I watched two versions of "Prince of Wails" today: the version decompressed in Topaz and Lanczos scaled to 1080p, and the version scaled from the DVD file to 1080p via Lanczos on its own.

The Topaz-enhanced version is sharper in medium and closeup shots, but for wide shots, there's a painful fuzziness, blurriness and vagueness to the shots that is frustratingly jarring when the medium and closeup shots look so much better.

The Lanczos version is not as sharp in the medium and close up shots compared to the Topaz version. It's a little blurrier and the wide shots have the same lack of definition as the AI cleaned version. However, it looks better than just scaling the DVD image bicubically. Lanczos has softened the compression artifacts and now it looks like film grain. There's a sense of definition and texture. The wide shots with this noise look better; the noise partially fills in the lack of detail with some texture.

I prefer the Lanczos version because even though it isn't as sharp in the close up and medium shots, it's consistent: the haziness and noise are on all the shots, so you can get used to it as opposed to being distracted by the inconsistency of the Topaz version.

It doesn't look as good if you're watching it on a computer screen and studying the image closely, but it looks good when played on my Android TV.

I watched them both on my 55 inch HDTV, played on my Android TV box.

The Topaz enhanced version looks okay/good enough in mediums and closeups, but poor/terrible in wide shots. The Lanczos version looks okay in all shots, fuzzier and with a heavy level of what seems like film grain (but is really compression noise). The wide shots are still blurry and undetailed, but the film grain look makes it look consistent with the mediums and wides. It is consistent whereas the Topaz version keeps jumping back and forth between not-bad to terrible. Even if the episode is only looking okay, at least it looks okay from beginning to end. You get used to the visual quality and then you can get into the story.

Well, I took a seven minute clip of "Prince of Wails" and dropped it into Topaz for upscaling. It says it'll take nine hours so I guess we'll see if it's worth it.

Let me congratulate you for your herculean effort in playing with the next-generation tools available to folks to rescue older, outdated content that deserves the royal treatment.

You tested, you theorized, you tried 1000 times.  This is exactly how science works. 

The professor?  Well, he would be very proud!

very interesting

to me (just looking at a computer online)

this one looks way better:  https://mega.nz/file/PsgGnZDQ#c6oLZfbYA … PUiN8JonfQ

I'm doing this blind so i have no idea which is which

Well, they've been upscaled in two different ways and Topaz is a crowdpleaser for a variety of reasons.

Lanczos upscaling operates on a specific mathematical algorithm that I don't actually understand except to say that the extra pixels added to make the image bigger are blended smoothly into the final version. Where bicubic scaling can add pixelated edges to the larger image, Lanczos scaling is softer, more forgiving, and also has the effect of mildly diminishing the noise in a grainy image.

Topaz AI upscaling, however, looks to me like it's composed of multiple algorithms for a wide variety of textures: skin, hair, grass, pavement, metal, sky, clouds, wood, fabric -- with specific models applied to specific portions of each frame. (The ability to recognize the texture and apply the right model is the AI part.) The models are dependent on using film grain as a guide to reconstruct the image at a larger size and to re-render that grain as pixel detail. As a result, Topaz upscaling is sharper than Lanczos because Topaz is actively re-rendering the film grain as a rebuilt pixel-based image. Most grain-reduction methods operate by blurring the image, but Topaz converts that grain to sharpen the picture.

In contrast, Lanczos is simply stretching, but with its single-algorithm serving to make the added pixels (mostly) smooth. It isn't re-rendering film grain, so the grain stays intact, just slightly faded out. It offers a mild level of sharpening with added contrast at the edges, but mostly in a way that offsets any new blurriness from stretching.

Topaz is generally better for closeup and medium shots. However, for wide shots, there's a smudged, painted look because the film grain is only serving to render an undetailed, distant image and the AI tries to sharpen detail that isn't present. The same wide shots in Lanczos, however, maintain the original blurriness and graininess so that it doesn't look like a watercolour mishap. Season 1 episodes stored on low-res analog videotape also do better with Lanczos because there isn't any film grain to use for sharpening.

Because SLIDERS was composed largely of closeups and mediums for television, most people will prefer a Topaz AI upscale for the digital videotape episodes.

Something I noticed: the Lanczos versions of my upscales are very hard to shrink down from their 4 - 6GB file sizes. Compression reduces the video noise that's alleviating the visual issues.

I was also running my Topaz upscales of THE DEAD ZONE through Handbrake to reduce the file sizes a bit, taking them from 3 - 4 GB a file to about 1 - 2 GB a file. The video quality went from being a high quality film scan to a medium quality 1080p Netflix videostream and some of the film grain texture left intact by Topaz was screened out by Handbrake.

Out of curiosity, I ran the very grainy blu-ray Turbine version of "As Time Goes By" through Handbrake at the "Super HQ" setting -- and Handbrake blurred out almost half the film grain and there was a slight loss of video quality. It's interesting: Universal's DVD compression in the early 2000s seemed to add compression noise; Mill Creek compression in 2016 added noise.

However, modern video compression codecs seem to screen out video grain and noise in the process of crunching a video file into a smaller container. I ran "As Time Goes By" but with the file size set to "Production Standard" and this time, all the graininess (and sharpness) remained.

I have to wonder if my technique for processing the DVD files has been a problem with SLIDERS, episodes 102 - 109. While I copy the file off the discs, I have to run them through Handbrake to deinterlace them or the final file has horizontal lines flashing across the screen.

I ran Seasons 2 - 5 through this preset and saved the video file at 'Super HQ' and the upscales turned out fine, but those files were already pretty sharp and Topaz had plenty to work with for increasing the resolution. I ran THE DEAD ZONE files through this same preset and the files definitely lost some grain as well, but again, the underlying image was good enough to upscale.

I wonder if my deinterlacing has been blurring the image and removing film grain from the Season 1 episodes. For other video sources, it was fine. But Season 1 was already blurry and lacking film grain; it couldn't have afforded to lose anything. I'm going to re-copy the disc files again and review the raw files.

**

I'd actually like to try another experiment.

Originally, I was only upscaling SLIDERS episodes to 720p and got okay results with Season 1 and great results with Season 2 onward. Season 1 episodes looked decent in mediums and closeups but like watercolour paintings in the wide shots.

After that, I read a bunch of articles from the DOCTOR WHO restoration workers upscaling old 1970s DOCTOR WHO episodes from 650 line analog videotape to 1080p for blu-ray releases. Their preference was to aim for 1080p so that the upscaling would be controlled in terms of their work on the video file and not by the television, so I began outputting my SLIDERS upscales to 1080p.

With episodes 102 - 109 of SLIDERS, however, I think that's proven to be a mistake because SLIDERS analog videotape episodes are 250 line videotape, 38.5 per cent the resolution of a 1970s DOCTOR WHO episodes. I haven't been happy with the AI upscales, but that's probably because I've been stretching the 250 line videotape files too far. These are 480p files; increasing them by 125 per cent via AI is too much whereas increasing them by a modest 50 per cent via AI is more achievable.

Originally, I upscaled the episodes from 480p to 720p using Topaz Artemis LQ for maximum compression artifact removal and sharpening, but I think that was also too far; there isn't enough to sharpen and it looks like a watercolour in wide shots. Topaz Artemis HQ doesn't try to sharpen too much; it's assuming that underneath the compression, the video won't benefit from further sharpening which is correct for SLIDERS' episodes 102 - 109 because they're too blurry for sharpening.

One of my subsequent experiments was running 102 - 109 through Topaz but leaving it a 480p file that had been deartifacted and denoised and then Lanczos scaling this version to 1080p. The results were good but inconsistent: wide shots looked fragmented or blurry while mediums and closeups looked decent.

When running the raw DVD files through Lanczos without Topaz, the 1080p versions were not as sharp in mediums and closeups and had compression noise, but the video quality was consistent across all shots instead of going from clear to blurry. The noise actually helped as a sort of filler data cross the blurry shots.

I'm wondering if using Topaz AI's Artemis HQ to remove the compression raise the resolution by 50 per cent (without trying to sharpen what isn't there like the Artemis LQ version) will yield at least a modest improvement on the initial 480p to 720p upscale.

I'm wondering if then using bicubic scaling to raise the resolution by another 50 per cent to 1080p will also be effective.

It will definitely add a little pixelation back into the image, but that will hopefully make all the wides, mediums and closeups look more consistent, with the pixelation serving a similar function to the compression noise.

I'm going to try it with "Luck of the Draw" and "Prince of Wails" which, to me, had the worst looking wide shots after AI upscaling.

It's indeed very strange that Season 5 goes for anywhere from $80 - $300 USD on some websites for the DVD set. My guess is that sales were good for Seasons 1 - 2, average for Season 3 and so bad for Season 4 that Universal Home Media grudgingly put out Season 5 but in an extremely low quantity anticipating low sales.

That said -- if you get the low bitrate Mill Creek DVD set, the Season 4 - 5 episodes don't really suffer very much from the overcompression.

Seasons 4 - 5, due to the reduced Sci-Fi Channel budget, clearly made a switch from 35mm film to 16mm film which is cheaper due to being smaller. 16mm film has extremely large film grains compared to 35mm film, so those film grains remain large and present even when downscaled to standard definition digital videotape and compressed for Mill Creek's DVDs.

The large grains that for the image keep it sharp and clear; the textures of all surfaces are defined and detailed (for 16mm reduced to standard definition, anyway). You can see all the grooves in Kari Wuhrer's lips. In the Turbine version, however, you can make out the pores in her skin.

In terms of upscaling, you can also get pretty much the same quality for S4 - S5 whether using Mill Creek or Universal. The reason is, once again, the 16mm film grains. AI upscaling algorithms are based on mapping additional pixels to the video image based on the film grain within the image. 16mm film grains, even under Mill Creek compression, are present and retained and AI has no trouble mining that film grain for pixel re-rendering.

DVDFab is interesting, but at the time it was available for a discount, I was in the middle of an upscale and couldn't run it. I'm still pretty deep in the world of Topaz, but it would be neat if a parallel upscaling project used DVD Fab and described what results they're getting. On my end, Seasons 2 - 3 look great after a Topaz Artemis HQ upscale to 1080p whether using the Turbine SD blu-ray or the Universal discs. (I only attempted to upscale "As Time Goes By" to 720p off the Mill Creek DVD and it actually looked okay but not as sharp and crisp the Universal or Turbine based upscales.) And Seasons 4 - 5 from Mill Creek look terrific after an upscale as well.

16mm film downscaled to standard definition digital videotape yields excellent results. It's bitterly ironic that SLIDERS was at its lowest point in Seasons 4 - 5 with casting and budget and stewardship but those two seasons have the best video quality.