This forum is for users to exchange information and discuss with other users about a TMPGEnc product.
In case you need official support, please contact TMPG Inc.
Pegasys Products BBS [ Sorted by thread creation date ]
I used Virtual Dub to save the Wav file from a movie and then i went into TMPGenc and loaded both the Video and the Wav File into the respective boxes. So i was converting the file to a Pal 25fps movie. However when it completes and it is an mpeg and i play the file, i cannot here any sound. Any Ideas?
System -
AMD Athlon XP 2000+
512 MB Ram
Latest Version of TMPGence Plus
Also i am getting this error on some movies -
"At Address 000004B4, read error occurred against address 000004B4"
Thanks for any help in Advance, i couldnt find the solution at the FAQ
I've been experimenting with some SVCD MPEG-2 creation using 2-pass VBR and CBR. Is there any real benefits with picture quality going with VBR? (Both are being encoded at 2520 kbps). The VBR file is slightly smaller but I can't notice any difference in the picture quality.
I presume 2520kbps is the average, what maximum and minimum settings are you using?
I think 2520kbps is the maximum recommended speed for an SVCD, so if your content is short enough to fit onto a single disc then you will derive no benefit from using 2520kbps VBR, since it cannot burst past that maximum. If you need to encode at say 1500kbps in order to fit everything, then VBR (with maximum set to 2520kbps) will generate a higher quality output when there is a lot of motion or fast change of detail. Remember for VBR to work that it must be able to burst above and dip below the average.
There is a BIG advantage to using VBR over CBR. VBR allows you to create higher quality movies while producing a smaller file size.
For instance you can fit only about 40 mins of CBR SVCD on an 80 min CDR, but with VBR you could achieve the same quality and higher yet fit almost an hour on one disk.
As for VBR it is generally accepted by most people on this board and others that Constant quality(CQ)is a better VBR method than 2 pass.
Here's a sample situation I run into. I have 2 MPEG2 files made for 80min CD-R and they're PAL. I do a conversion from PAL to NTSCFilm, with BeSweet, TMPGEnc, etc., all that I have down pat but im working on file size now. I have 80min discs but my DVD player will only play until the 74min mark then freezes (Baloney Sony) so I need to set them for 74min. These files will always end up being too large for 2 74-min discs even at 1600 kbps so I want to get it to 3 74min discs (each of the two mpeg2s at 1091.63 MB and i'll merge and cut after to break into 3 727.75).
I'm using the wizard display to gauge file size. Here is the comparison between CBR and VBR:
Should I assume the VBR method will output the same size file but in better quality? Or is the Est file size in VBR incorrect because it can't estimate until it finishes encoding and I should expect the file to be smaller?
When useing 2-Pass VBR you will get the Same File size as CBR when useing the Same Bitrate as the Average Bitrate But the Quality will Be much better with the 2 -Pass File and with the 2-Pass you can set the Bitrate much Lower than CBR but still get the same or better Quality, But the CQ Method is the Best for Reduceing the File size and still retianing Good Quality....
I'm taking a look at the CQ settings as you've advised. How do you adjust the output file size with this? When I change from 2P-VBR to CQ the file size jumps by 148MB. The same settings are intact from my above example 300min/2520max. It seems the only way to get the file size output where I want it in CQ is by going to 300min/2188max. If that's the case, won't 2P-VBR at 300min/2520max (2188 avg) output better quality since it has a higher max? Or is there a setting im missing? There's a "Quality" lever (currently at 65) but moving it doesn't change the projected file size. Thanks
>There's a "Quality" lever (currently at 65) but moving it doesn't change the projected file size
Of course it does unless you are encoding to MPEG1 and have not changed the stream type to 'MPEG1 VBR' which will cause TMPG to add unecessary padding.
Usually reducing the quality by 5% will reduce the file size by about 100mb for a regular movie.
It's not working on mine for some reason. Looking at project Wizard (screen 4/5), when I select CQ all the fields become locked and don't change as I move the quality lever in CQ settings. I'd rather not wing it and take a guess. It would suck after 6 hours of encoding if it comes out too large.
What do you mean become locked? You are making a mistake somewhere.
All you have to do is click the 'Expert' button on the 'Bitrate setting' page of the wizard then change the 'rate control mode' to Constant Quality(CQ). Next click the 'Setting' button right next to it and select the settings you want.
Whatever bitrate settings you choose will then be displayed in the wizard.
Yes it displays the bitrate at 2520 (the max setting) and the file is too big. Like 2P-VBR I want the min/max at 300/2520 but still output a file at 1077MB.
If I lower the max setting it will update in the wizard a new file size but now im going to get worse quality than the 2P-VBR setting at 300min/2520max 2188 avg.
If I lower the "quality" lever in the CQ settings it doesn't change the projected file size output.
When it comes to The CQ Method the Projected File size is Not going to be Accurate because there is No way to Project the File size when useing the CQ method, and useing the Quality slider should change the Real file size even though the Projected file size does Not change....
Yes if you have a strict File size requirements then CQ is Probably not the Best unless you have a Lot of experience with CQ and can aproximate the Files size Based on Previus Encodes of Simular Files..
I agree with minion. TMPG won't display a projected file size because it's impossible to determine the resulting file size, but it will change when encoding.
Once you get used to using the CQ method you soon gain a knowledge of what sort of size to expect and there is a way to approximate the final size.
Simply set your parameters and encode a regular part of the movie for 1 min.
check the size than multiply it by the minutes of the enitire movie to see if it will fit.
For it to be more accurate I would suggest encoding a 5min clip then do the math.
Even though this is a crude method and won't be as exact as 2pass it does give you something to aim for.
Having said this CQ is still a better method than 2pass as far as quality vs file size is concerned.
In my experience CQ always produces better quality movies than 2pass for the same file size.
Put it this way commercial DVD's use this method and so does one of best and the most expensive commercial encoders on the market.
While trying to encode Mulholland Drive pt.1, I can select video file no problem, but when trying to select audio, I get the "Abnormal Program Termination" error in the TMPGEnc.exe file. Does anyone know how to resolve this issue, or what I need to do to log a bug report (as I assume they would need a copy of the video and/or audio files). Also, is there any workaround I could try? I tried makeing a copy from VirtualDub without success. Please also note that the audio file I am trying to select is a wav generated by VirtualDub. But I get the error both if I select the movie as the audio source, OR if I use the generated wav. In addition I have also tried using DivFix on the movie, and I have also tried AviToWav to generate the audio wav without success.
When you say that you can"t extract the audio from the AVI file with other tools like AVI-2-Wav do you mean that those Programs Crashed or wouldn"t extract the audio??? If so then I would suspect that the AVI File is Corrupted or Broken in some way and that is why you are haveing Problems with it..What I would sugest is to Try Frameserveing the File to Tmpgenc with AVISynth or V-Dub and if that fails then maybe try re-downloading the file as Downloaded Movies are very often Corrupted and Because you can watch a File doesn"t mean that there isn"t anything wrong with it...
You don't understand, I can extract the audio successfully both from VirtualDub AND AviToWav. What I can't do is use that wav (or the movie itself) as the audio file in TMPGEnc, whenever I try it crashes.
The same happens for me as well. Often, I don't get an error though, TMPGEnc just stops functioning. (This is when I try to select the audio file). I can't do anything with TMPGEnc. I can't close it. I can't even crash it (i.e. ending its process via task manager). Thus, I can't even properly reboot my computer since TMPGEnc won't shut down. I can only get out of it by hardstarting my computer.
I have a svcd that is to big to fit on a cd. I had tried to convert the file to a smaller svcd file. But after 20 secends TmPgenc freezes. Had tested some other files and it worked perfectly. Does anybody know what i should do?
If I were you, I would use the source range function (I think it's called that) You'll find it when you select your encoding options. Choose the first half of your frames for disk 1, then the second half for disk 2. For example, encode the first file (CD 01) from frame range 0 through 91051 then when it's done encode the next one (CD 02) from range 91052 to the last frame in the movie. This will split up your movie into two files which will fit on two CDs.
I have found a bug in TMPGEnc DVD Author 1.0.10 (1.0.10.36)
It occurs when dropping an MPEG1 file into a project. The
audio sync starts to drift and by the time 45 min have gone
the audio leads the video by 3-4 seconds.
The MPEG1 file has Audio at 44.1 KSPS 224kbps
which is being resampled I guess to 48 KSPS for the DVD.
DVD bit rates of 192 kbps and 224 kbps have both been tested
and the bug is present. Looks like the sample rate conversion
is getting lost (round off error???)
Great product,
Will be getting work to buy it soon
Looking forward to the bug fix
I had this problem too - using tmpegenc to extract/reencode the audio (via the ES Audio Only) option and then pointing the audio at the created .mp2 file in TDA worked fine - I noticed that doing a demultiplex in tmpegenc produced an autio file that was a different length to the video file! Doing the reencode fixed the problem though.
BTW I am using TDA for creating DVD's of TV episodes in mpeg1 VCD-compliant format.
I am having trouble trying to working out Aspect Ration and Source Aspect Ratio in TMPGenc when encoding to Mpeg. Is there a Guide where I can relate to & work see which Aspect Ratio and Source aspect ration will look will and best suit the Mpeg file im encoding . I seem to get a Tall Picture and would try to change it so they head dont look slim and tall (best way to descibe it on various files)
I would assume a guide that show which ones looks like. Please Help ....
Yout problem is probably not your aspect ratio, but the arrange method you are selecting. If you want to maintain the same aspect ratio as the original then you should always select 'Full screen(keep aspect ratio) and select whatever input aspect ratio relates to that source.
For example if you have an AVI in 16:9 then select 16:9(PAL) or 16:9(NTSC) depending on the format of the AVI and 'Full screen(keep aspect ratio) under the 'Advanced' tab
If you have a standard 4:3 TV then the default output aspect ratio should be 4:3 under the 'Video' tab
If your SWF file come from the net, it is probabely locked to disallow any manipulation on it, there are various utilities to unlock such file.
You can try this one http://www.buraks.com/unlockswf/ but there are many others.
do you know if there is any free way to do convert swf to mpeg macromedia flash mx is outta my price range, i.e. i'm broke, if it can't be done for free nevermind
I don't know if this is a tmpgenc problem, or an issue elsewhere. If I record a program for 90 minutes on my DVD recorder, tmpgenc thinks it is 45 minutes in length. Preview, source range, mpeg tools all show the complete recording from the first to the last frame, but the end time is always half of the real length.
When I try to encode the motion freezes halfway through, displaying the same frame for the rest of the time, but the audio continues properly.
Is this anything to do with interlacing? When I view frame by frame, there are no jagged interlacing edges in frames of high motion. Is it possible that tmpgenc is interpreting the source as 50fps instead of 25fps? (This is just a wild guess, of course.)
If I encode without using "Source range", tmpgenc says there are 136501 frames total once I start encoding. If I enable source range and set the slider at the very end (which is really the end - verified by the content), then it says there are 83346 frames. This is even stranger, 83346 is not half of 136501. (????)
One other thing which may be causing this issue, the input file is larger than 2Gb. I am using FAT32 under 98SE. tmpgenc has no problems displaying to the end of the content in the preview and source range windows.
Stream was created by a DVD recorder, Panasonic DMR-E30 in PAL mode.
Tmpgenc gets all the Information Like the Amount of Frames and Length of Movie directly from the Header of the Mpeg file, so if Tmpgenc says that there are only a certain Amount of frames or says that the Movie is a certain length it is because the File is telling Tmpgenc this information even though the Information is incorrect..Anyways you are not doing this the Proper Way, Tmpgenc handles Mpeg2 files Much better if you Use DVD2AVI to Frameserve the Mpeg2 files to Tmpgenc, This is the Best way and fastest way to encode Mpeg2 files with Tmpgenc..
Hmmm.... tmpgenc is still doing something funny though, if you use a preview or source range screen it says there are X total frames, but once you start encoding it says Y frames. Regardless of the value in the file, it is saying two different things about the same file.
I ended up using dvd2avi, I was hoping not to have that extra step as I barely had enough space to fit it. It's a pity that the DVD recorder puts out 704x576 rather than 720x576; I was hoping just to cut the m2v file and save it directly, without having to re-encode. Instead of 20 minutes to cut the m2v and extract then encode the sound, it now takes 2+ hours and I do lose some quality... of course by re-encoding I do have the advantage of being able to use tmpgenc's ghost and noise filters. :)
I just re-read your reply, what is frameserve? Is that a way of linking the output of dvd2avi to the input of tmpgenc without needing an intermediate AVI conversion? (....my HD now has 600Mb free compared to 24Gb free when it started...)
>I ended up using dvd2avi, I was hoping not to have that extra step as I barely had enough space to fit it.
Please don't tell me you went and encoded the whole thing to an AVI with DVD2AVI. This not what minion meant!
He was trying to advise you frameserve the MPEG to TMPG by creating a DVD2AVI project file which only takes a matter of minutes and very little file size.
>It's a pity that the DVD recorder puts out 704x576 rather than 720x576
What do you mean by this? If you are intending to keep the MPEG in either DVD or SVCD format then there is no reason why you can't use a 704x576 frame size.
DVD will accept 720x576, 704x576 and 352x288 or 352x576.
Check out DVDrhelp.com for info how to frame serve with DVD2AVI.
>If I encode without using "Source range", tmpgenc says there are 136501 frames total once I start encoding. If I enable source range and set the slider at the very end (which is really the end - verified by the content), then it says there are 83346 frames. This is even stranger, 83346 is not half of 136501. (????)
I have a very similar problem here.
I worked with a big AVI file named 'output.avi'.
At first, I made some tests with TMPGEnc with this file, among others, I used the source range function, setting start at the middle, and the end at the very last frame of this big file.
Later, I replaced this file by a smaller one, but with the same filename.
Now, each time I open the new small 'output.avi', TMPGEnc handle the file as if he contained as many frame as the old big 'output.avi'.
The only way to have TMPGEnc working correctely, is to use the 'Source Range' filter.
>>I ended up using dvd2avi, I was hoping not to have that extra step as I barely had enough space to fit it.
>
>Please don't tell me you went and encoded the whole thing to an AVI with DVD2AVI. This not what minion meant!
>He was trying to advise you frameserve the MPEG to TMPG by creating a DVD2AVI project file which only takes a matter of minutes and very little file size.
Yes, I did decode everything to AVI. :) I also saved the dvd2avi project in case I needed to it again. This is how I discovered frameserve, when I tried to delete the original M2V. I couldn't delete it as it was locked by tmpgenc, so I did some further checking and worked out that it was actually reading from the M2V, rather than all of the AVI files. The hour of decoding time and 24Gb of HD space was completely redundant...
>>It's a pity that the DVD recorder puts out 704x576 rather than 720x576
>What do you mean by this? If you are intending to keep the MPEG in either DVD or SVCD format then there is no reason why you can't use a 704x576 frame size.
>DVD will accept 720x576, 704x576 and 352x288 or 352x576.
I didn't realise this, so far I have only dealt with commercial DVDs which are all 720x576. I have decided that re-encoding is better anyway, because the video is recorded in real time it can never be as efficient as dual-pass VBR. To get the same quality, you need to use a higher bit rate. For long term archiving it's better to have a 3000Mbit/sec VBR stream, than 5Mbit/sec CBR...
>For long term archiving it's better to have a 3000Mbit/sec VBR stream, than 5Mbit/sec CBR...
>Oops, of course that was meant to be 3Mbit/sec and 5Mbit/sec, or 3000kbps and 5000kbps.
Hmm...how do you work that one out?
How can a bitstream at 3000kb/s VBR be better than a CBR bitstream at 5000kb/s?
Surely for archiving purposes you want the highest quality posiible or do you mean it's better for the file size?
>I couldn't delete it as it was locked by tmpgenc, so I did some further checking and worked out that it was actually reading from the M2V, rather than all of the AVI files. The hour of decoding time and 24Gb of HD space was completely redundant...
This is what we were trying to tell you. you didn't need to create intermediate AVI's.
Also there is no point in saving the project file. The project file will only work with the original files with their original names in their original locations. Surely you didn't expect that you could delete the M2v file once you had created the project file. The project file is only a connection between the source and the output. You still need the source files for encoding.
Well if it is a Mpeg file it could allready be a VCD depending on the Resolution ,Frame rate,Bitrate and audio format...Go to http://www.dvdrhelp.com/ and read up on How to make and What a VCD is....
I have many vintage Mpeg-1 files that need noise reduction and pop and click filtering. I tried demuxing into .mv1(video) and .mp2(audio), then running through Goldwave. The problem is that you do not have the capability of saving the restored audio file in .mp2 format. Tmpgenc will not accept anything else when multiplexing the files back. Any ideas as how to proceed?
If that is the case then it probably has to do with the editing you have done to the original and not a problem with TMPG's encoding.
Is the wav file that Goldwave created the exact same length as the original audio and is the Mp2 that TMPG creates also the same length?
I have been trying to perform accurate cutting of a TMPGEnc encoded MPG2 PAL program stream. The GOP is 10 and I want to cut from 00:00:00 to 00:01:04 (PAL=25frames) so this is the end of the GOP. This works fine.
I then want to cut from 00:01:05 to 00:05.24. The cut works fine, but the first 2 frames of this cut are identical.
Does anyone have any suggestions ? I have opened this cut in both Ulead Media Studio pro and Premiere 6.5 and they both show an identical first frame.