ReClock 1.8.7.7

[JackCarver]

I'm a bigger jerk (23.97593).

wow that's a really good value! What kind of graphics adapter do u use? Nvidia or AMD? I use a Nvidia GTX 560 Ti.

What do u mean with that sacrificing audio for video? Is it that there is the possibility to get an asynchronous sound to the video or is it that there will be sound errors or sth?

I didn't realize anything of that in movie avatar but perhaps i didn't simply hear that or my loudspeakers (JBL) are too bad to make it audible

I will test this setup for further more movies.

 

[andy o]

Well, just use the resampler and output multi channel PCM. It works amazingly well. And you won't sacrifice (almost) anything.

Yeah, that's what I'm doing now. I ran it with bitstreaming when there was no easy DTS-HD solution, and then I decided that instead of worrying about the odd soundtrack that would give me pops, I'd just take a video frame error once every several tens of minutes.

 

[andy o]

wow that's a really good value! What kind of graphics adapter do u use? Nvidia or AMD? I use a Nvidia GTX 560 Ti.

What do u mean with that sacrificing audio for video? Is it that there is the possibility to get an asynchronous sound to the video or is it that there will be sound errors or sth?

I didn't realize anything of that in movie avatar but perhaps i didn't simply hear that or my loudspeakers (JBL) are too bad to make it audible

I will test this setup for further more movies.

If you don't notice anything, then do you really want me to tell you what to look for? (If you want you can read my previous post where I say exactly what happens.) In any case, the better way IMO is to decode everything, and let ReClock do its thing as it's intended.

I use an ATI 5770, and with other ATI cards 4670, 4550, 23.976 has been pretty accurate with me. Might depend on your AVR/TV too. I have a Pioneer KRP-500M.

 

[JackCarver]

I watched Star Trek 11 on Blu Ray (Dolby AC3 bitstream) and to sum it up I can say that i will further use bitstreaming. I don't realize any sound errors with my 5.1 system, the sound is clear and no delays or sth in soundtrack or speech or i can't hear them.
So IMO bitstreaming works with Reclock pretty good on some systems on other systems it doesn't.
I have a sandy bridge based system perhaps it's the reason that it works don't know but i would say if it works, Bitstreaming is the better choice cause sound output will be better as decoding is only be done by the AVR and there's no preencoding by the system (if i'm right)

 

[andy o]

I have a sandy bridge based system perhaps it's the reason that it works don't know but i would say if it works, Bitstreaming is the better choice cause sound output will be better as decoding is only be done by the AVR and there's no preencoding by the system (if i'm right)

That's right only if you're using SPDIF audio (which can also be sent through the HDMI cable). But with lossless audio, which many (most?) blurays use today, instead of bitstreaming the backward-compatible DD track or DTS "core", you can decode the lossless audio instead, let ReClock "fix" it, and then let ReClock encode it to DD. You'll get similar audio quality that you're getting bitstreaming the "core", but at the same time you'll get media correction by ReClock.

BTW, Sandy Bridge is not the reason it's working. You're just not noticing the errors, which is fine, I didn't most of the time either.

 

[JackCarver]

Yes i'm using an optical SPDIF conection to my AVR and i use LAV Audio Decoder with MPC-HC. There u can choose between Bitstreaming Dolby Digital (AC 3), DTS, DTS-HD, Dolby Digital Plus and True HD.
I can choose the first two as my AVR (Harman Kardon AVR 130) can't decode DTS-HD. I think it's simply too old.
If i will get sound errors which i can hear then i'll go back to the PCM way and i think i have to choose there "use ac3 encoding" to get surround sound.

 

[Zardoz]

Do not worry, this shows that reclock is working fine on your system. You have to know that James is not very fond of the bitstreaming option and due to some careless remarks some people now think that it does not work or is in some sense inferior to resampling. The reality is different - both approaches have pros and cons. Bitstreaming if used correctly will work great. Here is my take on the pros of bitstreaming:

- no degrading of sound quality due to unnecessary decoding (and even reencoding if reclock is set to ac3-encoding)
- most if not all receivers have limited bass management/delay/postprocessing capabilities if fed raw pcm instead of the original bitstream (f.e. on my flagship denon model some of the options are still available but do not work anymore)
- channel mapping is an issue if some receivers do not decode the stream by themselves
- Older receivers with only spdif inputs do not allow for multichannel pcm input, but work happily with the DTS/AC-3 cores of the HD formats

on some of the myths of packet drop/repeats:
- if the monitor refresh rate is very close to a multiple of the movie-framerate you will rarely get any drop/repeats. In fact I get none for the duration of a whole movie (except for the obligatory counted one at the start of playback)
- while repeats are audible and bad, drops are not. DTS and AC3 were designed to be error resilient in the case of corrupted or dropped packets
- so when using a monitor refresh that favors drops over repeats you will probably not even notice the adjustment happening
- packets are usually very small (e.g. DTS = 11ms) so that audio video delay is not an issue; delay of your projector, aero, postprocessing and other factors have a much larger impact


I have been using bitstreaming in reclock for years now (since the ogo days) and am quite happy with it.

Hi there,
it's strange but i think Reclock works for me in Bitstreaming mode or something else is going on.
Madvr stats show in movie Avatar following values when Reclock is on:

Display: 23.9716 Hz
Composition rate: 23.976
Clock deviation: -0.01865%
Movie: 23.976
1 frame drop every 13hrs

I watched the whole movie Avatar and Madvr stats says not 1 dropped frame over the movie
Sound was DTS Bitstream via LAV audio, the display of my Harman Kardon AVR showed 5.1 channel DTS and Video and sound was in sync over the whole movie.

When i disable Reclock then there are following values:

Display: same as above
Composition rate: same as above
Clock deviation: 0.00156%
Movie: same as above
1 frame drop every 3.40 minutes


So a very big difference in the frame drops. So does this mean that Reclock works or are there other reasons for this difference?

Thx

 

[Zardoz]

Yeah, that's what I'm doing now. I ran it with bitstreaming when there was no easy DTS-HD solution, and then I decided that instead of worrying about the odd soundtrack that would give me pops, I'd just take a video frame error once every several tens of minutes.

This makes no sense, if you "take a video frame error once every several tens of minutes" why use reclock in the first place? With or without bitstreaming reclock sole purpose is to sync the video stream to the monitors refresh rate in order to avoid frame drops and repeats. If you can live with these you do not need reclock at all. I think you should be careful of spreading misinformation to JackCarver, about Reclock supposedly not working for him, when in fact it is.

 

[andy o]

This makes no sense, if you "take a video frame error once every several tens of minutes" why use reclock in the first place?

It has other features, and not all formats are bitstreamable.

With or without bitstreaming reclock sole purpose is to sync the video stream to the monitors refresh rate in order to avoid frame drops and repeats. If you can live with these you do not need reclock at all. I think you should be careful of spreading misinformation to JackCarver, about Reclock supposedly not working for him, when in fact it is.

That's not true at all. See all the other checks and options in the ReClock settings panel? Those mean something, they're not just for show. One obvious example is getting WASAPI exclusive output with your DirectShow player of choice, most of which don't offer the option.

And if you must talk about a "sole" purpose, it would be to prevent both audio and video errors. Bitstreaming was added at the request of some people, when decoding wasn't all that viable.

 

[andy o]

Do not worry, this shows that reclock is working fine on your system. You have to know that James is not very fond of the bitstreaming option and due to some careless remarks some people now think that it does not work or is in some sense inferior to resampling.

It is, measurably, and perceivably. In some cases less so than others, to the point of being irrelevant (that depends on the user), but that's the cold hard truth.

Here is my take on the pros of bitstreaming:

- no degrading of sound quality due to unnecessary decoding (and even reencoding if reclock is set to ac3-encoding)

True in some cases, if you're using SPDIF, (which Jack Carver was, I know), but not with HDMI. The sound quality hit (if any) would come from the AC-3 re-encoding itself, so I don't get the "due to unnecessary decoding" bit. Also, lossy encoding is more likely to produce artifacts if you're lossy re-encoding an already lossy stream. If you're lossy-encoding a lossless or uncompressed stream such as most blu-rays offer nowadays, it's probably imperceptible. Lossy encoding gets a very undeserved bad rap when actually, when done properly, it's usually transparent (codecs like mp3 can be brought down by so-called "killer samples", but they're not regular music/movie soundtracks, and the listener needs to know what to look for - no "golden ear" nonsense here).

- most if not all receivers have limited bass management/delay/postprocessing capabilities if fed raw pcm instead of the original bitstream (f.e. on my flagship denon model some of the options are still available but do not work anymore)

Is an oft-repeated claim for which I haven't found any evidence, or even reliable anecdotes. Every time I've asked the person making the claim, either they don't own the receiver they're talking about, or they don't respond back. You're the first one to say you own one, so: which processing options are not available to you, and which model do you have? And how do you arrive from your one receiver, to most? I have tested at home Harman Kardon (AVR254), Onkyo (606) and Pioneer (VSX-01). I think one can safely assume that other models of the same brands behave similarly.

The only thing I can think of is for legacy DTS, if you have a 7.1 system, my Pioneer won't let you apply DPLIIx, but instead will force DTS-Neo:6 if you wish to expand the surrounds. DTS-HD doesn't apply Neo:6, but automatically duplicates the 5.1 surrounds into the rear-backs, which is worse (this one may vary by receiver, but I suspect it's on DTS's specs). DPLIIx or Neo:6 can be applied to PCM streams, thus giving you more choice, not less.

The audio in the receiver is decoded to PCM, then the DSP applied. I think the confusion (not in your case, but in others) arises by the fact that multichannel analog in many cases can't be DSP'ed.

- channel mapping is an issue if some receivers do not decode the stream by themselves

How so? WASAPI exclusive solves this.

- Older receivers with only spdif inputs do not allow for multichannel pcm input, but work happily with the DTS/AC-3 cores of the HD formats

If you already have an HD format then you have an equivalent (to the "core") AC-3 stream from decoding the HD format and re-encoding with ReClock.

on some of the myths of packet drop/repeats:
- if the monitor refresh rate is very close to a multiple of the movie-framerate you will rarely get any drop/repeats. In fact I get none for the duration of a whole movie (except for the obligatory counted one at the start of playback)

Most people don't have the refresh rate close enough that you don't get any during a whole movie. Not even me, and I'm one of the luckier ones. Which card are you using, and did you not pause or skip at all during that test (I'm assuming also that you're using ReClock's counter)?

- while repeats are audible and bad, drops are not. DTS and AC3 were designed to be error resilient in the case of corrupted or dropped packets
- so when using a monitor refresh that favors drops over repeats you will probably not even notice the adjustment happening

"error resilient" doesn't mean "error free". But I agree (and said above) that they are probably unnoticeable most of the time.

 

[Zardoz]

It has other features, and not all formats are bitstreamable.



That's not true at all. See all the other checks and options in the ReClock settings panel? Those mean something, they're not just for show. One obvious example is getting WASAPI exclusive output with your DirectShow player of choice, most of which don't offer the option.

And if you must talk about a "sole" purpose, it would be to prevent both audio and video errors. Bitstreaming was added at the request of some people, when decoding wasn't all that viable.

Think about why the program is called "Reclock" and not "wasapi renderer". I am not objecting that you use Reclock for whatever you like, and that it offers many useful functions besides it's MAIN point. But you are stretching it when you claim that reclock is not working for other people, when in fact it is working perfectly well and does more for smooth video (it's original intent) than your use case. You just use it as a WASAPI renderer - fine, but hardly the case reclock was originally designed for.

 

[Zardoz]

It is, measurably, and perceivably. In some cases less so than others, to the point of being irrelevant (that depends on the user), but that's the cold hard truth.

False, if you do not have drops or repeats the quality of decoding in a modern receiver most certainly beats software decoding, resampling, (optional) repitching and (optional) reencoding. If you cannot or do not want to use powerstrip and get some drop or repearts it is a matter of preference: some can live with occasional drops/repeats others cannot and live with decoding and resampling. In now way does either decision imply that something "is not working" as it should. You made this claim.

True in some cases, if you're using SPDIF, (which Jack Carver was, I know), but not with HDMI.

You were talking to Jack Carver - and telling him he was using "reclock" wrongly. It's good to see that you changed your mind here. As for HDMI it's still a matter of preference drops/repeats or software decoding and resampling. There is no WRONG or RIGHT here.

Is an oft-repeated claim for which I haven't found any evidence, or even reliable anecdotes. Every time I've asked the person making the claim, either they don't own the receiver they're talking about, or they don't respond back.

Simple, with my Denon AVC-A1SE with the 2nd gen upgrade board the audio delay does not work for split 6ch pcm input (no HDMI on this model). This is unpleasant.

Most people don't have the refresh rate close enough that you don't get any during a whole movie. Not even me, and I'm one of the luckier ones. Which card are you using, and did you not pause or skip at all during that test (I'm assuming also that you're using ReClock's counter)?

No luck involved. Simply use Powerstrip if it supports your card. I use it on my HTPC using a custom timing with a Radeon HD2600 and get virtually no drop/repeats. On my gaming machine I use a HD5970 where even the stock settings are reasonable accurate so that the catalyst setting of 23.976 you get 3-5 drops/repeats over a whole movie. I fail to see what you mean by "pause" and "skip".

"error resilient" doesn't mean "error free". But I agree (and said above) that they are probably unnoticeable most of the time.

Exactly. In order to end this discussion let me just say this: Jack Carver use of reclock is perfectly reasonable and reclock does what it is supposed to. Please be careful in the future as there are many more reasonable use cases than your own.

 

[andy o]

Think about why the program is called "Reclock" and not "wasapi renderer".

You said "sole" purpose.

I am not objecting that you use Reclock for whatever you like, and that it offers many useful functions besides it's MAIN point.

Is this not objecting"?: "This makes no sense, if you "take a video frame error once every several tens of minutes" why use reclock in the first place? With or without bitstreaming reclock sole purpose is to sync the video stream to the monitors refresh rate in order to avoid frame drops and repeats."

But you are stretching it when you claim that reclock is not working for other people, when in fact it is working perfectly well

No, it's not. And he's not using Powerstrip.

and does more for smooth video (it's original intent) than your use case.

It doesn't (did you even read my replies to him?). I'm getting a more accurate refresh than him.

You just use it as a WASAPI renderer - fine, but hardly the case reclock was originally designed for.

I don't use it just as that. I said it's one of its multiple uses, and that it doesn't have a SOLE purpose.

 

[andy o]

False, if you do not have drops or repeats the quality of decoding in a modern receiver most certainly beats software decoding, resampling, (optional) repitching and (optional) reencoding.

"the quality of decoding"? You have to be more specific here. Resampling in ReClock is transparent. Repitching is not necessary for 99% of content, and reencoding to AC-3 a lossless or uncompressed track will yield the same quality as bitstreaming the so-called "core" (for simplicity I'm counting the interleaved AC-3 track in TrueHD).

If you cannot or do not want to use powerstrip and get some drop or repearts it is a matter of preference: some can live with occasional drops/repeats others cannot and live with decoding and resampling.
In now way does either decision imply that something "is not working" as it should. You made this claim.

Are you sure I made this claim? I can only see "as intended" in my post. Not the same. If I missed my own words please lemme know.

You were talking to Jack Carver - and telling him he was using "reclock" wrongly.

I did no such thing. I told him that he wasn't missing anything if he decoded lossless and encoded to AC-3, compared to just bitstreaming the "core", but was gaining ReClock's main feature, which is to resample audio to match video.

It's good to see that you changed your mind here.

How did I change my mind? I told him exactly the same thing I said there in what you quoted. In cases where you have lossless/uncompressed available, bitstreaming is not an advantage.

As for HDMI it's still a matter of preference drops/repeats or software decoding and resampling. There is no WRONG or RIGHT here.

Except drops/repeats in video or audio are certifiably visible/audible. Resampling is transparent.

Simple, with my Denon AVC-A1SE with the 2nd gen upgrade board the audio delay does not work for split 6ch pcm input (no HDMI on this model). This is unpleasant.

You said "most". How do you extrapolate to "most" from one model of Denon with one gen of upgrade board, with what appears to be actually a bug? Delay works perfectly well for LPCM and bitstream equally in my Pioneer. You also said and I quote:"most if not all receivers have limited bass management/delay/postprocessing capabilities if fed raw pcm instead of the original bitstream".

No luck involved. Simply use Powerstrip if it supports your card. I use it on my HTPC using a custom timing with a Radeon HD2600 and get virtually no drop/repeats.

Most people don't know or care how to use PowerStrip, or can't. You're suggesting a more complicated "fix" than just let ReClock do its intended thing, all in the name of what could be (or not) an imaginary problem with resampling.

Exactly. In order to end this disussion let me just say this: Jack Carver use of reclock is perfectly reasonable and reclock does what it is supposed to.

That's what I said, and that's even how I just told James I was using ReClock in the past. But there's a more optimal way to do it.

Please be careful in the future as there are many more reasonable use cases than your own.

Again, suggesting a more complicated fix with Powerstrip is hardly more reasonable than just let ReClock do its thing.

And one more thing. I'm curious about this statement of yours, perhaps James can clarify something too.

I have been using bitstreaming in reclock for years now (since the ogo days) and am quite happy with it.

James added bitstreaming capability fairly recently, and with it, added the capability to not disable media correction. Are you suggesting that ReClock used to have this capability, and then James took it away, only to put it back in? (I honestly don't know the history of ReClock that far into the past.)

 

[SamuriHL]

James added bitstreaming capability fairly recently, and with it, added the capability to disable media correction. Are you suggesting that ReClock used to have this capability, and then James took it away, only to put it back in? (I honestly don't know the history of ReClock that far into the past.)

(I'm avoiding the rest of this conversation as I have definite opinions but am not going to stoke the fire.) However, James added HD passthrough support recently. It's, to my knowledge, supported AC3/DTS passthrough for quite some time now. The HD bitstreaming support was added because I whined a lot that I wanted to be able to use ReClock in a *VERY* specific use case involving TMT. The use case I wanted was to be able to bitstream HD audio, but, in the case of a 48/16 LPCM track like those found on the Kill Bill movies, I wanted ReClock to force exclusive mode to set the proper channels and not resample the audio. For me I didn't so much care about the media correction. Now that we have MC16 and the ability to decode everything lossless to PCM, the game has changed a bit. Nonetheless, there are many valuable use cases for ReClock and not all of them involve media correction. It's a very versatile tool. For that I'm quite grateful.

 

[andy o]

Yeah I suspected that was the case, that DTS/AC-3 had been able to be passed through before, but James did add the option to not disable media correction while doing it, I think? I.e. the option to choose audio errors instead of video errors.

 

[SamuriHL]

That was added when HD bitstreaming was added, yes.

 

[JackCarver]

Thx Zardoz and andy_o for ur explanations. I'm a HTPC newbie and it was not easy to understand many of the facts but thx to this forum and many really good tools i think i got what i wanted to.

I first followed SamuriHL's explanations to get things working with mpc-hc, LAV filters, LAV CUVID, ffdshow and madvr and it worked really good but the frame-drop problem was still there.
Then i installed Reclock and frame-drops were history Really I didn't get one in the movies i watched so far.

But up to this point i got the sound via my LCD TV and HDMI. Then i wanted to use my old AVR (it has no HDMI connection) and i connected it via an optical SPDIF connection.
But the display shows only a 3 channel PCM sound so my next problem was to get surround sound.
One solution was to enable Dolby Digital output in my Realtek sound driver but i can't choose there DTS (don't know if DTS is better)

Then i enabled this Bitstreaming feature in Reclock and used DTS Bitstreaming in LAV audio and at that point my AVR showed 5.1 channel DTS.

As i didn't recognize any sound errors i posted here as i wanted to know more about this "not recommended" mark in Reclock's Bitstreaming options.

And after watching many more movies i fully agree to Zardoz, Reclock makes an excellent job in avoiding Frame drops which were really annoying and the sound output is excellent too.

There may be good reasons for not using Bitstreaming with Reclock but for me it works nice or i can't hear the errors. I'm not skilled enough to understand many of this technical details as i'm a beginner but i think that most AVR's use better sound chips as every HTPC mainboard and so an AVR would make a better job in decoding than my mainboard soundchip.

 

[andy o]

i think that most AVR's use better sound chips as every HTPC mainboard and so an AVR would make a better job in decoding than my mainboard soundchip.

Well that's another can of worms, but that's off-topic. All I can say is, technology is great and cheap sound chips have gotten very far nowadays, better than audibility. (OK, if someone disputes this, I'll just leave it at that )

 

[Zardoz]

Are you sure I made this claim? I can only see "as intended" in my post. Not the same. If I missed my own words please lemme know.

Fact: It was intended - and if you do not agree with this, it was at least very reasonable to use Reclock the way Jack Carver did.

I told him that he wasn't missing anything if he decoded lossless and encoded to AC-3, compared to just bitstreaming the "core", but was gaining ReClock's main feature, which is to resample audio to match video.

Funny that he feels different after experimenting with reclocks options -there are pros and cons. You only provided a misleading one sided view.

How did I change my mind? I told him exactly the same thing I said there in what you quoted. In cases where you have lossless/uncompressed available, bitstreaming is not an advantage.

Again this is a one sided view - for one bitstreaming avoids resampling, does not need separate connections to the receiver or reencoding (if no hdmi). Funny that this is exactly his use case. If a newbie asks a question it is common courtesy to provide a more balanced view, even if oneself does not value some of the points very highly.

Except drops/repeats in video or audio are certifiably visible/audible. Resampling is transparent.

Again a matter of opinion. Some value occasional drops as being the better alternative, others do not mind resampling. But as it is possible to achieve zero to one drops per movie I find it hard to argue for the superiority of resampling. But that's only my opinion, I am fine with people using resampling and in a lot of cases it makes perfect sense.

You said "most". How do you extrapolate to "most" from one model of Denon with one gen of upgrade board, with what appears to be actually a bug?

You are getting childish here. You wanted an example, and you got one. You made it clear that you do not believe ANY hearsay. That's why i provided my personal first hand experience. Naturally, I do not own ALL receivers in the world. But all receivers I and a couple of HTPC nuts tested a while back exhibited problems in that respect (100% with N=5). As these were all flagship models from various brands I extrapolated "most". Some of these tests have not been performed by myself - granted - but I basically trust these people. At least everyone of them knows more about reclock than you do.

Most people don't know or care how to use PowerStrip, or can't. You're suggesting a more complicated "fix" than just let ReClock do its intended thing, all in the name of what could be (or not) an imaginary problem with resampling.

If you find Powerstrip complicated, no problem: do not use it. Out of the box catalyst settings are reasonably accurate - nothing complicated here. Again you mislead - I provided many more reasons than just resampling. For each use case non/some/all might be applicable. Case in point: Jack Carvers use case with a non-hdmi receiver.

That's what I said, and that's even how I just told James I was using ReClock in the past. But there's a more optimal way to do it.

I do not see the relevance about what you told "James". And optimal does only apply to your use case (maybe). For example on how diverse the opinions can be: Myself and the group of htpc afficiandos i am exchanging with would find any HTPC configuration (reclock or no) that allows for repeated frames occasionally (as you do) far from optimal.

Again, suggesting a more complicated fix with Powerstrip is hardly more reasonable than just let ReClock do its thing

Please stop misleading. It his striking how you omit completely that the out-of-the-box catalyst settings on the new cards work pretty good - as i have pointed out in the post you refer to. And I also would reckon that many people who dig as far as reclock, are also completely capable and willing to use powerstrip (even though its interface is pretty anti-intuitive).

James added bitstreaming capability fairly recently, and with it, added the capability to not disable media correction. Are you suggesting that ReClock used to have this capability, and then James took it away, only to put it back in? (I honestly don't know the history of ReClock that far into the past.)

The only thing in this paragraph that makes sense is that you do not know the history of reclock. But that does not prevent you from speculation.