I already have Titanic, so I'd be up for using that one, although since I have GShelly's deck, we might as well pick different frames than the one used in his shots. I don't think I have Flashdance, but I'd be happy to pick it up. Which release do we have so far, so that we all get the same version? If you guys have the UPC code, that'd probably be the best way to guarantee we all get the same version.

As far as EP/LP tapes, how about we focus on those later if the SP part goes well? We probably have enough to do gathering SP tapes. I can make a list of titles I have in that format in the meantime.

I'd be happy doing some macrovision tapes, although it doesn't sound like everyone will be able to participate with that, given the equipment out there.

I must say I'm unclear about the method of this project. Is simply the VCR output to be compared, or the capture techniques also?

What I mean is, it's not really an accurate comparison if proc amps, detailers or other optimization hardware is used. It should be just a straight capture from VCR to lossless, using only the settings available on the VCR. This would be to measure such things as colour, detail, tracking and filtering performance of each machine. But even that poses a problem, as to what capture card and how it is used. Even what TBC is used.

For the purpose of the test, it'd make sense to use common capture hardware and have a common, calibrated workflow. It goes without saying everything should be in optimal condition, including good cables and power to prevent interference and so forth. But if it isn't then it's a good way to find out too. All this stuff is no issue for the pros, but for anyone wanting to join in, it could be.

I suggest using the ATI 600 for capture, it's easily available, affordable and compatible. I think there should be some kind of guide or outline to follow in this project, otherwise there's just too many variables for an accurate representation of a VCR. It makes it harder that even if you use the same movie release, there's still the variable of condition. Many things to account for here. Anyway, just my ideas.

How about this: we must post calibrated test patterns from DVE (or the VideoHelp MiscTestPatterns stills disc) before we post our VHS caps.

I think forcing everyone to use the ATI 600 is too restrictive.

@robjv1, thanks for your detailed response on the 750. I'd like to give my thoughts on this, I hope people don't mind -

From what I've read, the 750 is identical to the 650 but without the MPEG hardware encoder. 650 was the first to be blamed for overactive AGC, and I believed this was due to its multistage/multispeed AGC instead of single stage as found in ATI 550 - and the 600 I'm assuming, given its behaviour (source). No difference between USB and PCIe models, according to AMD. I tried to find 750 specs but couldn't.

I think the 650 problem was first brought up on this thread at videohelp. My issue is that an ES15 was used, not a true TBC, so all the AGC info could be partially inaccurate. Perhaps it is just oversensitive to unstable analog signals, which isn't ideal but should be easily solved. Reports saying that some stabilizer hardware fixed or lessened the problem is pretty indicative of that. ES15 is just line TBC/visual correction mainly, so not the right tool.

But then, I'm sure you probably did test this stick without a TBC, so all I've written is kind of pointless. I don't know why the problem wouldn't be universal though, if the reason for it is the multistage AGC.

Yeah I agree with msgohan, I think this should be pretty open -- forcing everyone into the same capture card would be too restrictive and I doubt everyone is going to be enthused enough to go out and purchase another capture device/buy a new system to take advantage of it just to post in this thread (more power to you though if you do!). Luckily since a lot of us own the same VCRs but different capture cards, we'll be able to compare the same models and see differences most likely attributed to the cards themselves. Capturing losslessly will certainly help mitigate the differences. We'll just have to accept that there will be some variability. There is going to be variability from deck to deck even -- both due to slight problems with the decks and general variations. I have an SR-W5U that pushes a too hot picture (the gain can be adjusted with a screwdriver internally though) and another that doesn't, so they would display noticeably different pictures, even though they are of relatively of the same vintage.

I would like to include Titanic, even the same frames as we already have from the SR-W5U -- since we don't have those frames with different settings flipped. Nor could I find any info on how gshelley61 had it set up at the time. Upon comparison, the DNR feels over aggressive to me. At least I assume it was turned on for his shots. If not then I really don't know..

Yeah, true. I just thought that since ATI 600 is cheap, and commonly accepted as a good card, it wouldn't be a bad idea. Particularly since many members on here already own one. It'd be one less variable, and an important one IMO. But if it's too restrictive then so be it.

I wonder if we could find some place to do this conversation via chat, since a few of us seem to be online at the same time.

Lots of good analysis in your post. As far as testing for non-TBC tapes, I've done probably 15 or 20 hours of captures on a plain VHS Panasonic PV-9xxx series VCR and maybe a few hours less than that on a VHS Panasonic PV-7xxx series VCR and found no issues with the AGC. I've used it in conjunction with an ES-10 as well on my JVC SR-W5U with the TBC disengaged. 95% of those were Hi-fi EP tapes recorded off cable TV in 1997/1998. I suppose the triggers for the issue could be more complex though as for why some people seem to be hounded by it and others aren't. It's a mystery to me if it's due to the AGC specifically.

Sure, that's fine with me.

The SR-W5U is a TBC ON / DNR-ON, TBC OFF/DNR OFF system so it's likely engaged there. If I recall correctly though, he captured with a DVD recorder, which may have had it's own DNR on top of what the VCR does (and obviously not lossless). I did do a thread on Videohelp comparing the SR-W5U and SR-V101US, but that also used a JVC DR-M100 DVD recorder.

I'm going to walk outside right now, and see what we've got. Back in a bit. :)

If anybody wants to chat at any point, I set up an IRC channel -- it's available via this link: http://webchat.freenode.net/

The channel name is #VCRcompare

Alright, I went and dug Titanic out of storage. It's the two tape, full-screen version, ISBN 0-7921-5171-2 / UPC 09736-33481-39. There is Macrovision protection on both tape 1 and 2. I have to say, I've seen better VHS pictures too for a movie, it looks a little DNR-y out of the box. The SR-W5U has a full-frame TBC built in, so I don't have to use my DataVideo TBC to get a capture (the 750HD does not ignore MV).

Let me know which version of Flashdance we're using -- there appears to be at least four or five different releases on VHS.

Edit: Just for fun, I went ahead and made a quick screenshot, though I don't think this particular frame was part of GShelly's series in the one thread, he did use it later on in another thread (http://forum.videohelp.com/threads/1...=1#post1105897). My own personal commentary is that overall I like how it is rendered on my deck, but for this movie I think it's a touch overly sharp and could use a little bit of temporal filtering.

Equipment chain: JVC SR-W5U VCR (TBC ON) --> DIAMOND ATI Theater HD 750 USB

Edit: Oops, uploaded the wrong image, that one is with my PA-200 in-line, in unity mode. The direct image is attached.

Does anybody have a link to GShelly's thread on VH with the images we'll be using?

Also, to take advantage of the site's before and after feature, here is the original captured image as well as the original image with a little bit of temporal filtering (a more pleasing look IMO)

Edit: ... if I could figure out how to do it :p How do you upload an image directly to the site (versus creating a link to an image hosted on another site, per the guide) without adding it as an in-line image?

Alright trying this again:

Here we are!

Admin refused to unhide hidden content
nothing will help to see hidden content

Subtle difference, but I think it's a little bit more pleasing on the eyes, especially the cheeks, at least as a still.

Okay, here are the others from the Titanic VHS (not resized). Not 100% sure I grabbed the exact right frame for #1 and #3.

http://www.digitalfaq.com/forum/atta...1&d=1363827992
http://www.digitalfaq.com/forum/atta...1&d=1363827992
http://www.digitalfaq.com/forum/atta...1&d=1363827992

Looking at the GShelly61 pictures again and judging by the darkness of some of his images (especially that captain shot which is downright murky), I think he (or possibly I) may having issues with the MV slipping through with his combination of equipment. With the exception of the captain shot and the morse code, I noticed that he chose shots with very sudden, bright light sources, presumably to compensate for the very dark images otherwise in those scenes. I own three SR-W5Us, and I believe the one I have hooked up at the moment was his deck although I'm not so sure now. I'll do some captures with the other fully working one with the same material to see if it has any issues with the MV on the Titanic tape. I've had a few problems with tapes where I had to toggle the TBC on and off a few times to successfully defeat the MV. I'll probably redo all of these once I calibrate my card, but I thought I'd get us started.

Edit: It looks like he used a Sima Color Corrector in the chain for some of the images

I pulled out 8 tapes that I can work with:

• Teenage Mutant Ninja Turtles
• The Rocky Horror Picture Show
• Crocodile Dundee
• Lethal Weapon 2
• Hook
• Kindergarten Cop
• Home Alone
• Dick Tracy

These are what I can work with. I can't do Titanic. If we pick one of those, I'll be part of the test. :)

I'm gonna visit a thrift store nearby and see if they have any of those.

Don't think I have any of 'em but happy to locate them. Any barcodes so we can assure the versions/prints are the same?