Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Johnny B]

Now I do not know if this will embarrass me or not, but, "what the heck?" I'm beyond that...ha..ha...

I just got some kind of automated email notice for the NYC AES with all kinds of new product announcements on it.

And CAKEWALK, yes, that's right, CAKEWALK is making a claim that it is the first truly 64-DAW....that claim may be a bit exaggerated...

But more importantly how will 64 bit DAW's affect the "thruput" of the digital audio formats and signals? Will it slow it down, speed it up...it may be hard to tell at this point without more data...

Perhaps the people going to AES could press the "thruput" issue and get some definitive answers about "thruput impacts"



 

[danickstr]

hey combine this with Creative's E-MU 192K converters and voila! Overdone power in the low end package!  What more could little Junior want?

[Johnny B]

Yeah, maybe...maybe not...

The question still remains regarding the "Thruput Issue"...

If all the DAW's go to 64-bit, then what will happen with the "Thruput Issue?"

[PookyNMR]

In another forum an engineer from Apple stated how at this point there will be little benefit to the DAW software other than obvious things like addressable RAM.

Not sure on how things are going on the Windows side, but AFAIK the 64 bit Windows is not near done yet.  64 bit app on an incomplete OS?  No thanks.  Windows has enough problems at 32 bits...

Sound like another sales gimick to me to trick the ignorant into thinking their getting more than they really are.  Kind of like 192k converters.  

[Johnny B]

You may get some argument on both counts...

But let's say the OS already was 64-bit, just for purposes of discussion and that it was now stable...say Longhorn was stable and was now at a full 64-bits, OK.

Now address the thruput issue...

[Jon Hodgson]

Johnny B wrote on Mon, 03 October 2005 08:06

You may get some argument on both counts... But let's say the OS already was 64-bit, just for purposes of discussion and that it was now stable...say Longhorn was stable and was now at a full 64-bits, OK. Now address the thruput issue...

It's a hard one to be certain of, because in such complex systems you have more than one factor.

If the programmer (or the compiler) is not using the vector unit of the processor (SSE2) to do operations in parallel, then the processing should happen at pretty much the same speed, but if he makes extensive use of it then he can only process two samples at once rather than four.

With 64 bits moving around there's just more data to move, which is going to fill up your caches and slow things down there.

On the other hand, using the 64 bit mode of the processor gives you more registers, which can really speed things up in some cases (probably not the audio processing, but other aread of the code).

All in all I would expect throughput to be lower, but by how much I don't know, especially since I don't know how much actual audio processing goes on at 64 bits... for example if they just used it in the mix bus, then I would guess that the performance improvements from the extra registers would more than make up for the performance hit in the mixbus.

[PookyNMR]

Johnny B wrote on Mon, 03 October 2005 01:06

You may get some argument on both counts... But let's say the OS already was 64-bit, just for purposes of discussion and that it was now stable...say Longhorn was stable and was now at a full 64-bits, OK. Now address the thruput issue...

Since my friend from Apple deigned the 64-bit G5, I'll take his word over what any other person less qualified has to say.

It's pointless to event disucss "assuming there was a full release 64-bit version of Windows" because there is not such a version.  This is one of the things that makes this new Cakewalk release a pointless gimick.

As far as thruput goes..  I think there is a lot of misconceptions of what 64 bit actually is and what it does.  I'm certainly not a computer engineer or any kind of expert.  But when I hear people who are saying that there will be little increase for those of us processing audio.  I believe them.  

[Johnny B]

Jon,

Let's continue with our assumption about a stable full 64-bit OS, OK?

Now let's consider this factor, there are those who argue about bit depths, conversion, dither, truncation and whatnot, right?

Those are all intermediate steps. And those kinds of intermediate steps can create mischief, right?

What if you wasted some bits in the ADDA but you were able to bypass all or most of the intermediate steps as a result? That might actually result in some benefits: i.e., it could result in the elimination of some of the introduced errors, timing problems, and might even eventually speed up the thruput.

In more simple terms: Less steps = Less problems.



 

[Jon Hodgson]

Johnny B wrote on Mon, 03 October 2005 17:56

Jon, Let's continue with our assumption about a stable full 64-bit OS, OK? Now let's consider this factor, there are those who argue about bit depths, conversion, dither, truncation and whatnot, right? Those are all intermediate steps. And those kinds of intermediate steps can create mischief, right? What if you wasted some bits in the ADDA but you were able to bypass all or most of the intermediate steps as a result? That might actually result in some benefits: i.e., it could result in the elimination of some of the introduced errors, timing problems, and might even eventually speed up the thruput. In more simple terms: Less steps = Less problems.

If clock cycles were no object, then I'd simply process everything in 64 bit floats and be done with it. It certainly reduces the chances of problems. However processor speed is for the time being still an issue, so the brute force approach loses out to more careful algorithm design and implementation on commercial grounds for the time being.

However there aren't many places where it would lead to a reduction in processing steps, so compared to having to move and process twice as much data it is unlikely you'd see a throughput increase.

Compiling in 64 bit mode on the AMD and Intel processors will improve the speed of 32 bit calculations though, since the compiler has twice as many registers to play with.

Anyway there's no point in your ADC and DAC being more than 24 bit even if your processing is 64bit.

[Johnny B]

Jon,

You have some good answers. Still, I'm not so sure about some of it. This line of reasoning makes me wonder a bit about it.

I'm also assuming you will be seeing some super fast 64-bit CPU chips in the future, dual core, quad...who knows?...but extremely fast...

So I'm assuming a stable 64-bit OS with a super fast 64-bit CPU cranking along at blinding speeds, sufficient horsepower and adequate support chips...

And a great bus structure for the I/O....

I'm always amazed at how far we've come and how fast we got there...a long, long way from front panels, removable disk packs, and paper tape...that's for sure.














 

[seriousfun]

PookyNMR wrote on Mon, 03 October 2005 08:48

It's pointless to event disucss "assuming there was a full release 64-bit version of Windows" because there is not such a version.

Of course there is http://www.microsoft.com/windowsxp/64bit/default.mspx.

From Cakewalk's Ron Kuper http://www.digitalproducer.com/articles/viewarticle.jsp?id=3 0309:

Quote:

But the bottom line is on most of the benchmarks we’re running, there was an improvement, and it was signficant. It was like 20-30 percent more CPU efficient. Basically this is taking the same box, the same hardware, the same audio interface, running 32-bit native SONAR with a 32-bit OS, and then dual-booting and running 64-bit SONAR, 64-bit OS, the same box. And we saw a 20 to 30 percent performance gain. Now, that wasn’t on every benchmark -- I have to say your mileage may vary. In some cases, it was the same performance. The worst I saw was it was the same. But in almost all the cases it was better, and when it was better, it was significantly better.

We're certainly in a transitional period, but by all reports, better native DAW performance is just around the corner. Sonar/WXP is a stable, good sounding, full-featured DAW, as is Logic/OSX. In either case, let's look forward to embracing this 64-bit opportunity, as opposed to dismissing it before it arrives.

[Jon Hodgson]

seriousfun wrote on Mon, 03 October 2005 23:18

PookyNMR wrote on Mon, 03 October 2005 08:48 It's pointless to event disucss "assuming there was a full release 64-bit version of Windows" because there is not such a version.     Of course there is http://www.microsoft.com/windowsxp/64bit/default.mspx. From Cakewalk's Ron Kuper  http://www.digitalproducer.com/articles/viewarticle.jsp?id=3 0309: Quote: But the bottom line is on most of the benchmarks we’re running, there was an improvement, and it was signficant. It was like 20-30 percent more CPU efficient. Basically this is taking the same box, the same hardware, the same audio interface, running 32-bit native SONAR with a 32-bit OS, and then dual-booting and running 64-bit SONAR, 64-bit OS, the same box. And we saw a 20 to 30 percent performance gain. Now, that wasn’t on every benchmark -- I have to say your mileage may vary. In some cases, it was the same performance. The worst I saw was it was the same. But in almost all the cases it was better, and when it was better, it was significantly better. We're certainly in a transitional period, but by all reports, better native DAW performance is just around the corner. Sonar/WXP is a stable, good sounding, full-featured DAW, as is Logic/OSX. In either case, let's look forward to embracing this 64-bit opportunity, as opposed to dismissing it before it arrives.

Well the processor is more efficient in 64 bit mode because it has more registers (not because it is processing 64 bits at a time)... but are they actually doing the audio processing at 64 bits?

[Johnny B]

Jon,

Another good question...and something that should be tested.

We just saw that this Cakewalk guy was initially skeptical until he actually did it...but he did it anyway and got the the good results....

That's how it often works, you try stuff anyway regardless of what some people may say...and sometimes what you find after testing is that some stuff just works better...

You have to be willing to try new stuff..

[Jon Hodgson]

Johnny,

I'm not quite sure I see what you're saying.

The fact is that the extended (64bit) mode of the 64 bit x86 range is faster than the 32 bit mode, whether it is processing 32 bit samples or 64 bits samples.

But it will be slower processing 64 bit samples than 32 bit samples.

Basically if you use it to the max, the fastest processing on those processors will be 32 bit processing in 64 bit mode, that's a given, since you have twice as many normal and vector registers available compared to 32 bit mode, but twice the processing units and bandwidth compared to processing 64 bit values.

A simple recompile without changing a single line of source code will see a performance improvement, the processor is quite simple faster at doing the same things.

[Johnny B]

Emmm, the basic idea I'm trying to relate is that for every intermediate step you can eliminate, the less errors will be introduced and the faster the overall process should be. Just for the purposes of discussion, let's say you did everything in 64-bit words.

Sure, there would be some bits wasted in some of the process, but doing away with some of the conversions in the chain may yield some unexpected benefits. It's not enough to simply attack new or novel approaches based solely on theoretical analysis, you actually have to try new things to see what will happen.

Anyway, here's my major point...just look at the data flow and see how it might be improved...can we eliminate steps? What happens when we eliminate steps?

I hope you are getting the idea I'm trying to express. Speed up the data flow and make it more efficient on an overall system basis.

[Jon Hodgson]

Johnny B wrote on Tue, 04 October 2005 20:02

Emmm, the basic idea I'm trying to relate is that for every intermediate step you can eliminate, the less errors will be introduced and the faster the overall process should be. Just for the purposes of discussion, let's say you did everything in 64-bit words. Sure, there would be some bits wasted in some of the process, but doing away with some of the conversions in the chain may yield some unexpected benefits. It's not enough to simply attack new or novel approaches based solely on theoretical analysis, you actually have to try new things to see what will happen. Anyway, here's my major point...just look at the data flow and see how it might be improved...can we eliminate steps? What happens when we eliminate steps? I hope you are getting the idea I'm trying to express. Speed up the data flow and make it more efficient on an overall system basis.

Johnny,

Theoretical analysis?

Unlike you I am not guessing at these things, programming and optimizing DSP code on various processors is how I make my living. My experience is practical, not theoretical.

[Johnny B]

Jon,

Fine. So you see absolutely no areas that could be improved? You see absolutely no areas that should even be examined? Are you saying it's perfect? Maybe I'm misunderstanding you. That's easy for me to do, since I've been told by my wife that I have a small brain.

[Jon Hodgson]

Johnny B wrote on Wed, 05 October 2005 05:44

Jon, Fine. So you see absolutely no areas that could be improved? You see absolutely no areas that should even be examined? Are you saying it's perfect? Maybe I'm misunderstanding you. That's easy for me to do, since I've been told by my wife that I have a small brain.

Johnny,

Do you have selective memory, a reading impairment, or simply the attention span of a goldfish?

If none of those are the case, then how have you possibly missed the number of times I've stated on this forum that there IS room for improvement?

But some of us aren't just talking about these things, we're doing them every day, so we happen to know what we're talking about. This is not black magic, I know what the algorithms do, I know how the processors work, so when you drone on about analysing where performance improvements can be made by skipping steps due to processing in 64 bits.. I'VE ALREADY DONE THE ANALYSIS.


When Newton's equations of motion started being inadequate due to us doing things where relativity had an effect, can you imagine where we'd be if Einstein had listened to someone saying "Why don't you just step off the top of the building, you don't know, you might find that gravity doesn't work EVERY time"

[Johnny B]

Jon,

I have been told by some doctors that I suffer from ADHD, so I apologise for that defect.

Hmmm, IIRC, gravity laws work OK on Mother Earth, not so well in outer space. There are other areas that follow the same course.

I'm glad you feel there is room for improvement.

Given the trend towrd 64-bit, how about if you outline some of your improvement priorities for us?

[stuntbutt]

The pattern:

1) Johnny B starts thread with question

2) Someone with actual knowledge replies

3)  Johnny B implies person doesn't know what they are talking about


Why not try this:

1)  Start thread with question

2) Someone with actual knowledge replies

3) Thank poster for sharing

4) Incorporate new knowledge/insight into your own quest for understanding

-----------------------------------------------------------
John Katsafanas

[Johnny B]

I like his format better.

1. Start thread with question or link

2. Have debate or discussion

[Ronny]

Johnny B wrote on Wed, 05 October 2005 12:10

Jon, I have been told by some doctors that I suffer from ADHD, so I apologise for that defect. Hmmm, IIRC, gravity laws work OK on Mother Earth, not so well in outer space. There are other areas that follow the same course. I'm glad you feel there is room for improvement. Given the trend towrd 64-bit, how about if you outline some of your improvement priorities for us?

Gravity works the same in space as it does on earth, it's just that you don't notice it, in orbit you are still being pulled toward the earth by it's gravitational pull and the free fall gives the indication of no gravity. The larger an object is or actually the more mass it has, the more gravity it has. The farther away from the object, the less the gravity effects you. There is 240,000 miles of space between the earth and the moon and the earth is many times the mass of the moon, yet the moon still causes whole oceans to rise and fall several feet twice a day on the earth, a planet with much more mass. In reality, it is a physical law that works as far as we know, pretty much the same anywhere in the universe. There are variations, for example a black hole or star that has collapsed on itself that can bend and greatly effect lightwaves, but the physics of gravity are the same. It's one of the 4 major forces.  

[Johnny B]

Ronny, I do want to have this argument with about which kind of science works better in space and which works better on earth, however, I believe there is quite a bit of discussion going on about coming up with a "Unification Theory" and that "String Theory" is now in the spotlight. You may want to run some googles to learn more if you are that interested.


However, you may want to recall that the Bumble Bee was once thought to violate the laws of aerodynamics because its body weight was too heavy for its wing-span. And yet, this wonderful little creature still flies thru the air despite the old rules which said it was not possible.

The point is, and always has been, we need to keep pushing the research envelope if we ever want digital sound quality to advance.

Good scientists often admit their ignorance, bad scientists act like "know-it-all's."

[Jon Hodgson]

Johnny B wrote on Tue, 11 October 2005 17:05

Good scientists often admit their ignorance, bad scientists act like "know-it-all's."

And, pray tell, what would you call someone who acts like a "know-it-all", but doesn't even understand the science?

[Ronny]

Johnny B wrote on Tue, 11 October 2005 12:05

Ronny, I do want to have this argument with about which kind of science works better in space and which works better on earth, however, I believe there is quite a bit of discussion going on about coming up with a "Unification Theory" and that "String Theory" is now in the spotlight. You may want to run some googles to learn more if you are that interested. However, you may want to recall that the Bumble Bee was once thought to violate the laws of aerodynamics because its body weight was too heavy for its wing-span. And yet, this wonderful little creature still flies thru the air despite the old rules which said it was not possible. The point is, and always has been, we need to keep pushing the research envelope if we ever want digital sound quality to advance. Good scientists often admit their ignorance, bad scientists act like "know-it-all's."

I agree Johnny that we need to keep research going, but what you don't seem to understand is that it is going on as we speak. You act like all of the gear designers and digi R&D folks aren't trying to improve, when they "all" are. It's a continuous process and no one has given up on trying to improve. The problem lies with the area of improvement and the point of no returns. We've pretty much reached the limits of higher sample rates adding anything sonically and are close to the same with increased bit depths. What you are leaving out of the equation is the processing power and filtering that must be heightened when using higher sample rates. There is a point of diminishing returns and frankly most gear designers agree that 96k is sufficient. The only ones that don't are the ones that are selling 192k converters. Hat's off to Dan, he's a converter designer and he could play the same game and design 192k converters and market them, but fortunately for all of us, he's as honest as they come and rather than try to generate money with 192k ADC's, which he could no doubt design a top notch 192k converter, he's telling us that we have reached the point of diminishing returns at 96k. Actually IIRC, he said 60k was plenty, in the debate with Michal.

Johnny, I've run blind tests with digi consoles running at 44.1, 48, 88.2 and 96k, live mic and pre-recorded mic and instrument tracks. Particpants were accomplished musicians, a couple high profile producers and other engineer colleagues and friends of mine and not one person could consistently identify which sample rate was being used. That's between 44 and 96k and now you want to go to 384k. I'm quite confident that if you were to perform the same tests that I have, with experienced ears and long time industry participants, that your results would mirror mine and you would realize how futile it is to try to improve digital audio by raising the sample rates any higher than they are now. Any new improvements on converters, IMHO, will be on the analog sides of the converters, however how audible this will be, remains to be seen and heard. It's not that folks are giving up research, it's that they have already reached the point of diminshing returns with the higher sample rates and therefore research is going on more intensely in other areas.

[Johnny B]

Ronny I understand this position, truly I do.

Mind if I ask you a question? When you ran those tests, how many channels and instruments were you using?

I just don't think that digital sounds as good as it could. I think the CD and MP3 formats were poor design choices, and I can hear things in digital that I really don't like.

Right now, I still prefer the sound of analogue...so kill me.  

[Ronny]

Johnny B wrote on Wed, 12 October 2005 03:20

Ronny I understand this position, truly I do. Mind if I ask you a question? When you ran those tests, how many channels and instruments were you using? I just don't think that digital sounds as good as it could. I think the CD and MP3 formats were poor design choices, and I can hear things in digital that I really don't like. Right now, I still prefer the sound of analogue...so kill me.

Mics and instruments varied, however I used the same ADC's and DAC's and routed the ADC's to different channels in the digital domain, so there was no analog path or switches between the examples.  

[Johnny B]

Ronny,

Lemme clarify my question a bit, ok?

When you say you did some comparisons, did you just compare a single instrument recorded on a single channel, or did you compare 80 instruments recorded on 80 different channels. I know 80 is a big number, so maybe somewhere in the range of greater than one and less than 80...IOW, this is a single track vs a multi-track sort of question.

[Ronny]

Johnny B wrote on Wed, 12 October 2005 18:18

Ronny, Lemme clarify my question a bit, ok? When you say you did some comparisons, did you just compare a single instrument recorded on a single channel, or did you compare 80 instruments recorded on 80 different channels. I know 80 is a big number, so maybe somewhere in the range of greater than one and less than 80...IOW, this is a single rack vs a multi-track sort of question.

I've done several tests, but the one that I was referring to, just a few tracks, most of the live testing was 3 vocalists, but one vocalist at a time through a PA system. The recorded tracks were 3 mics on acoustic guitar and single vocalist track, same song recorded 4 times each at 44, 48, 88 and 96k and played back at each sample rate after all 4 examples were recorded. So no my tests didn't include what sampling rates do to 80 tracks, but over busy mixes aren't always condusive to these kind of tests where you are looking for subtle differences. My only regret is that I couldn't figure out a way to seemlessly A/B the singers on the live tests, because each wordclock change always took a few seconds.  

[Johnny B]

Well I'm not talking about overly busy mixes, merely ones with more than a few tracks.

Once the plug-ins are added to multi-track digital recordings is where I hear all sorts of problems.  You could blame it on bad code, I suppose, but I think there are some major underlying issues which make addressing these problems rather difficult with the current implementations of the currently popular technology.

In fact, if enough digital tracks are loaded up without any plugs whatsoever, I can still hear problems.  I think others hear these problems as well and have complained about it. Simply because they have not translated what they hear into precise scientific terms or math formulas, they have been subject to ridiclue and held up to scorn. I think this is grave mistake. Often the "ear people" and the musicians can hear things before the engineers get around to figuring out that these "ear people" are right about hearing the defects.

In my view, the "ear people" ought to be elevated in the process of moving digital technology forward. Although the descriptor "ear people" is somewhat of a misnomer because you are not limited to the people's ears, but rather, are looking at the extremely complex system that make up entire human beings, and that takes in the the body and the brain, among other things, the "ear people" moniker will have to do until a better descriptor comes along. In any event, this group should not be discounted and should, in fact, be elevated.

Perhaps another way to look at this would be to say, "When the 'ear people' are completely satisfied, then most human beings should be satisfied with the sound of digital. Satisfy the 'ear people,' and the analogue vs. digital debate should subside."











   

[Jon Hodgson]

Johnny B wrote on Thu, 13 October 2005 15:53

Well I'm not talking about overly busy mixes, merely ones with more than a few tracks. Once the plug-ins are added to multi-track digital recordings is where I hear all sorts of problems.   In fact, if enough digital tracks are loaded up without any plugs whatsoever, I can still hear problems.

What about for something simple, like bouncing a single track with a level change?

So if for example you did a bounce and dropped the level 5dB, and then another bounce where you raised the level 5dB, could you tell which was the original and which one had been bounced twice in a blind test?

Or does it require a bit more than that for you to start hearing problems?

[Johnny B]

Let's just say I don't think digital is "There" yet.

'nuf said....I'm out

[dcollins]

Johnny B wrote on Sat, 15 October 2005 22:39

Let's just say I don't think digital is "There" yet. 'nuf said....I'm out

Formidable.

DC

[Jon Hodgson]

dcollins wrote on Sun, 16 October 2005 07:07

Johnny B wrote on Sat, 15 October 2005 22:39 Let's just say I don't think digital is "There" yet. 'nuf said....I'm out Formidable. DC

He wants us lowly engineers to work with exhalted "Ear People" such as himself to improve the quality of audio, but any question which tries to actually identify and quantify the problems he hears, which is always the first step in making improvements, is evaded.

[Jonny Mumra]

Can we get back to the 64 bit please.

[Jon Hodgson]

Jonny Mumra wrote on Mon, 07 November 2005 12:53

Can we get back to the 64 bit please.

Sure, what would you like to know/say/suggest?

[Jonny Mumra]

Im a bit behind the 8 ball but im seriously considering going down this path of 64 bit with Sonar 5.

I saw the video from AES i think it was it seems great.
I just dont want to get sucked in like i did with high samplerates, but i suppose its worth a try.
My main concern is will the mix engine sound superior to the other DAW's and will processing plugins benefit from this 64 bit processing capability.


Also what will plugin counts be like.

[Jon Hodgson]

Jonny Mumra wrote on Mon, 07 November 2005 14:19

Im a bit behind the 8 ball but im seriously considering going down this path of 64 bit with Sonar 5. I saw the video from AES i think it was it seems great. I just dont want to get sucked in like i did with high samplerates, but i suppose its worth a try. My main concern is will the mix engine sound superior to the other DAW's

I'm finding it hard to imagine any real-world scenarios where the mix-bus would sound better, since under normal circumstances any error in a 32 bit system should be kept below the noise floor, though I could probably manufacture some artificial scenarios where it would.

Other processes (such as EQ, reverb etc) could benefit, it rather depends on what they are and how they are implemented.

Jonny Mumra wrote on Mon, 07 November 2005 14:19

and will processing plugins benefit from this 64 bit processing capability.

Depends on the plugin whether quality would be improved. However 32 bit plugins will execute faster if compiled for the 64 bit mode of the processor. To be honest I really can't see much advantage in passing data around between modules with greater than 32 bit float accuracy, even if processing internally at 64 bits yields advantages.

Jonny Mumra wrote on Mon, 07 November 2005 14:19

Also what will plugin counts be like.

I'm afraid I don't know as yet.

[Jonny Mumra]

Cool.

Your responses seem quite definitive which leeds me to believe im excited about nothing.

Thanks for your input.


How do you predict the future of 64 bit DAW.

Is 64 bit overkill or are in 10 years will we see 256bit DAW's

Huge figure i know but who would of thought 20 years ago over 100 gigs of accessible ram.

[Jon Hodgson]

64 bits across the board is already overkill in my opinion, 256 bit would be nuts, and anyway, unlike 64 bits which is already supported with good performance by general purpose desktop processors, greater resolutions almost certainly never will be, because their only real use would be in scientific analysis.

[maxdimario]

anyone tried a mix on a 64 bit program with a 64 bit computer and compared?

I am beginning to look around myself for a solution, and after 20 years of buying musical equipment, I find that the sonic end result is only relatively tied to what people SAY about most kinds of equipment.

I don't think this is intentional, but a result of people getting caught up in unimportant technical definitions.

the only real argument is: how does it sound?

[Jonny Mumra]

Im very excited about the future of digital audio though i must say.

I cant wait to give this 64 bit stuff a go.

[Ronny]

maxdimario wrote on Mon, 07 November 2005 10:29

anyone tried a mix on a 64 bit program with a 64 bit computer and compared? I am beginning to look around myself for a solution, and after 20 years of buying musical equipment, I find that the sonic end result is only relatively tied to what people SAY about most kinds of equipment. I don't think this is intentional, but a result of people getting caught up in unimportant technical definitions. the only real argument is: how does it sound?

"How it sounds" will always be an argument, as it's highly subjective to individual perception, experience and tastes. If someone records on a Roland VS880 and they move up to VS2480, they may say wow this thing sounds great, but you move on down the line to a VS2480 owner that just upgraded to Radar and he's may say the 2480 sucks. What I'm trying to say is, when you hear a good review on a product, you must take into consideration what the persons experience is that is giving the review and what the other gear is used as a basis for his comments.

[Jon Hodgson]

Ronny wrote on Tue, 08 November 2005 05:51

maxdimario wrote on Mon, 07 November 2005 10:29 anyone tried a mix on a 64 bit program with a 64 bit computer and compared? I am beginning to look around myself for a solution, and after 20 years of buying musical equipment, I find that the sonic end result is only relatively tied to what people SAY about most kinds of equipment. I don't think this is intentional, but a result of people getting caught up in unimportant technical definitions. the only real argument is: how does it sound? "How it sounds" will always be an argument, as it's highly subjective to individual perception, experience and tastes. If someone records on a Roland VS880 and they move up to VS2480, they may say wow this thing sounds great, but you move on down the line to a VS2480 owner that just upgraded to Radar and he's may say the 2480 sucks. What I'm trying to say is, when you hear a good review on a product, you must take into consideration what the persons experience is that is giving the review and what the other gear is used as a basis for his comments.

There is something else to consider, and that is how often people will swear blind they can hear something which blind tests then prove they do not. Suggestion is hugely powerful, to all human beings, and not just in audio.

On more than one occasion I've seen claims which I know were rubbish in magazines and online based on unscientific listening tests.