VCR/Hi8 capture tests; help evaluate?

[dianedebuda]

Polaroid 2001G - some folks were using this as a pass-thru at 1 time 'cause it had the "magic" chip. But mine won't even turn on now. It was not known to be have robust hardware. I could possibly replace the caps, but probably not worth it. I don't solder enough to maintain proficiency.

Some sage said only change Brightness and Contrast at capture time. I think it was you, sanlyn.

Mentioned AIW lack of Sharpening to note that I couldn't set it to 0 like I did the others.

[sanlyn]

Oh, I see about the sharpen setting. I don't know that the AIW you have would sharpen by default. Neither of my AIW's have sharpening turned on. However I have fooled with it in the past, saw the results, and never went there again. Note that sometimes it's the VCR that does the sharpening, and some consumer cameras do a bit of that during the original recording.

I see the following controls in VDub's link to my AIW capture drivers: Brightness, Contrast, Sharpen, Hue, Saturation. The default for each is 128. Hue is utterly useless for analog, which changes hue every couple of minutes. Sharpen is a no-no for VHS. I almost always end up with a 5 to 10% rise in brightness and a 5 to 10% cut in contrast (but, really, these depend on the tape and the VCR). If I see red or purples looking like blooming day-glo ready to ooze off the screen, it means the tape is oversaturated so I drop down about 2% to 5% there...and I had one really tacky looking retail VHS that required a 17% drop in saturation for about 20 minutes of tape because of the pumped reds. Saturation controls affect all colors, of course, but I was able to spruce up Green and Blue later without affecting red intensity, using some of those dandy filters I mentioned.

Usually, Brightness and Contrast are all you need. The others are either no-good or represent fixes that might work for 5 minutes of tape and suddenly look disastrous for the rest of the capture.

[jagabo]

I'd say the JVC deck is out of the running because of its bright crush problem.

The 550 and AIW are a mix. It looks like both can be adjusted (via their proc amps) to get levels within reason.The AIW has color artifacts from the less than ideal comb filter (2d vs 3d) and has more over sharpening halos even though its picture is not as sharp. The 550 and AIW are a mix. Both can be adjusted to get levels within reason. I don't know exactly what the colors should be but I think the AIW has more accurate colors and to the extent they are off they can be more easily fixed.

Here are frames from the two caps adjusted to similar levels and saturation:

550:


AIW:


I guess the 550 cap is still a little more saturated than the AIW cap overall. But the guitar is significantly more saturated in the 550 cap. If you test for valid RGB colors the guitar is mostly illegal. You have to reduce the saturation quite a lot (otherwise significantly under saturated) to get it to legal RGB values.

[dianedebuda]

See very little difference between the 2 pics. Girl's white shirt seems too bright in both, but more so in AIW. Similar for the guitar, but to a lesser extent. Maybe proc amp Contrast level still too high and detail lost there? Of course, it's possible that the camera "did it" and the capture is ok, I suppose.

I see the following controls in VDub's link to my AIW capture drivers: Brightness, Contrast, Sharpen, Hue, Saturation.

Should I have turned off all other proc amp levels except Brightness & Contrast? Didn't think about that. The Sharpen was set to 0 for the 550, disabled for the AIW. I left the others at default.

If you test for valid RGB colors the guitar is mostly illegal.

Don't know anything about legal RGB values at this point. Assume this relates to yuv -> rgb mapping.


I guess my next step really is to apply the improvements like these to the full 3-4 min captures, or at least the clips from them. Kind of hard for me to see with the pics alone. When I look at the actual captures with mpc-hc, I'm adjusting the widths to match, but the AIW one is faded compared to the 550, so I'm finding it tough to get beyond that. Also viewing the earlier pass-thru capture using the TRV480, just as a comparison.

Have installed avisynth 2.6 alpha5 w/o avs default to wmp. ok? Read a pdf on getting started with vdub, but it was mostly useless. Already learned more here about capture than it had; for processing, pretty much just listed plugins with not much more info than what's in the titles. On to the next...

[jagabo]

See very little difference between the 2 pics.

Yes, my point was that the two very different looking captures could both be adjusted to something like the full range of brightness and they would both look about the same -- neither is crushing brights or darks like the JVC deck was.

Girl's white shirt seems too bright in both, but more so in AIW. Similar for the guitar, but to a lesser extent. Maybe proc amp Contrast level still too high and detail lost there? Of course, it's possible that the camera "did it" and the capture is ok, I suppose.

The automatic gain controls on most camcoders (and other image/video devices) shoot to get ~95 percent of the image in the middle brightness range -- so small areas of darks and brights often get crushed. I could have used less gain on the brights making them less bright but there would still be little detail there. As it was, I just moved darks down to near the standard black level, and the brightest parts up to something near the standard white level. Of course, not every shot needs to use that full range. It's a bit of an artistic decision how dark/bright you want the extremes to be (and the gamma of what's in the middle).

Don't know anything about legal RGB values at this point. Assume this relates to yuv -> mapping.

Yes, it has to do with how YUV colors are converted to RGB for display. Many YUV combinations don't convert to legal RGB values, even if the individual components are withing their respective min/max limits. Here's a recent discussion of this issues:

https://forum.videohelp.com/threads/360935-Capturing-Correct-Chroma-and-Hue-Levels-From...ighlightbadrgb

And some general information on YUV/RGB (note the diagram of the RGB cube inside the YCbCr cube -- we often just use YUV here when we mean YCbCr):

http://software.intel.com/sites/products/documentation/hpc/ipp/ippi/ippi_ch6/ch6_color_models.html

Since you've now installed AviSynth and VirtualDub you should take a look at some of the scripts that I and others have uploaded.

[dianedebuda]

Just came back from reading the Capturing-Correct-Chroma... thead and, voila, it's mentioned here! Actually I'd tried to read it several weeks ago, but only was able to pick up tidbits here and there then. Kind of remember something there about YUV & RGB mapping. Makes a lot more sense now, but still quite a bit is over my head yet. I sure hope a great deal of the content there is "interesting" or academic and not mandantory.

A nudge here might be really helpful - in newbie terms. How do you determine what the first tweak should be? Is it that A, B, C and D usually have to be fixed in vhs and that B seems to be the worst in this sample, so do it first? Specifically in the guitar clip, what areas shout for improvement? What's the first? With LordSmurf's download, I have several filters already. Good place to start or elsewhere? At this point, stepping off into new territory again - so it's like not only having to figure out whether you want toasted white bread or a bagel, but how do you operate the toaster. Simple once you've seen it's just plop the bead in the slot, press the handle, wait, then magically the toast appears. But that first time...

[dianedebuda]

Back again after browsing oldretiredguy's thread yet again. Downloaded the orig script sanlyn did for the orig green cat clip. It feels pretty comfy to me; just function calls with parameters. Scary - probably famous last words. Is there a way to single-step? As I understand, you just open the avs from vdub's file menu. (push the button on the toaster?)

See the LordSmurf's selection of vdub filters. A lot to play with there. Where to start? Do you use the vdub filters at all or prefer only avs? Mix and Match? Depends on the video?

[jagabo]

Is there a way to single-step? As I understand, you just open the avs from vdub's file menu. (push the button on the toaster?)

Use the left and right arrows to step backward and forward by one frame.

Do you use the vdub filters at all or prefer only avs? Mix and Match? Depends on the video?

I usually do all my processing in AviSynth. Other people do some in AviSynth, some in VirtualDub. Some only use VirtualDub.

[dianedebuda]

Sorry, I meant single step the script. Or have breakpoints. Guess I'm thinking of these in programming or bat file terms.

Also found a couple links from a post on LS's site: Removal of chroma artefacts and Filtering VHS recordings in Virtualdub

[jagabo]

Sorry, I meant single step the script.

You can't really single step through the script like you could a program. But you can "Return(last)" at any point in the script to return the processing up to that point. You can comment out lines by putting a hash mark "#" at the start. You can stack videos side-by-side with StackHorizontal(v1, v2), or interleave them with Interleave(v1,v2).

[dianedebuda]

jagabo, is the avs that you used for the pics in post 93 available? That'd be a great nudge.

[jagabo]

jagabo, is the avs that you used for the pics in post 93 available? That'd be a great nudge.

It was something like this:

v1=AviSource("e vhs guitar sony composite Sh1-apc, 550 vdub B102-C148-Sh0 clip1.avi").Subtitle("550")
v2=AviSource("h vhs guitar sony composite Sh1-apc, aiw vdub B151-C78 clip1.avi").Subtitle("AIW")

v1=v1.ColorYUV(gain_y=20, off_y=-10)
v2=v2.ColorYUV(gain_y=30, off_y=-27, cont_u=100, cont_v=100)

Interleave(v1,v2)

I might have toned down the saturation with ColorYUV(cont_u=-50, cont_v=-50) before posting the images.

I recommend you also learn to read waveform monitor graphs.

Try adding TurnRight().Histogram().TurnLeft() to the end of the script:

v1=AviSource("e vhs guitar sony composite Sh1-apc, 550 vdub B102-C148-Sh0 clip1.avi").Subtitle("550")
v2=AviSource("h vhs guitar sony composite Sh1-apc, aiw vdub B151-C78 clip1.avi").Subtitle("AIW")

v1=v1.ColorYUV(gain_y=20, off_y=-10)
v2=v2.ColorYUV(gain_y=30, off_y=-27, cont_u=100, cont_v=100)

Interleave(v1,v2)
TurnRight().Histogram().TurnLeft()

That will add a traditional horizontal waveform monitor above the images.

You might find Tweak() to be more intuitive than ColorYUV(). Some rough equivalents:

ColorYUV(cont_u=-50, cont_v=-50) ~= Tweak(sat=0.8)
ColorYUV(cont_u=-256, cont_v=-256) ~= GreyScale()
ColorYUV(gain_y=50) ~= Tweak(contrast=1.2, coring=false)
ColorYUV(off_y=-50) ~= Tweak(bright=-50, coring=false)
ColorYUV(cont_y=50) ~= Tweak(cont=1.2, bright=-25, coring=false)

Since you've finally installed VirtualDub and AviSynth try downloading the attached "interactive" demo of ColorYUV(). Extract the two files to the same folder then open Interactive.avs in VirtualDub. Scrub through the video with the track bar or arrow keys.

[Brad]

Once upon a time, filtering in VirtualDub required you to convert to RGB (and then back again if you wanted to play your files on anything besides a computer). So I stayed away from that once I got into Avisynth. I know the built-in filters should support YUV now but with legacy addons I guess you don't know until you try.

[jagabo]

VirtualDub... I know the built-in filters should support YUV now

Very few of them work in YUV. You can turn on "Show image formats" in the Filters dialog to see what colorspace each filter runs in.

[dianedebuda]

Thank you. That's the sort nudge detail that I wanted - just a sample with known good values that I could run and validate that I was running it correctly. I couldn't find an example in oldretiredguy's thread for a "good" huffy avi. When I was asking for suggestions on the guitar clip, what I was really meaning was some "I see this problem, try that filter" type of things.

Vdub & YVU/RGB conversions. See why the lean towards AviSynth.

Since you've finally installed ... AviSynth

There is only just so much new info my noggin can handle at a time. Didn't want me to explode, did you? Actually, I have some open tabs "to do" yet about learning how to read the graphs that were mentioned earlier. And I understand that the AviSynth manual is fairly good, so I have some work for the weekend. Maybe next week for the hi8.

[Brad]

AvsPmod is a nice editor for Avisynth, with calltips for filters and a video pane integrated. (Press F5 to refresh video preview.)

Just be sure to go into the font preferences and change to monospaced instead of the awful Comic Sans look.

[dianedebuda]

Saw AvsPmod mentioned in oldretiredguy's thread & already downloaded it - to the "to do" list. Thanks for the reminder. Want to get into the improvements just enough to see differences in fixing the AIW vs 550 captures so I can make a final VHS capture equip decision. After that, want to do the hi8 equip/capture eval so I can actuallly get all the captures done - before I've forgotten all I've learned here

[lordsmurf]

Post a sample clip, then PM me when done.
I'll look at it, and tell you what I did.

[dianedebuda]

Visualize old Victor logo: dog (Nipper) listening to master's voice on phonograph record.

[sanlyn]

The attached mkv's are from clip "f" from post #75 and clip "h" from post #87. These look like the best captures and have about the same capture levels. Each attached clip used the same basic script and levels and color corrections. I think you can see that with the same basic corrections, both clips look alike. Part of the color correction was aroutine that adjust the color balance of Even and Odd fields so that are more alike. But there was no attempt to make both entire clips "look alike". The same corrections were applied to both clips.

The "A" and "B" versions used the same denoiser -- a modification of a routine using MVDegrain2 as the denoiser and LSFmod as a mild sharpener. That script is posted below. For the "C" versions I removed the MVDegrain routine and replaced it with another popular denoiser (dfttest.dll). For the "D" version I removed both the MVDegrain2 routine and LSFmod, replacing both with a low-powered NeatVideo.

The different denoising routines were adjusted so that they run at approximately the same filtering strength. All other corrections in the script remain as shown below. I tested all these clips in MPC-BE, MPC, an Windows Media players. As usual, VLC Player looks too clunky and "wrinkled" (I hate VLC. I've almost stopped using it altogether). The effect of the different denoisers can best be observed by looking at background areas such as the walls and the dark shadow areas. One could make the overall videos brighter or darker, depending on taste -- but remember that you have brightly lighted objects and objects with very little light. You have to decide which areas are more important; you can't have everything, as the original scene has a contrast range that won't "take" in video. The camera's AGC and autowhite didn't help either: the originals are brighter at the beginning and change color balance a few times during the scene. You could make the dark areas brighter, but it won't do any good: they will just appear as a lighter gray because there's much less detail down there. There is also some in-camera dark crushing.

To get an idea of what the camera was doing with motion, look at the electronic piano in the lower left of each clip. Observe that object during the camera zoom.

The basic script, which includes the MVdegrain2 routine:

vidpath="E:\forum\diane04\"
ppath="D:\Avisynth 2.5\plugins\"
vdubpath="D:\VirtualDub\plugins\"
Import(ppath+"FixChromaBleeding2.avs")
Import(ppath+"Vinverse.avs")
LoadPlugin("D:\AviSynth 2.5\plugins\plugins2\masktools.dll")
LoadPlugin(ppath+mvtools.dll")
LoadVirtualDubPlugin(vdubpath+"ColorMill.vdf","ColorMill",1)
LoadVirtualDubPlugin(vdubpath+"ccd_sse2.vdf","CCD",1)

AViSource(vidpath+"FileName.avi")

#---- Adjust basic levels for later color tweaks
ColorYUV(cont_y=30,off_y=-10,cont_v=-20)
ConvertToYV12(interlaced=true)
SmoothLevels(16,0.95,255,16,245,chroma=200,limiter=0,tvrange=true,dither=100,protect=6)
ContrastMask(enhance=3.5)

#---- Some chroma housekeeping
Cnr2("xxx",4,5,255)
FixChromaBleeding2()

#---- smooth some of the edge nosie on motion
Santiag()

#---- De-interlace
AssumeTFF().QTGMC(preset="super fast",lossless=2)
vinverse()

#---- modified MVdegrain2 routine from 2BDecided -- requires older mvtools + masktools -----
  source=last   # <-- save original, to be referenced later

#---- denoiser:
  backward_vec2 = source.MVAnalyse(isb = true, delta = 2, pel = 2, overlap=4, sharp=1, idx = 1)
  backward_vec1 = source.MVAnalyse(isb = true, delta = 1, pel = 2, overlap=4, sharp=1, idx = 1)
  forward_vec1 = source.MVAnalyse(isb = false, delta = 1, pel = 2, overlap=4, sharp=1, idx = 1)
  forward_vec2 = source.MVAnalyse(isb = false, delta = 2, pel = 2, overlap=4, sharp=1, idx = 1)
  source.MVDegrain2(backward_vec1,forward_vec1,backward_vec2,forward_vec2,thSAD=400,idx=1)

  clean=last    # <-- save cleaned version from above, name it "clean"

  diff1=subtract(source,clean).Blur(0.25)
  diff2=diff1.blur(1.5,0)
  diff3=subtract(diff1,diff2) #diff3 is high-ish frequency noise only
  sharpen(0.4,0.0)    # <--  sharpen cleaned version a little

#---- mix high frequency noise back in
  overlay(last,diff3.levels(128,1.0,255,0,127,coring=false),mode="add", opacity=0.7)
  overlay(last,diff3.levels(0,1.0,127,128,255,coring=false).Invert(),mode="subtract", opacity=0.7)

#---- put cleaned chroma back in with warp sharpening
  mergechroma(clean.aWarpSharp(depth=20.0, thresh=0.75, blurlevel=2, cm=1))
#--------------------------------------------
#---- END of modified MVdegrain2 routine ----
#--------------------------------------------

#---- Tighten edges a little more + mild sharpen.
DeHalo_ALpha()
LSFMod(strength=50)

#---- smoothing + add fine grain to counter banding and "clumps" of noise
GradFun2DBMod(mask=false)

#---- To RGB32 for VirtualDub plugins.
ConvertToRGB32(matrix="Rec601",interlaced=false)
AssumeTFF()
a=last
e1=a.SelectEven()
o1=a.SelectOdd()

#----- Correct tint in in Even and Odd frames ("e1" and "o1") to make them more alike.
e1=e1.ColorMill(25700, 25700, 25697, 25688, 25700, 25700, 25700, 25700, 25700, 25700, 25700, 25700, 25700, 1124, 5)
o1=o1.ColorMill(25700, 25695, 29794, 25700, 25711, 25700, 25700, 25700, 26468, 25700, 25700, 25700, 25700, 1124, 5)

#----- Re-assemble the frames in their original sequence
Interleave(e1,o1)

#----- Color tweak + denoise chroma for all frames
ColorMill(25702, 25698, 25697, 24416, 25704, 25700, 29044, 25713, 25700, 25700, 25444, 25700, 25700, 1124, 2053)
CCD(20,1)

#----- Re-interlace + new black borders
AssumeTFF().SeparateFields().SelectEvery(4,0,3).Weave()
Crop(16,0,-16,-10).AddBorders(16,4,16,6)

#----- To YV12 for encoding.
ConvertToYV12(interlaced=true)
return last

# ------------------- Contrastmask function used above -------------------

function ContrastMask(clip v, float "gblur", float "enhance")
{
enhance = default (enhance, 10.0)
gblur = default (gblur, 20.0)
enhance = (enhance>=0.0 && enhance<=10.0) ? float(enhance*0.1) : 1.0
v2=v.Tweak(sat=0)
v2=v2.invert()
v2=v2.gaussianblur(50.0,50.0+gblur)
photoshop_overlay=mt_lutxy(v,v2,"x 127.5 > y 255 x - 127.5 / * x 255 x - - + y x 127.5 / * ? ")
merged=overlay(v,photoshop_overlay,opacity=enhance)
return merged
}

The VCR's are one of your biggest obstacles. I happen to have owned a copy of your SONY at one time, now used by a relative.

I didn't post duplicates of "C" and "D". They used the same script, the same filters, and look almost exactly alike. I thought the AIW looked a little cleaner, but that's strictly subjective and I can't put numbers on it.

[dianedebuda]

Thank you, sanlyn, for the scripts. Soon as I can (that is, soon as I'm proficient enough), will check them out.

Have made it through reading all the vdub help - this time made a whole lot more sense. Loaded AvsPmod and starting on it. Like what I see, so far. First test problem is getting HFYU recognized by AviSynth - goggling now for solution.

LS looked at the last 2 Sony clips and he thought the AIW looked "more accurate" and the 550 was "more colorful and contrasty, but at the tradeoff for chroma noise and blooming". Not quite sure what he meant by "more accurate" yet.

Based on your experience with the Sony, sanlyn, how much improvement would I see with a panny? a younger AIW? frame TBC? "how much" = in your opinion, is it worth it to hunt any of these down or to send the "important" VHS off for capture (probably the cheaper of these 2 options)? At minimum, I'm still going to do a home capture so I'll have something - even if I send the tapes off. Remember that my target is not DVD - just pc players.

[sanlyn]

The SONY will likely suffice. It's a budget machine that doesn't work all that well with any tape.....note that there is enhanced "edge detail" but not much fine detail. That's a problem with all tapes to begin with. There are hundreds of posts by member orsetto that discuss a number of VCR's that you can find on the used market. Basically, anything made 2000 or later is at or near the lower portion of the barrel. The last SONY VCR I ever cared for was my original SLV-585HF from 1990, which I had rebuilt about 10 years ago. Made like classic VCR's of the early 90's, it weighs about 15 pounds. I don't think SONY has made as nice a VCR since the 585/595 series, and the 585 was much better with EP tapes.

You mentioned a "frame TBC". That type of TBC doesn't address the major problem with tape playback, which is mostly line timing errors within frames rather than frame timing itself. An explanation of the difference in operation between a frame tbc and a line tbc is in the oldretiredguy thread in post #284 (https://forum.videohelp.com/threads/360511-Vhs-DVD-clarity-question?p=2287513&viewfull=1#post2287513). That post also links to a tutorial video that I made years ago in the pleistocene era of my video processing days.

Post #285 (https://forum.videohelp.com/threads/360511-Vhs-DVD-clarity-question?p=2287517&viewfull=1#post2287517) in the same thread links to demo's that show a line tbc in action. The "corrected" version of the demo isn't the final version but is only the first stage of playback, capture, and first-run filters. Keep in mind that the correction demo still needed more work at the time. The "problem" demo was originally recorded off cable tv by a relative using a Panasonic PV-9662 from 1999. I played that tape on a used Panasonic PV-V8661 from 1998 with a Toshiba DVD recorder used for line and frame sync pass-through. You will notice in the correction demo that there is a slight grayish area at the bottom of the frame -- it was eliminated later by removing an external proc amp from the capture circuit (some tapes are affected that way by some proc amps, depending on the tape). As it turned out, the proc amp wasn't needed.

The "problem" tape in the demo was recorded at 6-hour speed for over 3 hours on SONY "HiFi" tape, one of the noisiest tapes around at the time.

Currently I use a Pansonic PV-4564 and PV-4664 (from 1995-96), a PV-S4670 SVHS (1996), a PV-8661 and PV-8664 from 1998, and a used/rebuilt AG-1980 that I just purchased (the latter being a bank-breaker, and not the "perfection" that some people claim, but pretty darn good). They all work pretty well with EP tapes and SP, but be prepared for some sharpened noise on long-play tape.

[dianedebuda]

Again thanks, sanlyn. I did learn the difference between a line and frame TBC from oldretiredguy's thread way back when. That info really presented well for a newbie there, BTW. When I was posting above, I was reciting the typical equipment recommended & briefly forgot that the Sony doesn't have an internal line TBC like my JVC does. Orsetto has made a believer out of me that these days it's not worth, for the limited number of tapes that I have, the crap shoot of finding another vcr. The question is more, how much improvement would I likely see with better equipment? After post processing, would I really notice much? Quite subjective, I know.

Going back to my fight with reinstalling HFYU. It's in SysWOW64 - install, darn it. Verbal commands don't seem to work

[sanlyn]

I don't know what gives with that huff version (I believe it's the same one I use on my capture PC's). Problem is, it only works correctly on those PC's. I have the same huff installed on my newer super-duper(!) processing PC's, but I still have problems if I transfer those huff videos to other PC's. What I do is open them with Avisynth and activate the ffdshow version of huffyuv, then fast-recompress the video to Lagarith. Sometimes I do that recompress on the capture machine before transferring. It seems to be a problem with the way huffyuv installs on specific machines.

The question remains: why not capture with Lagarith in the first place? Well, Lagarith is pretty fast, but huff is a tad faster -- just barely fast enough to avoid a few dropped frames during a long cap on my ancient 2GHz PC's. I suppose if my capture PC's were newer/faster I could avoid that, but no one makes a new motherboard that will accept my trusty AIW capture AGP cards.

ffdshow is a pain in the neck, but it does have boocoos of decoders that often come in handy. If I recall correctly, when it installs and finds a version of huffyuv on your machine, it disables its own version. That's the way it works on my machines, at least. I leave their huffyuv encoder disabled. But when I want their decoder, I go to their video decoder configuration dialog and enable their huffyuv encoder. Then I use this command in Avisynth:

DirectShowSource("Drive:\path\to\video\filename.avi")

Save that text as an .avs file and open it in VirtualDub. In VDub's "Video" menu, set "Color Depth" to YUY2. In the "Video" menu again, set "Compression" to "Lagarith lossless codec" and configure Lagarith for YUY2 output. Then under "Video" again, enable "fast recompress". Then "Save as AVI...". It will take several minutes. That's how I opened your AVI captures and made a new Lagarith AVI copy to work with.

VCR's: Yep, it's a hassle finding goodies that haven't been used to death. It took me a few years. I didn't spend all that much (except for the "AG"), but I did encounter a couple of cheap lemons. Par for the course.

[dianedebuda]

The huffyuv comments were just asides, really. Didn't think I had a problem on my Win7 4x processing box 'cause vdub worked there fine with the files from the capture box (like you, 2GHz boxes). When I had trouble with AviSynth, added MT and 64 - made a mess of it, really. Uninstalled all, in theory, but orig huffyuv won't install now. Probably registry leftovers I'll have to hunt down. Whatever.

After post processing, would I really notice much? Quite subjective, I know.
A lot

I meant my current equip + post processing VS better equip + post processing. Is that what you were answering?

[sanlyn]

Stay with 32-bit VirtualDub and Avsytnth. There are very few 64-bit plugins, and you can't mix 32 and 64.

I gave up on Avisynth MT for the time being. Many standby plugins in popular use don't work with MT modes and changing MT settings is a headache. When you're new to learning Avisynth, there are fewer hassles if you stay with the 2.5 versions for a while IMO.

[dianedebuda]

Was tired last night & shooting in the dark. Got a bigger hammer, hit the registry & huffYUV installed ok now and AviSynth sees it.

sanlyn, I'll try to ask another way. Was your "A lot" a reply to "difference between capture and post process version", which I know will be noticable or what I ws really trying to ask about - hardware differences? How "bad" is my hardware? IOW, will I be able to mostly make up in post processing the hardware deficiences? Would you expect it to be noticible enough to be worth the $ after I've done my home capture + post processing? It's just curiousity eating at me. My home videos aren't exactly a big budget, major release and I may be satisfied enough just to leave it at home processing only.

[sanlyn]

You should be able to make visible and satisfactory post-capture improvements using what you have now. Some of us are really picky, y'know. Obviously better hardware means better results. But as you've noted in other posts it's a big pot of of trouble. I wouldn't agonize over what you don't have or can't get. We've seen transfer projects around here that are far worse than your samples; some of those others are quite hopeless. At least your tapes appear playable and you're getting improved workable captures.

Even with better gear -- or even with better sources -- it's still work, and the processing won't be all that different. Technology is always improving and you might even find "better" later on. For that reason we always advise that you save your most valued tapes, for the time at least, and archive your captures onto portable hard drives. There are people who return to this forum several years after their first efforts and are surprised to find what they can accomplish later.

[dianedebuda]

You should be able to make visible and satisfactory post-capture improvements using what you have now

[dianedebuda]

Ok, think I've beat the VHS to death. Yesterday sanlyn's

The VCR's are one of your biggest obstacles.

got me anxious, but have recovered. At a point now where I can duplicate jagabo's and sanlyn's basic color/contrast tweaks using AviSynth. Time to move on the the hi8 tapes.

In an effort to not post more than needed this time:

TRV480 has a TBC and a DNR. Which off/on combos would you like to see posted?

2 conversion options: internal (firewire/DV using WinDV) and capture composite. Didn't see a lot of difference between 550 and AIW for VHS, so assuming it shouldn't make much difference here either. True? Don't see a need to try the VC500 or 650.