Sound

I think the issue is that the whistle on this new release is different from the original 2008 legacy GS4. 

The whistle on the 2008  is more comparable to the whistle of the real SP 4449. I have posted videos on the other threads supporting this.

I think if Lionel would have used the same whistle soundset as they used in 2008, like they did with the reruns of the NKP and PM berks folks would have not been questioning it.

No one's saying the whistle on the new is "bad" , just a bit off from the prototype.

As long as your happy Eliot that's all that matters but it doesn't hurt to have the "experts" keep Lionel on their toes. This certainly helped when it came to Lionel getting the cab forward sounds correct.

Just like the 2001 release of the Blue Comet 831... when they re-released it, it was totally wrong.

Jack, I am not questioning your expertise. I am aware that you know what the whistle sounds like and that no one that I am aware of knows it better. I am the first to acknowledge your railroading expertise. Everyone else does, too. My point is NOT that you are wrong. My point is that it seems to me that many of the initial responses to these locos is to point out a negative. The whistle is, undoubtedly, not the same. My point is that a model cannot possibly reproduce the sound the same way and that it is possible, the different tones from the same whistle may have been impacted by all types of conditions. Lionel states that the whistle tones, over time, were different. I have no clue.

It is my personal preference to avoid being upset by something wrong and to focus on what is good. I am surprised by the "hatch" issue and I do feel Lionel will somehow produce a great item and get something is wrong that should not be wrong. That is the nature of life, I'm afraid. C'est la vie.

Your opinion, freely expressed, is always valuable. I respect it. Personally, I think the whistle sounds amazingly good and, though it may not be as accurate as you would like, I love the sound. If it was up to me, you would be recorded in the cab. That would be authentic. My post is to help focus the Forum on how lucky we are to have such superb models. I owned the other GS-4 and, though I am certain you are correct about the whistle, I like this one better and I will tell you why. This whistle has superior resonance and echo. To my ears, it sounds more like a real steam loco even though it may not be the one others like.

By the way, I do consider myself an audiophile and I still think you are right about the whistle. You know, I tease you from time to time and you and I have not always agreed on things. I think you are great for the Forum and I hope to shake your hand one day and buy you a beer and a burger. Even if you think I am still a ****, it would be fun to talk to you about trains.

Eliot

I like big, fancy, not used very much, correctly spelled, words in a sentence.  Music to my eyes.  Thanks for that scrap.  

William 1 . . . . . Many tanks for youse kind woids. Hopefully, a literary agent will take notice. I need a publisher.

Scrappy

Crossword geek here.  Always like interesting words used properly in a sentence, regardless of the content.  This forum can get a little rough, in more ways than one.  Keep up the good work.

Ok, back to the regularly scheduled vitriol.

Cheers,       BK

First, I am prototypically challenged so I don't know what many of the real trains actually sound like and I couldn't identify one if I had to. On the brighter side they all sound good to me , but none have the sound of a real train. I have heard some very nice audiophile stereo systems over the years and as good as they may be they don't come close to sounding like a live band in a smaller club with good acoustics. I think I know exactly what you are saying here and I would certainly agree on that part. I do think that could have something to do with the sounds in our engines. I have no idea how to tell if an engine had the correct sounds or not, but they do sound different because of the reasons you stated above and I can tell a difference in sound in that respect. (Also, your grammar, spelling and vocabulary is much better than mine, I barely got through English class. I am certain my teachers all wanted me somewhere other than in their classrooms, sometimes that was out in the hall.)

I was coming home from my dad's house on a Saturday up to Lake Villa, Il on Wilson Rd., which crosses the tracks, this is Fox Lake, Il, must be UP, never seen anything but Metra on it.  I saw the smoke of a steam engine, I heard the whistle, and I immediately knew it was the 765.  From seeing / hearing it on the videos.  I just missed it by seconds.  There were cars by the side quickly dispersing.  I asked one guy who said I just missed it.  I asked was it the Nickel Plate 765?  He said, yeah, that sounds right.  It did sound right.  Exactly like every video of the engine I have seen.

Well Elliot it took them more than .03 seconds to tell you you were wrong. And the big words were great.  Are you sure you taught social studies and not vocabulary?

For me the sound is one of the most important factors.  For example I would like the horn or whistle to sound somewhat close to the real thing.  I understand that our ears hear things differently.  Take the NS Heritage SD70ace's the horn is absolutely atrocious and sounds nothing even remotely close to the real thing.  The sad part is Lionel already has the sound file with the correct horn.  I can only hope and pray that Lionel does not put the same sound as the NS Heritage SD70's for the next run in the current catalog.

Ah, just a couple of old farts having fun !!

I think I'd like to see MTH and Lionel offer the ability for the end-user to customize sounds.   Perhaps even going so far as to change out the whistle, horn or other engine features as it interests them.   Example:  I have an HO diesel with an ESU sound decoder.  I enjoyed the original sound for a bit but then swapped it for something else that I thought was more -- well interesting.  I know, this is probably against some law of nature... but seriously there's a bit of fun to it and I can always change it back.   MTH is sound file based and the sound files are right on their website.   There's some knowledge (not official) of the format...  And given that perhaps in theory with some tinkering one could change the sounds.  Still without direct MTH support, it's a lot of guess work.  Lionel is locked up completely in this area that I can tell ...   It seems to me that giving folks the ability to manipulate the sounds of the engines would enhance the appeal of their products overall ... 

You have a limited version of that with MTH, you can change the sound file.  You do have to change the whole file, but I've used this feature at times to "improve" the sounds of a unit.

gunrunnerjohn posted:

You have a limited version of that with MTH, you can change the sound file.  You do have to change the whole file, but I've used this feature at times to "improve" the sounds of a unit.

True, but if MTH delivers a model of a Premier steam locomotive, and the whistle is incorrect, they will issue a "corrected sound set" with the correct whistle. Such was done with the Union Pacific 9000 class 4-12-2 models, i.e. it was shipped with the 844 "deep steamboat" whistle instead of the UP freight Star Brass whistle. 

Some just like to pick I guess. They seem to forget all the years MTH engines had the same sound board. All they did was change the station announcements. Whenever I run them, they have become SOOooo boring. Be happy we don't have that anymore.

Eliot; you are becoming very philosophical as you prepare yourself for your new layout build. But you make some really good points!  Heck I really liked the sound of the original MTH PS-1 Allegheny back in the day; the newer releases don't sound anything like it did. But the newer releases are more accurate so I am told. Go figure. Anyway I am about half deaf so I am happy to hear them at all while I still can. By the way, what is a "cognoscenti"?

Rod

I have the JLC Allegheny and the JLC Legacy Big Boy, both with generic whistles burned in. That's pretty lame, especially for what they cost at the time. I think these two engines killed my enthusiasm for buying brand new steam years ago. I think my Legacy K4 was my last new steam purchase, and even though it cost half of what the JLCs went for, it has that raspy generic Legacy whistle they used in a bunch of other engines. 

In this age of techno wonder, why are we still getting stuck with hard coded sounds and features anyway? At least burn a few whistles in and make them selectable. 

Norm Charbonneau posted:

I have the JLC Allegheny and the JLC Legacy Big Boy, both with generic whistles burned in. That's pretty lame, especially for what they cost at the time. I think these two engines killed my enthusiasm for buying brand new steam years ago. I think my Legacy K4 was my last new steam purchase, and even though it cost half of what the JLCs went for, it has that raspy generic Legacy whistle they used in a bunch of other engines. 

In this age of techno wonder, why are we still getting stuck with hard coded sounds and features anyway? At least burn a few whistles in and make them selectable. 

You hit the nail on head Norm.  I am really surprised that the sounds are not more customizable.  Even a USB port to upload custom sounds or select different horns, whistles, prime mover, etc.... 

MLAT posted:

For me the sound is one of the most important factors.  For example I would like the horn or whistle to sound somewhat close to the real thing.  I understand that our ears hear things differently.  Take the NS Heritage SD70ace's the horn is absolutely atrocious and sounds nothing even remotely close to the real thing.  The sad part is Lionel already has the sound file with the correct horn.  I can only hope and pray that Lionel does not put the same sound as the NS Heritage SD70's for the next run in the current catalog.

I second this. I have the 2009 PRR Heritage SD70ACe and the horn is identical to a real thing. In fact I heard one in person one time. Knew exactly that it was A P5 on a SD70ACe from the sound. Let's hope Lionel brings that horn back. 

Well this thread looks like it may be going a little sideways, but on the plus side I did learn what a "cognoscenti" is:

"people who are considered to be especially well informed about a particular subject"

It's a great day when you learn something new!

Rod

Just sent a note of apology and hope that anyone offended by my comments or references knows I apologize for any offense. I tried to make a point about the complaining and I did so in a way that was offensive. I will be more careful in the future and I apologize for my words and tone.

Eliot

Thank you. I deleted my comments. I'm outta this thread.

Last comments and then I will go into the "Doghouse." Both Jack and Doug are perfect examples of why so many people here go to the Forum daily to learn about trains and discuss all the facets of railroading. Sometimes, we forget how valuable these contributions are. I really mean that. In letting my emotions get the better of me, I made something really fairly inconsequential become the focus instead of the trains. I hope both of these gentlemen will continue to provide us with the knowledge and perspective over the coming years.

Now, I will trundle off to 

You can never have too much cognoscenti.  That's what makes the world go around.  And a fun forum!

It seems simple to me, if they have the correct sounds available to them, especially the whistle, which is obviously the most cared about and scrutinized sound, why don't they put them in engines?  I thought that was part of  the gig.  The models sound like the real thing.  So use the real thing's sounds in the model.

For cry eye.  

That felt like a mini rant.   That's all could muster for now.  I do feel a tad better.  Have fun with your scale model trains.  Try to anway.

Some aspects of the MTH sound file (which contains more than sound) is understood and documented here: 

http://www.silogic.com/trains/ADPCM.html

We ran the latest version of the GS4 and the older TMCC GS2. Guess which whistle sounded better? 

The GS4 details were bar none better.  However, I tend to side with the 0.3 second crowd and challenge Lionel for the increased prices, please get it right!

Now don't get me wrong.  I'm still saving for one as I aimlessly drool all over the one in my LHS display case.

This is where mini 2" Bluetooth speakers come in handy. They sound great and can be added to a baggage car behind the tender and used to play sound clips of your own recorded whistle or from a youtube video.

Okay let's cover this whistle sound thing.  Please note that the real whistle is moving steam, not air.  I learned the difference from Robert Swanson who was the designer of the whistle for the CP Royal Hudson.  He explained how steam gives "bite" to the whistle as opposed to compressed air. The whistle he created is an A minor 7th chord which is currently the chord used by Nathan in their K5LA horn frequently heard on contemporary diesel locomotives.  

I also learned not to simulate whistles with sine wave generators because a whistle is a circular wave form much like a trumpet.

Hence I recorded the real whistles using a casette recorder and digitizing them with a 386 computer and an Antex board.  This was a 20+ years ago.  

So the point is, steam is hot and can cause the dimensions of a whistle to change enough that a recording cannot be looped.  Many of the whistles I used at RoW were the real thing on an active locomotive.  The environment can also cause minor changes in whistle pitch.  When RoW made the NKP berk I recorded this on a cool Sunday morning in Orrville Ohio just before an excursion.  I was able to record the NKP as well as a PRR whistle that Rich had installed.  These seemed to be perfect conditions.  

Other whistles were correct for the locomotive but were run on compressed air.  This required some clock speed adjustment in the software to get the pitch closer to what it would sound like with steam.

Also note that many real original whistles were single note and considered somewhat boring for excursion runs.  Many locomotive, Reading 2100 for example, had their whistles changed to a 5 chimes that sounded great to the average listener.

With all the videos and recordings available there should not be a generic sounding whistle on a popular loco.  Keep in mind however that the sound is not going to be absolutely perfect.  A lot of work does go into creating a realistic result for sounds.

Lou N

Lou, have you thought about modelling the steam whistles digitally?

Severn posted:

Lou, have you thought about modelling the steam whistles digitally?

Yes, I designed the first ever digital on board sound systems over 20 years ago.

Lou N

I meant simulate, but not by recorded sound.

Whats a whistle?

bluelinec4 posted:

Whats a whistle?

You might want to bone-up on railroad operating rules & regulations, not to mention FRA regulations. That sound making device, is STILL always referred to as a "Whistle".

Hot Water, are you saying we should capitalize the word whistle when we're referring to a train Whistle?

What?  You mean to tell me there are rules and regulations about all this stuff

Severn posted:

I meant simulate, but not by recorded sound.

Without the technical jargon, I can do more with a single voice IC than I can with the half dozen or so chips it takes to simulate a whistle. If you want the technical explanation, let me know.

Lou N

The processing and memory size of a smartphone, laptop, tablet etc. is hundreds if not thousands of times faster/larger than an O-gauge engine.  One approach would be to generate engine sounds externally using simulation, synthesis, recording or combinations thereof.  The "final" audio would then be sent wirelessly to an engine.  Engines would of course still have motor, smoke, lighting electronics and so on but the sound electronics would be what amounts to a wireless hi-fidelity Bluetooth (or whatever) speaker.

The imbalance is not just in the processing/memory but in the wireless communication throughput of any Wi-fi/Bluetooth smartphone vs. that of any O-gauge command method (TMCC/Legacy, DCS, DCC).  In other words, an Android smartphone App would have the ability to control/interact with multiple engines generating speed/lighting/smoke commands as well as the "final" audio.

MTH DCS has an albeit rudimentary version of this with their Protocast feature.  If you have a source of a different/superior whistle whether it be a recording or simulated on-the-fly you could mute an engine's built-in Protosound whistle and have the engine play your whistle sound transmitted to the engine digitally over the rails.  In other words, the engine's sound electronics becomes "just" a wireless speaker.  There are many "metaphors" for going this way such as streaming audio/video from the Cloud rather than being limited to whatever you have by way of DVDs or VHS tapes.

I was referring to using computational methods to simulate a train whistle...   I poked around and this is an area of research apparently -- well more generally overall that of generating realistic sounds for whistles or other similar instruments that require a rush of air against an edge ("edge tone") with some kind of resonant chamber.    It's all heady stuff, computational expensive and so on ... and also I think maybe reading between the lines, the results to do date don't work all that well ...

Anyway, I think it would be novel and fun to put a real solid CPU on a train engine but there's not much reason for it -- unless you wanted to do some heavy lifting/experimentation such as this...

But given all that,  this doesn't have much to do with the traditional aspects of the hobby, however some mind find it interesting. 

So here's one such paper... alas no free code to download to try out a little.

http://www.conforg.fr/isma2014.../articles/000051.pdf

Severn posted:

I was referring to using computational methods to simulate a train whistle...   I poked around and this is an area of research apparently -- well more generally overall that of generating realistic sounds for whistles or other similar instruments that require a rush of air against an edge ("edge tone") with some kind of resonant chamber.    It's all heady stuff, computational expensive and so on ... and also I think maybe reading between the lines, the results to do date don't work all that well ...

Anyway, I think it would be novel and fun to put a real solid CPU on a train engine but there's not much reason for it -- unless you wanted to do some heavy lifting/experimentation such as this...

But given all that,  this doesn't have much to do with the traditional aspects of the hobby, however some mind find it interesting. 

So here's one such paper... alas no free code to download to try out a little.

http://www.conforg.fr/isma2014.../articles/000051.pdf

 

I see where you are going with this.  Consider what makes an Allen Organ or a Wersi keyboad work.  The computer takes the waveform shape and generates it at the frequency dictated by the respective key.

If you go to York ask Dallas Gutaker (Dallee Electronics) about this.  He used to be an engineer at Allen Organ.

Go to youtube and search for Wersi.  You'll get the idea.

Since I have the emulator and burner for some 16bit sound ICs I think I'll stick to the single chip solution.

Lou N