|
Post by hsamwel on Jun 16, 2020 16:19:06 GMT -5
Volume seems lower than Dolby but not by 6db, no other symptoms Could you please tell us your setup, with either 7.x or if with height/top speakers. That goes for all that votes and comments, thanks. You will probably not notice anything by just using Neural:X. Also try to use a good recording with dynamic sound. Try switching between Neural:X and Surround mode. If you hear no apparent volume difference between the two your setting probably work fine in Neural:X. If it has the issue you will instantly notice a slight drop in volume, probably a slight muddiness with voices and generally not clear sound. This compared to Surround mode of course. In my tests 7.x.4 with tops sound fine, change tops to heights and I get the bad sound. Same if I disable the height/top speakers. I have not tried with wides or mixed height/top speakers.
|
|
|
Post by steelman1991 on Jun 16, 2020 16:29:56 GMT -5
Already voted.
|
|
|
Post by foggy1956 on Jun 16, 2020 17:11:49 GMT -5
Volume seems lower than Dolby but not by 6db, no other symptoms Could you please tell us your setup, with either 7.x or if with height/top speakers. That goes for all that votes and comments, thanks. You will probably not notice anything by just using Neural:X. Also try to use a good recording with dynamic sound. Try switching between Neural:X and Surround mode. If you hear no apparent volume difference between the two your setting probably work fine in Neural:X. If it has the issue you will instantly notice a slight drop in volume, probably a slight muddiness with voices and generally not clear sound. This compared to Surround mode of course. In my tests 7.x.4 with tops sound fine, change tops to heights and I get the bad sound. Same if I disable the height/top speakers. I have not tried with wides or mixed height/top speakers. 7.1.6 with one preset with tops and 1 with heights, both exhibit a small drop in volume but no muddiness or smearing
|
|
|
Post by PaulBe on Jun 16, 2020 20:45:59 GMT -5
Thanks for the tag. It's hard for me to vote on this issue. I don't experience major technical problems using either the Dolby or DTS upmixers with the RMC-1L. However, I don't perceive the differences I do hear in use coming from problems with the RMC. I don't like how either upmixer sounds compared to no upmixing. Both upmixers create sound perspectives that sound unnatural to me, So, I prefer the original multi-channel mixes, without synthesizing additional immersion channels. When I occasionally want to upmix a 2 channel recording, I play back in All Stereo. All Stereo still sounds natural and coherent. I save immersion channel use for recordings that are built with it. They sound good in Dolby Atmos or DTS:X My immersion channels are front and back heights, and mid tops. Setup is 7.2.6 I'd be curious as to what you think after changing all heights to Tops. It made a significant difference in what I hear. I still think the native DTS codec sounds like it is a hair "warmer", but I wouldn't say better now. Since there are Height, Top, and Dolby Enabled settings in the RMC menu, and in Dolby Atmos layouts, I expect EQ differences between them. I could speculate, but I don’t know what they are. My goal is to have similar voicing between all channels. If the Top setting has a different frequency response, I would still try to achieve similar voicing using EQ. My Mid-Tops, which are over my head at MLP, were treble heavy before EQ. After EQ, they voice similar to the Heights and floor channels, and the whole room sound field is more coherent. I think it’s best to use settings that correspond with the speaker layout, and voice the whole layout to be similar from channel to channel at MLP.
|
|
|
Post by megash0n on Jun 16, 2020 21:30:18 GMT -5
I'd be curious as to what you think after changing all heights to Tops. It made a significant difference in what I hear. I still think the native DTS codec sounds like it is a hair "warmer", but I wouldn't say better now. Since there are Height, Top, and Dolby Enabled settings in the RMC menu, and in Dolby Atmos layouts, I expect EQ differences between them. I could speculate, but I don’t know what they are. My goal is to have similar voicing between all channels. If the Top setting has a different frequency response, I would still try to achieve similar voicing using EQ. My Mid-Tops, which are over my head at MLP, were treble heavy before EQ. After EQ, they voice similar to the Heights and floor channels, and the whole room sound field is more coherent. I think it’s best to use settings that correspond with the speaker layout, and voice the whole layout to be similar from channel to channel at MLP. I don't disagree in general...or, I respect your opinion on it. That is assuming that these things are working as intended which is what I thought our concern was. Should I assume from your response that you haven't, or don't care to test with Tops because it doesn't associate physically with where your speakers are? If so, I guess I'm asking your opinion on if there is a sound different but testing whenever you have a free 10 mins.
|
|
|
Post by bluescale on Jun 17, 2020 2:08:48 GMT -5
My XMC-2 has been sent back to Emotiva for warranty service. I didn't know if it made sense to vote in this thread until I got it back and could properly retest it. It might be a moot point by then, but it feels like it'd make more sense.
|
|
|
Post by thezone on Jun 17, 2020 7:06:50 GMT -5
So I am only rocking a 5.1 set up but after having had some time to play around with my new XMC-2 and I have noticed this:
Upmixing 2ch using Dolby Surround is awesome, I actually think I prefer listening to 2ch this way (I know people don't like this but I find the separation truly remarkable and much better than I thought).
But when switching to Neural:X there is a significant drop in db, and a huge loss of sub gain, I say this because I turn up the volume to get back to where the Dolby level was and the sub channel is still lacking, and its as though someone has thrown a sheet over all the speakers. Not good, but I'm so happy with things in the Dolby camp that I'm not fussed.
For what its worth I always thought that PLxx Movies was better than DTS Neo in the old money.
Not sure its relevant to this thread and I certainly can't vote given the title.
|
|
|
Post by megash0n on Jun 17, 2020 8:18:44 GMT -5
So I am only rocking a 5.1 set up but after having had some time to play around with my new XMC-2 and I have noticed this: Upmixing 2ch using Dolby Surround is awesome, I actually think I prefer listening to 2ch this way (I know people don't like this but I find the separation truly remarkable and much better than I thought). But when switching to Neural:X there is a significant drop in db, and a huge loss of sub gain, I say this because I turn up the volume to get back to where the Dolby level was and the sub channel is still lacking, and its as though someone has thrown a sheet over all the speakers. Not good, but I'm so happy with things in the Dolby camp that I'm not fussed. For what its worth I always thought that PLxx Movies was better than DTS Neo in the old money. Not sure its relevant to this thread and I certainly can't vote given the title. I think it is relevant whether there is a vote or not. I'm not sure we have quantified what or where the issue really is, but the feedback is appreciated.
|
|
|
Post by hsamwel on Jun 17, 2020 11:31:37 GMT -5
Could you please tell us your setup, with either 7.x or if with height/top speakers. That goes for all that votes and comments, thanks. You will probably not notice anything by just using Neural:X. Also try to use a good recording with dynamic sound. Try switching between Neural:X and Surround mode. If you hear no apparent volume difference between the two your setting probably work fine in Neural:X. If it has the issue you will instantly notice a slight drop in volume, probably a slight muddiness with voices and generally not clear sound. This compared to Surround mode of course. In my tests 7.x.4 with tops sound fine, change tops to heights and I get the bad sound. Same if I disable the height/top speakers. I have not tried with wides or mixed height/top speakers. 7.1.6 with one preset with tops and 1 with heights, both exhibit a small drop in volume but no muddiness or smearing A small drop in volume is okey if it’s still same volume in Surround and Neural:X. There are differences between recordings. Many, especially movies, with DTS have lower volume than a normal Dolby or Atmos movie. But if there’s only lower in Neural:X, switch a couple more times and listen closely to a specific sound..
|
|
|
Post by hsamwel on Jun 17, 2020 11:51:39 GMT -5
I'd be curious as to what you think after changing all heights to Tops. It made a significant difference in what I hear. I still think the native DTS codec sounds like it is a hair "warmer", but I wouldn't say better now. Since there are Height, Top, and Dolby Enabled settings in the RMC menu, and in Dolby Atmos layouts, I expect EQ differences between them. I could speculate, but I don’t know what they are. My goal is to have similar voicing between all channels. If the Top setting has a different frequency response, I would still try to achieve similar voicing using EQ. My Mid-Tops, which are over my head at MLP, were treble heavy before EQ. After EQ, they voice similar to the Heights and floor channels, and the whole room sound field is more coherent. I think it’s best to use settings that correspond with the speaker layout, and voice the whole layout to be similar from channel to channel at MLP. Of course there can be differences, simply because objects can sound a little different depending on where the ”nearest” speaker is. But we ar investigating the bad upmixing some get with Neural:X. Within the same speaker setting from the same source there shouldn’t be any differences. But still there are. Depending on speaker layout, which is just crazy. The problem is that Atmos front/rear height does not exist in DTS:X. The tops do, but are named otherwise and have slightly different angle and azimuth. IMO a possibility to configure separate Atmos and DTS speaker settings would helt alot. For the immersive speakers anyway. Or better still, correct the difference automatically within RMC/XMC..
|
|
|
Post by PaulBe on Jun 17, 2020 13:22:40 GMT -5
Since there are Height, Top, and Dolby Enabled settings in the RMC menu, and in Dolby Atmos layouts, I expect EQ differences between them. I could speculate, but I don’t know what they are. My goal is to have similar voicing between all channels. If the Top setting has a different frequency response, I would still try to achieve similar voicing using EQ. My Mid-Tops, which are over my head at MLP, were treble heavy before EQ. After EQ, they voice similar to the Heights and floor channels, and the whole room sound field is more coherent. I think it’s best to use settings that correspond with the speaker layout, and voice the whole layout to be similar from channel to channel at MLP. Of course there can be differences, simply because objects can sound a little different depending on where the ”nearest” speaker is. But we ar investigating the bad upmixing some get with Neural:X. Within the same speaker setting from the same source there shouldn’t be any differences. But still there are. Depending on speaker layout, which is just crazy. The problem is that Atmos front/rear height does not exist in DTS:X. The tops do, but are named otherwise and have slightly different angle and azimuth. IMO a possibility to configure separate Atmos and DTS speaker settings would helt alot. For the immersive speakers anyway. Or better still, correct the difference automatically within RMC/XMC.. More on DTS:X from link above - “DTS:X supports other immersive formats’ speaker layouts, giving consumers freedom to choose their favourite layout” Neural:X will have to make do with whatever layout you choose, and it won't be DTS:X. Perhaps you just don't like what Neural:X does. I don't. Nor do I like the Dolby upmixer. I was tagged to come vote here. I stated why I couldn't vote. I have no interest in what you are investigating here. Good Bye.
|
|
|
Post by hsamwel on Jun 17, 2020 14:12:44 GMT -5
Of course there can be differences, simply because objects can sound a little different depending on where the ”nearest” speaker is. But we ar investigating the bad upmixing some get with Neural:X. Within the same speaker setting from the same source there shouldn’t be any differences. But still there are. Depending on speaker layout, which is just crazy. The problem is that Atmos front/rear height does not exist in DTS:X. The tops do, but are named otherwise and have slightly different angle and azimuth. IMO a possibility to configure separate Atmos and DTS speaker settings would helt alot. For the immersive speakers anyway. Or better still, correct the difference automatically within RMC/XMC.. More on DTS:X from link above - “DTS:X supports other immersive formats’ speaker layouts, giving consumers freedom to choose their favourite layout” Neural:X will have to make do with whatever layout you choose, and it won't be DTS:X. Perhaps you just don't like what Neural:X does. I don't. Nor do I like the Dolby upmixer. I was tagged to come vote here. I stated why I couldn't vote. I have no interest in what you are investigating here. Good Bye. The difference between DTS:X and Atmos speaker layout and the problem it creates is a common issue not specific to Emotiva. You could find info on it on the net. I know there are threads on AVS about it. The problem is not that DTS:X/Neural:X can’t handle whatever speaker position. It’s that it gets the wrong position from the processor and because of this puts out an incorrect mix. I never use any upmixers. Actually don’t have an opinion on which is better. My interrest is only to have a prepro that works with any feature. But may start using when DTS:X Pro arrives with the updated Neural:X. It’s fine if you don’t want to contribute to this subject. No one is forcing you.. But for the sake of RMC-1/L and XMC-2 it would be good if as many as possible report their setup and if they have issues. Makes it easier for Emotiva to find the cause of this.
|
|
|
Post by PaulBe on Jun 17, 2020 16:15:33 GMT -5
More on DTS:X from link above - “DTS:X supports other immersive formats’ speaker layouts, giving consumers freedom to choose their favourite layout” Neural:X will have to make do with whatever layout you choose, and it won't be DTS:X. Perhaps you just don't like what Neural:X does. I don't. Nor do I like the Dolby upmixer. I was tagged to come vote here. I stated why I couldn't vote. I have no interest in what you are investigating here. Good Bye. The difference between DTS:X and Atmos speaker layout and the problem it creates is a common issue not specific to Emotiva. You could find info on it on the net. I know there are threads on AVS about it. The problem is not that DTS:X/Neural:X can’t handle whatever speaker position. It’s that it gets the wrong position from the processor and because of this puts out an incorrect mix. I never use any upmixers. Actually don’t have an opinion on which is better. My interrest is only to have a prepro that works with any feature. But may start using when DTS:X Pro arrives with the updated Neural:X. It’s fine if you don’t want to contribute to this subject. No one is forcing you.. But for the sake of RMC-1/L and XMC-2 it would be good if as many as possible report their setup and if they have issues. Makes it easier for Emotiva to find the cause of this. I made my report. I said I don’t like either upmixers. They both have problems, and sound unnatural to me. I have no other reference to show me if they are doing what they are supposed to do or not. Don’t presume to know what I want to contribute. I have no interest in what you are investigating here. Since there are only 21 votes so far, 12 who have problems, among 10,000 owners, (so I read), I’d say there aren’t many other owners who have an interest in what you are investigating here. Carry on.
|
|
KeithL
Administrator
Posts: 9,945
|
Post by KeithL on Jun 17, 2020 22:35:29 GMT -5
I've got to make what amounts to an editorial comment here...
I've been listening to "upmixers" since the 1970's - when I got my first "SQ-4 four channel receiver". If you play a recording that was SQ-4 encoded through an SQ-4 decoder then it acts as a decoder (although not a very accurate one). If you play a two channel recording through an SQ-4 decoder it acts as an "upmixer" (which is another name for "channel synthesizer").
Dolby Digital, and DTS, and later Dolby Atmos, and DTS:X, are true decoders. Specific content is encoded by the encoder and, when you decode it, you get back those original channels. In contrast... "upmixers" are just faking it. They look at the incoming signal in certain ways, analyze certain details, then guess what content should go where.
Dolby PLIIx was sort of like SQ-4. You could encode content with the idea that, by encoding it a certain way, you could encourage the decoder to decode it a certain way. It wasn't exact for the same reason - because things like phase shifts in the original content could cause the results to vary from what you planned.
We then progressed to encoders like Dolby Digital and DTS - which actually save each channel separately - and retrieve it exactly. And then to encoders like Dolby Atmos and DTS:X - which store the locations of individual objects - and recreate the original when they decode it.
Both the current Dolby Surround Upmixer, and the DTS:X Neural-X Upmixer, do essentially the same thing as SQ-4 and PLIIx. They analyze the signal very carefully and, based on what they figure out, they make a guess about where various sounds should be directed. The new ones are much smarter, and so do a better job of faking it, in ways that are more often pleasing to more people....
However, and here is the point I was getting to, they are still faking it.
Because of this simple fact any discussion of which is more accurate constitutes a sort of surrealism... The only question is of which one sounds better, to you, on your system, with the particular content you happen to be listening to...
Personally, I would rather listen to Dolby Atmos movies in Dolby Atmos, and stereo movies in stereo... (And I still prefer The Maltese Falcon in original black and white...)
I know plenty of people who find the results of various upmixers to be quite pleasant... However, personally, I agree that they all sound "unnatural"...
The difference between DTS:X and Atmos speaker layout and the problem it creates is a common issue not specific to Emotiva. You could find info on it on the net. I know there are threads on AVS about it. The problem is not that DTS:X/Neural:X can’t handle whatever speaker position. It’s that it gets the wrong position from the processor and because of this puts out an incorrect mix. I never use any upmixers. Actually don’t have an opinion on which is better. My interrest is only to have a prepro that works with any feature. But may start using when DTS:X Pro arrives with the updated Neural:X. It’s fine if you don’t want to contribute to this subject. No one is forcing you.. But for the sake of RMC-1/L and XMC-2 it would be good if as many as possible report their setup and if they have issues. Makes it easier for Emotiva to find the cause of this. I made my report. I said I don’t like either upmixers. They both have problems, and sound unnatural to me. I have no other reference to show me if they are doing what they are supposed to do or not. Don’t presume to know what I want to contribute. I have no interest in what you are investigating here. Since there are only 21 votes so far, 12 who have problems, among 10,000 owners, (so I read), I’d say there aren’t many other owners who have an interest in what you are investigating here. Carry on.
|
|
|
Post by ttocs on Jun 17, 2020 22:51:02 GMT -5
I've got to make what amounts to an editorial comment here... Personally, I would rather listen to Dolby Atmos movies in Dolby Atmos, and stereo movies in stereo... (And I still prefer The Maltese Falcon in original black and white...) I know plenty of people who find the results of various upmixers to be quite pleasant... However, personally, I agree that they all sound "unnatural"... As I've stated in other threads, I too enjoy the theatrical release version of all movies, especially when it comes to originally released in B&W. That said, I wonder if you can help with regard to the drop in volume a lot of us experience when Neural:X is engaged? It's approximately 6dB lower volume level. This is the most consistent issue and very easy to demonstrate. The other issues are more difficult to pin down and replicate on every system, although there does seem to be a trend.
|
|
|
Post by markc on Jun 18, 2020 3:19:07 GMT -5
I've got to make what amounts to an editorial comment here...
I've been listening to "upmixers" since the 1970's - when I got my first "SQ-4 four channel receiver". If you play a recording that was SQ-4 encoded through an SQ-4 decoder then it acts as a decoder (although not a very accurate one). If you play a two channel recording through an SQ-4 decoder it acts as an "upmixer" (which is another name for "channel synthesizer").
Dolby Digital, and DTS, and later Dolby Atmos, and DTS:X, are true decoders. Specific content is encoded by the encoder and, when you decode it, you get back those original channels. In contrast... "upmixers" are just faking it. They look at the incoming signal in certain ways, analyze certain details, then guess what content should go where.
Dolby PLIIx was sort of like SQ-4. You could encode content with the idea that, by encoding it a certain way, you could encourage the decoder to decode it a certain way. It wasn't exact for the same reason - because things like phase shifts in the original content could cause the results to vary from what you planned.
We then progressed to encoders like Dolby Digital and DTS - which actually save each channel separately - and retrieve it exactly. And then to encoders like Dolby Atmos and DTS:X - which store the locations of individual objects - and recreate the original when they decode it.
Both the current Dolby Surround Upmixer, and the DTS:X Neural-X Upmixer, do essentially the same thing as SQ-4 and PLIIx. They analyze the signal very carefully and, based on what they figure out, they make a guess about where various sounds should be directed. The new ones are much smarter, and so do a better job of faking it, in ways that are more often pleasing to more people....
However, and here is the point I was getting to, they are still faking it.
Because of this simple fact any discussion of which is more accurate constitutes a sort of surrealism... The only question is of which one sounds better, to you, on your system, with the particular content you happen to be listening to...
Personally, I would rather listen to Dolby Atmos movies in Dolby Atmos, and stereo movies in stereo... (And I still prefer The Maltese Falcon in original black and white...)
I know plenty of people who find the results of various upmixers to be quite pleasant... However, personally, I agree that they all sound "unnatural"... Thank's Keith for demonstrating that you read these posts at least. That is very clear about differences between decoders and upmixers roles. I think the consensus response from us will be that we do appreciate that the upmixers are extrapolating known (the specific channel encoding of audio and objects) to unknown and "guessing / calculating / faking" which of the user's channel(s) it will mix audio signals into, using a variety of algorithms. Our issue is that many of us do not believe that DTS Neural:X is being fed information it can reliably understand and interpret about WHICH speakers a user has, based on what is set in the speaker configuration. We believe this is causing odd results depending on which speakers are specified as being present. I have confirmed that the effect we are describing does not occur in 7.1.4 setups on a Marantz 8805 so the Marantz is doing something better, and doing it reliably. The XMC-2 and RMC-1 are not necessarily doing it as expected for all speaker configurations with height vs top vs Dolby vs no upper speakers. DTS's big "sell point" for DTS;X and Neural:X is that it does not matter where the user's speakers are. The decoder/upmixer will adapt. However, this relies on the DTS decoder and upmixers getting information relayed to them by the hardware (Emotiva's programming and the user's interface) about a users speakers in a way that it can interpret and understand and adapt to. I prefer Dolby's prescriptive speaker placement recommendation approach, but Emotiva and HT users have to work with DTS. We. as users. require that the DTS Neural:X is fed information it can appropriately use to give us the up-mix that the DTS upmixer "thinks" it is giving us. This does not seem to be the case. Neural:X can sound fantastic on other systems (As good as Dolby and in some situations, better, as it doesn't elevate as much "general" audio to the heights as Dolby but gives more dramatic height specific audio and leaves most of the sound at the normal 7.x and 5.x speakers) but on the Emotiva's it is not reliably doing what it should be doing for all users despite doing nothing different than for the people who don't get problems.
|
|
|
Post by megash0n on Jun 18, 2020 4:41:10 GMT -5
I've got to make what amounts to an editorial comment here...
I've been listening to "upmixers" since the 1970's - when I got my first "SQ-4 four channel receiver". If you play a recording that was SQ-4 encoded through an SQ-4 decoder then it acts as a decoder (although not a very accurate one). If you play a two channel recording through an SQ-4 decoder it acts as an "upmixer" (which is another name for "channel synthesizer").
Dolby Digital, and DTS, and later Dolby Atmos, and DTS:X, are true decoders. Specific content is encoded by the encoder and, when you decode it, you get back those original channels. In contrast... "upmixers" are just faking it. They look at the incoming signal in certain ways, analyze certain details, then guess what content should go where.
Dolby PLIIx was sort of like SQ-4. You could encode content with the idea that, by encoding it a certain way, you could encourage the decoder to decode it a certain way. It wasn't exact for the same reason - because things like phase shifts in the original content could cause the results to vary from what you planned.
We then progressed to encoders like Dolby Digital and DTS - which actually save each channel separately - and retrieve it exactly. And then to encoders like Dolby Atmos and DTS:X - which store the locations of individual objects - and recreate the original when they decode it.
Both the current Dolby Surround Upmixer, and the DTS:X Neural-X Upmixer, do essentially the same thing as SQ-4 and PLIIx. They analyze the signal very carefully and, based on what they figure out, they make a guess about where various sounds should be directed. The new ones are much smarter, and so do a better job of faking it, in ways that are more often pleasing to more people....
However, and here is the point I was getting to, they are still faking it.
Because of this simple fact any discussion of which is more accurate constitutes a sort of surrealism... The only question is of which one sounds better, to you, on your system, with the particular content you happen to be listening to...
Personally, I would rather listen to Dolby Atmos movies in Dolby Atmos, and stereo movies in stereo... (And I still prefer The Maltese Falcon in original black and white...)
I know plenty of people who find the results of various upmixers to be quite pleasant... However, personally, I agree that they all sound "unnatural"... Thank's Keith for demonstrating that you read these posts at least. That is very clear about differences between decoders and upmixers roles. I think the consensus response from us will be that we do appreciate that the upmixers are extrapolating known (the specific channel encoding of audio and objects) to unknown and "guessing / calculating / faking" which of the user's channel(s) it will mix audio signals into, using a variety of algorithms. Our issue is that many of us do not believe that DTS Neural:X is being fed information it can reliably understand and interpret about WHICH speakers a user has, based on what is set in the speaker configuration. We believe this is causing odd results depending on which speakers are specified as being present. I have confirmed that the effect we are describing does not occur in 7.1.4 setups on a Marantz 8805 so the Marantz is doing something better, and doing it reliably. The XMC-2 and RMC-1 are not necessarily doing it as expected for all speaker configurations with height vs top vs Dolby vs no upper speakers. DTS's big "sell point" for DTS;X and Neural:X is that it does not matter where the user's speakers are. The decoder/upmixer will adapt. However, this relies on the DTS decoder and upmixers getting information relayed to them by the hardware (Emotiva's programming and the user's interface) about a users speakers in a way that it can interpret and understand and adapt to. I prefer Dolby's prescriptive speaker placement recommendation approach, but Emotiva and HT users have to work with DTS. We. as users. require that the DTS Neural:X is fed information it can appropriately use to give us the up-mix that the DTS upmixer "thinks" it is giving us. This does not seem to be the case. Neural:X can sound fantastic on other systems (As good as Dolby and in some situations, better, as it doesn't elevate as much "general" audio to the heights as Dolby but gives more dramatic height specific audio and leaves most of the sound at the normal 7.x and 5.x speakers) but on the Emotiva's it is not reliably doing what it should be doing for all users despite doing nothing different than for the people who don't get problems. Thanks. It annoys the hell out of me to read a dissertation to only find deflection. Sometimes I feel as if I've bought some scrap metal off the back of a truck. I would rather have silence than this. There are either serious levels of denial, or some have been directed to not comment on certain issues publicly. I'll let you be the judge of that.
|
|
|
Post by PaulBe on Jun 18, 2020 6:44:49 GMT -5
I've got to make what amounts to an editorial comment here...
I've been listening to "upmixers" since the 1970's - when I got my first "SQ-4 four channel receiver". If you play a recording that was SQ-4 encoded through an SQ-4 decoder then it acts as a decoder (although not a very accurate one). If you play a two channel recording through an SQ-4 decoder it acts as an "upmixer" (which is another name for "channel synthesizer").
Dolby Digital, and DTS, and later Dolby Atmos, and DTS:X, are true decoders. Specific content is encoded by the encoder and, when you decode it, you get back those original channels. In contrast... "upmixers" are just faking it. They look at the incoming signal in certain ways, analyze certain details, then guess what content should go where.
Dolby PLIIx was sort of like SQ-4. You could encode content with the idea that, by encoding it a certain way, you could encourage the decoder to decode it a certain way. It wasn't exact for the same reason - because things like phase shifts in the original content could cause the results to vary from what you planned.
We then progressed to encoders like Dolby Digital and DTS - which actually save each channel separately - and retrieve it exactly. And then to encoders like Dolby Atmos and DTS:X - which store the locations of individual objects - and recreate the original when they decode it.
Both the current Dolby Surround Upmixer, and the DTS:X Neural-X Upmixer, do essentially the same thing as SQ-4 and PLIIx. They analyze the signal very carefully and, based on what they figure out, they make a guess about where various sounds should be directed. The new ones are much smarter, and so do a better job of faking it, in ways that are more often pleasing to more people....
However, and here is the point I was getting to, they are still faking it.
Because of this simple fact any discussion of which is more accurate constitutes a sort of surrealism... The only question is of which one sounds better, to you, on your system, with the particular content you happen to be listening to...
Personally, I would rather listen to Dolby Atmos movies in Dolby Atmos, and stereo movies in stereo... (And I still prefer The Maltese Falcon in original black and white...)
I know plenty of people who find the results of various upmixers to be quite pleasant... However, personally, I agree that they all sound "unnatural"...
I made my report. I said I don’t like either upmixers. They both have problems, and sound unnatural to me. I have no other reference to show me if they are doing what they are supposed to do or not. Don’t presume to know what I want to contribute. I have no interest in what you are investigating here. Since there are only 21 votes so far, 12 who have problems, among 10,000 owners, (so I read), I’d say there aren’t many other owners who have an interest in what you are investigating here. Carry on. Thanks for your editorial. I tried to state some of what you said in an earlier and first post at this thread: I too have listened to surround formats since the 70's. I was never an early adopter of these technologies. The best I heard in the early days of multi-channel sound was 4 ch. 8 tracks. Channels were discrete. It wasn't till about 15 years ago when I made my first multi-channel purchase - an HK AV receiver. Tried it's upmixing capabilities and didn't care for them, even when the sound was pleasant. The current crop of upmixers is an evolution. It is my opinion - based on what I read at this thread, the predominant problem here with upmixers is unrealistic expectations, encouraged by over enthusiastic marketing from both Dolby and DTS. Dolby being the marketing champs. The real argument about upmixers, and formats, is between Dolby and open sourced DTS, and is about politics, power, and money. Sound is a secondary consideration. DTS is the underdog. DTS is the peoples's format. DTS has the overall edge in sound quality and versatility. Follow the money.
|
|
|
Post by megash0n on Jun 18, 2020 6:51:59 GMT -5
I've got to make what amounts to an editorial comment here...
I've been listening to "upmixers" since the 1970's - when I got my first "SQ-4 four channel receiver". If you play a recording that was SQ-4 encoded through an SQ-4 decoder then it acts as a decoder (although not a very accurate one). If you play a two channel recording through an SQ-4 decoder it acts as an "upmixer" (which is another name for "channel synthesizer").
Dolby Digital, and DTS, and later Dolby Atmos, and DTS:X, are true decoders. Specific content is encoded by the encoder and, when you decode it, you get back those original channels. In contrast... "upmixers" are just faking it. They look at the incoming signal in certain ways, analyze certain details, then guess what content should go where.
Dolby PLIIx was sort of like SQ-4. You could encode content with the idea that, by encoding it a certain way, you could encourage the decoder to decode it a certain way. It wasn't exact for the same reason - because things like phase shifts in the original content could cause the results to vary from what you planned.
We then progressed to encoders like Dolby Digital and DTS - which actually save each channel separately - and retrieve it exactly. And then to encoders like Dolby Atmos and DTS:X - which store the locations of individual objects - and recreate the original when they decode it.
Both the current Dolby Surround Upmixer, and the DTS:X Neural-X Upmixer, do essentially the same thing as SQ-4 and PLIIx. They analyze the signal very carefully and, based on what they figure out, they make a guess about where various sounds should be directed. The new ones are much smarter, and so do a better job of faking it, in ways that are more often pleasing to more people....
However, and here is the point I was getting to, they are still faking it.
Because of this simple fact any discussion of which is more accurate constitutes a sort of surrealism... The only question is of which one sounds better, to you, on your system, with the particular content you happen to be listening to...
Personally, I would rather listen to Dolby Atmos movies in Dolby Atmos, and stereo movies in stereo... (And I still prefer The Maltese Falcon in original black and white...)
I know plenty of people who find the results of various upmixers to be quite pleasant... However, personally, I agree that they all sound "unnatural"...
Thanks for your editorial. I tried to state some of what you said in an earlier and first post at this thread: I too have listened to surround formats since the 70's. I was never an early adopter of these technologies. The best I heard in the early days of multi-channel sound was 4 ch. 8 tracks. Channels were discrete. It wasn't till about 15 years ago when I made my first multi-channel purchase - an HK AV receiver. Tried it's upmixing capabilities and didn't care for them, even when the sound was pleasant. It is my opinion, based on what I read at this thread, that the predominant problem here with upmixers is unrealistic expectations. You may be correct. I had a very different response, but decided to keep it civil and move on from the temptation. It is unrealistic to expect everything to work as it was intended. Very few people care what you, Keith or I "prefer" to listen to. What some of us care about is that these processors work correctly %99 of the time. Then, and only then, can we determine what our ears prefer.
|
|
|
Post by PaulBe on Jun 18, 2020 7:22:09 GMT -5
Thanks for your editorial. I tried to state some of what you said in an earlier and first post at this thread: I too have listened to surround formats since the 70's. I was never an early adopter of these technologies. The best I heard in the early days of multi-channel sound was 4 ch. 8 tracks. Channels were discrete. It wasn't till about 15 years ago when I made my first multi-channel purchase - an HK AV receiver. Tried it's upmixing capabilities and didn't care for them, even when the sound was pleasant. It is my opinion, based on what I read at this thread, that the predominant problem here with upmixers is unrealistic expectations. You may be correct. I had a very different response, but decided to keep it civil and move on from the temptation. It is unrealistic to expect everything to work as it was intended. Very few people care what you, Keith or I "prefer" to listen to. What some of us care about is that these processors work correctly %99 of the time. Then, and only then, can we determine what our ears prefer. I made edits. I saw your first response. Moving on from temptation was a good choice.
|
|