Ubuntu Studio Linux

[jcpatter]

Ubuntu Studio is a good choice for music production free of charge (https://ubuntustudio.org/ ), esxpecially for musicians on a budget.  I have been a loyal Line 6 consumer for many years but I have  been a linux user for even longer. 

 

I have owned the Line 6 Pod 2.0, the Tone Port UX1, The Flextone amp, The Pod HD500, The Firehawk 1500, the James Tyler Variax 89F, and have now orderd the Helix Floor with Powercab +.  I love my line 6 products but I don't love that after all these years Line 6 is still unwilling to support Linux.  I am forced to own a computer with an expensive, proprietary operating system, prone to viruses just so I can run my line 6 gear. 

 

Please someone from Line 6 please explain why you are unwilling to support linux?  There is an opportunity for growth here as linux users have been steadily rising over the same period of time.  People are waking up and want more from their technology.   

 

[phil_m]

Well, in the personal computer market, Linux represents 2-3% of users... I'm sure it's simply a matter of not being able to justify the cost of investing resources into such a small number of users. It's the same reason Line 6 doesn't currently make left-handed Variaxes.

[cruisinon2]

Why? 

 

For the same reason that McDonald's doesn't have garden slugs in an deer-tick cream sauce, or chilled ostrich brains on the dollar menu....

[rd2rk]

People use MACs (OSX) because they think that they're easier than PCs. People use PCs (Windows) because they ARE easier than LINUX.

The reason that LINUX is not now and never will be ready to challenge those markets is that, while the open-source model is idealistically wonderful, the vast majority of people with computer skills want to be PAID for their work! A wise man once said:

 

"The problem with communism is communism. The problem with capitalism is capitalists."

 

I've been revisiting LINUX every few years since the mid 90s. Twenty years later there doesn't seem to be that much progress toward making LINUX consumer friendly. Some people might call that a failed experiment. YMMV. 

[foxmeister]

27 minutes ago, rd2rk said:

Twenty years later there doesn't seem to be that much progress toward making LINUX consumer friendly. Some people might call that a failed experiment. YMMV. 

 

 I know I'm nitpicking here, so please don't feel the need to tell me so, but to say that consumer-friendly Linux is a failed experiment is just a nonsense.

 

Is it a failure at the desktop, yes, I'd probably agree with you for the most part, but the Linux kernel powers the world's most popular mobile OS (Android), and a huge percentage of the world's web servers, along with set-top boxes of all sorts.

 

Consumers are often using often using Linux, even if they don't know it, and that makes it far from consumer-friendly.

 

All that being said, as much as I'd like to see a Linux version of HX Edit, I can also see why it doesn't behoove Line 6 to expend resources on developing one.

 

[rd2rk]

13 minutes ago, foxmeister said:

the Linux kernel powers the world's most popular mobile OS (Android), and a huge percentage of the world's web servers, along with set-top boxes of all sorts.

 

Consumers are often using often using Linux, even if they don't know it, and that makes it far from consumer-friendly.

 

Yes, agreed, these applications are LINUX's strong point. I always wonder why someone as brilliant as Linus Torvalds can't come up with a way to solve the desktop conundrum. While I'm (blissfully) ignorant of the details of Penguin Politics, hopefully now that he's chosen to reenter the fray some progress can be made?

[gunpointmetal]

Linux is fun. Unless you want to use any readily available commercial audio hardware, lol. 

[codamedia]

3 hours ago, jcpatter said:

Please someone from Line 6 please explain why you are unwilling to support linux?

 

I certainly cannot speak for Line 6... but most likely it starts with the tiny market share it holds, and secondly because you cannot support Linux in general.... you must support the various distributions of Linux. That 2% - 3% market share that is tiny to begin with is also fractured into Centos/RHEL/Fedora, Debian/Ubunto, Gentoo/Chrome/OS etc... etc... 

 

Linux is a great developer platform... but it's the developers that put out the usable interfaces and obscure the inner workings. The TV Box might be easy to use.... the smart phone might be easy to use, and the Chromebook might be easy to use... but a white box computer loaded with a Penguin and no recognizable tech support is utter chaos to the vast majority. 

[cruisinon2]

2 hours ago, codamedia said:

...but a white box computer loaded with a Penguin and no recognizable tech support is utter chaos to the vast majority. 

 

Which is precisely why Linux will never be adopted by the masses...

When it comes to computing technology, the average mouth breather just wants his daily dose of e-bliss delivered quickly, effortlessly, and thoughtlessly (with special emphasis on thoughtlessly). Linux is all but 3 of those things... it appeals to tinkerers who actually understand computers, coding, and problem solving. Whereas Joe Average, who divides his time fairly evenly between professional wrestling and Budweiser, couldn't problem-solve his way out of a paper bag....

 

I gotta hand it to the Linux crowd though... they are nothing if not devoted, and fanatical devotion to anything is always funny to observers outside the cult. Unfortunately (for them), they're devoted to a "revolution" that's been just minutes away for the last 20+ years...;)

[rd2rk]

5 minutes ago, cruisinon2 said:

When it comes to computing technology, the average mouth breather wants his daily dose of e-bliss delivered quickly, effortlessly, and thoughtlessly (with special emphasis on thoughtlessly). Linux is all but 3 of those things... it appeals to tinkerers who actually understand computers and coding. This will forever rule out the vast majority of society, most of whom seem to divide their time fairly evenly between professional wrestling and Budweiser. 

 

While my personal interests and preferences include neither professional wrestling nor Budweiser, I'll take plug'n'play over Rube Goldberg all day long. Mouth breather?

 

14 minutes ago, cruisinon2 said:

This right here is precisely why Linux will never be adopted by the masses...it appeals to tinkerers who actually understand computers and coding.

 

It's not that LINUX is unable to be everything we'd want it to be, but that it's idealistic devotees spend most of their time on the esoteric underpinnings of the OS, and very little time on practical applications. I attribute this, as noted earlier, to the lack of financial incentive built in to the open source model. Idealism is lovely, but Junior wants a Les Paul......

[cruisinon2]

31 minutes ago, rd2rk said:

 

While my personal interests and preferences include neither professional wrestling nor Budweiser, I'll take plug'n'play over Rube Goldberg all day long. Mouth breather?

 

Just trying to be emphatic ;)....I want stuff that's easy to use, too... and I certainly don't claim to be a genius. But the sad truth is there are a whole lot of people roaming around out there, for whom using a Linux system would produce nothing but a blank stare. That's why it'll never sell, and is also what Linux proponents don't seem to grasp... they think that because they understand it, and enjoy the process of making machines bend to their will, that everyone else is similarly capable,  and willing to spend the time doing so. That just ain't the case...

 

[rd2rk]

24 minutes ago, cruisinon2 said:

 

Just trying to be emphatic ;)....I want stuff that's easy to use, too... and I certainly don't claim to be a genius. But the sad truth is there are a whole lot of people roaming around out there, for whom using a Linux system would produce nothing but a blank stare. That's why it'll never sell, and is also what Linux proponents don't seem to grasp... they think that because they understand it, and enjoy the process of making machines bend to their will, that everyone else is similarly capable,  and willing to spend the time doing so. That just ain't the case...

 

 

Some years ago, my sister's computer died. Since they only used a computer for internet/email and I had an older box running Linux sitting around, I donated it. When I tried explaining the difference between Linux and Windows, my brother-in-law wanted to know why the OS was named after the character from Peanuts......

[cruisinon2]

48 minutes ago, rd2rk said:

 

Some years ago, my sister's computer died. Since they only used a computer for internet/email and I had an older box running Linux sitting around, I donated it. When I tried explaining the difference between Linux and Windows, my brother-in-law wanted to know why the OS was named after the character from Peanuts......

 

I rest my case... lol

[DunedinDragon]

Having been a software developer that grew up starting with the IBM 370/MVS through the Unix age of System V and BSD and finally into the age of development on the Windows and Apple platforms I've always found Linux to be a curiosity and somewhat of a throwback to the earlier days of computer system design and I wondered where it would fit in today's market.  Ironically it appears to have found it's space in the closed, turnkey solution spaces such as Android and in the areas of dedicated server apps that don't require a lot of overhead or complexity and therefore can take full advantage of the slimmed down OS such as web servers.  But trying to do anything more complex than that would take away the particular advantage that Linux has in the marketplace of being a small footprint, tiny overhead, highly efficient and flexible OS.  And that's the greatest impediment Linux faces in establishing itself on the desktop or in bigger and more complex backend server solutions to the extent that it makes it cost effective to develop desktop apps or underlying support facilities (middleware) for it in any commercial sense.

 

In short, to become anything more than that it has to bloat and become Unix, and we all saw how that went....

[Thark]

I’ve been a Linux user for many years now (Slackware only) and would agree that for the new user it’s pretty daunting. Most folks want their computer to “just work” and rightly so. However, being the computer nerd that I am, I really love tinkering with software right down to rewriting some of the OS code.

 

That said, I agree that Linux is seriously lacking support for most major audio devices and I can understand why, based on the arguements above concerning user base. What Linux *is* good for is networking. I have my household network set up using couple of Linux boxes for various networking purposes. You just set them up and let them go. Can’t beat it.

 

I have a Windows machine because sooner or later you need one for one reason or another but my main audio machine is an iMac. I’ve not had it for long and I’d have to say that it’s performance is good but it’s a bit quirky. Icons keep moving on the desktop and default programs (or is it “apps” in the Mac world) keep changing. I might change over to the Windows computer at some point but that could be a chore and a half since there’s a lot of licensed audio related software on the Mac that would have migrate. 

[sparkyERTW]

Okay, moving past the OS fights and getting back to the OP's point...

 

No, Line 6 will almost certainly not be supporting Linux. Things like desktop apps, games, etc., have enough cross-platform libraries and frameworks to allow developers to "write once, run anywhere", for the most part. But when you're talking hardware that needs drivers, those layers are so drastically different that it doubles the work. Yes, companies like Intel, NVidia, etc., do make that effort, but Line 6 doesn't have that level of resources. One could argue that they could let the open source community handle it, but the legal work and supplying technical details would be almost as much work. As someone who enjoys using Linux OSes at home, I do feel your frustration.

 

What I can tell you - though not ideal - is that the USB extensions for VirtualBox work well enough that having Helix connect to HX Edit running in a Windows virtual machine has posed no problem. Pretty sure I ran the last firmware update through that as well and had no issue. Likewise, Workbench works fine through the USB extensions, and I would suspect Powercab's editor as well. Yes, it does mean you'd still have to run Windows, but you would have the benefit of Ubuntu for your day-to-day use and only use Windows for HX Edit. 

 

As far as audio interfacing goes, if I recall correctly, Helix is a class-compliant USB audio device. Because there are class-compliant USB audio drivers in Linux, that means its ins and outs should be recognized if you plug it in.

 

Windows in VirtualBox for patch editing/updating on your PC, native audio device in LInux for recording. Not perfect, but it could be worse.

 

 

[jonandtice]

4 hours ago, sparkyERTW said:

As far as audio interfacing goes, if I recall correctly, Helix is a class-compliant USB audio device. Because there are class-compliant USB audio drivers in Linux, that means its ins and outs should be recognized if you plug it in.

 

Windows in VirtualBox for patch editing/updating on your PC, native audio device in LInux for recording. Not perfect, but it could be worse. 

 

 

No, it doesn't work as an audio interface.  It is supposedly class compliant but not really.  Someone with some knowledge of the kernel and audio interfaces could work thing out I think.  This is why it doesn't work (last two lines):

[74449.448191] usb 1-1: new high-speed USB device number 8 using xhci_hcd
[74449.651241] usb 1-1: New USB device found, idVendor=0e41, idProduct=4244
[74449.651245] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[74449.651247] usb 1-1: Product: HELIX   
[74449.651249] usb 1-1: Manufacturer: LINE 6
[74449.651250] usb 1-1: SerialNumber:    2804389
[74449.673547] usb 1-1: parse_audio_format_rates_v2(): unable to retrieve number of sample rates (clock 16)
[74449.677222] usb 1-1: parse_audio_format_rates_v2(): unable to retrieve number of sample rates (clock 16)

 

[cruisinon2]

20 minutes ago, jonandtice said:

No, it doesn't work as an audio interface.  It is supposedly class compliant but not really.  Someone with some knowledge of the kernel and audio interfaces could work thing out I think.  This is why it doesn't work (last two lines):

[74449.448191] usb 1-1: new high-speed USB device number 8 using xhci_hcd
[74449.651241] usb 1-1: New USB device found, idVendor=0e41, idProduct=4244
[74449.651245] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[74449.651247] usb 1-1: Product: HELIX   
[74449.651249] usb 1-1: Manufacturer: LINE 6
[74449.651250] usb 1-1: SerialNumber:    2804389
[74449.673547] usb 1-1: parse_audio_format_rates_v2(): unable to retrieve number of sample rates (clock 16)
[74449.677222] usb 1-1: parse_audio_format_rates_v2(): unable to retrieve number of sample rates (clock 16)

 

 

Perfect! This illustrates my point very nicely, lol. Anybody who genuinely thinks that Everyday Joe Computer User can decipher any of that, much less be willing to trade their Macbook for a system that will require them to parse through reams of gibberish just to play guitar...is out of their mind. It might as well be written in sanskrit.

[zolko60]

(BEWARE IT IS A JOKE)
All you nedd to do is to write a quirk, put the following lines into quirk-table.h and recompiled. :D :D :D

/*
 * Line6 devices
 */
{
    USB_DEVICE(0x2466, 0x8003),
    .driver_info = (unsigned long) & (const struct snd_usb_audio_quirk) {
        .vendor_name = "Line6",
        .product_name = "Helix",
        .ifnum = QUIRK_ANY_INTERFACE,
        .type = QUIRK_COMPOSITE,
        .data = "" (const struct snd_usb_audio_quirk[]) {
            {
                .ifnum = 0,
                .type = QUIRK_AUDIO_STANDARD_MIXER
            },
            {
                .ifnum = 1,
                .type = QUIRK_AUDIO_FIXED_ENDPOINT,
                .data = "" (const struct audioformat) {
                    .formats = SNDRV_PCM_FMTBIT_S24_3LE,
                    .channels = 2,
                    .iface = 1,
                    .altsetting = 1,
                    .altset_idx = 1,
                    .attributes = 0,
                    .endpoint = 0x02,
                    .ep_attr = 0x05,
                    .maxpacksize = 0x002a,
                    .rates = SNDRV_PCM_RATE_CONTINUOUS,
                    .rate_min = 48000,
                    .rate_max = 48000
                }
            },
            {
                .ifnum = 2,
                .type = QUIRK_AUDIO_FIXED_ENDPOINT,
                .data = "" (const struct audioformat) {
                    .formats = SNDRV_PCM_FMTBIT_S24_3LE,
                    .channels = 4,
                    .iface = 2,
                    .altsetting = 1,
                    .altset_idx = 1,
                    .attributes = 0,
                    .endpoint = 0x86,
                    .ep_attr = 0x05,
                    .maxpacksize = 0x0054,
                    .rates = SNDRV_PCM_RATE_CONTINUOUS,
                    .rate_min = 48000,
                    .rate_max = 48000
                }
            },
            {
                .ifnum = 3,
                .type = QUIRK_MIDI_STANDARD_INTERFACE
            },
            {
                .ifnum = -1
            }
        }
    }
},
https://www.spinics.net/linux/fedora/alsa-user/msg10927.html

[sparkyERTW]

27 minutes ago, jonandtice said:

No, it doesn't work as an audio interface.  It is supposedly class compliant but not really.  Someone with some knowledge of the kernel and audio interfaces could work thing out I think.  This is why it doesn't work (last two lines):

...
[74449.673547] usb 1-1: parse_audio_format_rates_v2(): unable to retrieve number of sample rates (clock 16)
[74449.677222] usb 1-1: parse_audio_format_rates_v2(): unable to retrieve number of sample rates (clock 16)

 

Perhaps I'm mistaken; I'll try to remember to try it on my own machine and see if the results are the same.

 

13 minutes ago, cruisinon2 said:

 

Perfect! This illustrates my point very nicely, lol. Anybody who genuinely thinks that Everyday Joe Computer User can decipher any of that, much less be willing to trade their Macbook for a system that will require them to parse through reams of gibberish just to play guitar...is out of their mind. It might as well be written in sanskrit.

Except that it's not written in Sanskrit. It's stating in plain English that the driver is having trouble retrieving the sample rate from the device, information that could passed on to Line 6's development staff that might indicate there's a bug in their class-compliant implementation (which the link in zolko60's post actually confirms) - which IS something they might actually be willing and have time to fix.

 

The vast majority of contributors to this post has been civil, thoughtful, and respectful during this post, and if anything has potentially uncovered something that could legitimately improve Line 6's product and help it fulfill some people's needs. I would suggest that if that's not of interest to you, please leave it to the people that do want to continue that discussion.

 

(though admittedly zolko60's cheekiness actually provided us with ammo as a result :) )

[psarkissian]

Don't forget to read the notifications at the top of the window about 3rd party patches and such.

 

Be aware of warranty implications, wouldn't want someone's warranty voided over this. Go in with

eyes wide open.

 

Be careful, would hate to see someone's gear crash because of a "used before defined" error.

@sparkERTW,...  yeah and watch out for those segmentation and parse faults too.

 

Line 6 doesn't support modification to its gear that might affect the function of the product.

Aside from that, luck and happy hunting.

 

[sparkyERTW]

5 minutes ago, psarkissian said:

Don't forget to read the notifications at the top of the window about 3rd party patches and such.

 

Be aware of warranty implications, wouldn't want someone's warranty voided over this. Go in with

eyes wide open.

 

Be careful, would hate to see someone's gear crash because of a "used before defined" error.

@sparkERTW,...  yeah and watch out for those segmentation and parse faults too.

 

Line 6 doesn't support modification to its gear that might affect the function of the product.

Aside from that, luck and happy hunting.

 

Absolutely, Partev; I am in no way suggesting hardware modification (or software modification for that matter), and I do appreciate that use cases like communicating via USB through a virtualized hardware layer does add an extra layer of risk. So perhaps I should have prefaced my suggestion with, "while fairly safe, don't expect Line 6 to come running if things go wrong", which is a completely understandable position, as per my comments in previous posts.

 

That said, as per the previous couple posts, there's no harm in letting you guys know that the firmware might be short a function or two in the class-compliant spec.  If the knowledge base and documentation indicate it's class-compliant, it's not unreasonable to ask your team if they would be able to find time to make it meet the spec in a future firmware update, time permitting.

[psarkissian]

Compliance,... we have a couple of people here for that.

A number of hoops to jump through regarding such things.

 

[sparkyERTW]

Of course; I wouldn't condescend and suggest that "it's easy", as I know as a software engineer how much it irritates me when someone who has no understanding of the complexities of producing and releasing software makes that claim. But if the team does look into it and discovers it wouldn't be too much trouble, there's clearly a few users of DAWs for Linux and Android (which it may also solve) who would be grateful.

[jonandtice]

14 hours ago, sparkyERTW said:

Of course; I wouldn't condescend and suggest that "it's easy", as I know as a software engineer how much it irritates me when someone who has no understanding of the complexities of producing and releasing software makes that claim. But if the team does look into it and discovers it wouldn't be too much trouble, there's clearly a few users of DAWs for Linux and Android (which it may also solve) who would be grateful.

Being pessimistic, it could still turn out that the problem is actually in the Linux kernel's implementation and not the firmware.  And the chances are slim that Line 6 will even look at it so our best chance is still for someone to merge a quirk either into the kernel or package it into a module.

[hideout]

On 12/5/2018 at 1:35 PM, rd2rk said:

 

Yes, agreed, these applications are LINUX's strong point. I always wonder why someone as brilliant as Linus Torvalds can't come up with a way to solve the desktop conundrum. While I'm (blissfully) ignorant of the details of Penguin Politics, hopefully now that he's chosen to reenter the fray some progress can be made?

Because as Steve Jobs once said, making something that’s very complicated easy to use is very very difficult.  This is also compounded by how engineers think. They have zero clue, nor do they seem to care about how non-engineers might use the things they design. 

[psarkissian]

sparkyERTW,... know what you mean, part of the Csound email group, so I get a small taste of it.

 

[franz]

It's just a matter of time: https://www.zdnet.com/article/microsoft-releases-its-first-linux-product/

Be patient, first come first served.

[christroiani1]

I'm with you, JCPAtter.  I had lots of latency in Windows and tried everything. I reinstalled the OS, disabled all unnecessary services and startup processes, etc. and then gave up.  About five years ago, I did two things. I bought a Zoom R24 for direct recording and also went to Ubuntu Studio with the Ardor DAW.  I run it on a crappy old laptop and I have no noticeable latency.  Windows users, give up your fruitless and frustrating struggles to reduce latency and use Ubuntu Studio :)

[zolko60]

As far as I understand, making  Helix Class Complaint (in Linux means) would be a way to get to Class Complaint (OSX) Round Trip Latency timing.
Both systems are Unix based, right?
You can use your OSX/Win vst plugins on Linux, right?

[rd2rk]

1 hour ago, christroiani1 said:

I'm with you, JCPAtter.  I had lots of latency in Windows and tried everything. I reinstalled the OS, disabled all unnecessary services and startup processes, etc. and then gave up.  About five years ago, I did two things. I bought a Zoom R24 for direct recording and also went to Ubuntu Studio with the Ardor DAW.  I run it on a crappy old laptop and I have no noticeable latency.  Windows users, give up your fruitless and frustrating struggles to reduce latency and use Ubuntu Studio :)

 

Gee, I use W10 and have no noticeable latency. AND I use anything I want to use, DAWs, VSTs, whatever. I guess I'm just not as sensitive to latency as some folks. Just lucky I guess!

[zolko60]

30 minutes ago, rd2rk said:

Gee, I use W10 and have no noticeable latency.

Can you then comfirm the reported RTLs values in that thread? 

 

[rd2rk]

Upfront, I'll admit that the subtleties of all this escape me. That said, this is how I ran the tests, using the CEntrance Latency Test Utility v3.71 (I couldn't get the Oblique utility to work):

 

HP i7 laptop, 16gb, W10

Focusrite 18i20 (2nd gen)

Helix v2.70

Line Input 64/48k

RTL Utility settings -6db (differences between -6/-12 negligible).

 

The utility sends various sample sizes on successive tests. On the 18i20 it consistently sent 372-378 samples. On the Helix it was all over the place, below are the min/max samples sent.

 

18i20

372=7.75ms

378=7.78ms

 

Helix

65=1.35ms

226=4.71ms

 

DAW reporting was interesting.Using empty (no VSTs or tracks) projects for testing.

 

RTL:

 

Reaper

18i20=3.9ms

Helix=8.6ms

 

Sonar (Bandlab)

18i20=7.9ms

Helix=12.6ms

 

This could be a result of the number of samples sent by the different DAWs. I didn't test using LIVE, but I suspect that the results would be similar.

 

The point of all this being, not really noticeable latency. I'll stick with W10 and all the flexibility it gives me over LINUX.

[zolko60]

I do not want to mess in the thread which can end up with successful Class Compliancy (in Ubuntu terms) of Helix but...
Centrance measurements in case of Hx are total bulllollipop. It also shows me readings several samples higher than driver (CPU) buffer. I don't take it seriously.

[rd2rk]

15 minutes ago, zolko60 said:

I do not want to mess in the thread which can end up with successful Class Compliancy (in Ubuntu terms) of Helix but...
Centrance measurements in case of Hx are total bulllollipop. It also shows me readings several samples higher than driver (CPU) buffer. I don't take it seriously.

 

It sounds like you're saying that our results disagree, therefore you must be right?

"My tool's better than your tool, my tool's better than yours..." doesn't fly. Document your claims that the CEntrance tester is defective, or get off it.

The comments following the referenced post on facebook seem to support the idea that his test results are the ones that shouldn't be taken seriously.

 

As I demonstrated, results obtained by direct hardware tests differ from results obtained using a DAW, which also differ from one DAW to another. If I could get the Oblique tester to work, it would probably give a whole other set of results.

 

Methinks there's more to this latency thing than anybody really understands. Anybody's apple compared to anybody's orange is always going to be bulllollipop. The only TRUE answer is 42.

 

But seriously, MY ONLY POINT here is that there's no NOTICEABLE (YMMV) latency attributable to use of W10 vs whatever other OS happens to be the particular favorite of any individual poster on this forum. I only took the time to run those tests because YOU ASKED me to.

 

 

[zolko60]

"Noticeable" - "no noticeable" depends on personal feeling, experience. This is the thread about Ubuntu. You can easilly prove me wrong in RTL thread I liked.

[rd2rk]

10 minutes ago, zolko60 said:

"Noticeable" - "no noticeable" depends on personal feeling, experience. This is the thread about Ubuntu. You can easilly prove me wrong in RTL thread I liked.

 

That's what "YMMV" means.

The thread diverged into a latency discussion with the post by christroiani1.

I don't care about proving you wrong, nor can you prove that you're right.

 

Bottom line, use whatever OS serves your purpose.

 

Meanwhile, MANY THOUSANDS of people are using VSTs on Windows platforms without latency problems.

 

YMMV!

[zolko60]

8ms Class Complaint RTL vs 12ms on ASIO driver at the same buffer means 50% better latency performance. YMMV!

[rd2rk]

50% of a latency undetectable by human ears is not what I call bragging rights.

If it takes 10,000 years to get to the nearest inhabitable planet, and that can be cut to 5,000 years, that's a 50% improvement, but you're still DOA! ;-)

[duncann]

On 12/5/2018 at 1:13 PM, cruisinon2 said:

... chilled ostrich brains ...

 

Lol. Chilled monkey brains anyone?