TBC in audio-video A/V receivers?
 

I am curious if anyone have tried using TBC in audio video receivers. I came across a few models from Denon and Marantz that have TBC. The Marantz 7500 I believe has line TBC. Also I have Onkyo TX-SA805 receiver that can transcode from composite video to S-Video and Component video. I decided to look into what is inside the receiver and downloaded service manual for it. Tracing the composite video signal I can see that the signal goes through video processor (fli8125-lf-bc) for monitor output (it could also bypass it).
Looking into data sheet of fli8125-lf-bc I can see there is a TBC in the block diagram.

https://postimg.cc/SjXtJ1rh
https://postimg.cc/SjXtJ1rh

I understand the fact that it exists doesn't mean much. Are there any known good TBC's in the audio video receivers that would be worse considering? Some of the old receiver cost nothing nowadays and perhaps could be used as TBC device. :D

Agree it'd be interesting to see how these perform. Since you've got one, have you tried it for capture and are you getting many dropped frames?

I have tried it today. Connected to my capture card from S-Video and the VCR was feeding AV receiver with composite video. I have analyzed the capture after performing deinterlacing and noise reduction in hybrid. There is an improvement to picture stability. Going frame by frame I have been concentrating my attention on vertical objects (poles/buildings) and I can see small distortion with direct capture in some parts (but have to look really hard for those going frame by frame). Though there is one big drawback. The picture routed through receiver (TBC) has additional frames on right and bottom. With bottom also having horizontal line across the entire image. I can remove that by cropping image, but figured something is not right. I have tried rebooting PC and using another VCR, but it made no difference. With another VCR I cannot even get stable image through AV receiver, I see periodic tearing happening. The black lines are not visible when connected to TV (though only tested with composite and component video since TV doesn't have S-Video input). I hope it helps.

What you must understand is that the term "TBC" (time base corrector) is wide, encompassing many technical uses. In some cases, it's outright distorted, the meaning abused.

The term "vehicle" is about as descriptive as "TBC". Car, boat, plane, unicycle? :laugh:
Further definition required.

Also, please attach images to post -- do not hotlink other sites. :2cents:

It's always interesting.
It's almost never impressive.

A lot of these "also includes" type TBCs are weak to do-nothing. This is a known known, and has been for decades. Every few years, somebody gets excited to discover that a device "also has a TBC", but then that conversation turns very quickly.

Receivers, like DVD recorders, simply are not TBCs. :o
No more than my toaster is a TBC.

Can you post a screenshot of the image artifacts that it is adding? "The line at the bottom" could be some sort of masked head switching noise which may not technically matter in terms of viewable picture area. I'm not sure how to visualize "extra frames at right and bottom" either, but guessing a screenshot would help.

Uploading images to the site option is not working for me. When I select the file and click upload I get this error. "Sorry, you have been blocked
You are unable to access digitalfaq.com"
I have tried Safari and Chrome browsers. Link to the screenshot of the error message I am seeing. I am uploading screenshot in png format. The one I just tried was 78kB.

TBC on this receiver certainly brings some benefit and can be seen. The video is more stable for lack of better word, though additional boarders added to the image are not great. I can certainly remove them in post processing but it could introduce other issues. The best way to test would be to record some test patterns from test DVD on VCR and then play them for comparison purposes. I may get to it one day.
I understand your skepticism in relation to TBC bargains. LOL. It is just I haven't seen many discussion on TBC subject related to AV receivers. Is that because there is nothing to discuss or simply the area hasn't been explored... Looking up service manuals and chipset data sheets should not be difficult, but it doesn't gurantee the correct implementation was enabled in FW.

-- merged --

one of the 3 VCRs I have is doing a pretty good job of not showing any major issues that I could see without cherry-picking scenes and doing slow motion (frame by frame). The improvement would be visible if I was to make a short video but not possible to show on the still images (unless several images of each frame were taken and you were switching between them back and forth).
Here is the link to the sample screenshot with boarders added. Once again sorry for providing links to third party image hosting sites as I am having problems uploading images.

Hmmm. Looking at the image, I don't really see the boarders you're describing. The black vertical columns on the extreme right and left are considered normal for most SD formats as about 16 pixels of total horizontal space represent some of the horizontal blanking area and are expected to be there. The green line at the very bottom I suspect is related to some sort of head switching noise that the receiver doesn't know what to do with and most SD captures you'll find will have that to some degree, though usually it's more of a black and white static-like lower few lines.

Here's a bit of an example of what head switching noise tends to look like - would just look at the extreme bottom of the video frames:
https://www.avartifactatlas.com/artifacts/head_switching_noise.html

Here is the screenshot of the same video captured bypassing the receiver (VCR to capture card via composite video). There is no green line on the bottom and the boarders are smaller. There is static black/white line at the bottom though with the direct capture.
Using crop view in hybrid I have to cut 4 on bottom and 4 on the right side to get rid of black boarders on direct capture (cannot do odd numbers in crop preview). With TBC capture using receiver I have to cut 8 and 6 on bottom and right side to get rid of the boarders.
Another observation is when watching VCR through receiver on the TV with component video transcode done by receiver colors look more saturated. There is more noise in the video as well and I have to use more aggressive noise reduction on the TV. I am using Samsung TV to watch this on and it has "digital clean view" noise reduction. With composite video direct VCR connection noise reduction in low mode reduces all visible noise. It is impressive how well the vhs looks and TV doesn't have any issues with interlaced signal :salute: In order to achieve similar noise reduction results I have to use MC Temporal Denoise in hybrid set to very high.
I am curious to explore ways to bypass deinterlacing all together and try re-encoding raw captures(HUffYUV) to mpeg2 to make them compatible with TV build-in player, but not sure what program would be suitable to eliminate the noise. The raw image is simply unwatchable.

Since I was intrigued by this post, I did end up grabbing a Marantz SR8500 which has the same claimed line TBC in the marketing materials as the SR7500, but also has DVI output, so technically it can act as an analog to digital converter while potentially doing line-tbc-things - maybe.

You'd think it'd need to have pretty stable digital output in order to offer DVI as an output option and the back does label a lot of it's ins and outs as being for VCRs, so you'd think it should handle unstable (VCR) inputs. The fact that it mentions "TBC with line memory" in the marketing materials does suggest that they knew this was a problem that users might notice and be willing to pay a premium for - or it could just be a marketing shenanigan.

Another interesting feature is that it appears all inputs have passthroughs, so as long as you have two displays or capture devices that are the same, you could see if there's any visual improvement in picture stability in realtime versus going through the device versus not without needing extra devices to split the signal.

Lastly, seems at least some models had a "lip sync" feature which suggests that it might be able to add audio delay of up to 200ms (1/5th of a second) if necessary. That's a handy feature, though always something you could do in post when it comes to capturing.

Will it be any better than something like an ES10/15 perhaps without as much posterization? Probably not, but no way to know without trying.

That's great. Keep us in the loop on your findings.
Onkyo 805 also has a bypass. Transcoding only works for Monitor out only, not other video outputs (at least that is what it said in the instruction manual if I am not mistaken, but it appears that video processing is still possible as there are switches for video outputs as well with signal routed through video processing board).
You can see how the composite and s-video signals are routed here at the high level.

Here's some preliminary pictures of the inside of the SR8500 - at least it appears just has a few SMT capacitors on the DVI output board only. Not exactly sure if the SAA7115HL chip (swear I've seen this in some other devices?) is what actually is doing the "line TBC" things, but the data does mention a FIFO line buffer (first in, first out).

This particular unit has a bulged capacitor on the power supply, so I'm going to try to swap that first and hopefully I have a similar one around. I'll also swap out the 6 or so SMT caps while I'm in there whether they are bad or not.

From the manual:

8.3.3.1 Line FIFO buffer (subaddresses 91H, B4H and C1H, E4H)
The line FIFO buffer is a dual ported RAM structure for 768 pixels, with asynchronous write and read access. The line buffer can be used for various functions, but not all functions may be available simultaneously.
The line buffer can buffer a complete unscaled active video line or more than one shorter lines (only for non-mirror mode), for selective repetition for vertical zoom-up.
For zooming up 240 lines to 288 lines e.g., every fourth line is requested (read) twice from the vertical scaling circuitry for calculation.
For conversion of a 4 : 2 : 0 or 4 : 1 : 0 input sampling scheme (MPEG, video phone, Indeo YUV-9) to ITU like sampling scheme 4 : 2 : 2, the chrominance line buffer is read twice or four times, before being refilled again by the source. It has to be preserved by means of the input acquisition window definition, so that the processing starts with a line containing luminance and chrominance information for 4 : 2 : 0 and 4 : 1 : 0 input. The bits FSC[2:1] 91H[2:1] define the distance between the Y/C lines. In the case of 4 : 2 : 2 and 4 : 1 : 1 FSC2 and FSC1 have to be set to ‘00’.
The line buffer can also be used for mirroring, i.e. for flipping the image left to right, for the vanity picture in video phone applications (bit YMIR[B4H[4]]). In mirror mode only one active prescaled line can be held in the FIFO at a time.
The line buffer can be utilized as an excessive pipeline buffer for discontinuous and variable rate transfer conditions at the expansion port or image port.

You can download the service manual for Marantz 8500 here. I would not try to trace the signal manually.
SAA7115 is used in some video capture cards, based on what I have seen. I have no idea what performs TBC in that receiver. Possible SA7115, but looking at the block diagram I don't see TBC specifically being called out.

Reviewing the service manual, agree a separate TBC is not called out (don't have HiFiengine access, so found it elsewhere), so I'd have to assume the SAA7115 is doing the line TBC function described. I can't recall what other devices I've seen it in, but perhaps some Snell and Wilcox format converters/TBCs?

But that function is not in the datasheets. And I call BS on it being programmed in for a mere consumer-level stereo AV reciever.

This is why use of the term "TBC" is so bastardized.

... and why my toaster can probably have a TBC, too! :no1:

Given the discussion above, I would bet that "TBC" was used to describe the "buffer" -- which is hogwash. A buffer is not a TBC.

Now, those Philips/NXP chips can indeed be used to extract data for latter timebase correction (inside actual TBCs), but the chips themselves do not provide that function. Such a setup has costs, and is wildly outside the scope of an AV receiver. So odds of it existing are 0%.

Attached is the 7115 datasheet. The HL may be slightly different, but it's mostly physical spec differences.

The root cause is always PLL, and people not understand that it's not synonymous with TBC. PLL in itself can have the side effect of nominal line correction, but it's not true nor reliable. PLL is very basic, infantile even, in that is operates at a signal level with no understandings of visual outcome. Whereas line TBC is specifically aware of visual outcomes.

.............

^ FYI, I actually ran that by AI, "analyze this writing for accuracy", just to make sure I didn't make any pre-coffee mistakes. I did not.

And I think I'll throw in a "bullshit" just to round it out. :laugh:

HiFiengine account is free. Very helpful resource for sourcing instruction manuals, service manuals and various catalogues...

Have you done an actual test though? In my case the results are obvious as once you know what to look for you cannot miss it. When you run a test make sure to test with your capture card as you may not notice the difference if you use your TV (I haven't done in depth evaluation this is a pure speculation on my side). When displaying more or less static images look for vertical objects. Also what is quite apparent in my case are the shots with slow panning motion. The building/s in the background is/are simply more stable vs being jittery with direct capture. The effect is similar to using image stabilization in post processing, but without "digital" artifacts. I bought JVC GR-DV1 camcorder in 1998, most of my recordings are still on DV tapes. That camera didn't have 1394 output. The tapes cost around $15 a pop for 60 minutes of video. I have recorded over some tapes and archived recordings I didn't care much for on VHS tapes. In some cases I have both DV and VHS copies and it is easier to compare the difference.

I am skeptical as much as you are:
1) does it do anything?
2) if it does, what are the side effects?
3) are the tradeoffs worth it?
On another side, considering Marantz SR7500 was released in 2005, why would marketing rave about feature that could only benefit VHS playback when VHS was practically dead in 2005. Perhaps this was the time when everything had to be upconverted to DVI (dying at that time)/HDMI and the lack of TBC was apparent during the playback...

I am open to suggestions what test material should be used for proper evaluation. TBC aside, does the processing affect resolution/color accuracy... Will need to confirm whether composite output on my receiver goes through TBC as it would make comparison easier. I could record test patterns on VCR. I may already have macrovision free copy of the calibration DVD I made back in the days when I recorded test patterns on VCR Pioneer PRO-730HDi TV required a special approach to calibration as composite video input offsets were throwing off component video input calibration. That is if you calibrated component video first and then decided to make changes to composite video, the component video calibration was messed up...

Agree HifiEngine is supposed to be free - but if you try to make a new account, it won't let you, most likely because people probably started selling manual PDFs from there on ebay or similar when they were meant to be easily findable for free.

I've currently got the unit all apart since I saws that obviously bulged power supply capacitor and I've now got the 4 main output capacitors ordered and on the way from Mouser, so probably won't be able to get it back together until early next week. The unit was also sold in untested condition, so there's always a chance that it doesn't work. We shall see.

The FIFO (first in first out) line memory that the Phillips chip describes could be considered a line TBC since that's really what a line TBC does, buffers one or more single lines at a time on a rolling basis and then outputs the stored data at a digitally precise rate. Near as I can tell, there's no separate RAM for this function, so I don't expect it to be particularly robust, but it could be possible that it starts all lines at the same spot on the left which is the main critical function of that. I wouldn't be too surprised if this chip is inside of many TVs of that vintage and perhaps why those TVs don't have visible line errors on unable signals whereas capture cards do.

No, that is not correct.

That's an easy tell for "no TBC".

But that's going to help those sellers sell more because people can't get the manuals for free, If this is their strategy then it's failing. Manuals are by the members for the members, no one should be barred from accessing them, it's not like they are paying for them.

What is the definition of a line TBC if not:

"A rolling buffer of one or more lines and then releasing each line aa digitally precise rate (hence also replacing the original horizontal blanking interval length)"

If only storing data for a line or two, it is possible that the memory for a couple lines could be integrated into the chip (rather than having separate RAM) and that seems to be what the manual for the Phillips chip describes.

buffer = buffer
digitally replaced = digitally replaced
Still not a TBC.

Processing is what makes a line TBC not a mere digitizer/buffer. Zero mention of the processing.

That chip digitizes into a buffer, but that's all it does. The PLL isn't TBC. See also why the Analog Devices chips are not TBC. That's why the AD's function was disabled on most devices, as it tends to make more problems, not correct anything reliably. Again, basic, infantile.

Use JVC as a blueprint. You can see a difference in responsiveness between 2mb and 4mb early HR. Anything less than 2mb was likely deemed subpar. I can't imagine that IC has enough integrated, at least for the age/era of the chips discussed here.

I still have to call BS. These terms are far too liberally used. Everything doesn't have TBCs.