{"@context":"http://iiif.io/api/presentation/3/context.json","id":"https://arsc.aviaryplatform.com/iiif/nc5s75724t/manifest","type":"Manifest","label":{"en":["Expert Transfer Techniques: A Special Focus on Mechanical Discs"]},"logo":"https://d9jk7wjtjpu5g.cloudfront.net/organizations/logo_images/000/000/019/original/ARSC_Full_Logo_RGB_K.jpg?1605438091","metadata":[{"label":{"en":["Agent"]},"value":{"en":["Stefano S. Cavaglieri (Presenter)","Maya Lerman (Chair)","Michael Biel (Videographer)","Leah Biel (Videographer)"]}},{"label":{"en":["Date"]},"value":{"en":["2017-05-11 (Created)"]}},{"label":{"en":["Format"]},"value":{"en":["Video","Audio"]}},{"label":{"en":["Description"]},"value":{"en":["\u003cp\u003eThe presentation leads through the problem of transfer, digitization, and restoration of historical obsolete disc formats. Starting with the possibilities, advantages, and limitations of a conventional mechanical transfer, the discussion will outline some of the most proven and tested optical transfer methods and technologies and their special usability with broken/delaminated/damaged discs. The different approaches will be presented, including various audio examples.\u003c/p\u003e"]}},{"label":{"en":["Language"]},"value":{"en":["English"]}},{"label":{"en":["Publisher"]},"value":{"en":["Association for Recorded Sound Collections"]}},{"label":{"en":["Rights Statement"]},"value":{"en":["\u003cp\u003eCopyright Association for Recorded Sound Collections\u003c/p\u003e"]}},{"label":{"en":["Video Editor"]},"value":{"en":["Amanda McCabe"]}}],"summary":{"en":["\u003cp\u003eThe presentation leads through the problem of transfer, digitization, and restoration of historical obsolete disc formats. Starting with the possibilities, advantages, and limitations of a conventional mechanical transfer, the discussion will outline some of the most proven and tested optical transfer methods and technologies and their special usability with broken/delaminated/damaged discs. The different approaches will be presented, including various audio examples.\u003c/p\u003e"]},"requiredStatement":{"label":{"en":["Attribution"]},"value":{"en":["\u003cp\u003eCopyright Association for Recorded Sound Collections\u003c/p\u003e"]}},"provider":[{"id":"https://arsc.aviaryplatform.com/aboutus","type":"Agent","label":{"en":["Association for Recorded Sound Collections"]},"homepage":[{"id":"https://arsc.aviaryplatform.com/","type":"Text","label":{"en":["Association for Recorded Sound Collections"]},"format":"text/html"}],"logo":[{"id":"https://d9jk7wjtjpu5g.cloudfront.net/organizations/logo_images/000/000/019/original/ARSC_Full_Logo_RGB_K.jpg?1605438091","type":"Image"}]}],"thumbnail":[{"id":"https://d9jk7wjtjpu5g.cloudfront.net/collection_resource_files/thumbnails/000/097/535/small/open-uri20200922-6764-agydyt_1600816290.jpg?1600801911","type":"Image","format":"image/jpeg"}],"items":[{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535","type":"Canvas","label":{"en":["Media File 1 of 2 - open-uri20200922-6764-agydyt.mp4"]},"duration":2913.984,"width":640,"height":360,"thumbnail":[{"id":"https://d9jk7wjtjpu5g.cloudfront.net/collection_resource_files/thumbnails/000/097/535/small/open-uri20200922-6764-agydyt_1600816290.jpg?1600801911","type":"Image","format":"image/jpeg"}],"items":[{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/content/1","type":"AnnotationPage","items":[{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/content/1/annotation/1","type":"Annotation","motivation":"painting","body":{"id":"https://aviary-p-arsc.s3.wasabisys.com/collection_resource_files/resource_files/000/097/535/original/open-uri20200922-6764-agydyt.mp4?1600801881","type":"Video","format":"video/mp4","duration":2913.984,"width":640,"height":360},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535","metadata":[]}]}],"annotations":[{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074","type":"AnnotationPage","label":{"en":["AUTO_TRINT_Expert Transfer Techniques: A Special Focus on Mechanical Discs [Transcript]"]},"items":[{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/1","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"And the first presentation is expert transfer techniques. A special focus on mechanical discs. And our presenter is Stephane. Sergio Cavalieri and Stefano is the chief technology and information officer of the Swiss National Sound Archives. Now a department of the Swiss National Library in Lugano, Switzerland. His career started back in the 70s and the audio engineering field for a number of years. He was assigned to all kinds of audio productions, culminating his professional career with a series of World-Class recordings and events. In the late 80s, he brought and his interests to include computer science starting dealing with networking and systems management. He's very soon got got involved in software design and all things digital storage. His expertize and analytical skills led him to be now recognized as one of the highest technical authorities and the multimedia archives community. In 2011, he was awarded the James A.. Lynn Lindner Prize for his contributions to research in the field of the technology of preservation of recorded sound. The final holds a degree both in electro acoustics and in computer science. He served as an active member of Yassa Technical Committee Alé A.D.s and ask. He's helped me walk on Stefano. Good morning. Thank you very much for the introduction. We are going to talk about transfer techniques, traditional mechanical transfer techniques versus some things that people call emerging technologies such as optical replay of mechanical records. Maybe a first question, how many of you are. Have a have an engineering or technical background here? OK, that's good. It's just just. It's good. It's good to know because I can get very, very technical in my presentation or I can just skip some some part of the microphone. All right. All right. OK. Let's take this to the rerecording principles.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=11.86,170.31"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/2","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"They say those SCHAFT ensure the health and safety of the recording. So this is the most important part or the most important sentence when you ever rerecord for archival purposes. In order to do so, we should extract the best possible signal from the original recording, which translates into carefully adjusting mechanical as well as the electronic components to match the specs of the original recording. Because on the copy corrections are technical, limited, if not impossible. A rerecording is kind of an art and science. It has a very scientific part in it, but it's connected to an artistic part because there is with the usual means, there is always some kind of an interpretation of what we get out of the off the groove. So in usual terms, when we rerecord from a record from anything that has a groove on it, we have to think about groove geometry. We have we have to select the stylus shape and size. We have to check hey it adjustment of the torn arm, the lateral balance as a mode up the cartridge. We have tracking adjustments to take care of. And we have a bias adjustment on the electrical adjustments we have. Of course, the level and the DNA forces or turnover and Roloff kind of equalization. But we also have things like typical mechanical issues like speed choice or speed selection centering of the record many of records on off the center. And we have to deal with or being. Now, I will skip all the part concerning the different adjustments need for a mechanical playback with a turntable, because the focus of these presentation is to make a comparison between the mechanical and the optical means. So the first thing that comes to my mind is the transducers.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=173.61,350.24"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/3","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"We have mechanical transducers when we talk about acoustical records. No electricity in between. We have air pressure that causes physical displacement. At the end of the chain, when we write for it, when we when we cut the groove into the record. And on the other end, when we replay the record, the physical displacement turn in to air pressure. By the end, all of this, as you can see, is black because it is amplitude responsive. Now, if we go to the electoral map, mechanical means, which is what we use, what we use usually most of the times. We have on the cutting. And so at the very beginning, we have either voltage that turns into a magnetic flux, that turns into a physical displacement, which would be amplitude responsive if we will use an open loop cutting hard or we have the same thing, but in red, because it becomes a loss of the responsive if we use a closed loop. Cutting head, which is how most of the cutting has to work. And when we replay the record, our physical displacement turns into a magnetic flux that turns into a voltage can be then listen to it or when when it's amplified. And this also is the loss of the response. Then we have another part, which is not a we can not talk about transducers, but in the optical world, we haven't the the image contours that are translate that into the audio signal signal because the image contours represent the audio signal as it is. Now, velocity, velocity is probably the major issue when we compare two different systems. Here you can see an explanation of what were lost. He hears it depends, of course, on the amplitude, but it also depends on the frequency.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=352.89,531.46"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/4","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"So the distance, which is travel divided by time by the stylus, is known as the velocity. What are the issues with the law city and why? Why do we care about velocity and all related problems? If you look at the left hand side. You see it here, a radical drop in Digby's per octave. If we would cut a record at constant velocity so we would have a six theby per octave drop from the very lowest frequency to the very highest frequency. If we take up the whole eight and a half octave, all nine octaves, we would have more than 50 Didi's drop or 50 for it to be dropped, which at the replay would have to be compensate. In fact, the picture on the right hand side shows the red line, which is the theoretical in real life. It doesn't work this way. But this is just the practical response of a magnetic cartridge. So somebody which was very clever, invented re-emphasis pre emphasis, turns of the lost city curve in to an almost amplitude sensitive curve. Which makes things a little bit easier when it comes to replay the fact. If we look at this. It says very clearly why we need a pre emphasis and because otherwise we would have low frequencies that would require such a such a big such a huge displacement that wouldn't fit on a record. And on the other hand, we need better or we try to get a better signal to noise ratio, which is obtained by raising a little bit the high frequencies while cutting. But this helps us limiting the deviation of our groove. From the very low frequencies up to the high frequencies within certain certain levels. Now. The next big point or peak, it's not it's not the real issue, it's just it's just the that's just the point.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=534.67,743.68"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/5","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"Are the analog to digital converters. We all know what an analog to digital converter does. It turns an electric signal into numbers. This is basically what it does. The choice of the analog to digital converter. For a given application is defined for the requirements you have. And this is also theory because when you're buying an analog to digital converter, you don't know what's in it. You don't know that the chip that does the conversion, what kind of converter it is, is a second with the. But it is something else you don't know. So here we have different architecture of analog to digital. So we have flash, we have pipelined, et cetera, et cetera. And at the end, we have to signal, they have to convert, which is widespread today because it is easy to build. It is small. It is quite accurate. But if we look at some earlier digital machines, such as the Sony cash machines, for example, or does it still with cash machines, they had sahi an analog to digital converter and not seeking a. We have a coding scheme. If we look at the characteristics of the converter, we may have an offset ever made. We may have a case in error. We may have non-linearity errors. And of course, we end up having resolution requirements and conversion speed requirements. If we look at the AAA this year, Kersee, you see some figures on the bottom because the main point is a convertor, let's say a sixteen beats convertor to be cold that way. It should be capable of addressing every single bit, not just the top 12, four or that. So what the what? What that means is that if we feed the convertor with a reference voltage of five volts, then the less significant bit or that the smallest difference involved, that would be of 76 micro volts, which is not too much.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=744.85,936.26"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/6","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"It's a very little if you talk about 24 beats conversion. We go down because we don't talk about micro volts. Nanomoles, which is divided by thousand. Again. So the point is that 80 CS or analog to digital converter do not sound all the same. We all know that, but we have no power on that. We have no choice. The only choice we have is use these or these or these because we like it better. Now, if we. Try a broader view. We may ask ourselves how accurate this mechanical playback, if there are any other options, such as imaging, for instance. And I can you know, the sentence that's there. Some say that imaging is an objective means for transferring the physical representation of a sound recording to another media while retaining all. And DCT, a tent that has an archive is to jump into this kind of technology because the if I if I trust the sentence, this would be the only way to make an exact copy from the original record to something else. However, several scientists, engineers, audio files that said the right half a start that several projects, most of the projects simply failed last year. The Yarza established and Emerging Technologies subcommittee to try to shed some light on it. This is a schematic overview of one of the most well-known optical retrieval systems for records for all kinds of records, visual audio and presenting visual audio, because it is by design. Our development from my organization. But I'm not saying that it is better or worse than everything, for example, which is well known here, especially here in the US. In some aspects, it is maybe worse, but we'll see after. Let's take a comparison. A comparison between the equipment required for mechanical replay versus the visual audio equipment on the mechanical side.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=937.15,1124.45"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/7","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"We have a turntable to preemptively fire a Navy, convert our digital workstation. That is well, it's a summary. On the other side, we have a photo camera, we have a film scanner, and we have a digital imaging and audio workstation because we deal with images before turning them into audio. If we compare that to two flows for required for producing an accurate copy, we see that the list on the mechanical side is a little bit longer. All. Wait longer. But probably the most important thing is that. All almost every point remains in the analytic domain. So all what we do is, is is analog. So, of course, clean the record, we'll replace the sleeves. We inspect the grooves, choose the stylus and so on. We had just the playback level. We said get polarization. If there is an equalization or for this, we can listen to the to the copy or the signal when it comes out of the top of the pre amplifier before entering the analog to digital converter. And then we can, of course, listen carefully also off there. So in order to make a real comparison between what we get of the out of the our our tools, the biases and what is then converted to numbers on the visual audio side. We have a certain number of procedural steps that happen digitally because once we have the digital picture, which is what we scan or what we derive from the from the analog picture is and remains in the digital world. So we we we determine and said the right speed. For example, we said the equalization and just the playback level digitally. We cannot do it in the analog domain. Whoops, sorry. Was the wrong. OK, now, if we make a comparison on timings typical replayed time for one side, 78 R.P.M.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=1124.91,1299.19"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/8","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"record is 60 Minutes, including steel, phylis choice, mechanical and electrical adjustments and so on. Typical replay time with and with an optical system. It's much longer, but it is divide that in different steps that can be also separated and performed and different by different people. Nine different stuff, different times. So what we can compare is actually producing the the the the the digital copy with a mechanical means and producing the photography with with the optical means. If you look if you only look at this photographies that a little bit quicker. Again. I should switch off. This is. All right. Sorry for that. No. We have some sort of bitmap to wave, convert to deal with when we talk about optical means, typically on a two dimensional imaging technology system. We simply read the roster. Imagine representing the signal, which is some divide, and then pixel and each pixel has plan or coordinates that tell the tell you apart about the amplitude. This is it. There is no magic. If we make a comparison between the analog to digital converter characteristics and the input and beat map to wave converter characteristics, there are some things in common, but just a few of them. The coding scheme, of course, the resolution and the conversion speed. And that's it. So we theoretically, we don't have offset gain ever and all the kind of errors in the optical world. So the coding scheme is a staircase. Typically, if you would draw this for on a D converter, it will be you wouldn't have this cream box, but just a stair, a staircase in the optical domain. You'll have boxes because the boxes represent the pixels. The currency is the key element. Off the beaten up to bleep wave converters so that they mention of a pixels are a power of paramount importance.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=1299.7,1494.69"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/9","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"Let's have a look at some some figures if we want to image at 16 beat. I'm not talking 24 beats, I'm talking 16 beats. A core groove. Record. So it's 78 r.p.m. record. Assuming a maximum groove. Displacement of seventy five. Micrometer at hundred. Mm. From the center of the disks and the rotational speed of San Diego r.p.m.. In order to be accurate. It would require a pixel size in the amplitude direction of one point fourteen nanometer. OK. While in the time direction we can leave with an eighteen point five micrometer pixel width which is big enough. If we take a microbrew record, of course we have a different this group displacement that turns these numbers a little bit down so that the smallest number I see here would be zero point for three nanometer indeed amplitude direction. Well, I don't even mention these numbers, because here we are talking about people meter, which is something you can't even imagine how small it is. So and but this is what we what we need for imaging at 24 bit. Ninety six kilohertz, a corkscrew fracture on the top or a microbrew of record on the bottom. Now, I've got these specs from Carl Haber. So which is which is the developer or design or off the Irene system? I didn't ask him if I could just scheu freely, but I take my chance because these these figures are have never been published. This is just something to sort. Irene uses a line science or with four thousand nine and ninety six pixels of seven micrometer square size each and an optics of roughly 10 times. So it emerges theoretically zero points and seven micrometers on the disks sort of face by using high contrast illumination and some other stuff in identifying the grayscale, he says with Irene.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=1499.24,1688.25"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/10","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"They can get a resolution close to zero point twenty five micrometer on the sort of face off the desk in the time direction. No problems. If we take visual audio, which is our system, we shoot an analog picture on a on a midsize and all film on an actual license film, which, as you can see, all the numbers there. So they halide Crystal from LA Radio zero point two microns. That forms like clouds. It's a little bit different. If we compare these two up to a digital picture, but we scanned the film using line sensor with 10 micrometers, square size pixels. We use optics to magnify that. And we use coarse edge detection and reconstructions. So we come close theoretically to the same figures as Irene, which is zero point twenty five micrometers for a pixel. Same thing in the time direction. No problems. And just to show you that we are not faking anything. Of course it is magnified. But the pixels, you know, the little squares you see on the screen are the real world pixels that you get on the final image before it is transformed into audio. Okay. So we have up here, we can measure eighty three pixels times 2.5. It gives two hundred seven point five, which means if you do some divisions and then you. And you. He'll lower it a little bit. I hate to say this, these group calls for a two point three meals. Files, more or less can be a little bit less, a little bit more. But it's about this. Now things get worse, unfortunately. If we look at the amplitude. I did that on a on an A-S calibration record, one kilohertz. Zero to be a reference level. And all the other numbers that are mentioned on the record itself.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=1688.82,1851.89"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/11","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"Well, I get I get twelve pixels, which means 30 micrometers, which means it's true. Because if I refer to that up to the table we have earlier with that, with the pre emphasis curve. The numbers correspond. So that's true. If you look at the bottom, we see a correlated representation, a wave lab sample by sample. Same record when we replay that using a turntable. So. We have constraints. We have sampling is it does not represent an issue, but the quantization is the real issue. And we can we can do nothing about that at this time. And with the means, we have lack of information, least processing limitations, such as for the equalization, for example, and for all the kind of restoration. Now, just to just give us some more numbers, quantization issues. How bad is that? This is the resolution we get. Worst case and best case with visual audio and. And with Irene on course. Girl of record. On the top, on micro growth records on the bottom. So as you can see, the maximum resolution we get on cause group report with either system way of assuming that we are really capable of spotting a zero point twenty five microns pixels on the surface, which is something that, you know, just leaves in the number sometimes. But we can get no more than nine resolution. That's it. That's it. No way. And this is for maximal a full scale grooves, displacements. So it's full signal. It's another minus 10 or something. So this is just the Auri. If we take a music piece, for example, it will never be like that. It will be much, much smaller than that. So the resolution will be much smaller than that. Well, this is what happens at one kilohertz.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=1853.31,2013.45"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/12","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"Why? Simply because at one kilohertz, we have a smaller group of maximum group displacement. We don't have the same with. We do not cut the grooves in constant amplitude. We are not doing that. So this is the result. And we can say at 70 hertz on a course growth record. In order to be accurate, we will need pixels or sub pixels. Two hundred and nineteen times smaller than we had than the best case we had. Not the worst case. And this is this is the problem. The come off the storm. I have some I have a couple of audio samples just at the end. But before I would say, well, photography or picturing or optical retrieval systems still have benefits. You know, the photo person is a very, very, very accurate copy of the original signal. If if it would be the right resolution, I mean, the optical resolution, which is not with the with with the current means, we have that field. One thing we have that you can not beat with mechanical with or you can even match with mechanical means is we can replay broken records. The laminated records, you know, all these acetate records that starts falling apart. We can't take a picture of that and we can work on the picture in order to reconstruct the audio. So this is probably the only application, the only real strength of optical technology today. So there is room for improvement on the image captured. There is room for improvement on the image processing and on the signal processing is the same sentence as before. Mm hmm. Just to say that imaging is still a good thing. I want to thank you all these people for the information that gave to me and has a very, very last thing I will.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=2017.61,2188.34"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/13","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"This is just a little extract of our shelach record. Right. With a turntable. And this is the same record with an offical flat with no, I mean, we know it by ization, the first one where well, with the emphasis this is without any equalization. Of course, you have your you are missing some high frequency. It's not there. A resolution is not high enough. And you have too much low frequency is why you'll have more too much more frequency. Because here we are reading in constant amplitude while we wind up with a car CRH. We are reading in philosophy, philosophy, science. That's it. Thank you very much. So much questions, George, I'm scared. So in your example, looking at the comparing the risk. Should. Yeah. That. It seems that there is a convergence of the image data representation of the other night of the audio signals. My understanding is that what you're seeing is that what you're saying is that that's not the way it works, is it uses the words and then takes a derivative of the boundaries to convert it to, unfortunately. Yes and no, you're not. If you talk about the derivatives, it is the only, let's say, one way to get a better resolution instead of only reading the pixels. You know, if you're the net because in a pixel, you don't have a black or white information. You'll have different information. So you can say if this pixel is completely black, I'm on the edge. If this is a little less than black and you'll have a gradings. Great. Great. You can assume you can assume that it is a little bit lower. So that makes it possible to transform things into what they called sub pixel mode. Well, it makes sense.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=2198.28,2414.17"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/14","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"So we don't only take that plan our coordinates, we take the plan recording it, plus an information coming from the luminance. That's the. This is how it works. This is how different is that from the resolution of a PCM single Blow the LSP. So she's afraid she's damages. Yes, of course not. But the point is, with the pixels, we have to determine where the grooviest where the edge of the groove of the groove is. OK. Because the the movements of the groove, the displacement of the groove are much smaller than the pixels capability. They turn in two shades of gray. When you read it. Yes. Fifty six. And that's it. And that's it. See, I don't have any other means. You can artificially you can produce out of one pixel. You can produce several order smaller pixels. Just by interpreting the luminance off of the pixel. We'll take this outside. Yes. See this with her? All right. Well, of course, if we talk about two dimensional imaging, which is what we what we saw here, we need a lateral representation. I mean, if it's a vertical cut, which is not straight, I mean, that causes added a variation of the weight of the groove. It can be erect, but only in this case. Otherwise, you'll need a 3D system such as Irene, 3-D, Iris, Irene, 3D completely. It's something completely different. But it is usually used for cylinder playback in the case of Syria, in case of stereo. Girl, you'll have enough you'll have enough information in the lateral displacement because you have because the stereo grooves are always cut with the stylus like this. So it means when it goes down or either on one side or the other side, it causes a different thing.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=2415.1,2581.31"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/15","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"We live off the growth. That can be campy, right? Here on the front row of. I'll speak loudly and far. Based on the amount of change and the advancement, this taking place with optical sensors nowadays. What if your original capture was done electronically in the analog environment? The point is that the resolution that we achieve even today is not. It's not time and it's not higher than the analog on it. Simple as that. You know that the science of the of the pixels into the sensors doesn't matter what kind of sensor. It's still, let's say, big. So are we having or getting them sensors with a pixel smaller than one micron? It's very difficult to take accurate statements, right. I'm not talking about the sensor. You're having a smartphone or in a different way. Choices. The decision. Oh, sorry, was the first one. Sighs I don't have that one with Irene because if we can, I have the one with with visual all you want. Yeah. These shows are the first ones. One was a surprise. OK, so this one was taken with a stylus. OK? And the second one was taken with visual audio. There is a major differences. This is a stereo or two channel signal. And this is a normal signal. That was when I was coming to do. What should we do from this? What? Sorry. Well, you choose to sign off. It's actually. Well, it's actually the same sample. The one the first version was reproduced with a turntable, with a stylus, with the proper equalization. It's a very worn record. But OK, with a proper equalization and with the and the second sample was reproduction with the optical system, with no equalization. So that means just flat, flat, but indeed amplitude domain.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=2581.97,2772.97"},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/16","type":"Annotation","motivation":"transcribing","body":{"type":"TextualBody","value":"I understand what the conclusion is that with the optical means, we can get quite close, quite close. I'm not saying it's exactly the same thing to the replay with a turntable, with an equalization, without using the magnetization. OK, so wait, wait. We cannot. What I'm saying is that they cannot apply the same deemphasizes to an optical readout as we deal with with a mechanical readout. OK, well, let me first place go see the third experiment conservancy. If that first US boy, apparently Jehu being played by and to certify my son on that basis, currency would go probably simple as the noise. I'm not sure that it is acceptable to all sides with a very good option that says. Well, it's the same record. It's the same. I mean. Yeah, I know. I just I. I understand. I understand what you say. Of course I am. I accept you or your note. And yes, we could we could either use a mono cartridge or. Yeah. But I tell you, there is a difference and there will be a difference at the end. In any case. So I mean. Thank you very much.","format":"text/plain"},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535#t=2774.59,2892.68"}]},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074","type":"AnnotationPage","label":{"en":["English [Transcript]"]},"items":[{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/97535/transcript/19074/annotation/17","type":"Annotation","motivation":"subtitling","body":{"type":"TextualBody","value":"https://d9jk7wjtjpu5g.cloudfront.net/file_transcripts/associated_files/000/019/074/original/open-uri20200924-1401-1t4bubt?1600957159","format":"text/vtt","language":"en"},"target":"https://d9jk7wjtjpu5g.cloudfront.net/file_transcripts/associated_files/000/019/074/original/open-uri20200924-1401-1t4bubt?1600957159"}]}]},{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/255833","type":"Canvas","label":{"en":["Media File 2 of 2 - ARSC_conf_2017_Cavaglieri_audio.mp3"]},"duration":2903.81356,"width":640,"height":360,"thumbnail":[{"id":"https://d9jk7wjtjpu5g.cloudfront.net/public/images/audio-default.png","type":"Image","format":"image/png"}],"items":[{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/255833/content/1","type":"AnnotationPage","items":[{"id":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/255833/content/2/annotation/1","type":"Annotation","motivation":"painting","body":{"id":"https://aviary-p-arsc.s3.wasabisys.com/collection_resource_files/resource_files/000/255/833/original/ARSC_conf_2017_Cavaglieri_audio.mp3?1730761575","type":"Audio","format":"audio/mpeg","duration":2903.81356,"width":640,"height":360},"target":"https://arsc.aviaryplatform.com/collections/1144/collection_resources/29704/file/255833","metadata":[]}]}],"annotations":[]}]}