Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. |
26th May 2004, 19:17 | #481 | Link | |||
Registered User
Join Date: May 2004
Location: Montreal
Posts: 729
|
Quote:
Quote:
Quote:
Last edited by ursamtl; 26th May 2004 at 20:09. |
|||
26th May 2004, 19:49 | #482 | Link | |
Banned
Join Date: Jun 2002
Location: The Netherlands
Posts: 326
|
Quote:
We rewrote his method in a bidule and we haven't had any negative comments, except for one. We wouldn't have done that if we didn't see the potential of Kpex' method !!!! So I cannot see where you get that statement from...... or I must misunderstand you ! EoH |
|
26th May 2004, 23:22 | #483 | Link | |
Registered User
Join Date: Oct 2001
Location: Switzerland
Posts: 225
|
@ ursamtl / All:
Quote:
- Soundfield B-Format - Farina's OSER-recordings (public) You might have to resample to 32/44 or 16/44 (using CoolEdit or similar), but you then can use Gerzonic's B-Player and decode using Emigrator as Pentagon or Square. This will give you an idea of what you are can expect from Ambisonics. Sure these samples have been properly recorded (Soundfield Mic), but you can get similar (good) results from Stereo (but 2D only). Good luck, Andreas |
|
26th May 2004, 23:34 | #484 | Link | |
Registered User
Join Date: May 2004
Location: Montreal
Posts: 729
|
Quote:
Your point is valid but only somewhat. After all, if my system were skewing my impressions of the different bidule results, it would do likewise to my impressions of movie soundtracks, DVD-As, etc. I don't get what you and a couple of others mean when you talk about the "Dolby stuff." Do you mean movie sound effects? If so, then that's not at all what I'm aiming for. I want something like Kevin Shirley's wondeful mix on the Led Zeppelin DVD last year. He mixed the surround to sound as if the listener were sitting about five rows back from the front center of a Led Zeppelin show. The rears mainly provide the ambience of the hall and the audience and serve to reinforce the front, which is presented as an accurate soundstage of the band from the vantage point I just mentioned. If this were done with the Ambisonics bidule as I've heard them so far, the entire band would be cramped into a thick middle that's repeated in the rears. It would sound something like the old Motown remasters that have been released in mono. They might sound punchier than a stereo mix, but only 1 speaker is necessary for such reproduction and 4.1 speakers are being completely wasted on such material. I'll retry some Ambiophonics testing this evening to see if the results are better, but certainly Ambisonics by itself seems to be an utter disappointment. Ursa |
|
27th May 2004, 01:01 | #485 | Link | |
Registered User
Join Date: May 2004
Location: Montreal
Posts: 729
|
Quote:
I'll try some more tests tomorrow. I have some domestic stuff to take care of now. |
|
27th May 2004, 10:08 | #486 | Link | |
Banned
Join Date: Jun 2002
Location: The Netherlands
Posts: 326
|
Quote:
Ever thought of that it can be yourself ??? I have had friends listening here to the Ambi mix of War of the Worlds. 6 out of 7 : I didn't know this was available in surround. The 7th reacted like you : I don't hear it ! OK, as you probably know : I did a listening test. I took 10 stereo songs with all kind of styles of music. I converted these 10 songs with 10 different methods to surround. I made a poll online where people could fill in which version they liked the most. I didn't say before the poll what the used methods were. End result : 70 % preferred the conversions with Ambisonics ! So obviously a. you're doing something wrong (BTW did you convert the stereo first to 32 bits ?) or b. you just don't (want to ?) hear it. But keeping on saying how thrash Ambisonics is without adding a few "IMHO" gives me indeed the impression you're behaving like a troll. Eespecially since your own published bidule didn't give 3 people a good result. And believe me : if it was better than anyting already published, people sure would let it know here ! So come up with something better (your MM bidule was definitely worse). If not, than please stop all this whining and groaning ! Thanks. EoH |
|
27th May 2004, 13:05 | #487 | Link | |
Registered User
Join Date: May 2004
Location: Montreal
Posts: 729
|
Quote:
I repeat what I said in my earlier post, if Ambisonics is so great, then why are there so many alternatives presented in this thread? Why did you yourself say there's lots of work to do? If Ambisonics is the great method you pretend, then surely your work is done. Why did you say you were working on ways to improve separation? Surely if Ambisonics were so wonderful, then there would be no need to "improve" anything! Obviously, if my bidule is so bad, then so are the AC3Filter and Matrix Mixer on which it is based. Those seem to be quite popular. As well, my bidule's front generation is a greatly simplified version of the LCR Gerzon method that kempfand posted. His method is actually much better than mine in generating a solid soundstage (and far better than encoding traditional stereo signals with Ambisonics), but I didn't include it in my bidule because I wanted to honestly present it as what I said it was, an interpretation of the MatrixMixer DirectX filter. I repeat what I said in an earlier post: I never said it was the answer, just that I got better results for my listening taste with it than with your method. Someone asked me to post it, so I did. You seem to treat this whole thread as if it were some kind of competition and if someone comes along and criticizes Ambisonics, you pounce on them as if they were attacking you personally. I don't think this is the first time I've seen you berate others in threads on the Doom9 forum. It's extremely unpleasant to read and it certainly adds nothing to the spirit of community present here. Please restrict yourself to a discussion of the methods and stop the name calling and personal attacks. |
|
27th May 2004, 13:41 | #488 | Link | ||||
Banned
Join Date: Jun 2002
Location: The Netherlands
Posts: 326
|
Quote:
When you write in almost every post how bad the results are from Ambisonics and seldom add IMHO, yes, this is whining and groaning. This has nothing to do with a personal attack, but simply with the fact that you are constantly saying how bad Ambisonics is for you in this originally Ambisonics thread. I consider your comments about Ambisonics offensive, because 7 out of 10 have another opinion, so you are insulting 7 out of 10 ! Quote:
All those remarks like "Ambisonics is so wonderful" in the above and other contexts, are suggestive and I think you meant it that way. So who is offensive here ? Quote:
I can take good criticism without any problem. Otherwise Kempfand and me wouldn't have added so many extensions to the original bidule. I never criticized others bidules too. But...... when presenting your bidule with the attitude you spread out on Ambisonics, is IMHO not the way to discuss things here. You can present your bidule without that constant nagging about Ambi !! Thanks ! Quote:
Now you make another suggestive remark. Come with the facts ! The only one with his constant nagging about Ambisonics and who seems not to get the spirit of this forum is not me : it's you ! When I reacted to your remark about Kpex, you didn't come back with a "sorry". Not that I am waiting for that, but you set the tone ! Not me, I only react. You stated clearly that your method gave you a better result. The first one you posted was completely wrong. The second should be fine. Well perhaps for your ears, but till now I have heard 3 negative experiences and not one positive one (except your own of course). When I say something about that you feel attacked ? I would consider it a remark to stimulize you to come up with something working Till now I sense more pretentions than inventions ! You put down Ambi, every time, you would come up with something better, but didn't. On this forum we respect everyone's effort, but not by first putting down on the hard work of others first ! EoH |
||||
27th May 2004, 16:21 | #489 | Link | ||||||
Registered User
Join Date: May 2004
Location: Montreal
Posts: 729
|
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Last edited by ursamtl; 27th May 2004 at 16:24. |
||||||
27th May 2004, 19:42 | #490 | Link | ||
Registered User
Join Date: Oct 2001
Location: Switzerland
Posts: 225
|
Quote:
1) I think you (and others) are "victims" of what you (others) are used to in terms of 5.1 music. Note that this is not meant as a 'personal attack' or criticism. It's just how I see things. I think the discussion in the previous few posts of the threat are not so much about "Ambisonic or Other Method". It's more about how the recording was done (mics used etc.) and how it was mixed before it came to the stores. In detail: We are used to what we are used to, and I think the enormous (marketing) power of commercial recording companies and patent holders (e.g. Dolby & DTS just to name two) is often not considered enough as influencing factor on our behaviours. After all, if we are made to believe that what we buy is good, this means $$$ for the seller. 2) The same applies to how music is recorded and mixed before it ends on 5.1 DVD's or DTS CD's. Most 'modern' "Tonmeisters" (German expression for "mixing masters") have their style, which is often similar, i.e. music soundfield in the Fronts (incl. C), Ambience in the Surrounds. That's it. Full stop. In the worst case, there is but some echo and some audience applauding in the Surrounds. Horrible for my taste ... More so: Professional listening tests (Verdie -> Listening Test) raised the criticism, that Ambisonics gives too much room sound, and that isn't what the "Tonmeisters" want (even if it is correct!). It is also so that most people are so used to Stereo so hearing to much of the room won't sell (even if it is right !). 3) I agree with your point about "coloration" with Ambisonics, but this can be avoided by doing it right. Note that the most of the Ambisonics methods here (such as the guide at the beginning of this threat) are suited for the masses. Real Ambisonics uses signed filters (real-world impulse repsonses) to creat the B-format, as well as creating the decode to xyz speakers. See Conversion between UHJ and B-format for an outline on howto use signed filters for the B-format. 4) Quote:
Regards, Andreas Last edited by kempfand; 27th May 2004 at 19:45. |
||
28th May 2004, 00:21 | #491 | Link | |||||
Registered User
Join Date: May 2004
Location: Montreal
Posts: 729
|
Quote:
Quote:
Don't take my word for it, check out Ralph Glasgal's site (which you mentioned yourself when you presented your Ambiophonics bidule): Quote:
Quote:
Quote:
Anyway, time to do some more testing. Regards Last edited by ursamtl; 28th May 2004 at 00:47. |
|||||
28th May 2004, 03:58 | #492 | Link |
Registered User
Join Date: May 2004
Posts: 30
|
feed & queries
first off a big thanks to all the people here who have worked hard on this whole 2.0 to 5.1 issue. i've been playing around with several of the posted bidules and have settled on the SAD5.1 for now as being the better of the bunch for my tastes.
1) i'm certainly no expert when it comes to audio though but i was wondering if it was possible somehow to integrate a way of making use of dolby surround (and i think pro logic) if it exists in a 2.0 stream? from my understanding both dolby surround and prologic were multichannel signals mixed into a 2.0 format and re-expanded by the amp at the playback stage. i have a dolby surround test CD which i use to test the bidules (mostly using the channel identifier to make sure i ended up with all 6 channels going to the correct speakers and by far the result after processing isn't as good as the original. so i don't know if its even possible or not to detect if any such surround encoding is already existing in the 2.0 WAV file and once identified be able to make use of that knowledge to get an even cleaner 5.1 seperation? 2) also it seems the main use for people in this thread is for turning music into 5.1. do people feel the same bidules are just as good for tv/movie audio or would a bidule layout specific for non-music sources provide a better end result? 3) my setup is a computer with sound audigy and cambridge soundworks 5.1 amp/speakers and i connect my dvd players audio to the same amp. with the audigy card it plays 2.0 sources to all 5.1 channels already (but i'd like to put things onto DVDR so i can do other work with my computer). does anyone know how the bidule approach compares to the audigy? or even making use of the audigy hardware somehow if it is better to do hardware processing into 6 channel wav files? 4) ideally what is the end goal people here are aiming to achieve? complete sound seperation with for example voices staying in the centre channel instead of "leakage" into all channels. its a great thread so far, the only problems i've ever come across have been related to 2/4GB filesize limitations with various apps. i'm certainly happy with the current results when compared to boring old stereo but i hope it'll be possible to improve and get results closer to proper 5.1 source results thx |
28th May 2004, 12:07 | #493 | Link | ||||||||||||
Banned
Join Date: Jun 2002
Location: The Netherlands
Posts: 326
|
Quote:
Although I can understand your point here, you should be aware that this is an international forum ! Every now and then making clear that it is your personal opinion, doesn't harm, but would make it much more clear to users who don't have English as native language. TO ME it sounded as if you condemn the whole Ambisonics thing and when the majority of users is very satisfied with the results, I feel like you're nagging. A bit more IMHO would lead ME to the conclusion : hey, for this guy Ambi doesn't seem the solution !, instead of : hey, this guy is saying Ambi is thrash ! These are only nuances, which make it easier for non English speakers. I hope you understand that ! Quote:
Quote:
Quote:
Quote:
Hey, even we have records where we have the same opinion !! That's why we are still developing new methods. I said it many times before and also in this thread : use the method that gives YOU the best results. Listen and do not just base your opinions on theory ! I never said it was the ultimate method ! Quote:
Quote:
Quote:
For you I only suggest you ask me how I meant something, instead of take it for granted that your interpretation of what I wrote is exactly how I meant it .It's easier for you to adjust your English than it is for me ! Quote:
Quote:
Of course : gimme a few thousand dollars and I will buy the C3 soundanalyser, buy Wavelab 5 and Nuendo and a few other goodies and I can make a perfect mix with the output from C3. But that's not our goal ! As soon as you have to taylor anything, we think our bidule is not good enough !!! And that's a very different apporach. Don't you agree ? And : we will try to keep closer to the original stereo. I think that what we want is : when you play a surround CD and put your receiver into stereo mode, it should sound the same as the original stereo. In other words : We try to change the mix, but not the sound and when possible in an "convert only" bidule. If the stereo is narrow, we will leave it narrow ! That's the intention ! On a sidenote : every receiver will do DPL. Why bother at all ? Why trying to re-invent the wheel ? a. because it's just so much fun to experiment b. because we're not always satisfied with DPL Quote:
Quote:
But these exchange of posts should clear things up and for me it did ! Did it for you ? kind regards, EoH |
||||||||||||
28th May 2004, 12:27 | #494 | Link | |||||
Banned
Join Date: Jun 2002
Location: The Netherlands
Posts: 326
|
Re: feed & queries
Quote:
I am not 100% sure, but wouldn't Softencode give that information when you load in a WAV with integrated Dolby ? Worth to try ! Quote:
Quote:
I found out that a simple listening test says more than the opinion or theory of others :-) Quote:
A lot of people are very enthusiastic about the SAD51inBidule we made. I like it very much on electronical music, but not on music with singers ! It's also a matter of taste and personal preferences. Quote:
kind regards, EoH |
|||||
28th May 2004, 13:00 | #495 | Link |
Registered User
Join Date: May 2004
Posts: 30
|
EoH,
by "proper 5.1" i basically meant what a professionally produced 5.1 track can sound like. like what you'd expect on a good special effects movie with things audibly moving clearly throughout the room. so while atm we may be able to take a 2.0 copy of say the battle sequence from Master and Commander it might sound "big" when we turn it into 5.1 but it is still a long way from the original 5.1 soundtrack. i guess there's no such thing as a "surround sound benchmarking" system or something. something that could analyse a 5.1 track and show you a visual "spread" of the soundstage...maybe something like a colourspace graph. it could look at how much each channel has in common at any given instant and graph it over time. so when you put a single mono track into all 6 channels it would show you a dot in the middle of the room because each channel is identical. something along those lines anyway i was hoping might exist, it might be a way of helping to benchmark bidules and fine tune them? or am i just fantasising? would also help people like myself who dont have a very good surround setup nor a good "ear". Also being able to use an app to test a file i'd probably trust more anyway since the audigy spits everything i play out to 5.1 channels anyway so its hard to tell if i'm hearing the results of the upmixing or the audigy's own processing. so far i've been using a channel seperation dolby surround track for my testing. it is a voice speaking each channel in turn (left, centre, right, surround) followed by a few seconds of (pink? white?) noise. then i load up each of the 6 mono wavs at the end and "measure" the height of the noise sections for each channel mentioned by the audio. from that i can see how much cross channel leaking there is easily. since i can't subjectively trust my hearing for small differences between techniques i tend to use this one track as my benchmark to compare a new bidule layout to what i've used already and go from there. its probably the best testing method i'm going to be able to use. does anyone know if there are prologic channel seperation tests available free online in a two channel format? my dolby surround test cd rips to a 2 channel wav and that wav clearly plays in each designated speaker as the original cd does. from memory prologic should be the same, just it gives more channels but i dont have a prologic test cd but something like that i'd have thought would be available free online to help developers. i've been having a lot of fun playing around and exploring this process anyway since i recently got a dvd burner. i originally stumbled across this thread, it was not even something i'd considered possible but its certainly given me some inspiration to put a bit of extra effort into my vhs to dvd conversion so i end up with something that will last a long time to come. can't wait to see what new tricks people come up with! jsut wish i understood it all thx btw.. SMILE!!! its the weekend!!! |
28th May 2004, 13:28 | #496 | Link | |
Registered User
Join Date: May 2004
Location: Montreal
Posts: 729
|
Quote:
Just to follow up on this, let me say that I retried your Ambiophonics dipole last night and was quite impressed with the results. First of all, for some reason the wave files in the www.app.demon.nl download would not load into SIR. However, the version on Needfulthings did work. While the separation was still lacking and the rears were disappointing, the biggest thing I noticed about the front sound was a significant enhancement to the bass. There was much more clarity, definition, and punchiness without it sounding over-EQed. On some tracks it was actually quite stunning. In addition, the overall mix seemed to possess the kind of clarity and richness I normally associate with 5.1 mix. What was missing of course was the sense of width and depth to the rear. First of all, even if theoretically the speakers are supposed to be spaced at +-10°, this just collapsed the width of the soundstage too much for me. Returning my speakers to +-25° (the way my PC system is setup this is the maximum front width I have to work with), gave me a better sense of width. Still, on some music, it seemed too dry. For example, I have a beautifully remastered CD of Elton John's Madman Across the Water. Running "Tiny Dancer" through this bidule resulted in the vocals sounding incredibly dry, virtually devoid of any ambience. While one might argue that this is a more realistic recording of his actual voice, the effect was not as pleasant. I tried inserting a stereo expander VST and this improved the image significantly. There is a sweet spot at approximately 120-130% width. Anything over that improves the sense of ambience but starts to negatively affect that glorious bass response. There remains the problem of the rears. I don't want to just slap some reverb on it as that can get tedious. But the Ambisonics rears in the original bidule just don't work for me. In fact, I find that they detract from the enhanced fronts. I have a couple of ideas for both the front and rear that I'll try to work on this weekend. Nonetheless, this is truly an encouraging result and definitely a step in the right direction. Ursa Last edited by ursamtl; 28th May 2004 at 13:33. |
|
28th May 2004, 13:29 | #497 | Link | ||||||
Banned
Join Date: Jun 2002
Location: The Netherlands
Posts: 326
|
Quote:
Starting with 6 mono wavs recorded separately would be ideal. But we're using stereo and need to extract as much as possible from only 2 channels ! I don't think we can get the same results, but with music we can get close ! Quote:
OTOH what would you exactly do with such software ? I mean : as you say with a single mono the spot is in the middle, but this doesn't represent the ideal situation for the listener. Music can be loud sometimes in a song from one speaker only and than with the same benchmarking you're thinking of, that spot won't be in the middle. And that doesn't indicate the mix is wrong then !! Quote:
Can you upload that track to alt.binaries.sounds ? It would be very interesting to use as test sound for the new bidules we're developing (Kempfand and me). They are already very nice and give more separation than the things I tried till now, but such a track would be very helpful to finetune even further ! Quote:
I will do a search online too and let you know if I could find anything..... Quote:
I am not the writer of the bidules made ! I am more the guy with the (sometimes very weird) ideas behind bidules and I try to translate every technobabble into a useable guide. Some of the techniques used to get the results are too difficult for me to understand. The outcome however is for me the only thing important !! Quote:
kind regards, EoH |
||||||
28th May 2004, 13:41 | #498 | Link | |
Registered User
Join Date: May 2004
Location: Montreal
Posts: 729
|
Re: feed & queries
Quote:
If you haven't done so already, run (don't walk) all the way to www.kxproject.com and see if your Audigy model works with the kx drivers. These are free ASIO-compatible drivers for the Creative Labs cards based on either of two EMU DSPs (most internal Soundblaster Live's, Audigy's and Audigy 2's) and they can turn your card into a much more powerful tool for music. Their DSP applet is very similar to Plogue Bidule in that you can customize routing and connections graphically to suit your needs. With your system, you could easily create bidules with multichannel ASIO outputs for realtime monitoring of your surround mixes. In Plogue, if you're sample rate is set to 44.1k, just insert a kX ASIO output device (15 channels), or if at 48k, a kX ASIO Duplex input/output device (15 channels each way). Then check your routings in the kX DSP to make sure you're routing the 5.1 channels properly. After that in Plogue, you can connect each channel in real time while playing back and experimenting with different approaches, settings, etc. |
|
28th May 2004, 13:42 | #499 | Link | |
Banned
Join Date: Jun 2002
Location: The Netherlands
Posts: 326
|
Quote:
I will have another look into these wavs, because they should work in SIR, because we used them too..... I also found this method giving a good sound, but my objection : a. I cannot move my speakers and b. I am not in the position to buy 2 extra ones :-) (I'm sure you will like our new ones...... but they need some more work on them before we publish them.) Just curious : do you convert your stereo to 32 bits first ? kind regards, EoH |
|
28th May 2004, 14:20 | #500 | Link | |||
Registered User
Join Date: May 2004
Location: Montreal
Posts: 729
|
Quote:
Quote:
Quote:
By the way, yes, your earlier post did clear things up between us. I'm at work and don't have time for a detailed reply, but yes, there are surely cultural nuances to our communication that can cause misunderstandings. For example, I once worked in an office with some Swiss and German personnel who were working on a contract for the Canadian govt. here. They would get into discussions that seemed to me like they were ready to haul out weapons and kill each other, yet walk out of the office at the end of the day smiling at each other! Here I was the poor Canadian fellow shocked into thinking I was about to witness murder! These folks told me this was simply their European temperament and nothing I should be concerned about. Perhaps we Canucks are just too polite! Plus, having worked with them for awhile I found I was adopting some of their attitude. Have a great weekend! Ursa Last edited by ursamtl; 28th May 2004 at 14:24. |
|||
|
|