Please ensure Javascript is enabled for purposes of website accessibility Jump to content

IR length discussion


MLSoundLab
 Share

Recommended Posts

So today I saw a video of Pete Thorn talking about his new IR loading Suhr amp and he said his amp supports 20.5ms IR's. Helix supports IR's up to 42ms which is twice as long. Now why does someone like ML Sound Lab (that's me) have IR's up to 500ms in the Cab Packs? I wanted to show you what this is all about and here's a video so you can find out too:

 

 

 

Now I understand that we're not all used to listening to the snap that the IR makes but this is the raw data that you'll find in the IR and it's the best way to demonstrate this. Playing with a shorter IR will feel "tighter" while a full length IR will feel "smoother" and the poll results I got years ago were 50/50 meaning that half of the people actually preferred the tight feel of a short IR vs the more realistic and smoother full length IR.

 

Have you guys ever tried an IR loader that uses more than that 42ms of IR length?

  • Like 2
  • Upvote 1
Link to comment
Share on other sites

This is just one guy's opinion, but if we all continue to fixate on the tiniest of technical details, dissecting everything down to the musical "subatomic level", we will completely sterilize the creative process, and eventually suck every last ounce of fun out of playing the guitar. Yes, the engineers who design all our fancy toys need to understand and be cognizant of this stuff, but what does it do for the average user? Not much.

 

I can't diagram an amp circuit, don't understand what tube bias actually is, or the electrical reason (is there one?) that makes an EL34 equipped amp sound different than one with 6L6's in it...But none of that prevented me from successfully using tube amps for decades. There's already more than enough ways to go down the rabbit hole with digital gear. The absolute last thing on earth I want/need/intended to do, is to start obsessing over milliseconds of IR length. If I like how it sounds, then I "win", as it were. And if Helix allows me 42ms IR's, then that's what it allows.... anything beyond that will disappear no matter what I do. Can't hear what ain't there. Doesn't matter if I bought an IR that can be timed with a calendar... if it's gone, it's gone. 

Link to comment
Share on other sites

I'm interested to watch this when I get home....

 

While I agree with the premise of 'if it works for me, I don't care about the details'....I also like understanding why something does or doesn't work.  It doesn't mean I obsess over every detail with every tone or if I want a certain tone it 'has' to be a certain way....but knowing that your swiss army knife has a toothpick accessory sure makes it easier than using the knife to make one out of a branch. 

 

Your mileage may vary...

  • Like 1
  • Upvote 3
Link to comment
Share on other sites

34 minutes ago, cruisinon2 said:

This is just one guy's opinion, but if we all continue to fixate on the tiniest of technical details, dissecting everything down to the musical "subatomic level", we will completely sterilize the creative process, and eventually suck every last ounce of fun out of playing the guitar. Yes, the engineers who design all our fancy toys need to understand and be cognizant of this stuff, but what does it do for the average user? Not much.

 

I can't diagram an amp circuit, don't understand what tube bias actually is, or the electrical reason (is there one?) that makes an EL34 equipped amp sound different than one with 6L6's in it...But none of that prevented me from successfully using tube amps for decades. There's already more than enough ways to go down the rabbit hole with digital gear. The absolute last thing on earth I want/need/intended to do, is to start obsessing over milliseconds of IR length. If I like how it sounds, then I "win", as it were. And if Helix allows me 42ms IR's, then that's what it allows.... anything beyond that will disappear no matter what I do. And if it ain't there, I won't hear it. Doesn't matter if I bought an IR that can be timed with a calendar... if it's gone, it's gone. 

I agree and disagree. More and more people in music are wearing both the studio engineer and composer/performer hats, and a lot of this stuff is of interest to that engineer brain. I know I geek out just as much on the minutiae of IR length, reverb decay, envelope slopes, etc as I do about where in the upper mids my "grind" in distorted rhythm tones sits, so I think this is a fun conversation to have. I just have to remember that none of that stuff matters when I'm actually playing. My personal opinion in the IR arena is that I don't want to hear any, or at least as little as possible, of the room the IR was captured in. I want the sound of the speaker in the box, then I want to be able to control any type of ambient sound that comes after that. Especially for live use. Maybe in the studio having that extra length that gets some of the "vibe" of the room might be nice, but for my uses (mostly hard rock/metal/death metal/core styles), I'd rather have a "room" buss to send everything through as glue instead of trying to deal with several different "rooms" in a mix. 

 

27 minutes ago, themetallikid said:

I'm interested to watch this when I get home....

 

While I agree with the premise of 'if it works for me, I don't care about the details'....I also like understanding why something does or doesn't work.  It doesn't mean I obsess over every detail with every tone or if I want a certain tone it 'has' to be a certain way....but knowing that your swiss army knife has a toothpick accessory sure makes it easier than using the knife to make one out of a branch. 

 

Your mileage may vary...

100% 

More information only ever hurts people who don't know when to apply it. It's the same conversation that is always regurgitated about music theory. Of course you CAN write a good song or piece of music without knowing what you're doing, but if having the extra information is somehow hindering your creative process, I'm going to assume that you were not very creative in first place. This technical information, like music theory, is only necessary when its applied in situations that may require it, or to solve a problem. 

  • Like 2
Link to comment
Share on other sites

Being an engineer by trade I can't help but be intrigued by technical minutiae.  I've noticed that as I've gotten older, I've also gotten much more pragmatic about it.  The fact is we can convince our brains to hear anything that we logically think is the case.  But the only relevant thing is when it comes to music and performance, our audience isn't that discerning.

 

Think about it this way.  As much as is made over stereo effects, how many people in the audience even notice or care that you did or didn't use stereo in your live performance?  Play the wrong chord or the wrong note and they'll notice for sure.  So while much attention is given within the musician community over matters like IR length and other technical details, those things will never have any effect over whether or not you'll play the right notes.  Stuff like this is more about marketing to the ego of musicians than it is about providing a memorable and satisfying performance for our audience. 

Link to comment
Share on other sites

Your mileage may vary... Your mileage may vary...
At the end of the day 1 mile is 1 mile and 10 miles are 10 miles no matter if something nice happend during 1 mile and nothing during 10 miles ;)
Longer IR just means better resolution just like higher sample rate and bit depth. 1024 and 2048 IR loaders are used in Helix because of very limited DSP power hardware modelers provide.
If you can not hear it you can see what FR plots shows with FIR filters truncation.

One can ask - what maximal IR lenght should be - where natural cab "reverb" ends and room reverb starts. Noone knows. You can make distant miced 4-6s IRs in churches but no present hardware modeller will make use of them.
I can understand Pete Thorn point. While he is the Suhr endorser and Suhr makes 1024 samples IR amp so this is enough. If he was Fractal endorser 8192 would be enough.  ;)
Most of hardware modelers users have no chance to hear the difference so your mileage is just your mileage. :D
 

 

  • Like 1
  • Upvote 1
Link to comment
Share on other sites

Lol... Some may find it interesting, for them they'll read, listen and contribute constructively. Some may not find it interesting or necessary, for them there's other things to read, listen and contribute to. I don't understand negative contributions to a topic one doesn't think will be beneficial to them.

  • Like 3
Link to comment
Share on other sites

16 minutes ago, DunedinDragon said:

Being an engineer by trade I can't help but be intrigued by technical minutiae.  I've noticed that as I've gotten older, I've also gotten much more pragmatic about it.  The fact is we can convince our brains to hear anything that we logically think is the case.  But the only relevant thing is when it comes to music and performance, our audience isn't that discerning.

 

Think about it this way.  As much as is made over stereo effects, how many people in the audience even notice or care that you did or didn't use stereo in your live performance?  Play the wrong chord or the wrong note and they'll notice for sure.  So while much attention is given within the musician community over matters like IR length and other technical details, those things will never have any effect over whether or not you'll play the right notes.  Stuff like this is more about marketing to the ego of musicians than it is about providing a memorable and satisfying performance for our audience. 

 

I'm definitely of the attitude that most of what we guitarists obsess over, particularly in regard to live sound, isn't noticed by the average audience. Stereo?...Check...Pure tube hand wired amps?...Check...Broken in perfectly matched to the amp speaker cabs...Check...Exotic custom effects?...Vintage classic electric guitar?...Check...Oxygen free, gold plated guitar cable?...Check. Guitar player experience with all of this? Nirvana. Average listener experience with all of this? Same as listening to the same guitar player plugged into a Peavey combo playing an Epiphone SG with a tube screamer wah and a chorus. Not that there's anything wrong  with that. It's mostly about the notes you're playing. Give a Stradivarius to an average 8 year old...nuthin' but noise. Give a world class orchestra violin player a "Dandy-Note" violin. Nice.

  • Like 2
  • Upvote 1
Link to comment
Share on other sites

Your mileage may vary... Your mileage may vary...

 

I agree, I'm just referring that some people get more use out of that information than others.  While I find it interesting to know, but will probably never hear, nor dial a tone based on that information.....some may dial in a tone specifically for that information you speak of. 

 

 

Link to comment
Share on other sites

3 hours ago, cruisinon2 said:

This is just one guy's opinion, but if we all continue to fixate on the tiniest of technical details, dissecting everything down to the musical "subatomic level", we will completely sterilize the creative process, and eventually suck every last ounce of fun out of playing the guitar. Yes, the engineers who design all our fancy toys need to understand and be cognizant of this stuff, but what does it do for the average user? Not much.

 

I can't diagram an amp circuit, don't understand what tube bias actually is, or the electrical reason (is there one?) that makes an EL34 equipped amp sound different than one with 6L6's in it...But none of that prevented me from successfully using tube amps for decades. There's already more than enough ways to go down the rabbit hole with digital gear. The absolute last thing on earth I want/need/intended to do, is to start obsessing over milliseconds of IR length. If I like how it sounds, then I "win", as it were. And if Helix allows me 42ms IR's, then that's what it allows.... anything beyond that will disappear no matter what I do. Can't hear what ain't there. Doesn't matter if I bought an IR that can be timed with a calendar... if it's gone, it's gone. 

 

This is well put except for the case of what if you're writing a song about the millisecond length of an IR?

  • Haha 1
Link to comment
Share on other sites

6 minutes ago, brue58ski said:

 

I'm definitely of the attitude that most of what we guitarists obsess over, particularly in regard to live sound, isn't noticed by the average audience. Stereo?...Check...Pure tube hand wired amps?...Check...Broken in perfectly matched to the amp speaker cabs...Check...Exotic custom effects?...Vintage classic electric guitar?...Check...Oxygen free, gold plated guitar cable?...Check. Guitar player experience with all of this? Nirvana. Average listener experience with all of this? Same as listening to the same guitar player plugged into a Peavey combo playing an Epiphone SG with a tube screamer wah and a chorus. Not that there's anything wrong  with that. It's mostly about the notes you're playing. Give a Stradivarius to an average 8 year old...nuthin' but noise. Give a world class orchestra violin player a "Dandy-Note" violin. Nice.

 I agree...in my area, there are some bands for sure that have phenomenol instruments, boutique amps on stage, pay $600 for sound/lights in a night and....they get the same audience response to the same song my band plays.  Using my $600 guitar, Helix direct to FOH, that we run ourselves on a pair of powered single 15" tops and single 18" powered subs.  The difference...between them and us is that they pay 15% booking fee, $600 sound/lights and cost the bar a lot more money.....which in turn raises drink prices on band nights and prevents people from coming out...

 

sorry, I rambled a bit, lol.....but ultimately those patrons will not notice the difference between that band and my band regardless of gear and IR technicalities.

 

 

I'm not disagreeing with ideas of the OP or how much the technicalities matter....I just think in 'my' scenario of live use I don't get as much mileage (to keep with my vocabulary earlier) out of the information as someone in a studio or recording environment.

Link to comment
Share on other sites

Interesting video @MLSoundLabthanks for posting. One question I have is "How consistent is the duration of meaningful and audible information across IRs created by different sources?". If most people are creating cab IRs where there is still significant data contained in the first 80ms, as the video above indicates, then it would appear to me that that is an ideal minimum starting point for manufacturers to set the max length of an IR. If the above observation is consistent across the bulk of the IRs intended for guitar, any multi-effect or DAW striving to provide maximum quality sound and authenticity from an IR should probably plan on allowing at least 80ms.

 

Resolution matters right up to the point at which it is no longer audible, or in the case of visual media up until it is no longer visible. For example 4k television clearly provides a superior image to lower HD formats and we certainly shouldn't have stopped developing TVs before we reached that because "the image was good enough". I have to assume that the same applies, perhaps to a less dramatic extent, in the audio realm with longer IRs. I don't think this is a slippery slope. The time to stop development is when you can no longer perceive a meaningful or positive difference. Although there is a point of diminishing returns where the perceptible changes to audio or sound are not significant or too expensive, this change to an 80ms capacity does not sound like one of those.

 

RAM and DSP are getting cheaper and faster all the time. I won't beat around the bush, it would appear to me that shooting for longer max IR lengths like the Fractal's Ultra-Res standard is worthwhile.  At least as demonstrated by this video. Longer IR times are also demonstrably more desirable for non guitar amp/cab IRs such as those made for reverb.

Link to comment
Share on other sites

Great video!

Yep, 1024 (18.5ms) sounds fine to me.

One more techie question I don't have to care about anymore.

 

Only.... when an IR company has "L6/Helix" length IRs, what, EXACTLY, does that mean? 1024/18.5ms? 2048/50ms?

 

Link to comment
Share on other sites

6 minutes ago, rd2rk said:

Only.... when an IR company has "L6/Helix" length IRs, what, EXACTLY, does that mean? 1024/18.5ms? 2048/50ms?

It means you buy something "meant for Helix". While Hx does it's truncation to 2048 it does not matter if it is 26, 200 or 600ms
500ms does not also mean much. You can have 200ms of usable tail and the rest is just the noise floor. 

Link to comment
Share on other sites

I think this just adds more tools to the tool box, which can help creativity when utilized properly with understanding.   

In the end your ears should tell you what was best for X and Y.  

Thanks for posting. 

 

 

Link to comment
Share on other sites

3 hours ago, gunpointmetal said:

but if having the extra information is somehow hindering your creative process, I'm going to assume that you were not very creative in first place.

 

So not the point I was trying to make. Nobody with an IQ larger than their shoe size is gonna argue that knowledge hinders creativity. However, there are only so many hours in the day, and the tendency to over- tweak these devices is already very real. The last thing I need is something else to micro-analyze, especially when the difference is likely to be barely discernable for practical use. That's all I meant... it's hard to be creative if one spends all their time doing blind taste-tests between 42ms, 35.2ms, and 20.5 ms IR's. Just not time well spent, imho... but at the end of the day, that's all it is...an opinion. I'm no more "right" or "wrong" than anybody else. 

  • Like 2
Link to comment
Share on other sites

1 hour ago, duncann said:

 

This is well put except for the case of what if you're writing a song about the millisecond length of an IR?

 

If you have reached that point, it means all possible songs have been written, and we all may as well sell our gear and learn to knit...;)

Link to comment
Share on other sites

6 minutes ago, cruisinon2 said:

 

So not the point I was trying to make. Nobody with an IQ larger than their shoe size is gonna argue that knowledge hinders creativity. However, there are only so many hours in the day... and the tendency to over- tweak these devices is already very real. The last thing I need is something else to micro-analyze, especially when the difference is likely to be barely discernable in the first place for practical use. That's all I meant... it's hard to be creative if one spends all their time doing blind taste-tests between 42ms, 35.2ms, and 20.5 ms IR's. Just not time well spent, imho... but st the end of the day, that's all it is an opinion. I'm no more "right" or "wrong" than anybody else. 

I wasn't saying that what was you're saying, its just another example of a situation where people who don't care or have a use for the information argue that it's not necessary. And that's fine, don't need it? Don't use it. There's a lot of "this technical mumbo jumbo gets in the way of makin' music" whereas my point was that it only does so if you allow it to. Much like people who find constant arguments on social media about stuff they seemingly don't care about, yet take the time to comment negatively in regards to. Don't need it, don't use it, ignore the topic completely.

  • Thanks 1
Link to comment
Share on other sites

1 hour ago, duncann said:

 

This is well put except for the case of what if you're writing a song about the millisecond length of an IR?

Every millisecond counts when your band covers “You Suffer” by Napalm Death!

Link to comment
Share on other sites

9 minutes ago, gunpointmetal said:

I wasn't saying that what was you're saying, its just another example of a situation where people who don't care or have a use for the information argue that it's not necessary. And that's fine, don't need it? Don't use it. There's a lot of "this technical mumbo jumbo gets in the way of makin' music" whereas my point was that it only does so if you allow it to. Much like people who find constant arguments on social media about stuff they seemingly don't care about, yet take the time to comment negatively in regards to. Don't need it, don't use it, ignore the topic completely.

 

And one of the many problems with social media is that expressing a difference of opinion is automatically labeled as "being negative". This is just a conversation. Had I said "you're all fools if this matters to you"... then you'd have a gripe, but that's not what I said. Why must anyone be obligated to shut up and move on just because they disagree? This is exactly why everything is so relentlessly polarized today. Anytime someone comes across an idea that isn't in lock-step agreement with their own, the default reaction is to claim offense and tell the "oppressor" to be quiet. It's baffling to me. But never mind, I'll move along. 

Link to comment
Share on other sites

1 minute ago, cruisinon2 said:

 

And one of the many problems with social media is that expressing a difference of opinion is automatically labeled as "being negative". It's just a conversation. Had I said "you're all fools if this matters to you"... then you'd have a gripe, but that's not what I said. Why must anyone be obligated to shut up and move on just because they disagree? This is exactly why everything is so relentlessly polarized today. Anytime someone comes across an idea that isn't in lock-step statement with their own, the default reaction is to claim offense and tell the "oppressor" to be quiet. It's baffling to me. But never mind, I'll move along. 

Again, nobody said that. I'm saying that if someone's response to a topic is a genuine "Who cares?" why would they feel the need to generate response? And if they are interested in the topic, they'd generate a response other than "Who cares?" or "This doesn't matter." Conversations are good. Declaring a topic of conversation as unnecessary for no other reason than a person doesn't have an interest, is a pointless and "look at me not caring" move. 

  • Upvote 1
Link to comment
Share on other sites

6 minutes ago, gunpointmetal said:

Again, nobody said that. I'm saying that if someone's response to a topic is a genuine "Who cares?" why would they feel the need to generate response? And if they are interested in the topic, they'd generate a response other than "Who cares?" or "This doesn't matter." Conversations are good. Declaring a topic of conversation as unnecessary for no other reason than a person doesn't have an interest, is a pointless and "look at me not caring" move. 

 

Ok you win...it was a horribly misguided "look at me" moment. 

  • Haha 1
Link to comment
Share on other sites

Just now, cruisinon2 said:

 

Ok you win...it was a horribly misguided "look at me" moment. I'll remember my place from now on. 

You made your point, and it was countered with reason. No need to take it personally. I only quoted your original post to point out that not everyone in every gear/guitar forum is strictly here to "shut up and play guitar". Some people are recording engineers (professional or hobby), some people are ACTUAL electronics engineers, some people just like having the information. 

  • Like 1
Link to comment
Share on other sites

1 hour ago, zolko60 said:

It means you buy something "meant for Helix".

 

Well, DUH, I get that the "HELIX" folder means that the contents are "Made for Helix".

 

That should, therefore, mean that the IRs therein are already at 2048 (50ms), and you can CHOOSE to further truncate them within Helix to 1024 to save on DSP.

 

However, the Ownhammer readme says:

 

"These files are no different than those in the Wav-200ms\48000-Hz Mpt
folders outside of being placed here for organizational convenience."

 

That would indicate that they ARE NOT actually "Made for Helix" , since the Helix manual says:

 

".....WAV type IRs of up to 2,048 samples are natively supported.......The imported IR is automatically shortened (or lengthened) to 2,048 samples. You may optionally choose a 1,024-sample version from the model list to save DSP, which fades out the IR halfway through."

 

I also read somewhere that the algorithms from various mfrs that are used for that truncation are not always as efficient/accurate(?) as they could be, so it COULD make a difference? 

Whether or not humans could hear the difference is another matter.

 

So, clear as mud!

Link to comment
Share on other sites

I'll try and steer this conversation back on topic. :D

 

When posting this I didn't even think about the video being taken as a "Helix doesn't support long enough IR's" type of thing. I'm cross-platform with my gear so I'm always just thinking about technology in general and what can be done to make it better. @brue58ski said it best. We as guitar players enjoy gear and technology and that's always been a part of being a guitarist whether we're talking about advancement in tube amps, pickups, modelers or even guitar strings. Choosing to ignore this is a subjective choice which you can make but I think I was at least capable of proving that there is semi-important information that's missing if you're using IR's shorter than about 200ms.

 

The limitation in modelers is processing power. I'm sure that with Pete's Suhr amp this is also what it came down to - having more power to run a longer IR would've made the amp more expensive. Same thing with the Helix and 42ms - how much more expensive would the Helix be if it needed to be powerful enough to run 8,192 sample 170ms IRs to get most of the IR sound. I'm more of a problem solver so I'd rather dive in and play around with the tools available like trying to find a reverb setting that simulates a longer IR tail.

  • Like 1
Link to comment
Share on other sites

46 minutes ago, rd2rk said:

 

That would indicate that they ARE NOT actually "Made for Helix" ,

Welcome to the real world. If there were not Helix folder they would probably have to answer a question "Can I use them with Helix?" ten times a day. ;)

 

41 minutes ago, MLSoundLab said:

I'm more of a problem solver so I'd rather dive in and play around with the tools available like trying to find a reverb setting that simulates a longer IR tail.

This is not the "reverb tail" problem only. Each FIR sample is a coeficient - FFT point. Complex, hi resolution EQs also require large number of FFT points regardles there have no reverb tail. 
There is something wrong with hardware DSP chips not to be able to keep up to CPUs performance. 
I think for live use 2k loader is just enough. If you are recording and want to take advantage of high resolution FIRs you can use software IR loader and truncate IRs to your taste.

Link to comment
Share on other sites

4 hours ago, zolko60 said:

It means you buy something "meant for Helix".

 

42 minutes ago, zolko60 said:

Welcome to the real world. If there were not Helix folder they would probably have to answer a question "Can I use them with Helix?" ten times a day. ;)

 

 

I misunderstood your first response - I thought you were saying that the IRs in the Helix folder were actually MADE FOR HELIX (2048/50ms), as opposed to "OK to USE with Helix, because Helix will fix it for you".

 

My bad.

Link to comment
Share on other sites

In many ways this whole discussion reminds me of the difference between listening to an .mp3 versus a .wav file.  Lossy compressed file formats provide the potential to store much larger numbers of songs while offering a compromise in quality that most listeners find acceptable. The .mp3 uses a "lossy" algorithm that uses compression algorithms to determine where there is redundant data as well as which data can be dropped with minimal impact. This is roughly akin to the choices for IR length made by device manufacturers. Right now most IRs are in essence rendered/truncated by a lossy algorithm when loaded into a modeler. The extent of the data loss is determined to some extent by the max  IR length.

 

By contrast a .wav file is lossless and retains more audio data. The .wav file format still remains the gold standard in many respects.  That gold standard also includes formats that can provide compression and max audio quality such as .ogg(can be lossless when compressed with FLAC) and Neil Young's efforts with FLAC files. The equivalent gold standard for an IR would be one that allowed longer IR lengths. That does not appear to be cost-effective at the moment as the resultant improvement in sound is simply not substantial enough for many users and device manufacturers. They would prefer those resources be used elsewhere where they perceive more bang for the buck.

 

Manufacturers and consumers pick their compromises based on current hardware/cost constraints for storage space and processing resources. On a computer,  if you are an audiophile,  you can throw in a five terabyte hard drive to store .wav or .ogg files, as well as more RAM and a faster CPU but that is generally not an option on a modeling device.  Modeling devices tend to have a fixed limit on storage and DSP.

 

As others have pointed out qualitative differences for live applications are going to be more difficult to detect whereas studio applications can be more demanding and qualitative differences in sound may be more pronounced. Different users have different priorities based on their focus, ears, and how they will be using their gear. So, would I love to have a device that allowed 200ms+ IRs? Sure, why not. Not however at the cost of way fewer IRs, processing bottlenecks and glitches, and/or a potentially substantially higher price tag.

 

Rock & roll and guitar sounds are a totally different animal from Hi-Fidelity. I am not looking for my guitar sound to mimic a $3000 stereo.  Horrors! I am however definitely interested in an IR format that does a better job of capturing all the low-fi details, quirks, decay characteristics, woody greatness, etc. of a top notch guitar cabinet. If getting there requires that we allow a longer max IR length I say bring it on. I'm not gonna be sitting around wringing my hands until it comes but I see no point in discouraging it either.  If, as it appears, there truly is significant data in the average IR beyond the approximately 40ms max length provided by Helix I will welcome a higher limit in the next Helix hardware generation.  For now this thing still sounds great and has many strengths in other areas that competing devices lack.

  • Like 1
  • Upvote 2
Link to comment
Share on other sites

And way back in the day, we listened to cassettes in our car which was acceptable at the time since we couldn't play records in a car. Noisy, compressed and they're making a come back!!! The worst way to listen to music in my opinion. So whatever works for ya.

  • Like 1
Link to comment
Share on other sites

1 minute ago, brue58ski said:

And way back in the day, we listened to cassettes in our car which was acceptable at the time since we couldn't play records in a car. Noisy, compressed and they're making a come back!!! The worst way to listen to music in my opinion. So whatever works for ya.

The only good thing about the cassette revival is we've made a killing selling old dual-cassette decks on line the last few years. Cassettes are/were awful and I know a whole bunch of underground music snobs who have MASSIVE collections, but can't listen to them anywhere because, ya know, dead technology.

Link to comment
Share on other sites

17 hours ago, HonestOpinion said:

In many ways this whole discussion reminds me of the difference between listening to an .mp3 versus a .wav file.

Yes. There is nice online test if that difference is important for you..
http://abx.digitalfeed.net/?fbclid=IwAR29my8UYktEJtujTwCWQP-MrNmWQW4dzvnSXJPqZ0YUN2bU64bM2w_BIiM#numTrials/10/sample/2
You can also make a test of how you percieve IR lenght difference but Helix hardwere is useless for such a test because it uses only low lenght FIR filters. Happily you can always use Helix Native and some free IR loader.
Rock & roll electric guitars are indeed different animals. 48kHz sampling rate and 32 bit floating point math is such a waste of that precious DSP power of two old 450Mhz Sharc procesors. 48/32 It is too Hi-Fi for the purpose. 32kHz, 16 bit would be just enough. ;)

 

  • Like 1
  • Upvote 1
Link to comment
Share on other sites

On 5/15/2019 at 10:58 AM, zolko60 said:

Your mileage may vary... Your mileage may vary...
At the end of the day 1 mile is 1 mile and 10 miles are 10 miles no matter if something nice happend during 1 mile and nothing during 10 miles ;)
Longer IR just means better resolution just like higher sample rate and bit depth. 1024 and 2048 IR loaders are used in Helix because of very limited DSP power hardware modelers provide.
If you can not hear it you can see what FR plots shows with FIR filters truncation.

One can ask - what maximal IR lenght should be - where natural cab "reverb" ends and room reverb starts. Noone knows. You can make distant miced 4-6s IRs in churches but no present hardware modeller will make use of them.
I can understand Pete Thorn point. While he is the Suhr endorser and Suhr makes 1024 samples IR amp so this is enough. If he was Fractal endorser 8192 would be enough.  ;)
Most of hardware modelers users have no chance to hear the difference so your mileage is just your mileage. :D
 

 

 

Probably belaboring the issue but these videos and the comments here were definitely educational. Unfortunately as usual just like in the Helix versus Fractal tests I find differences hard to detect on my PC speakers. It appears that for many players the difference between the shorter and longer IRs will be imperceptible. Then there is the point Pete Thorn makes in his video that some of the data contained in the longer IRs such as early reflections is arguably better controlled with a reverb block, implying that a longer IR, at least in this respect, may not be more desirable anyway.

 

On the other side of the debate as referred to earlier in this topic is sort of the Neil Young point about .mp3s which is that listeners have unfortunately gotten used to hearing lossy, less than optimal versions of music such that they no longer even detect, know, or care about the difference. The tyranny of low expectations. Part of that argument is that even though many people can't tell the difference in an A/B test, that there is an adverse cumulative effect over time from listening to lossy signals that is almost subconscious. A lot of people think that is BS though or at least not worth the extra cost and hardware resources to remedy.

 

Good as they sound already IR development and capture is still an evolving art/science and I think many of us have no doubt IRs will sound even better in another five years and not just due to larger IR files. Doesn't seem like the technology has quite topped out yet (if ever).  In these situations I always look forward to that point where  affordable technology strikes the best compromise with optimal configuration and sound.  Ultimately though, like so many other incremental or sometimes infinitesimal improvements, although the optimal provides us with a goalpost to work towards, getting caught up in or concerned about it as @cruisinon2 points out may ultimately just detract from the creative process or your enjoyment of the equipment.

  • Upvote 1
Link to comment
Share on other sites

The educational side is exactly what this is about and it's a great time to be a guitar geek when reaching "professionals" is relatively easy. There's one thing I think we need to discuss that I'm a little bit afraid of talking about but here goes: much of the IR length being good/bad has to do with the quality of the IR itself.

 

Bare with me - I'm not hear to bash other IR producers but rather talk about the tests that I've made and the findings. A full length IR f.ex. 500ms in this example contains raw data of a mic up in a room. Now let's say that the room is in many ways no ideal for miking cabs and there are many reflections and almost like a really loud reverb in it... this would happen if you shoot IR's in a small room with no treatment. This kind of an IR will most likely sound better when you're only hearing that 20-40ms instead of the full 500ms. This does support what Pete said that it's better to control the tail of the IR with a reverb.

 

However the situation with IR's that f.ex. I make is that my room is built to be the perfect size for a cabinet and treatment has been added where needed but not so much that the "air can't move naturally". So I pay a lot of attention to make that tail very nice. Remember that the highest quality reverb plugins are IR based reverbs like Waves IR-1. This is what we're essentially talking about here - you're cutting away that super realistic room sound and replacing it with a reverb effect that will not be the same quality.

 

Yes we're talking about a very small thing here. Still I believe that the only way to get to the next level of modeling is to do multiple small enhancements everywhere you can and in the end the enhancement will be much bigger.

Link to comment
Share on other sites

23 hours ago, MLSoundLab said:

...

Yes we're talking about a very small thing here. Still I believe that the only way to get to the next level of modeling is to do multiple small enhancements everywhere you can and in the end the enhancement will be much bigger. 

 

Very much agree that these small changes accumulate and can sometimes become substantial. I like to see the big changes keep coming as well. My interests definitely align with IR creators, as well as musicians who have produced a recording, in that I want to see their creative output as well as my own rendered and delivered with the maximum quality possible as it was intended. All of that has to happen unfortunately within the reality of cost constraints which is why I am always happy to see storage and processing costs go down while their capabilities increase.

  • Like 1
Link to comment
Share on other sites

On 5/17/2019 at 12:51 PM, HonestOpinion said:

On the other side of the debate as referred to earlier in this topic is sort of the Neil Young point about .mp3s which is that listeners have unfortunately gotten used to hearing lossy, less than optimal versions of music such that they no longer even detect, know, or care about the difference. The tyranny of low expectations. Part of that argument is that even though many people can't tell the difference in an A/B test, that there is an adverse cumulative effect over time from listening to lossy signals that is almost subconscious. A lot of people think that is BS though or at least not worth the extra cost and hardware resources to remedy.

 

I think you brought up a great analogy to audio compression trends of the last two decades. As a lover of true lossless formats (such as Flac) it has always driven me nuts that the great compromise of the iPod era was that people embraced crappy . . . in many cases EXTREMELY crappy, recordings of their favorite songs in order to have the convenience of portability. My ears could never stand anything less than 320 AAC format and mp3 was absolutely out of the question. What's funny is that people who embraced bad mp3's staunchly argued with me that it "did not matter." I could get the emotional baggage they were carrying, as no one wants to feel like their method has problems. What's interesting is if they'd let me show them the differences through an A/B comparison - once heard, it could never go UNHEARD (and I almost invariably felt kind of bad for showing them), and in real live studies (not done by me) people do show that subconsciously they DO care because they often choose to listen to better recorded audio for much longer than trashy mp3 recordings when subjected to them in blind tests. Now that memory just keeps getting cheaper and cheaper, we're seeing a push towards higher quality audio.  Is that good?  I think so. Quality DOES matter, at least to the point that our human body becomes the biggest obstacle to experiencing better versions of things (and who knows, maybe then we'll rip out our original eyes and ears and start implanting them with higher quality synthesized stuff). 

 

However, I get cruisinon2's point - look, the Helix does not sound like trash. It sounds beautiful. Nobody realizing longer IRs bring more to the table should look at their Helix, feel threatened, and never want to turn it on again. At the same time though, I do not expect the Helix to be THE last, or even MY last, modeler. I do expect Line 6, or Fractal, or whoever comes next to one day pull the next big rabbit out of their hat and it will no doubt embrace, and SHOULD embrace MANY advancements. It will have a FAR better processor (I mean, come on, processors advance so much just year over year and the old stuff gets cheaper and cheaper), FAR better memory and ram . . . honestly it better have crap I can't even dream up yet. 

 

Those engineers are going to need to be aware of the modest improvements that can come from things like the videos above show. I demand that of them and so should you. In the mean time though, those of us not involved in the design of these things can choose how interested we want to be in knowing about what those improvements could be. In cruisinon2's case, that probably won't be all that much. Me, I DO want to know about them because I just think it's cool lollipop. Still, just like cruisinon2, at the end of the day my biggest improvements are going to come from my practice time, not what modeler I buy. 

  • Like 2
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...