Any thoughts on Retrotink based on new profiles?
 

This is a follow-up to the previous thread.

RetroRGB recently released a new video discussing the passthrough profiles for the Retrotink

I did find this Dave’s Archive video that somewhat compares the Retrotink to a Key West Big Voodoo BVTBC8, but the details on what mode(s) were used are unclear. Additionally, he doesn’t compare them side by side; instead, he feeds the TBC into the Retrotink. It is something I guess.

Has anyone here done (or seen) a direct head-to-head comparison? What would be a good test to see how this thing really fares?

I have more to reply, but for the moment, I'll leave here what I wrote in a reply there:

I do plan to do that test eventually along with testing every other often recommended method and I will be testing both the 4K and 5x.

The whole color accuracy and luma/chroma clipping will be actually very easy to see and test. You just have to look at an analog waveform monitor/vectorscope of color bars as it enters the RT as S-Video (or Composite) and then use an HDMI to SDI converter with SDI waveform monitor/vectorscope to look at the signal coming out of it. They should match. You probably don't need a hardware SDI vectorscope/waveform monitor since you're capturing that signal which is already digital - so you could just look at the captured file and pull up vectorscope/waveform monitor for that file.

Converting from SDI to HDMI or vice versa is a lossless digital to digital conversion, so adapting the HDMI to something a SDI scope can use will not result in conversion losses.

The Retotinks do have proc-amp settings, so if for some reason the luma/chroma is always too hot, you could dial those in just by looking at known good color bars and adjusting levels you see on the captured file. The question is what it does when it receives a signal that is already too hot and if it has any automatic gain control to automatically get luma levels back into legal ranges. It's very common, particularly with commercial releases, for luma levels to be 130IRE or more (when max is supposed to be 100IRE). Depends a lot on the specific tape and VCR though. Heck, my DVD player outputs whites at 105-110 IRE, and that's using a digital source to base those levels on. Some capture cards will just clip values above 100IRE and they'll all show as being at 100IRE which leads to detail loss and blown out whites.

SDI and HDMI do have legal ranges as well and there are devices out there that are called "legalizers" that I suppose are their own form of automatic gain control. Or it could be that they also just clip values that are too high so that the signals being broadcast are legal, but they could still be clipped. I have at least one SDI legalizer that I'll be testing for how that works as well.

Odds are if the community pointed out to Mike Chi (the developer of Retrotink) that AGC needs to be implemented better, he would probably add that feature if the hardware can support it. It's basically a realtime variable proc amp, so that's probably a lot less complicated than a lot of the other things the Retotinks can already do.

I forgot to reply to this earlier (got buried in tabs), but I did reply to a similar post in Dec last year.

Specifically:

Do not use a video game capture card for video tapes. Both are "video", but not the same. No more than a cat is a dog.

That's a terrible comparison, because I don't think he has the BV TBC setup correctly. I would wager money that he has the comb or pedestal set wrong. Or more likely, he never set anything, just plugged it in, didn't RTFM.

Dave says he got his BV10s for $20 dollars a piece in 2017 at 7:45 in this video https://youtu.be/9W37qO4fnjQ?si=D-h7HMKUbZetRJB1. I hear the wrong gen of BV10 can be really bad. Who knows what kinda shape his are in.

Well based on this tread -- it sounds like I need to do my own Pepsi Challenge. :wink2:

You can loan it to me when you're done. :)

I'm about tapped out on buying test/dev gear right now. All funds are being put into the new forum updates currently.

eBay algorithm has been learning a lot, Now anything has a TBC word in it even if it's a T shirt will be listed for hundreds if not thousands of dollars.

I take it as the Keywest BVTBC10 is a 10bit machine and the BVTBC8 is a 8bit machine, and I believe the 10 has SDI digital output.

No, it does not. :old:

In the beginning, KW had multiple BV 10 variations, with 95%+ of them being baseline models. No SDI. There was the base, mid-range, and pro/premium models. Only the pro models had SDI, and it cost almost double. At the time, mid 2000s, SDI was losing favor. The production run on those was tiny. By the end, they mostly just cranked out base models with no real QC.

I accidentally sold an SDI model to a member here, but let him keep it. (I forget who off-hand, enjoy!) I have enough for myself, and would only consider selling one off the the right price (as I'll probably never get another one). And preferably to a fellow archivist.

I've met current/former DataVideo and Cypress (mostly TVOne) employees, but have yet to find somebody that worked at Keywest. Also met former SignVideo/Studio1/Vidicraft employees (and I believe the company founder, too). It was such a tiny operation that I probably never will. (Note: Keywest the company still exists, but it had nothing to do with the company of the 2000s. DataVideo is actually quite comparable in this regard, having largely trashed all TBCs and most related internal documents by the mid 2010s. I spoke to the TPTB there around 2016, and he said all "TBC era" employees were long gone.)

BTW, Gary34, if you're reading this, this is why a 5th gen 3000 model was such a surprise to me. Nothing in conversations ever suggested such a late production run. (DataVideo, unlike BV and Cypress, got better as time went on. Mostly.)

I can't believe those conversation were almost 10 years ago. :huh1:

This 2 hour review of RetroRGB's video on the Retrotink is hoot.

https://www.youtube.com/watch?v=6PMRHDFBeQw

I prefer direct comparisons done on the same source material with the device being reviewed in-hand and tested rather than a review of a review. There are some valid points though for sure, so it is still an interesting watch and I'd recommend anyone to watch it for the perspective alone.

It is true that the retrotink is not a capture card, it's an analog to digital converter, so one does still need an HDMI capture card that can handle 480i which does add additional cost to something that is already fairly expensive which is a fair point. The advantage of such HDMI capture cards is that most are compatible with modern OS's, so you can use it with any decently spec'd modern PC.

What I have not seen is a video showing that, for example, the TBC-1000/3000 is superior as a frame TBC in terms of dropped frames or audio sync problems. I'm not sure if the retrotink processes at 8 bit or 10 bit, but we know for sure that the recommended TBCs process at 8 bit color depth, so that shouldn't be hard to at least tie if not beat. The highest bit depth and processing chain I've seen would be the BrightEye75 which has 14 bit ADC/DACs and 12 bit processing, though I haven't seen a side by side compared to the recommended TBCs for that device either. How good/accurate the DACs are don't really say how well they handle unstable video though, so those specs on paper don't tell the whole story of "best device". 12 bit is overkill for analog video in my book, but there can be visual advantages to at least 10 bit in certain scenes where they are mostly the same color like very dark scenes or sky/sunset where you'll get banding of similar color values that are near each other end up rounding to the same value in an 8 bit palette.

I am personally undecided where the retrotink falls in terms of best/worst devices/chains one could use for analog capture. I suspect it is in the realm of the 80th percentile or greater of what's out there, though that's not particularly hard to do given most people buy modern devices off Amazon that do not preserve interlacing or have sufficient bitrate.

My observations:

- Sometimes you overthink things. For example, bit depth is mostly just "measurbating" (as spewed by the product marketing dept), and not actually anything useful. With VHS, you're dealing with a source that's on par with 6-bit dithered (sub-8), so 8/10/12/14/10000 means nothing.

- Sometimes you put far too much faith in "videos" (Youtube) from random strangers. They have no background in video, no perspective, no understanding of concepts involved. It's simpleton grade-school "seeing" without comprehension. Monkey-see, monkey-do. Anything they can't "see" is unfathomable. Those are the kinds of people that are easily misled (ie, fooled by optical illusions, think magic is real, etc), and the mouthy ones spread the FUD/BS/myths (mostly for the sake of appeasing a Youtube algorithms, for money, which has zero care about facts/reliability/etc -- and NOT to actually help others, as we all do here). And FYI, I often actually lose money by having this site online. So not only do I not get paid by anybody, it's costing me funds. I could easily be a Youtuber jackhole that just spews BS so I can grab myself some Google $$$$. But that's not who I am.

Very often, the difference between a Youtuber, and a gear user, is the silly idea that something can be "reviewed" on a weekend afternoon. They gather half-ass information, make a half-analysis, then smile at the camera. Whereas I, and many others, have literally been using the gear discussed for decades. Decades. We've seen more scenarios than we can even remember.

RetroTINK is fine at what it was made for. It wasn't made for VHS.
So how well does it fair, you ask?
- For me, salad was a good supper tonight.
- For my cat, Purina was a good supper.
- For me, Purina would be awful.
- For my cat, salad would be awful.
- RetroTINK for VHS is like me eating cat food. Wrong tool.

If only I had 2 hours. I've started to "rip" Youtube videos as scripts with Gemini (also telling it to add punctuation), so I can skim the text.

But even the first few minutes, the title, and the description, make me want to watch at least part of it. :)

... but seriously ... 2 hours? :huh1:
I really do think a lot of the Youtube/podcast/Tiktok types have zero concerns in life. No job, good health, no family, etc. I cannot imagine watching a Youtube video that is literally longer than a good movie (which I also don't have time to watch; I still haven't seen all of the Harry Potter or John Wick movies, to name a few).

I like it a lot. I’m glad you helped persuade me not to get a TBCish setup. It’s nice having an actual TBC. I don’t see flaws in this myself. The generation is soo rare that I don’t worry about the generation part now. What I say about it is my opinion because no one that has a reference has used it but my opinion is really high.

If a good BV10 is that rare and the difference between good and bad is as drastic as I hear it is then the eBay prices are way too high.

I watched 15 minutes. The part at the beginning where the retro RGB guy says if that’s still not enough timebase correction then you can get one of these and it’s the TBCs that have been sitting on eBay forever that everyone has passed up on. There should be some kind of warning with that I think. All that stuff has different generations and the condition is unknown. The TBC 3000s that are marked for NTSC can be preset to PAL or vice versa. You can’t trust the sticker. There’s more to it than just grabbing one real quick from eBay I think.

RetroTINK is fine for VHS depending on your goals and expectations. If you already own one because you're a vintage console gamer, and you also happen to have a VCR with some tapes you want to digitize, you could do far worse. Many of with devices sold on Amazon specifically marketed for VHS capture are a visual catastrophe.

Will it work as good as high-end video signal processing equipment tuned by experts to professionally capture a VHS tape in lossless codecs for the United States Congressional Archives? No. It all depends on what you're trying to do.

I'm just sharing old commercials from deteriorating tapes, so teh RetroTINK 2x Pro hits the price/performance tier I'm happy with (for now). For my personal home videos, I might wait to capture those with higher-end devices.

I remember we had a thread about that. That’s why I was getting the Deja Vu with the Dave’s archive video. https://www.digitalfaq.com/forum/vid...-dmr-es15.html At least you are getting better gear for your home videos. Hope it works out well.

To me it just seems like a bunch of false claims.

For my purposes, my modest setup is probably overkill.

Which claims are false?

The time base corrector claim.

Yeah - I have no way of knowing what sort of video voodoo is going on with the RetroTINK 5X and 4K... in terms of TBC. I'm using their budget 2X Pro purely as an analog-to-digital converter in my setup. Most every thing on the channel I just launched was captured with the stuff - https://www.youtube.com/@Retro480i ...but again, this stuff is more for fun than saving cherished memories.

My plan is to build a video workstation desktop this summer and switch over to a higher-end capture card. For now, I'll make do.

Sounds like a good plan.

FWIW, your videos look pretty good to me.. What's your current workflow (besides the Retrotink)?