Go Back   Cockos Incorporated Forums > REAPER Forums > REAPER Compatibility

Reply
 
Thread Tools Display Modes
Old 07-31-2015, 12:20 PM   #1
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default Win 10 obsolescence through erosion?

Something that has just dawned on me. Having had compatibility issues in Win10 and having to roll back.

Windows 10 is meant to be the last ever version of Windows, soon it will just be Windows. Windows as a service; endless updates, especially Home versions. It will have to evolve dramatically over time. Compatibility issues inevitably rearing their ugly head.

That being the case at some point the software & devices you have purchased could become incompatible with your existing copy of Windows 10 at any time. As soon as they make a big enough update that keyboard, soft synth, Sampler, whatever could become unusable on our ever evolving software platform.

That being the case coupled to the home version forcing you to take any and every update.......... food for thought.

Maybe best to have a back up of Win7 (Pro version) on a spare unused SSD.
Softsynth is offline   Reply With Quote
Old 07-31-2015, 12:23 PM   #2
ivansc
Human being with feelings
 
Join Date: Aug 2007
Location: Near Cambridge UK and Near Questembert, France
Posts: 18,889
Default

Remember what I said about cloning your current install to a fresh disk? Disks are cheap.
ivansc is offline   Reply With Quote
Old 07-31-2015, 12:28 PM   #3
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default

Quote:
Originally Posted by ivansc View Post
Remember what I said about cloning your current install to a fresh disk? Disks are cheap.
An upgrade to pro may be a good idea too.
Softsynth is offline   Reply With Quote
Old 07-31-2015, 12:32 PM   #4
Lannister
Human being with feelings
 
Lannister's Avatar
 
Join Date: Jan 2014
Location: Norway
Posts: 111
Default

People using OSX have been living with this for years.

Windows isn't about to drop compatibility over-night. That's up to the hardware vendors, and whether they think they can extract money from consumers because "X doesn't work with Y" all of a sudden.
Lannister is online now   Reply With Quote
Old 07-31-2015, 01:02 PM   #5
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default

Quote:
Originally Posted by Lannister View Post
People using OSX have been living with this for years.

Windows isn't about to drop compatibility over-night. That's up to the hardware vendors, and whether they think they can extract money from consumers because "X doesn't work with Y" all of a sudden.
Agreed, that's the worst thing about it. If a company goes bust, or gets bought out and doesn't update an otherwise good product, or they simply cannot be bothered, bean counters dictate the direction of their resources.

Imagine for instance if our VSTs hadn't been compatible with Win10?
Will the likes of Native Instruments work hard on creating patches or new versions of perfectly decent software? or will they just say "oh well we made that years ago for Windows XP/7/8 and just patch their newest titles?

Imagine you are working on your virtual instrument magnum opus in your DAW, months of work down the line and an update comes in and makes half your existing software stop working correctly.

I guess this is worst case scenario stuff.
Softsynth is offline   Reply With Quote
Old 07-31-2015, 02:19 PM   #6
babag
Human being with feelings
 
Join Date: Nov 2009
Posts: 1,630
Default

wait a minute. are you telling me my zip drive could become unusable?!?!?!?
babag is offline   Reply With Quote
Old 07-31-2015, 02:52 PM   #7
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Version numbers aren't full rewrites so it's moot anyway. We do realize a new OS version doesn't mean all new code I hope, typically far from it. IOW, the only thing changing is the way updates ar described and their frequency.
__________________
If it requires a null test to find it, it is by definition minuscule.
karbomusic is online now   Reply With Quote
Old 07-31-2015, 02:58 PM   #8
bluzkat
Human being with feelings
 
bluzkat's Avatar
 
Join Date: Jun 2007
Posts: 6,919
Default

Quote:
Originally Posted by babag View Post
wait a minute. are you telling me my zip drive could become unusable?!?!?!?
Oh man, a blast from the past. I still have mine packed away somewhere.


__________________
Peace...
bluzkat
bluzkat is offline   Reply With Quote
Old 07-31-2015, 03:01 PM   #9
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
Originally Posted by Softsynth View Post
That being the case at some point the software & devices you have purchased could become incompatible with your existing copy of Windows 10 at any time. As soon as they make a big enough update that keyboard, soft synth, Sampler, whatever could become unusable on our ever evolving software platform.
Any company writing drivers and selling you said devices can and should be far aware of such things long before changes go public. If some update inadvertantly breaks their driver they should fix it, because they can assuming it agrees with their own end-of-life policy.... but the question is do those vendors want to when all their cash it tied up in the current device they are working on.

That being said, no matter what compelling story or example we spin, legacy = risk unless everything is kept legacy. That risk isn't a vendor or OS problem. It's a techncal issue where there is simply no way for legacy code to be aware of code that hs been written after the legacy code was compiled. The only other choice was to have stopped all computer progress 15 years ago, never moving forward because that's the only way to escape this risk. That also means every hardware technological advancement since then would be moot and we'd all still be on 1gHz, 32 bit machines with 2GB memory in the name of non-obsoletion.
__________________
If it requires a null test to find it, it is by definition minuscule.
karbomusic is online now   Reply With Quote
Old 07-31-2015, 03:09 PM   #10
xpander
Human being with feelings
 
xpander's Avatar
 
Join Date: Jun 2007
Location: Terra incognita
Posts: 4,785
Default

Quote:
Originally Posted by bluzkat View Post
Oh man, a blast from the past. I still have mine packed away somewhere.
Ha, three of those sitting right in front of me and still working. But they will never touch the computer world here.
xpander is online now   Reply With Quote
Old 07-31-2015, 04:42 PM   #11
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default

Quote:
Originally Posted by babag View Post
wait a minute. are you telling me my zip drive could become unusable?!?!?!?
Noooooooooooooooo!!!
Softsynth is offline   Reply With Quote
Old 07-31-2015, 05:03 PM   #12
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default

Quote:
Originally Posted by karbomusic View Post
Any company writing drivers and selling you said devices can and should be far aware of such things long before changes go public. If some update inadvertantly breaks their driver they should fix it, because they can assuming it agrees with their own end-of-life policy.... but the question is do those vendors want to when all their cash it tied up in the current device they are working on.

That being said, no matter what compelling story or example we spin, legacy = risk unless everything is kept legacy. That risk isn't a vendor or OS problem. It's a techncal issue where there is simply no way for legacy code to be aware of code that hs been written after the legacy code was compiled. The only other choice was to have stopped all computer progress 15 years ago, never moving forward because that's the only way to escape this risk. That also means every hardware technological advancement since then would be moot and we'd all still be on 1gHz, 32 bit machines with 2GB memory in the name of non-obsoletion.

I'm no luddite. Since my childhood I've gone from Z80 at 3.25 MHz to todays multi core power houses. Well aware of the benefits.


There is a major difference today with operating systems and software (locked to security keys) working within ever evolving platforms.

Old software & hardware didn't have that problem. Okay in most cases you wouldn't want to go back to those machines other than nostalgia but at least you can if you want to. However we have reached a point with the sophistication in music production software where many people are buying large sample libraries that I am sure they expect to be able to use potentially for a very long time, perhaps decades, such as:

https://vsl.co.at/en
http://8dio.com/
Softsynth is offline   Reply With Quote
Old 07-31-2015, 05:34 PM   #13
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
Originally Posted by Softsynth View Post
I'm no luddite. Since my childhood I've gone from Z80 at 3.25 MHz to todays multi core power houses. Well aware of the benefits.


There is a major difference today with operating systems and software (locked to security keys) working within ever evolving platforms.

Old software & hardware didn't have that problem. Okay in most cases you wouldn't want to go back to those machines other than nostalgia but at least you can if you want to. However we have reached a point with the sophistication in music production software where many people are buying large sample libraries that I am sure they expect to be able to use potentially for a very long time, perhaps decades, such as:

https://vsl.co.at/en
http://8dio.com/
None of ^that changes or discounts a thing I said nor is it what you previously stated at least what I'm replying about. �� nor am I debating you but rather mentioning that the change in how versioning is labeled changes nothing here.

It's fine if you make it sound like more but it isn't when talking newer code with legacy code. I don't really get the Luddite remark lol but drama rains down for every release so we'll all just have to suffer through until it blows over. ��
__________________
If it requires a null test to find it, it is by definition minuscule.

Last edited by karbomusic; 07-31-2015 at 05:49 PM.
karbomusic is online now   Reply With Quote
Old 07-31-2015, 05:51 PM   #14
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default

Quote:
Originally Posted by karbomusic View Post
None of ^that changes or discounts a thing I said. �� nor am I debating you but rather mentioning that the change in how versioning is labeled changes nothing here. It's fine if you make it sound like more but it isn't when talking newer code with legacy code. I don't really get the Luddite remark lol.
Okay, but I'm not discounting what you said anyway.
Did you write this on a phone?, that paragraph is hard to understand.

luddites symbolize a hatred of progress. You mentioned being stuck with 1ghz machines. Of course even without the internet processor development would have continued as before. Computers would have continued to get faster and better. For sure the increased computer sales accelerated development and mass adoption of computers made them more affordable than ever due to economies of scale.
Softsynth is offline   Reply With Quote
Old 07-31-2015, 06:01 PM   #15
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
Originally Posted by Softsynth View Post
Okay, but I'm not discounting what you said anyway.
Did you write this on a phone?, that paragraph is hard to understand.

luddites symbolize a hatred of progress. You mentioned being stuck with 1ghz machines. Of course even without the internet processor development would have continued as before. Computers would have continued to get faster and better. For sure the increased computer sales accelerated development and mass adoption of computers made them more affordable than ever due to economies of scale.
Yes, phone... in a restaurant before flying back home. Maybe better if I just say I don't think there is that much more to worry about just because they changed to a more agile update format. That and I get frustrated with driver devs not updating their own drivers then blaming the OS, either support it or be very clear they don't care about you.
__________________
If it requires a null test to find it, it is by definition minuscule.
karbomusic is online now   Reply With Quote
Old 08-01-2015, 01:03 AM   #16
ivansc
Human being with feelings
 
Join Date: Aug 2007
Location: Near Cambridge UK and Near Questembert, France
Posts: 18,889
Default

Quote:
Originally Posted by bluzkat View Post
Oh man, a blast from the past. I still have mine packed away somewhere.


(grin) I last used mine on my Amiga 1200 setup - along with the 2 Jaz drives I had at the time!
ivansc is offline   Reply With Quote
Old 08-01-2015, 02:42 AM   #17
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default

Quote:
Originally Posted by karbomusic View Post
... Maybe better if I just say I don't think there is that much more to worry about just because they changed to a more agile update format. That and I get frustrated with driver devs not updating their own drivers then blaming the OS, either support it or be very clear they don't care about you.

Neither of us are really arguing with each other.
Which ever way around, lack of development or operating system incompatibility the result for us is the same, an unusable product.

I wish Microsoft could make the compatibility mode work better, it almost never works.
Softsynth is offline   Reply With Quote
Old 08-01-2015, 03:56 AM   #18
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
Originally Posted by Softsynth View Post
Neither of us are really arguing with each other.
Which ever way around, lack of development or operating system incompatibility the result for us is the same, an unusable product.

I wish Microsoft could make the compatibility mode work better, it almost never works.
I don't see us arriving at an increase of unusable comparatively just because win10 arrives. For example all my hardware and software works without issue on 10. We'll likely see the same 'percentage' as before (probably far, far less percentage than Vista). We are over the Vista hump and we have class compliant drivers to take care of the rest so by default things will fare much better than the days of Windows 98 for example where drivers were a nightmare comparatively.

Compat mode is yet another attempt to fix what devs won't do themselves which is an absolute disregard for secure practices (read that as the dev only cares that the software runs at the expense of your machine's security). Case in point, compat mode deals with devs who like to write to program files and other protected locations which was advised as a no-no starting around Windows 2000. It's been 15 years, and compat mode is simply trying to deal with some devs not following clear standards - There's a saying that goes with that practice "If it compiles, it ships" LOL.

Reaper fixed this FYI around late V3 or early V4 by properly using %AppData% for writes. Much of this I know from being a dev and we get bombarded with notices very early on with about as much information that is humanly possible to disseminate long, long before release in order to keep any of this from occurring.

Lastly, some of what I was saying before (I'm home now) deals with the point that hardware improvements and technological advancements always cause legacy code to be adjusted in order to continue that progress because the advancements were simply not possible before. There is no question that we all enjoy those advancements. In order to guarantee 100% compatibility, we'd near literally have to halt those advancements which leaves us still using single core machines running 32 bit and a couple GB of memory as a loose comparison.

Thusly, since hardware can't be recompiled, a reasonable percentage is going to go by the wayside as those vendors create new hardware which is able to consume those advancements and new or updated protocols. It makes little sense to me to blame any OS (Apple, Win, Linux etc.) for that natural hurdle for lack of a better term. Not that you are saying that, just noting that there is always this drama that surrounds any new OS release which much of the time is overblown and riddled with misinformation.

For example, much of the Vista debacle was because one hardware maker had a huge inventory of substandard hardware they wanted to sell through. MS had a hard-stop policy on Vista requirements. The vendor wouldn't budge and the compromise was to label those machines "Vista Capable" instead of "Vista Ready". This can be found in the email threads that were used as evidence in the court case that ensued and should be in PDF form somewhere on the net which is where I found it. Translation: Vista would have not had much of the issues that existed had the vendor not demanded deploying it on hardware that simply could not properly run it, go figure.
__________________
If it requires a null test to find it, it is by definition minuscule.

Last edited by karbomusic; 08-01-2015 at 04:40 AM.
karbomusic is online now   Reply With Quote
Old 08-01-2015, 05:00 AM   #19
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default

Karbomusic,
Thanks for the comprehensive reply. Your latest explanation makes much more sense.

I'm not playing the blame game. I can understand Microsoft's frustration with what you describe. I can also sympathize with beleaguered programmers, pulled in every direction, having to botch something to make it work.

I have friends and relatives in I.T and I'm aware how much companies lie through their behinds to clients regarding what stage of development their software has got to, agree to functionality they can never supply with given technology, and lie about how much development has been going on. Also the woefully inadequate I.T departments in companies with many, many more Salesmen than developers.

It is frustrating when a layer of software has stopped me from using the Roland A800Pro MIDI keyboard when others have had no problems with much older products. I bet there are some poor disillusioned folks out there with a home studio full of Roland components (all bought recently) that do not work with Win10.

Current state of play:
http://www.roland.com/support/support_news/250024
Softsynth is offline   Reply With Quote
Old 08-01-2015, 06:33 AM   #20
kenz
Human being with feelings
 
Join Date: Aug 2013
Posts: 339
Default

Quote:
Originally Posted by karbomusic View Post
In order to guarantee 100% compatibility, we'd near literally have to halt those advancements which leaves us still using single core machines running 32 bit and a couple GB of memory as a loose comparison.
Explain in detail because I fail to see that at all, especially when we're talking about software (OS)?!? Twice as many cores (or even physical CPUs) or twice the speed is incompatible how exactly? Unless you mean change in instruction set mode (like x64) but well that works with the other modes as well and is the SINGLE example in hardware which even comes close to prove your point.

The whole point of OS APIs is to abstract the hardware from software, so the latter "just works" and is not locked down to one system.

Doesn't matter what the obvious Microsoft fanboys say: they like to blame the app devs for not "updating their apps to work with Windows 10", as if that's supposed to be a given following some Operating System Bible or something.

It's the other way around. If the app stops working, it's Microsoft's fault, because what worked doesn't anymore. How is it the fault of others that MS decided to "upgrade" their OS and breaking stuff in the process? It's solely MS' decision. They can always just improve it without breaking the damn API and fanboys still don't give them at least a blame. This doesn't apply just to Microsoft, it's for all devs who break/stop supporting a given feature. Seeing people passing this blame onto other devs is what irks me.

If Steinberg stopped supporting Microsoft Operating Systems at all, would you blame Microsoft for it for not changing their OS to work with Cubase?!?? What nonsense is that?!

So if Microsoft decide to break their compatibility with something or stop supporting a given API or whatever, how is it NOT microsoft's fault here? So much nonsense from some people here, and I'm not talking about you obviously. But sure, in this case it's Steinberg's fault for not updating Cubase to work on Windows 10 right?

Because clearly Microsoft's decisions are the holy book and others should follow and never question/blame on them but instead blame the rest for failing to adhere to the book's wisdom!

The only exception was apps made for Windows 95/98/ME and earlier because those usually used low-level access and stuff to hardware (since those OSes allowed it), although not all of them... but any app made for Win2000 and up is most likely not at fault.


And of course I'm only giving out examples with Steinberg and Cubase etc... don't take it literally.

Point being, progress is perfectly viable and should be done as it says: progress. That means things keep progressing in steps. Not erasing what's old and replacing with new, but improving it. Running old software on newer hardware is the simplest example of progress: the same action you did now takes half the time as example. Simplest. Software can progress as well but I can't call "replacement" as progress. If you remove feature X and replace it with Y, that's not progress in the literal sense, just replacement. Progress is when you add to it.
kenz is offline   Reply With Quote
Old 08-01-2015, 10:00 AM   #21
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
So if Microsoft decide to break their compatibility with something or stop supporting a given API or whatever, how is it NOT microsoft's fault here?
If someone can provide examples and prove that ^this is more prevalent in Windows 10 than previous versions, I'm all ears. There is a shit ton of legacy code to support very old APIs. Point being, this particular OS would/could likely be an order of magnitude leaner if it didn't support the vast majority of legacy APIs that it does. It's fine to leave it there, just noting that we are speaking about how much stuff gets broken in this release compared to the past and that inevitably some improvements cause obsolescence by nature.

Quote:
The whole point of OS APIs is to abstract the hardware from software, so the latter "just works" and is not locked down to one system.
Absolutely but the point is, that cannot always be the case 100% of the time due to advancements, new protocols and security type changes. We're only talking about the times that is the case. I'm pretty sure I never said OS makers never screw up, I said that the percentage of obsolescence isn't going to suddenly change with Windows 10 and result in this massive number of ousted users. When protocols improve and new ones come into existence, calling conventions eventually change. The vast majority of the time obsolescence is avoided but it can't always be avoided so again, I urge anyone to show us where the sky is actually falling here.

Lastly, if new security practices are rolled out via code changes and devs ignore it after a decade of warning... The OS, again after a decade of warning begins enforcing it... If the dev ignores it for that long, placing your machine at risk in the name of their software, it is absolutely their fault. People bitched about sandboxing and insecurities, the OS added them with years of warning in some cases, developers ignored it, then people blamed the OS. That isn't fanboy, that's bullshit. Or we could just keep running our apps with elevated privileges so we can blame the OS. Devs have a high responsibility; they should live up to it just like the OS has to. Giving devs as much control as they have, creates that responsibility. I have personally ran into this, I complained a little, then I FIXED IT.

I don't think we disagree that much really, I simply wanted to share that the world isn't coming to an end here and progress inevitably causes some obsolescence by its very existence. We shouldn't blindly blame the OS (any OS for that matter) every single time. Sometimes it is absolutely the OS' fault but many other times it isn't and is instead devs who do what I call "developing via F5 key" where the moment it compiles and runs, that's all they care about, leaving you in the cold and using the OS as a scapegoat. There are a number of VSTs where that is the case as I type this who knew or should have been aware nearly 15 years ago.
__________________
If it requires a null test to find it, it is by definition minuscule.

Last edited by karbomusic; 08-01-2015 at 10:24 AM.
karbomusic is online now   Reply With Quote
Old 08-01-2015, 10:12 AM   #22
reapercurious
Human being with feelings
 
reapercurious's Avatar
 
Join Date: Jul 2007
Posts: 1,661
Default

are there any advantages to the structure of the windows operating system versus Unix/BSD?

is it faster or more streamlined or somehow better?

i wonder what things would be like if microsoft stuck to making apps for Unix/BSD/Apple
reapercurious is offline   Reply With Quote
Old 08-01-2015, 10:47 AM   #23
brainwreck
Human being with feelings
 
Join Date: Jul 2006
Posts: 20,840
Default

Quote:
Originally Posted by reapercurious View Post
are there any advantages to the structure of the windows operating system versus Unix/BSD?

is it faster or more streamlined or somehow better?

i wonder what things would be like if microsoft stuck to making apps for Unix/BSD/Apple
1) Yes. Windows 10 provides Microsoft and partners with a better way to extract users' personal data and identify users, including but not limited to an audio bugger, key logger, and Ad ID. *sarcasm*

2) Yes again. Instead of coaxing users to visit websites and enter their personal information there, they are doing it right in the os.

3) Richard Stallman would have a heart attack.

On a less dark note, the -nix based os's are as capable as anything out there, even more so in some respects. The catch being that there is a monopoly on which os is installed on new pc's by default (Windows), therefore much less desktop development and a much smaller user base.
__________________
It's time to take a stand against the synthesizer.
brainwreck is offline   Reply With Quote
Old 08-01-2015, 12:17 PM   #24
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default Reaper OS

Reaper OS.
I think it could be great if someone could develop a VST compatible skeleton Operating system come DAW, dedicated to that sole purpose but working on any traditional PC hardware platform.
This would create no limitations on the PC since it could be operated in a dual boot system with Windows if required.
Softsynth is offline   Reply With Quote
Old 08-01-2015, 12:35 PM   #25
paulheu
Human being with feelings
 
Join Date: Apr 2013
Posts: 635
Default

Quote:
Originally Posted by kenz View Post
It's the other way around. If the app stops working, it's Microsoft's fault, because what worked doesn't anymore. .
If a developer uses shortcuts or loopholes to gain 2% performance and then the OS is updated where the official API is intact but the loophole closes how is that the OS developer's fault exactly?

If one driver keeps working just fine through the existing updated API but another driver bombs then how is the crashing driver not at fault?

Microsoft is by far the best example of a company going through a tremendous amount of work to keep backwards compatibility in place.

Windows 10 runs fine on hardware 7-10 years old. I installed it on a friends eeePC running a N270 Atom and compared to the original WIN7 starter OS it flies..

It is _not_ the responsibility of Microsoft to ensure decade old hardware, often discontinued and/or abandoned by the company which developed it continues to work in every new version of their OS. The notion of this or even suggesting they are is IMO utter BS.
paulheu is offline   Reply With Quote
Old 08-01-2015, 12:46 PM   #26
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
If a developer uses shortcuts or loopholes to gain 2% performance and then the OS is updated where the official API is intact but the loophole closes how is that the OS developer's fault exactly?
Actually that is a point I failed to mention. Many developers completely ignore APIs. They then find some functionality due to exploiting the API and assume it is a supported method out of ignorance. They then build their entire app on top of that exploited functionality. Later on, via some update, that unknown and exploited functionality gets removed or broken because it was never intended for use, usually unintentionally precisely because it was never published for use to begin with. This is why APIs are written, to describe how to access it, what is supported and shouldn't break. The dev cries foul when they should have read the APIs and only used the supported methods to begin with. Vicious circle. I see the exact same thing with RFC compliance where SHOULD and MUST become the final decision makers. Of course that affects the OS developer too so he/she should code against the RFC when a protocol applies.
__________________
If it requires a null test to find it, it is by definition minuscule.

Last edited by karbomusic; 08-01-2015 at 12:52 PM.
karbomusic is online now   Reply With Quote
Old 08-01-2015, 12:47 PM   #27
paulheu
Human being with feelings
 
Join Date: Apr 2013
Posts: 635
Default

Quote:
Originally Posted by reapercurious View Post
are there any advantages to the structure of the windows operating system versus Unix/BSD?

is it faster or more streamlined or somehow better?

i wonder what things would be like if microsoft stuck to making apps for Unix/BSD/Apple
Windows now is a very sleek, streamlines and resource efficient OS and has been since Windows 7 was released.

While I am sure Linux would be able to develop it's core functionality to equal Windows and it is excellent for certain applications, it's simply inadequate for others and IMO low-latency, audio being one of them still.

Despite usually very loud advocates Linux has no relevance in the desktop/workstation market. That simply makes it not viable for any commercial enterprise to develop native applications for it. And that minute installed base is simply too fragmented to have any serious movement towards getting core stuff done.

Bottom line is that there is no money to be made in the Linux Desktop market.
paulheu is offline   Reply With Quote
Old 08-01-2015, 02:06 PM   #28
brainwreck
Human being with feelings
 
Join Date: Jul 2006
Posts: 20,840
Default

Quote:
Originally Posted by paulheu View Post
While I am sure Linux would be able to develop it's core functionality to equal Windows and it is excellent for certain applications, it's simply inadequate for others and IMO low-latency, audio being one of them still.
Where are you getting that information. It is false, as far as I can tell.

Quote:
Originally Posted by paulheu View Post
Despite usually very loud advocates Linux has no relevance in the desktop/workstation market. That simply makes it not viable for any commercial enterprise to develop native applications for it. And that minute installed base is simply too fragmented to have any serious movement towards getting core stuff done.
Most of the desktop software running on linux distros is free software (free source), developed by users. I'm sure that does scare away some commercial developers. Much of the available software is not ulta-polished, but plenty of it gets the job done, nonetheless.

On fragmentation, yes, there are hundreds of different linux distros, but the majority of them are running one of two desktop environments - Gnome and KDE - for which most desktop applications are targeted. And linux users are free to install additional desktop environments and switch between them at will.

Quote:
Originally Posted by paulheu View Post
Bottom line is that there is no money to be made in the Linux Desktop market.
We couldn't possibly know that without linux distros being installed on a significant percentage of new pc's by default. If linux came installed on say, 1/4 of all new pc's, do you really think that no commercial developers would be interested in linux?
__________________
It's time to take a stand against the synthesizer.
brainwreck is offline   Reply With Quote
Old 08-01-2015, 03:12 PM   #29
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default

Quote:
Originally Posted by Softsynth View Post
It is frustrating when a layer of software has stopped me from using the Roland A800Pro MIDI keyboard when others have had no problems with much older products. I bet there are some poor disillusioned folks out there with a home studio full of Roland components (all bought recently) that do not work with Win10.

Current state of play:
http://www.roland.com/support/support_news/250024
Did it say this before, maybe I didn't notice?:
It looks like Roland have NO plans to update their drivers. The guide offers three possibilities for driver compatibility:
As follows:

"Legend

1) Yes: Compatible with Windows(R) 10 64-bit Edition / 32-bit Edition

2) Limited: Limited compatibility Compatible with Windows(R) 10 64-bit Edition / 32-bit Edition with limited operation

3) No: No plan for compatible version We are sorry but we have no plan to make it compatible with Windows(R) 10 64-bit Edition / 32-bit Edition "



"No: No plan for compatible version We are sorry but we have no plan to make it compatible with Windows(R) 10 64-bit Edition / 32-bit Edition "

If this happens to the A800 Pro I will certainly never buy any Roland products ever again. This is a current product, available to buy in the shops. How the hell can they be trusted with anything!
Softsynth is offline   Reply With Quote
Old 08-02-2015, 08:21 AM   #30
SmajjL
Human being with feelings
 
SmajjL's Avatar
 
Join Date: Nov 2013
Location: Sweden
Posts: 2,045
Default

If may use my imagination..
Lets say Karbo just came in and wrote that he knew for sure! that fixing a driver IF it was not broken from the ground, just updating it so it would work from 8.1 to 10 and then said, it would take 5 minutes to an hour for a dev.
That! would trigger a pad or two over here.
__________________
:: REAPER / Bitwig / Komplete 12 & stuffs :: :)
SmajjL is offline   Reply With Quote
Old 08-02-2015, 09:43 AM   #31
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default

"Under testing"

What exactly does that mean?
Their statement is woefully imprecise and missing an extra legend:

"landfill"
If you are a Roland retailer we are very sorry but we won't be updating the drivers on your current stock. Feel free to take it to a recycling centre.
We are very sorry but we won't be reimbursing you. Have a nice day.

Maybe they work on Apple still? Not relevant to most people.

Even if they get my product working there is a lot of items in the list showing "NO" for which there will be no new drivers according to that page. This is pathetic.
Softsynth is offline   Reply With Quote
Old 08-02-2015, 10:39 AM   #32
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
Originally Posted by SmajjL View Post
IF it was not broken from the ground, just updating it so it would work from 8.1 to 10 and then said, it would take 5 minutes to an hour for a dev.
That!
Possibly. It really depends on just what broke. Once clue is how many it breaks. Since everyone is using the same APIs (generally) and only one or two or three vendor's driver stops working. There is a very high chance they could fix it. Often it is something incredibly simple, sometimes it is the OS fault (bug introduced), sometimes it is a design change that the developer would have had access to beforehand; sometimes it is a corner case based on some unknown difference on that user's machine. And sometimes it really is a breaking change made by the OS that the vendor cannot fix because it would take a firmware update or chip replacement that may not be possible. AKA, we can't generally recompile the entire piece of hardware.

However, in the driver/hardware/vendor's defense... When they create a product, there is some inference as to how long they support that product regardless of how easy it is. If they are always coming out with newer products, there will be a point where they stop reaching back to legacy gear and maintaining it. Supporting old devices becomes very expensive over time and every moment spent keeping those current is time away from making the new product even better. T

^This why I mentioned earlier that it would be nice if vendors stated in some form just how they (on an individual vendor basis) support their products up front, so that you can make the best decision when purchasing. It is possible that they fully intended to support the device for ~n years, across ~n OS builds before they stop doing so. It is obvious that I have less driver issues because I've been fortunate enough to either use gear where the vendors keep up with it better and/or fortunate enough to replace outdated hardware with more current versions.
__________________
If it requires a null test to find it, it is by definition minuscule.
karbomusic is online now   Reply With Quote
Old 08-02-2015, 10:43 AM   #33
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
"Under testing"

What exactly does that mean?
The honest explanation is that they (whomever) are trying to test every scenario they can to ensure it doesn't break further or cause regressions. These days "unit testing" , which is part of "Smoke testing", is all the rage where hundreds of code functions are coded to catch all the unexpected issues.

However, unit testing isn't perfect because it requires someone to know up front what might break, it can rarely cover the stuff the dev had no idea could actually occur; especially when those are time/event based where multithreading could result in a deadlock or access violation but only under very specific and rare conditions. It also can't take into account the vast myriad of differences on machines in the wild.
__________________
If it requires a null test to find it, it is by definition minuscule.
karbomusic is online now   Reply With Quote
Old 08-02-2015, 10:47 AM   #34
DBMusic
Human being with feelings
 
DBMusic's Avatar
 
Join Date: Sep 2009
Location: Illinois
Posts: 1,195
Default

Quote:
Originally Posted by paulheu View Post
Windows now is a very sleek, streamlines and resource efficient OS and has been since Windows 7 was released.
You've got to be kidding, right? +2GB's of "services" running in the background at startup is hardly a definition of sleek and streamline.
__________________
My Stuff
DBMusic is offline   Reply With Quote
Old 08-02-2015, 10:56 AM   #35
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
Originally Posted by DBMusic View Post
You've got to be kidding, right? +2GB's of "services" running in the background at startup is hardly a definition of sleek and streamline.
What other plans did you have for that memory? The vast majority of unallocated memory is cached (looks like it is being used) for fast access. If something else needs it gets it because actually and technically, free memory is wasted memory. This is precisely the reason OS's are using more resources yet seem faster and snappier because they don't have to allocate memory constantly which is very expensive on the CPU.

The minimum system requirement for the entire machine is 2GB for 64 bit. Additionally, we can't keep holding on to this windows 95 idea of 64MB for the OS to use. 10 years from now your machine might have a terabyte of memory with half of that being cached because the bar/floor/ceiling moves when more is accessible and available.

Lastly, the OS has to cover millions of desires AKA features you may not feel you personally need that someone else does. If that 2GB means that much to you and you don't have enough memory to function otherwise or it doesn't give enough of that back when needed (see caching above), either add more or disable all those services.
__________________
If it requires a null test to find it, it is by definition minuscule.

Last edited by karbomusic; 08-02-2015 at 11:03 AM.
karbomusic is online now   Reply With Quote
Old 08-02-2015, 11:19 AM   #36
kenz
Human being with feelings
 
Join Date: Aug 2013
Posts: 339
Default

I'm sorry but cached memory does not show up in the Task Manager as "used" memory. If it shows up there, it's memory that can't be freed because it's in use. Every OS caches reads/writes from disk, even Windows 95, that's beside the point.

Yet Linux, even with fancy graphics, takes no more than ~500MB on my other box, and it has a much more superior caching because you can even use its filesystem cache as a filesystem in itself (that only exists in RAM, in Windows it's called 'ramdisk' but ramdisks have a fixed size so it's not quite the same).

Unless Win10 task manager is any different, I'm pretty sure it does not include cached memory in the "available memory".

Obviously Linux is not as compatible as Windows for windows applications, that's a given unfortunately. But we're on the subject of caching here, not available software for the OS. Linux is actually much faster than Windows 7 and boots in 15sec even from hard disk. (I mean 7 uses a lot of needless resources as well)
kenz is offline   Reply With Quote
Old 08-02-2015, 11:28 AM   #37
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
Originally Posted by kenz View Post
I'm sorry but cached memory does not show up in the Task Manager as "used" memory. If it shows up there, it's memory that can't be freed because it's in use.
Task manager does not show the details between commit charge, working set, free. cached, modified, standby etc... (though I admit it is better than it used to be). ProcExp on top, TM on bottom...



Nevertheless, in 2015 2GB shouldn't be unheard of even if it is committed unless you wish, as I have already stated to disable those services that bug you so much.
__________________
If it requires a null test to find it, it is by definition minuscule.

Last edited by karbomusic; 08-02-2015 at 11:49 AM.
karbomusic is online now   Reply With Quote
Old 08-02-2015, 11:49 AM   #38
richie43
Human being with feelings
 
Join Date: Dec 2009
Location: Minnesota
Posts: 8,550
Default

This thread has turned into an "lol" thread.....

Older OS's were 'sleek" because they didn't know how to have them do more AND they had to run on what is not "weak" hardware. Do you think it is a coincidence or an accident that, for instance, Intel has developed the i7 monsters and can have enormous amounts of RAM available? What was once considered "bloat" or "unnecessary" resource usage IS the new efficient and sleek!! Let this technology work for you instead of fighting it and holding it accountable for what is not outdated and almost worthless tech.
__________________
http://soundyaudio.com/ The Sounds of the Hear and Now
richie43 is offline   Reply With Quote
Old 08-02-2015, 12:09 PM   #39
Softsynth
Human being with feelings
 
Softsynth's Avatar
 
Join Date: Jun 2015
Posts: 5,453
Default

Quote:
Originally Posted by karbomusic View Post
The honest explanation is that they (whomever) are trying to test every scenario they can to ensure it doesn't break further or cause regressions. These days "unit testing" , which is part of "Smoke testing", is all the rage where hundreds of code functions are coded to catch all the unexpected issues.

However, unit testing isn't perfect because it requires someone to know up front what might break, it can rarely cover the stuff the dev had no idea could actually occur; especially when those are time/event based where multithreading could result in a deadlock or access violation but only under very specific and rare conditions. It also can't take into account the vast myriad of differences on machines in the wild.
Actually working at all would be nice. I don't know how they are testing it. The drivers wont finish installing, it just hangs. Other Roland devices are having exactly the same install issue.
Annoyingly I have since learned that they used to work in an earlier (pre release) build of Win10 and an update broke it.
Indeed Microsoft's website is falsely claiming that the driver works:
https://www.microsoft.com/en-gb/wind...d=Windows%2010
Softsynth is offline   Reply With Quote
Old 08-02-2015, 12:14 PM   #40
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 24,536
Default

Quote:
Originally Posted by Softsynth View Post
Actually working at all would be nice. I don't know how they are testing it. The drivers wont finish installing, it just hangs. Other Roland devices are having exactly the same install issue.
Annoyingly I have since learned that they used to work in an earlier (pre release) build of Win10 and an update broke it.
Indeed Microsoft's website is falsely claiming that the driver works:
https://www.microsoft.com/en-gb/wind...d=Windows%2010
Help me out... Did you upgrade and the existing driver stopped working post upgrade and you are reinstalling or did you do a clean install and attempting to install the driver?

Lastly, I assume you went through that convoluted procedure Roland posted? There may be some hope because the MS link may be somewhat accurate, meaning it "may" work fine if it would actually install. Btw, which download/file is the actually frigging driver?
__________________
If it requires a null test to find it, it is by definition minuscule.
karbomusic is online now   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 03:02 PM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2019, vBulletin Solutions Inc.