I've gotten to do a very quick test of a Panasonic AG-1980. I only had on hand a small LCD TV that only has composite input and no BNC composite cable on hand and so ran the VCR on S-Video out to Green TBC and the composite output of that to the TV. I wanted to see the settings shifts at larger scale than the smallish VirtualDub capture window and wasn't booted into that at the moment anyway. Will test it also that way.
My quick findings are that the edit setting seems to provide less detail, like a slight smear almost (which seems to go against some I've read but confirm one thing I recently saw on a webpage where some place was saying that you could see detail fall on an oscilloscope when you switch it to edit mode). This is very unlike the edit setting on say a Canon A1 Digital/L2 Digital Hi8 camcorder where the edit mode appears to give quite more detail (at the expense of more noise) and some other slight changes.
Switching the TBC on seems to slightly change contrast/brightness and possibly slightly lowers resolution (except where something is line to line all wiggly and then it can easily make things line up for true resolution) and noise?
I was reading that some were saying that the picture soft-sharp slider even in the neutral center position still adds sharpening so it should be to the left (where even there it was said that sharpening is still applied even) and so it should be to the left and full soft setting to be most natural and neutral possible on it. I did NOT find this to be the case at all. As soon as I move it away from the middle notch toward the left (soft) I start noticing stuff looking weird and it starts smearing out some edges and stuff in a ghost smear like a simple blur filter. OK, maybe one can be tricked BUT the thing is it sure appears to erase true real details so I don't think it is going more neutral and do think it really is smearing (as it looks to be). For instance, on a quick test cartoon tape there were some tiny little evergreen trees on the hillside horizon and if you go full left they all smear into a solid blob and you can't make out any of the trees any more and along the top of the hillside it's just a solid dark green smear, no individual trees, there is no way you could ever bring that detail back and it is real detail. On neutral position you can clearly see actual tiny little tree top next to tree top and their individual shape (AND if you compare to the blu-ray disc you clearly see the same little distinct trees and shapes so again it is NOT some sort of fake detail being 'created' to make the little trees, they are real detail and setting slider all the way to the left simply smears them all into one big blob and anything starting to go left of neutral starts appearing to smear off true actual detail). As soon as you move the slider to the left this odd almost smearing and spreading and inverse ghosting of details out happens and it looks weird and after moving it enough it flat out erases real detail 100% away. Left of neutral does NOT at all to me seem to be bringing the machine back closer to a real neutral setting but does seem to be for real simply smear softening things like with a simple blur filter.
OTOH moving it the right does make things looks crisper and appear more detailed, perhaps adds a little contrast to details with less contrast and maybe also some halo sharpening. Anyway, you also see it emphasizes halos and starts doing artificial things and looking less neutral. So it seems likely to be just a sharpening filter more likely than a reducing of a NR filter to give more details or something, certainly once dialed up stronger.
The center notch neutral position seems to me like it is true neutral neither a radial blur filter nor an old school simple haloing sharpening filter. Maybe it is still doing something even at neutral, would have to do lot more tests and comparisons but as far as that slider goes neutral definitely seems to be the most neutral and least processed setting.
I'd have to test detail vs. norm more to see if there is any difference. Perhaps a very, very subtle shift in NR or something else?
Edit bypasses the sharpness slider. Edit in a 1980p ag isn’t recommended for playback. That was meant for connecting two 1980p ags together for editing.
Norm is recommended for playback.
Quote:
the thing is it sure appears to erase true real details
Quote:
it is NOT some sort of fake detail being 'created' to make the little trees, they are real detail and setting slider
The sharpness slider is dealing with the analog signal before it becomes the digitized picture. It amplifies the high frequency parts on the signal which adds detail.
Quote:
The center notch neutral position seems to me like it is true neutral neither a radial blur filter nor an old school simple haloing sharpening filter. Maybe it is still doing something even at neutral, would have to do lot more tests and comparisons but as far as that slider goes neutral definitely seems to be the most neutral and least processed setting.
I like to middle notch it also. It adds some sharpness but I can look at high contrast spots and I don’t see hallowing.
Some captures, or even screenshots of the same frame might help, but my guess is part of the equation could be if it was recently refurbished and by who. You should be able to notice significant softening in normal vs edit mode which I agree does bypass the sharpness slider and puts it to full sharp essentially. So if you were to have the sharpness slider set all the way to sharp, moving between normal and edit won't look very different, but I've never seen a situation where normal looks significantly "more sharp" than edit mode.
I do think the middle sharpness notch is fairly true to what I'd expect for VHS. The degree of halos seems to vary quite a bit with content if adding sharpness, but it's a nice option to have that you really don't have with many other VCRs on a linear adjustment scale which is nice. Sharpness will usually amplify noise too, but even that noise can be perceived as a sort of "film grain" and can give the perception that something is higher resolution than it is. I do think the AG1980 generally has a smoother/less noisy output than some of the competition at default settings and has a unique look to it. Some may prefer that, others may prefer JVC's implementation instead.
Given that each TBC is an analog to digital and then back to analog stage in your chain (as well as for the internal TBC), you should expect to have some slight detail loss with each TBC stage, though usually the benefit of having the stabilization is considered worth the tradeoff. Have you tried capturing without the frame TBC (sounds like you have a green AVT-8710) to see if that is causing some of your detail loss?
Edit Mode is curious. When I switch it on, with many tapes, I see a LOT more noise. It seems like it probably turns off all NR or at least all of one type. It can appear sharper and more detailed at first glance with some tapes but looking closely with one quick set of tests, it seems it's just the noisy grain that can make things seems more crisp. I'm not as yet really able to find any actual more real details at all compared to DETAIL mode (which mostly seems rather similar to NORM mode, have to look into it more).
On the first tape, a commercial tape (cartoon) the noise difference was more minor so you didn't get the extra 'detail' effect and I saw on some fine text and details it seemed to almost spread out a touch more.
Not so sure that Edit Mode turns sharpness to full. It seems to maybe turn it off, totally totally actually? It doesn't seem to mimic turning sharpness to full on NORM/DETAIL (but would have to test more to make sure didn't mix something up).
Curiously the DETAIL/NORM settings seem to show a touch more actual real details, more often, than on EDIT MODE but with anywhere from a little to tons less fine noise.
EDIT MODE is maybe turning off EVERYTHING, from NR to sharpness to possibly also some circuits that actually generally always improve things and maybe altering boosts to transfer to another magnetic tape better but maybe not ideal for other usage?
It seems to not really do the same things that EDIT mode on say Canon A1 Digital/L1/L2 Hi8 Camcorders do. There EDIT mode gives some more noise and stuff (maybe not quite as radical of a difference but also, in initial quick test, seems to give more real detail not the same or less).
I wonder if the AG1980 has some sort of secondary NR system that looks across a couple frames or across both fields and brings some detail back at the same time as reducing noise?? Or maybe EDIT mode is just shifting some boost or removing a trace of sharpness from DETAIL/NORM default position and then having tons more noise.
These are not the best test pics below since the detail on various bits can change frame to frame although I did try to pick overall best from screenshotting the sequences a bunch of times. And the bwdif de-interlacer can be doing various tricks and maybe getting confused by noise and improving things less. So I should probably also try a non-deinterlaced look (although with jaggies all over it's often hard to get a sense of images in that form).
But FWIW, if anything, you can see a lot more noise with EDIT mode pics below but also the non-EDIT seem both less noisy and actually same to more real details, as far as I can tell, basically across the board with no real details lost. Although the EDIT one with all the noise 'grain' can at some glances "feel" more detailed. But then look at some actual details and text readability and so on seem better on the non-EDIT mode.
-- merged --
It's easier to notice the differences if you load the pics into a full screen viewer and then do like a 2x ZOOM view to make it jump out more.
-- merged --
Also EDIT mode seems to make a huge difference whether the TBC is off or on. The above samples had the TBC off though.
(I might have had it feeding into external full frame TBC when I did those, can't recall, think so. One of the "Green" TBCs.)
It can appear sharper and more detailed at first glance with some tapes but looking closely with one quick set of tests, it seems it's just the noisy grain that can make things seems more crisp.
I've stated this for years.
People are often fooled by this. Kudos on not being fooled.
Quote:
the DETAIL/NORM settings
EDIT MODE
I don't have time yet, but I'll give a full analysis with my 4 refurb'd AG-1980P decks.
The added problem with AG-1980P is that about 5 version exist, and there are some observed differences. At least one model really cannot be refurb'd, nobody wants to touch those. So you have variances even when all things are equal condition-wise, and refurb/non-refurb is a monkey wrench.
You probably feel more lost now. Welcome to my world.
Hah! Thanks.
I was almost about to say I wonder if the AG1980P didn't have some silent changes over time since a few things seemed to be reacting differently than some posts on various forums made it appear.
-- merged --
Hmm my images came out small. I'll try to download the ones that uploaded here but they seem way downscaled so not sure if what I mention will be apparent or not (still greater than VHS res though so maybe it'll be apparent but need more zoom to make clear assuming the downscaler didn't do too much damage).
Attach larger images inside zip/rar/7z files. (Flicker/whatever compresses your images for web browser display, too. Maybe not size, but definitely quality.)
On that image with the cooler on the dresser, notice how the noisier "more detail" image messes with the text logo. The cleaner "softer" image shows it properly. Great example to show how people are fooled by grain.
Attach larger images inside zip/rar/7z files. (Flicker/whatever compresses your images for web browser display, too. Maybe not size, but definitely quality.)
Ah, OK thanks.
Quote:
On that image with the cooler on the dresser, notice how the noisier "more detail" image messes with the text logo. The cleaner "softer" image shows it properly. Great example to show how people are fooled by grain.
And how you can see the groves on the top of the case in the 'smooth/soft' image. To be fair that does change a bit frame to frame (although I tried to pick better frames for both) but viewing back the video you (where you see all the frames, it does hold, you do see that the groves are overall more distinct for sure when EDIT mode is off despite the image seeming smoothed up). The text on the TV too.
Again it does change frame to frame on both captures but the one on DETAIL/NORM does seem a touch easier to be able to sort of read and seems to overall show a touch more real detail.
I guess using Edit is really only best if you want more 'detail' to work with for restoration purposes. That's why I do anyway, I've been capturing my tapes in EDIT mode now for a while before needing to clean them up during the restoration phase. Granted I use a JVC VCR, so they might handle the differences between Edit and Norm differently than the Panasonic 1980.
Keep Hi8/Video8 captures at Norm since I can not spot any difference with those tapes being captured with Norm and Edit mode.
I should also add, for any who see the images and are confused, thinking it doesn't seem like VHS with the aspect ratio and detail levels. Well it's SVHS (thus detail seeming different than VHS horizontally) and it's a recording off an old letterboxed DirecTV widescreen 720p recording (thus the 1.78:1 aspect ratio letterboxed within 4:3 and the loss of vertical resolution compared to a full screen VHS P&S version of this commercial) probably off an S-Video out (although maybe the box only had composite? I'd assume it had S-Video but who knows, it's been too long now to recall).
I guess using Edit is really only best if you want more 'detail' to work with for restoration purposes. That's why I do anyway, I've been capturing my tapes in EDIT mode now for a while before needing to clean them up during the restoration phase. Granted I use a JVC VCR, so they might handle the differences between Edit and Norm differently than the Panasonic 1980.
Keep Hi8/Video8 captures at Norm since I can not spot any difference with those tapes being captured with Norm and Edit mode.
I think this all suggests that yeah different brands/machines (perhaps even different versions of the 'same' model) do different things with their edit mode and recommendations for one brand/model/sub-model might not universally apply for everything. You get more, apparently, more real detail on JVC in edit mode but same on your 8mm player. I get, at least tentatively so far, seemingly less for the former and more for the latter on my machines.
Tgrantphoto.com used to have a page on the detail edit and norm switch on the 1980s that was very informative. I don’t know if he still does business anymore.
Quote:
I've been capturing my tapes in EDIT mode now for a while before needing to clean them up during the restoration phase. Granted I use a JVC VCR, so they might handle the differences between Edit and Norm differently than the Panasonic 1980.
Edit it different on the 1980s than it is on the JVCs.
Gotcha, from the JVC VCRs I use I realize I can get a bit more detailed extracted out of using Edit mode, wouldn't hurt to capture in Edit, and have a bit more to work with during the restoration process.
Though this "extra noise makes us think there is more detail when in reality there isn't" reminds me of the decode option. I guess it's true that you are getting the raw unfiltered messy signal, but just how much of that added noise is actual detail, the answer is seemingly none.
Per the AG-1980P Operating instructions the EDIT position disables the PICTURE (Sharp/Soft) switch. I provides the most accurate copy if the information read off the tape, not necessarily the most watchable on a TV. When copying to another tape it is generally the best position to use. The Soft/Sharp position cuts/boosts image signal components above 1 mHz or so.
(IMG_0001 is an extract from the AG-1980P manual on this subject)
This test is uses a B&W signal (where image sharpness driven) and does not address color or noise performance.
The VHS-test-1980 and SVHS-test-1980 documents provide screen captures from an oscilloscope views of the S-VIDEO luma signals output by the AG-1980P. For this test I used a mulit-burst test signal that shows the resolution performance. It contains bursts a 0.5, 1, 2, 3, 3.58, and 4.2 mHz - these correspond to the nominal bandwidth of the NTSC broadcast signal; more that VHS but less than SVHS.
One will note that the VHS record/playback performance is decent up to 2 mHz but tails off to near nonexistent a by 3 mHz - corresponding to the nominal 240 line limiting resolution of VHS. The SVHS performs well beyond 4.2 mHz as expected.
Thanks for quoting the manufacturer's Operating Instructions on this question. Perhaps we should see more of this on forums.The manufacturers just might know a thing or two about their own product.
Per the AG-1980P Operating instructions the EDIT position disables the PICTURE (Sharp/Soft) switch. I provides the most accurate copy if the information read off the tape, not necessarily the most watchable on a TV.
It seems more unclear than that in that it simply says it disabled the soft/sharp section of the circuits but it doesn't really get into anything more there. It obviously does a lot more than just that since there are tons of differences between edit mode look and picture slider set to ANY position and that part of the circuit wouldn't put back to much noise, etc.etc.
Quote:
When copying to another tape it is generally the best position to use. The Soft/Sharp position cuts/boosts image signal components above 1 mHz or so.
(IMG_0001 is an extract from the AG-1980P manual on this subject)
test is uses a B&W signal (where image sharpness driven) and does not address color or noise performance.
The VHS-test-1980 and SVHS-test-1980 documents provide screen captures from an oscilloscope views of the S-VIDEO luma signals output by the AG-1980P. For this test I used a mulit-burst test signal that shows the resolution performance. It contains bursts a 0.5, 1, 2, 3, 3.58, and 4.2 mHz - these correspond to the nominal bandwidth of the NTSC broadcast signal; more that VHS but less than SVHS.
One will note that the VHS record/playback performance is decent up to 2 mHz but tails off to near nonexistent a by 3 mHz - corresponding to the nominal 240 line limiting resolution of VHS. The SVHS performs well beyond 4.2 mHz as expected.
Thanks!
Although it is a little hard to judge the edit mode comparing to only the extreme soft max sharp and max soft and nothing to picture soft/sharp in center position for NORM and DETAIL. It's certainly not overdoing things like max sharp nor hyper softening like max soft but maybe still a little soft compared?
